Home

UI-SimCor Manual - No website was found at this address

1. Whole ABAQUS with imperfection 3 4 Whole ABAQUS No Imperfection 5 4 Displacement in X mm IS 64 4 8 Time sec 250 Whole ABAQUS with imperfection Whole ABAQUS No Imperfection 200 150 100 50 4 Displacement in Y mm Time sec J A i o a o a L 1 15 4 Displacement in Z mm E bs Whole ABAQUS with imperfection Whole ABAQUS No Imperfection 25 4 Time sec Figure 49 Comparison between w and w o imperfection Figure 50 shows the 3 DOF system for PSD test and Figures 51 53 are results of PSD test As shown in figures the PSD test results show good agreement with whole model 104 o UI SIMC GOR Z X Dummy element for initial loading y Translation mass 800 kg in X Y and Z direction So 3 DOF system Direction for Ground Motion Figure 50 3 DOF system for PSD test 0 10 0 00 oa Fr E M e ee ae E FLUTE EE E Or eee eee 0 30 BAG PM NEUEM at ile aoe ade C oo ee ee ee ee ee ee eee ee eee eee splacement in Y mm rnc MM MM EMEN EE Initial Stiffnes FormulationSiage Stage 0 90 Load Step 0 01 I 0 00 AN l I 0 5 10 l 15 20 25 30 35 a a PSDTest ABAQUS PU e EEEE educ Co
2. 5 10 15 Step 20 25 a Vertical displacement history 30 0090 Dynamic Stage Static Stage 5 10 15 Step 20 25 b Translational displacement history 30 Figure 33 Displacement history for initial loading problem OpenSees 68 o UI SIMC GOR 0 01 0 02 0 03 0 04 0 05 L Static Stage Dynamic Stage Vertical displacement mm 4 s 4 0 06 L o_o 0 07 i 0 5 10 15 20 25 30 Step a Vertical displacement history 0 e a E 0 05 L E o 501r o a Qa L 2 0 15 F s ce o N S 0 2 Static Stage Dynamic Stage 0 25 0 5 10 15 20 25 30 Step b Translational displacement history Figure 34 Displacement history for initial loading problem ABAQUS 69 o UI SIMC OR 15 Whole Model ZEUS PSD Test ZEUS Displacement mm 10 F 15 L 0 0 5 1 1 5 2 2 5 3 3 5 4 4 5 5 Time sec 15 Whole Model OpenSees IW H aside PSD Test OpenSees 5 J Displacement mm 10 F 0 0 5 1 1 5 2 2 5 3 3 5 4 45 5 Time sec Whole Model ABAQUS PSD Test ABAQUS Displacement mm 0 0 5 1 1 5 2 2 5 3 3 5 4 4 5 5 Time sec Figure 35 Pseudo dynamic test result Horizontal displacement of top node 70 o UI SIMC OR 8 3 Inelastic Cantilever Column Concrete Material The
3. Displacement Limit Displacement Limit 45 Displacement b wNEES at UCB SDSC 0 015 0101 0 01 0 015 Displacement Limit Displacement Limit Displacement c MiniMOST 1 at UIUC Figure 83 Hybrid test results input case 1 149 0 0 Displacement 0 0 Displacement 45 L 2 5 Displacement 0 5 015 o UI SIMC R UIUC 005 p eo 005 015 Displacement Limit JA AAA AN a Displacement Limit 25 Time a MiniMOST 1 at UIUC UCB 15 k 05 ou Displacement Limit ahah ho AN 1 a STATUTA WS Displacement Limit 0 015 0 01 0 005 L o 0 005 0 01 0 015 Time b wNEES at UCB SDSC Displacement Limit Wi AMA AD UN Displacement Limit Time c MiniMOST 1 at UIUC Figure 84 Hybrid test results input case 2 150 Force Force o UI SIMC OR 0 015 0101 0 01 Displacement Limit Displacement Limit 0 015 Displacement a MiniMOST 1 at UIUC UCB Force 1 5 Displacement Limit Displace 19 Limit N a Displacement b UNEES at UCB SDSC 0 015 0 01 Displacement Limit o Displacement Limit 0 015 Displacement c MiniMOST 1 at UIUC Figure 85 Hybrid test results input case 2 151 o UI SIMC OR
4. o UI SIMC OR Now ready to run the three site hybrid test Experiment Conduct the hybrid test UIUC Site 16 Click Establish Connection button in UI SIMCOR control window You will see some messages indicating that the connection is established in a MALTAB command window a UCBMon 5 xj UCB Simulation Monitor Additional simulation monitor of GUI for UCB site will be displayed in the uNEES control computer as shown in Figure B 12 This GUI helps UCB site seeing current statues of experiment SDSC site does not have this GUI for current test The status of NEES SAM window for ZEUS NL model is changed to Initializing from Standby for connection UCB 1 Coordinate system UCB Site Axis values x1 Step b 1 TDisp Node 01 Comp 1 gt X2 sten E Y2 Im Disp Node 01 Comp 1 Qa v1 2 Y2 05 0 05 4 O08 06 04 02 02 04 05 0s 1 Coordinate system UCB Site Axis values X1 T Disp Node 01 Comp 1 Y1 M Forc Node 01 Comp 1 7 X2 MDisp Node 01 Comp 1 Y2 M Forc Node 01 Comp 1 Qa Y1 X2 Y2 A E 08 06 04 02 0 02 04 05 08 1 Status Message Figure B 12 Monitoring window for UCB site 17 Click Stiffness Evaluation button in UI SIMCOR control window UI SIMCOR will read the predefined stiffness matrix from files MDLO1 K txt MDL 02 txt MDL 03 txt and MDL 04 txt located in C SIMCOR 03_Example
5. oe end oe Relaxation check If this parameter is 1 UI SimCor send commend to retrieve data and check relaxation just before the execution of proposed command If it s 1 the checking criteria needs to be provided for i 1 Sys Num RF Module MDL i CheckRelax 0 Module i if MDL i CheckLimit 1 define following variables Variable size should be number of control nodes 6 array oe oe oe oe oe o Displacement variation ratio not increment MDL i MES D inco abcde f ns IE Force variaiton ratio not increment MDL i MES_F_inc abcdef ds 1 o o oe end oe Check displacement and force limit At every steps check if the displacement or force are approaching to the limitation of the equipments stroke or force capacity for i 1 Sys Num RF Module MDL i CheckLimit 0 Module i if MDL i CheckLimit 1 define following variables Variable size should be number of control nodes 6 array oe oe o o9 of o Displacement increment limit not ratio MDL i TGT D inco abcdef oe oe l Displacement limit MDL i CAP D tot abcdef oe oe 1 Force limit TA o UI SIMC OR MDL i CAP F tot abcde f E a 1 Displacement toleranc ratio MDL i TOL D inc L ab ode L EJ 13 end oe Loading and Boundary Condition Box LBCB case If it s 1 the coordinate transformation matrix needs to be provided This can be also used f
6. 8 1875 0 0 0 0 0 Sys Node_Mass 14 0 0 0 0 0 01 oe oe oe Restoring force module configuration oe oe Create objects of MDL RF DL 1 MDL RF DL 2 MDL RF DL 3 MDL RF oe Name of each module DL 1 name Joint Module ID of this module is 1 DL 2 name Column Module ID of this module is 2 DL 3 name Frame Module ID of this module is 3 oe URL of each module Format IP address port number ex http c nsp4 cee uiuc edu 11997 oe oe 86 o UI SIMC GOR for local machine 127 0 0 1 11997 MDL 1 URL 127 0 0 1 11997 MDL 2 URL 127 0 0 1 11998 MDL 3 URL 127 0 0 1 11999 oe Communication protocol for each module NTCP communicate through NEESPOP server TCPIP binary communication using TCPIP LabViewl ASCII communication with LabView plugin format Propose Query Execute Query LabView2 same as LabViewl but Propose Query OpenFrescolD OpenFresco only 1 DOF is implemented now 5 NHCP NHCP linear 1 DOF simulation mode Mini MOST 1 and 2 at oe UIUC or SDSC MDL 1 protocol LabView2 MDL 2 protocol LabView2 MDL 3 protocol LabView2 Module 1 Joint MDL 1 node 9 10 Control point node number MDL 1 EFF DOF 1 1 000 1 Effective DOF for CP 9 11000 1 Effective DOF for CP 10 Module
7. ds 1 Force variaiton ratio not increment MDL i MES F inco abcdef ues 1 end oe o o 88 o UI SIMC OR oo Check displacement and force limit At every steps check if the displacement or force are approaching to the limitation of the equipments stroke or force capacity for i 1 Sys Num RF Module MDL i CheckLimit 0 Module i if MDL i CheckLimit 1 define following variables Variable size should be number of control nodes 6 array oo oe o oe oe o Displacement increment limit not ratio DL i TGT D inc abcdeef oe oe oe Displacement limit DL i CAP D tot abe d F oe oe oe Force limit MDL i CAP F tot abcdeef E H Displacement toleranc ratio MDL i TOL D inc abcdeef oe end oo Loading and Boundary Condition Box LBCB case If it s 1 the coordinate transformation matrix needs to be provided This can be also used for any other actuator which has diffrence number of DOF coordinate with those of UI SIMCOR oe oe oo for i 1 Sys Num RF Module MDL i LBCB 0 end for i 1 Sys Num RF Module MDL i LBCB TransM end oe oe Auxiliary module configuration AUX 1 MDL AUX AUX 1 URL 127 0 0 1 12000 AUX 1 protocol labviewl AUX 1 name Camera Module ID of this mdoule is 1 AUX 1 Command displacement z 3500 Variable Description Sys xi 1 S
8. Current displacement limit for UCB speicmen 6 00e 000 If you want modify the displacement limit please open API Main m j Reset Figure 59 Warning message for the displacement limit 121 a o UI SIMC OR If the displacement limit is confirmed click Okay button then you will see the Waiting for connection from client through port 11999 in a MATLAB command window IP address and port number are dependent on experimental configuration If you want modify the displacement limit click Reset button then the API Main file is opened automatically Modify the Disp Limit variable at line 17 If you want to try the pushover test by numerical simulation then type API Main sim instead of API Main Now ready to run the pushover test This test will run automatically 3 4 Run another MATLAB for the pushover test and change the directory to C ASIMCOR 03_Examples UCB PushoverTest 00_Coordinator Type PushOver and enter in MATLAB command window to run the pushover test a b One GUI for monitoring window as shown in Figure 60 will popup Another simulation monitor of GUI as shown in Figure 61 for UCB site will be displayed in the MATLAB window for APL Main The default value of maximum displacement for pushover test is 2 If you want to modify the maximum value open the PushOver and modify the disp_scale variable at line 66 This will run automatically so make sure the displacement limit E lolx
9. oe Number of initial static loading steps When ther xist static constant loading i e gravity forces apply then in Zeus NL or OpenSees as a incremental loading with n steps In this file SimConfig m specify the number of static steps in the following variable Sys Num Static Step 0 oe oe oe o Number of dynamic analysis steps Sys Num Dynamic Step 500 Dynamic analysis time steps Sys dt 0 01 Rayleigh damping xi 1 and xi 2 Damping ratio Tn 1 Tn 2 Target period Sys xi 1 0 00 Sys Tn 1 0 00 Sys xi 2 0 00 Sys Tn 2 0 00 oe Number of Stiffness test If stiffness is evaluated through experiment th valuation need to be don several times and the average of the results are used as the initial stiffness This parameter is used when Sys Eval Stiffness 1 Sys Num Test Stiffness 1 oe oe oe oe Enable GUI for SimCor Yes 1 nable the GUI for SimCor o 0 disable the GUI for SimCor Hybrid simulation will be run automatically oe oe oe 109 o UI SIMC R Not recommended for the experiment Sys EnableGUI 1 Use GUI for SimCor Number of restoring force modules Sys Num_RF_Module 5 Number of auxilary modules Sys Num AUX Module 0 Total number of effective nodes Effective nodes are interface nodes betw modules and nodes where lumped masses are defined Sys Num Node 7 5 Lumped mass assigned
10. 48 o UI SIMC OR Variable Description Port This variable defines the port number that the NEES SAM will open and waiting for connection MDL Type This variable defines the module application 1 for ZEUS NL and 2 for OpenSees SC Node This variable defines the effective node number in simulation coordinator The order of node number should be identical to that specified in the simulation coordinator configuration file MDL Node This variable has same role with MDL MDL Node in the configuration file for FEDEAS Lab It identifies which nodes are related to the node number in simulation coordinator See following figure 1 n2 gt Pax SIMCOR ZEUS NL Figure 18 Module node number definition for ZEUS NL EEE DOF i This variable has a same function with MDL EFF DOF i in the configuration for FEDEAS Lab MODEL ZEUS NL model file name without extension MDL Dim This variable has a same function with MDL MDL Dim in the configuration for FEDEAS Lab TH_MONITOR Time history monitoring By defining this variable real time monitoring window pops up from the beginning of the test Every DOFs displacement and force can be monitored in real time Instead of this monitor there are also monitoring windows of GUI in UI SIMCOR see MDL i EnableGUI in the simulation configuration file 49 o UI SIMC R Real time Monitoring xj Monitoring Window X disp n2 0 20 40 60 80 100 120 140 160 180 Step Nu
11. 7 8 Install the coupon at specimen Restart the xPC Real time target Log on to the uNEES control computer and run a MATLAB for communication with xPC Real time target and change the directory to C SIMCOR 04_API 01_UCB Type API_ Main then you will see the message box confirming the displacement limit as shown in Figure B 11 a Check Displacement Limit lal x 5 Current displacement limit for UCB speicmen 6 00e 000 If you want modify the displacement limit please open API_Main m Y Reset Figure B 11 Warning message The displacement limit in the API Main is 6 however the displacement limit in the SimConfig of UI SIMCOR is 2 So if the target displacement which will be sent to xPC Real time target is greater than 2 the test will be paused with warning message in UI SIMCOR control window If the displacement limit is confirmed click Okay button then you will see the Waiting for connection from client through port 8090 in a MATLAB command window If you want modify the displacement limit click Reset button then the API_Main file is opened automatically Modify the Disp_Limit variable at line 17 188 o UI SIMC R SDSC Site For MiniMOST I SDSC 10 Turn on the MiniMOST 1 system 11 Log on to the MiniMOST 1 computer and open the lv programs folder located on the Desktop 12 Double click on the daemon programs icon in this folder It should bring up a list of LabVIEW programs Choos
12. June 2004 June 2005 October 2006 o UI SIMC R NEES SAM Update History Version 1 0 release Version 1 5 release Interact with the latest version of OpenSees version 1 6 2 which include soil material model and can make reaction force output Hence it isn t necessary anymore to add additional nodes with stiff springs on the control points to measure reaction forces Minor interface updates Version 2 0 release Develop three NEES SAM for TCP IP LabVIEW1 and LabVIEW2 protocols Minor interface updates o UI SIMC OR 1 Introduction MOST Multi Site Online Simulation Test experimental example has been extensively used to verify the NEESgrid infrastructure The simulation coordinator for the MOST experiment http it nees org software ntcp index php is designed only for the MOST experiment Eventually more than three sites are going to be involved in the earthquake simulation and various structural configurations need to be tested The UI SIMCOR is developed for this purpose UI SIMCOR can control distributed pseudo dynamic PSD test in several sites The number of sites and number of control points are not limited Using this software in together with various communication protocols we can test from a single degree of freedom system to highly complicated structures which need to be tested in various sites The simulation can be either all experiments combination of experiments and analyses or all analyses
13. MATLAB based simulation coordinator for distributed hybrid simulation and testing Project Admins Kyu Sik Park Lawrence Miller s Development Status 5 Production Stable Oh Sung Kwon e Environment Platform Independent Paul Hubbard e Intended Audience Developers End Users System Administrators e License Public Domain e Programming Language Other s Topic Large Scale Structural Engineering DEDE Creare Registered 2006 01 20 15 09 Request ta join Activity Percentile 2596 View project activity statistics Package Version Date Notes Monitor Download UI SIMCOR UI SIMCORv2 5 December 1 2006 G Download View All Project Files TT Announcing UI SimCor 2 5 fit Project Home Page Oh Sung Kwon 2007 01 09 11 59 0 Comment Read More Comment tracker News archive orgy IIT D Bine Figure 12 UI SIMCOR project website on NEESforge Click on any Download on the right side of the page All links load the page shown in Figure 13 28 o UI SIMC GOR d NEESforge UI SimCor Simulation Coordinator Project Filelist Microsoft Internet Explorer File Edit View Favorites Tools Help Osak x 2 Sy Favorites Advanced Log In NEESforge Search the entire project lt Sa E Account UI SimCor Home My Page Project Tree Code Snippets Project Openings Simulation Coordinator Summary Forums Tracker lists Tasks Docs Surveys News SEM Below is a l
14. UI SIMC R User Manual and Examples for UI SIMCOR v2 6 Multi Site Substructure Pseudo Dynamic Simulation Coordinator NEES SAM v2 0 Static Analysis Module for NEESgrid by Oh Sung Kwon Narutoshi Nakata Kyu Sik Park Amr Elnashai Bill Spencer DEPARTMENT OF CIVIL AND ENVIRONMENTAL ENGINEERING UNIVERSITY OF ILLINOIS AT URBANA CHAMPAIGN URBANA ILLINOIS FEBRUARY 2007 UI SIMC DR TABLE OF CONTENTS 1 oeeie al lt T 1 2 Simulation Framework eee 1 3 Pseudo Dynamic Test Procedure 9 4 Sub structuring of a SIruCure sese 5 4 1 Concept of Static Condensation and Sub structuring 5 4 2 Effective DOES for Dynamic Analysis sse eee eee 6 4 3 MOST Experiments eee 7 5 Static Analysis Module Interface sees 8 6 Protocol Specification ee reo eee 10 6 1 NEESgrid Teleoperation Control Protocol NTCP 10 6 2 TCP IP Protocol 11 6 3 LabVIEW1 and LabVIEW2 Protocols for NTCP Server 13 6 4 OpenFresco1D Protocol 16 6 5 NEES Hybrid Communication Protocol NHCP 17 7 Installation of UI SIMCOR eee eee 27 7 1 System Reduirementis eee eee 27 7 2 Installation of UI SIMCOR and Interface Application 27 7 3 Installation of Structural Analysis Software 30 7 4 Downloading and Updating UI SIMCOR Source Code 32
15. o UI SIMC OR UIUC Site 3s Click Establish Connection button in UI SIMCOR control window You will see some messages indicating that the connection is established in both MALTAB command windows and another simulation monitor of GUI for UCB site will be displayed in the MATLAB for API Main sim UCB site as shown in Figure A 10 2 UCBMon aigi x UCB Simulation Monitor UCB Coordinate system k 2 v ml emn ual os Status Mertoge Figure A 10 Monitoring window for UCB site a If the UI SIMCOR run in the other computer or the other site 1 e UIUC then this GUI helps to UCB site seeing current status of experiment Click Stiffness Evaluation button in UI SIMCOR control window UI SIMCOR will read the predefined stiffness matrix from file MDLO1_K txt located in C SIMCOR 03_ Examples UCBYSDOF 00 Coordinator Click Apply Static Loading button in UI SIMCOR control window In this stage the gravity force is applied however there is no plan to apply gravity as this exemplary test is to demonstrate the efficacy of hybrid simulation framework Click Start PSD Test to run test by simulation The target displacement and feedback displacement and force will be displayed in the monitoring window and UCB monitoring window Click Disconnect Modules after finishing hybrid simulation to disconnect UI SIMCOR from xPC Real time target You will see some messages indicating the communication is disconnected in both
16. y Figure 78 MiniMOST 1 at SDSC 142 o UI SIMC R Remote site Force N 0 015 0 015 Displacement m Figure 79 Stiffness of Mini MOST 1 at SDSC 9 4 4 Scale factor in UI SIMCOR In the configuration file of UI SIMCOR SimConfig m there is a variable MDL i ScaleF related to scale factor for displacement rotation force and moment Experimental specimens are not always in full scale These factors can be used to apply scale factors The displacement and rotation scale factors are multiplied before they are sent to module or experimental specimen Measured forces and moments are divided with scale factors before used in the PSD algorithm The scale factor can also be used to adjust the target displacement within the limit of experiment specimen For this exemplary test the scale factor is used to insure that the target displacement which is calculated by UI SIMCOR and will be sent to each experiment site is within the displacement limit of each site Based on the results of PSD test by simulation with original modules shown in Section 9 4 2 stiffness maximum displacement and force of each module can be obtained as shown in Table 3 Table 3 Properties of each module Module Stiffness Drax Fmax Module 1 1 584X10 0 0228 36 12 Module 3 1 584X10 0 0228 36 12 Module 4 1 584x10 0 0228 36 12 Furthermore based on the results of test which was recentl
17. 2 6 6 1 0 1 1 1 1 0 1 2 2 1 0 1 6 6 1 0 6 1 1 1 0 6 2 2 1 0 10 1 1 1 0 10 2 2 1 0 14 1 1 1 0 14 2 2 1 0 5 1 1 1 0 5 2 2 1 0 5 6 6 1 0 90 o UI SIMC OR This order should be the same as in the AbaqusCFG m file MDL MDL Node 48 12 1637 11 15216 10 14 5 MDL EFF DOF 1 1 10000 MDL EFF DOF 2 1 10000 MDL EFF DOF 3 1 10000 MDL EFF DOF 4 1 10000 MDL EFF DOF 5 1 10000 MDL EFF DOF 6 1 1 0000 MDL EFF DOF 7 1 10000 MDL EFF DOF 8 1 1 0000 MDL EFF DOF 9 11000 1 MDL EFF DOF 10 11000 I MDL EFF DOF 11 11000 0 MDL EFF DOF 12 11000 0 MDL EFF DOF 13 11000 0 MDL EFF DOF 14 11000 1 8 5 4 Running simulation The running procedure of SAC example is very similar to that of MOST example so further explanation is not give in this section 8 5 5 Result and verification In this example Rayleigh damping is used as shown in Figure 43 The results of PSD test using UI SIMCOR are very good correspondence with those of whole model run as shown in Figures 44 and 45 0 01 L Ist mode 2nd mode 0 5 10 15 20 25 30 35 40 45 50 Natural frequency rad sec Figure 43 Rayleigh damping 9T o UI SIMC GOR 0 15 Whole Model FEDEAS B E perma PSD Test FEDEAS E E t o E D o S Qa E a 0 0 5 1 1 5 2 2 5 3 3 5 4 4 5 5 Time sec 0 1 Whole Mode
18. 8 Numerical Examples 35 8 1 Cantilever Column sese eee EU cee 35 8 2 Inelastic Cantilever Column with Initial Loading Steel Material 61 8 3 Inelastic Cantilever Column Concrete Material 71 8 4 MOST Example sese 73 8 5 SAC Three Story Steel Structure 84 8 6 LBGB Example po ker ten e eee deen us etum i nui ibm EE 94 8 7 Buckling Example using ABAQUS sse 103 8 8 7 DOF Model using Five Protocols sss 108 9 Specific Experimental lasues sse 117 9 1 One site Experiment with NEES at UGB 117 9 2 Two site Experiment between UIUC and UCB 132 9 3 Two site Experiment between UIUC and SDSC 138 9 4 Three site Experiment among UIUC UCB and SDSC 139 9 5 Two site Experiment between UIUC and SDSC using NHCP 158 uuu CS rnEeE 161 Appendix Experimental Procedure 162 A Two site Hybrid Test between UIUC and UCB 162 B Three site Hybrid Test among UIUC UCB and SDSC 178 June 2004 June 2005 October 2006 November 2006 e o UI SIMC OR UI SIMCOR Update History Version 1 0 release Version 1 3 release Renamed SIMCOR to UI SIMCOR Renamed FedeasMDL to NEES FL for consistency Inclusion of NEES MW Development of GUI for UI SIMCOR Include the latest version of OpenS
19. 9 4 7 Experiment results The three site hybrid test among UIUC UCB and SDSC using recently enhanced UI SIMCOR were conducted at October 6 2006 Two steps were conducted to verify the communication and hybrid test Hybrid test by simulation 100 steps The main purpose of this step is to verify the communication among UIUC UCB and SDSC The UI SIMCOR communicates with control computer through LabVIEW plugin and LabVIEW plugin communicates with MiniMOST 1 through NTCP server as shown in Figure 86 The uNEES specimen at UCB communicates with xPC Real time target PC through SCRAMNET and UI SIMCOR communicates with xPC HOST PC through API as shown in Figure 58 UIUC or SDSC NEESPOP UI SIMCOR LabVIEW Plugin NTCP Server Ethernet HOST PC LabVIEW System Collector amp Distributor Control Deamon t Control Program DG Simulation Controller Figure 86 Configuration of HSF at UIUC or SDSC with UI SIMCOR Figures 87 89 show the simulation results The Expected in figures means the results simulated in one computer whereas Actual means the results simulated with three site i e the UI SIMCOR MiniMOST 1 analytical model and ZEUS NL model were simulated at UIUC uNEES analytical model was simulated at UCB and MiniMOST 1 a
20. DL 1 DEL t 0 DL 2 DEL t 0 DL 3 DEL t 0 DL 4 DEL t 0 DL 5 DEL t 0 DL 1 DEL r 0 DL 2 DEL r 0 DL 3 DEL_r 0 DL 4 DEL r 0 DL 5 DEL r 0 Enable GUI for Translation Del_r Rotation in radian 0005 0005 0005 0005 0005 0002 0002 0002 0002 0002 each module GUI for each module can only display the data GUI for each module can not control the hybrid simulation 1 enabl Yes the GUI for each module 0 disable the GUI for each module EnableGUI EnableGUI EnableGUI EnableGUI EnableGUI How gd PPR PR 1 Advanced modular parameters Thes parameters need to be redefined for following situations 1 Different coordinate system between UI SIMCOR and static module 2 When scale factor needs to be applied either in experiment or simulation 3 To define force and displacement criteria for tolerance and safety 4 To trigger camera modules or DAQ system 111 o UI SIMC OR oe 5 When LBCB at UIUC is used for experiment 6 When NHCP protocol is used oe oe oe URL of remote site and NHCP mode for NHCP for i 1 Sys Num RF Module if strcmp lower MDL i protocol nhcp MDL i remote URL 127 0 0 1 99999 MDL i NHCPMode simld end end Stiffness for NHCP Only valid if NHCPMode SimlD for i 1 Sys Num RF Module if strcmp lower MDL i NHCPMode simld MDL i NHCPSimK
21. OpenSSL http www openssl org are used for TCP IP both plaintext and encrypted OpenSSL allows to leverage the considerable investments in code infrastructure and hardware acceleration that others have made with confidence that the underlying code is reliable and well tested Most of part of this section is based on the technical documents of NEESit which can be found in http it nees org weblog hybridsim p 67 6 5 2 Abstractions and paradigms The various systems differ in how they split up the work Here is a pretty common setup from a physical viewpoint Simulation Coordinator 9B igr 7 Hardware Server Figure 6 Common setup of hybrid simulation The functional view looks like as follow NEES comms library NEES comms NEES comms E NEES comms Figure 7 Functional view of hybrid simulation 18 o UI SIMC OR Here is a diagram of the second design Block in green represent TCP IP server process where the program in question is responsible for accepting a TCP IP connection and light brown represents the corresponding TCP IP client This is revised quite a bit as follow Security Coordinator NEES comms library NEES comms library plugin Log file Figure 8 Functional view of hybrid simulation using NHCP The major difference is that the server is back to running multiple plugins As before a coordinator can and will talk to multiple ser
22. Reaction forces and displacements at converged state are saved amp Disp for ID 1 The received displacements are applied to structure Reaction forces and displacements at converged state are saved amp Disp for ID 1 The received displacements are applied to structure Reaction forces and displacements at converged state are saved amp Disp for ID 1 bd Ready Request processed MMT UIU Connecte Zeus NL 7 Figure B 7 NEES SAM window for MMI UIUC module during PSD test 11 Click Disconnect Modules button in UI SIMCOR control window after finishing hybrid simulation to disconnect UI SIMCOR from each site You will see some messages indicating the communication is disconnected in a MALTAB command window and NEES SAM API of each site E NEES SAM LabViewl File View Help Disp for ID 1 The received displacements are applied to structur Reaction forces and displacements at converged sta Disp for ID 4 1 The received displacements are applied to structur Reaction forces and displacements at converged sta Disp for ID 1 FORTRAN Runtime Error 124 TL0 amp ASISteLICI z b z eady Disconnected IMMi UlU Clased Zeus NL 7 Figure B 8 NEES SAM window for MM1 UIUC module after disconnecting 12 Backup the results files and send to UIUC a ForUIUC site e MDLOI recv txt MDL02 recv txt MDL 03 recv txt MDL04 recv txt 185 o UI SIMC R Netwklog txt and NodeDisp txt located in C S
23. ZeusNL PSD OpenSees with TCPIP 0 025 0 025 0 02 5 Whole Mode ZeusNL PSD FedeasLab with LabMew2 0 015 4 0 01 0 01 4 0 015 4 0 02 4 0 025 25 3 Time Sec Figure 55 Pseudo dynamic test result Horizontal displacement 115 o UI SIMC R Whole Model ZeusNL PSD ZeusNL with LabView1 Whole Mode ZeusNL PSD OpenFresco with OpenFresco1D Time sec Whole Mode ZeusNL PSD NHCPSim1D with NHCP Figure 55 Pseudo dynamic test result 25 Time Sec 116 3 3 5 o UI SIMC OR 9 Specific Experimental Issues Every experimental site may have their own configuration of the actuators measurement systems and data formats As a result the global data format that is sent through communication protocol may not directly adaptable to each experimental site In these specific situations it is necessary to develop an API which acts as an adapter between experimental sites and UI SIMCOR strictly speaking communication protocol In this section various application examples using UI SIMCOR are explained Various experiments including one site two site and three site hybrid test using UI SIMCOR with various specimen were conducted at UIUC UCB and SDSC to verify and demonstrate the efficacy of UI SIMCOR 9 1 One site Experiment with NEES at UC
24. and SDSC by conducting a numerical hybrid simulation Step 2 is to conduct the real three site hybrid test using the actual experimental components Each step is divided in a Prepare the communication among UIUC UCB and SDSC b Prepare the UI SIMCOR for hybrid simulation and c Conduct the hybrid test B 1 Step 1 Hybrid test by simulation 100 steps This is simulation so it will not send any signal to actuator The purpose of this step is to verify the communication among UIUC UCB and SDSC through running the three site example by simulation UI SIMCOR MiniMOST 1 analytical model and ZEUS NL model UIUC uNEES analytical model UCB and MiniMOST 1 analytical model SDSC will be simulated For this test following IP address and port number of each site will be used For MM1 at UIUC 130 126 240 138 44000 For uNEES at UCB 169 229 203 152 8090 For MMI at SDSC 137 110 118 95 44000 Table B 1 Displacement limit of each site UIUC UCB SDSC 0 01 2 0 0 01 The displacement limit of each site for this experiment is shown in Table B 1 However the actual displacement limit of each site is larger than theses values For this test the displacement limit of each site is assumed smaller than actual one for safety issue If the target displacement which will be sent to each actuator is greater than the displacement limit of each site the test will be paused with warning message The expected results of this t
25. click Reset button then the API Main sim file is opened automatically Modify the Disp Limit variable at line 17 Now ready to run the test by simulation 5 10 11 Click Establish Connection button in UI SIMCOR control window You will see some messages indicating that the connection is established in both MALTAB command windows and another simulation monitor of GUI for UCB site will be displayed in the MATLAB for API Main sim as shown in Figure A 5 iion nx UCB Simulation Monitor Uca 1 Coordinate system C UH sae al Ars values G Status Merzoge Figure A 5 Monitoring window for UCB site a If the UI SIMCOR run in the other computer or the other site 1 e UIUC then this GUI helps to UCB site seeing current status of experiment Click Stiffness Evaluation button in UI SIMCOR control window UI SIMCOR will read the predefined stiffness matrix from file MDLO1_K txt located in CASIMCORYO3 Examples UCB SDOF 00_ Coordinator Click Apply Static Loading button in UI SIMCOR control window In this stage the gravity force is applied however there is no plan to apply gravity as this exemplary test is to demonstrate the efficacy of hybrid simulation framwork Click Start PSD Test to run test by simulation The target displacement and feedback displacement and force will be displayed in the monitoring window and UCB monitoring window Click Disconnect Modules after finishing hybri
26. 2 Column MDL 2 node 14 Control point node number MDL 2 EFF DOF 1 1 00 0 1 Effective DOF for CP 14 Module 3 Fram MDL 3 node 1 2 3456789 10 11 12 13 14 Control point node number MDL 3 EFF_DOF 1 0000 Effective DOF for CP 1 14 1 0000 n 0000 1 0000 als 0000 n 0000 1 0000 1 0000 1 0001 Hi 0001 l 0000 1 0000 1 0000 ab 000 1 Dismplacement for preliminary test for each module Del t Translation Del r Rotation in radian MDL 1 DEL t 1e 5 MDL 2 DEL t 1e 5 MDL 3 DEL t 1e 5 x MDL 1 DEL r 1e 5 MDL 2 DEL r 1e 5 MDL 3 DEL r 1e 5 z Enable GUI for each module GUI for each module can only display the data GUI for each module can not control the hybrid simulation Yes 1 enable the GUI for each module No 0 disable the GUI for each module MDL 1 EnableGUI 1 MDL 2 EnableGUI 1 87 o UI SIMC OR MDL 3 EnableGUI 1 oe oe oe Advanced modular parameters oe oe These parameters need to be redefined for following situations 1 Different coordinate system between UI SIMCOR and static module 2 When scale factor needs to be applied either in experiment or simulation 3 To define force and displacement criteria for tolerance and safety 4 To trigger camera modules or DAQ system E 5 When LBCB at UIUC is used for experiment 6 When NHCP protocol i
27. ABAQUS 6 00E 03 4 00E 03 2 00E 03 0 00E 00 2 00E 03 Displacement m 4 00E 03 6 00E 03 8 00E 03 0 0 5 1 1 5 2 2 5 3 3 5 4 4 5 5 Time sec Figure 40 Pseudo dynamic test result Horizontal displacement at control point 1 continued 83 o UI SIMC OR 8 5 SAC Three Story Steel Structure 8 5 1 Structural configuration A three story steel structure model is presented in this section The structure is consists of three bay and three story Whole structural model is given for FEDEAS Lab ZEUS NL and ABAQUS Then the whole structural model is subdivided into three modules one external beam column joint one internal column and remaining frame Procedure for configuration of network and running the simulation are same with the previous examples The whole structure is three story and three bay as shown in Figure 42 Lumped masses are assigned to every beam column joint One external beam column joint and one internal column will be physically tested In the whole model there is a lumped mass at the external beam column joint that will be tested As discussed in Section 4 all nodes with mass or with DOFs of interest should be defined in the SimConfig m file and those nodes will be displacement controlled The external beam column joint however will be tested using only two control points Thus the lumped mass in the tested beam column joint is shifted to connection point as shown in Figure 42 T
28. COCOS l l 0 03 J y E 000 d l l 0 04 X Dynamic Loading Stage Initial Stiffness I I Formulation Stage b TORREN Static Loading 1 Tum Stage Displacement in Z mm Load Step Figure 51 Displacement in Y and Z during PSD test 105 Displacement in Z mm Displacement in Y mm Displacement in X mm 250 200 150 100 50 o UI SIMC R Whole ABAQUS No Imperfection PSDTest ABAQUS Time sec 0 5 1 Whole ABAQUS No Imperfection PSDTest ABAQUS 2 5 uckling in Z direction Time sec Whole ABAQUS No Imperfection PSDTest ABAQUS gt Mild buckling in X direction Time sec Figure 52 PSD test results displacement 106 o UI SIMC R 40000 PSDTest ABAQUS pink eina ke 2 8000 CEE EEEE gt E 8 E5000 T ais hte fi a et a o LL NM t EP E mum E E c o 50 250 K 006 K a a K R K e e 4000 Measured Displacement in Y mm 1500 PSDTest ABAQUS z E IEEE IEEE o 2 o LL 8 2 5 7 p r X
29. DOF case is supported by NHCP Simulation Server A generalized version of NHCP will be implemented in UI SIMCOR when the development of NHCP in NEESit is completed o UI SIMC OR 6 Protocol Specification 6 1 NEESgrid Teleoperation Control Protocol NTCP 6 1 1 Background and introduction NTCP is the first communication protocol for the hybrid simulation developed by NEESit Recently NEESit has been developing new communication protocol NHCP but it is not officially released yet The further explanation about NHCP will be given latter UI SIMCOR uses the NTCP Toolbox for MATLAB developed by NEESit The setup procedure of NTCP is very complicated compared to other protocols The detailed explanation of NTCP including NTCP Toolbox for MATLAB can be found in NEESit website http it nees org In this section a simple explanation how NTCP works with UI SIMCOR will be given 6 1 2 NTCP server connection with UI SIMCOR Two types of plugin are used for the NTCP server to connect to the backend client such as FEDEAS Lab ZEUS NL OpenSees or experimental equipment Currently OpenFresco and NHCP Simulation Server do not support the NTCP The plugins are MATLAB plugin and LabVIEW plugin As FEDEAS Lab is running on the MATLAB the connection to FEDEAS Lab is made using MATLAB plugin For the communication through MATLAB plugin NTCP server acts as a server for both ends Thus the NTCP server waits for the connection from UI SIMCOR
30. FORC M MOM NHCP MM2 run T DISP T ROT where run for running the simulation and t_p1sp and r sor are target displacement and rotation which are sent from UI SIMCOR and double float respectively For simip and w 1 modes there is one return variable i e M_FORCE measured force and UI SIMCOR assumes that the measured displacement 1s same with target displacement For mm2 mode there are four return variables i e M prsP measured displacement m mor measured rotation w FORC M MOM measured moment MiniMOST 1 in UIUC or SDSC has one DOF in translational direction i e x direction and MiniMOST 2 in UIUC or SDSC has two DOFs in translational direction and rotation 1 e x and rz direction For examples M FORC NHCP Sim1D run 10 M DISP M ROT M FORC M MOM NCHP MM2 run 10 0 1 Syntax NHCP NHCPMode close where close for disconnecting UI SIMCOR from NCS and SimServer or Mini MOST 1 or 2 computer There is no return value during close action instead there will be error messages in the NCS if there are some errors during close action For example NHCP MM1 close 26 o UI SIMC OR T Installation of UI SIMCOR All the files necessary to run UI SIMCOR including NEES SAM are posted on the NEESforge website https neesforge nees org projects simcor After installation is complete the user should be able to run the examples in
31. In this example it is assumed that there is only x directional mass MDL RF Class MDL i name This variable define the name of each module MDL i URL MDL i URL directs the address of server and port number for each module For local machine MDL i URL 127 0 0 1 port number MDL i protocol This variable defines communication protocol for each module Four protocols are available in the current version of UI SIMCOR as follows NTCP communicate through NEESPOP server TCPIP binary communication using TCP IP Lab View ASCII communication with LabView plugin format Propose Query Execute Query LabView2 same as LabViewl but Propose Query OpenFrescol1D OpenFresco only 1 DOF is implemented now NHCP NHCP linear 1 DOF simulation mode Mini MOST 1 and 2 at UIUC or SDSC When FEDEAS Lab or ABAQUS is used as a static analysis module only LabView2 protocol is available until now When OpenFresco or NHCP Simulation Server is used as a static analysis module only OpenFrescolD or NHCP protocol is available until now MDL Z node This cell array defines the control points that are connected to module i Since there is only one control point in simulation coordinator and there is one module it is naturally 1 From latter examples such as MOST example or SAC frame you will get more explanation about node definition for each module MDL i EFF DOF In three dimensional space there are 6 DOFs for each node If the stru
32. MALTAB command windows 10 Backup the results files MDLOI recv txt Netwklog txt NodeDisp txt located in C SIMCOR 03_Examples UCB SDOF 00_Coordinator for UIUC site and another results files NetwkLog txt UCB_Results txt located in C SIMCOR 04_API 01_UCB for UCB site 11 Send results to UIUC 170 o UI SIMC R Step 7 Pushover test for UCB specimen Purpose To calculate the scale factor for UI SIMCOR and to make analytical model for comparison Most of procedure is same with Step 5 except the running PushOver instead of UI SimCor The default value of the displacement limit is 6 If the target displacement which will be sent to xPC Real time target is greater than the displacement limit then the simulation will be stopped with error message If you want to modify the displacement limit open the API Main sim for simulation or API Main for experiment and modify Disp Limit variable at line 17 1 Run a MATLAB for communication with xPC Real time target and change the directory to C SIMCOR 04_API 01_UCB 2 Type API Main sim for simulation or API Main for experiment then you will see the message box confirming the displacement limit as shown Figure A 11 Check Displacement Limit i ml xj Current displacement limit for UCB speicmen 6 00e 000 IF you want modify the displacement limit please open API_Main_sim m Reset Figure A 11 Warning message a If displacement limit is confirmed click Ok
33. OpenSees Tcl Tk interpreter also should be installed to run OpenSees and OpenFresco This 1s also needed for NEES SAM when OpenSees is used as static analysis module OpenFresco including user s manual can be downloaded from the following link https neesforge nees org projects openfresco For the NHCP the OpenSSL is required for communication between UI SIMCOR and NCS OpenSSL can be downloaded from the following link http www openssl org Folder List after Installation The folder list of installed UI SIMCOR is shown in Figure 14 3T o UI SIMC R CASIMCOR 101 SIMCOR MDL_AUX E MDL_RF parmatlab 102 NEES ABAQUS 102 NEES FL FedeasLab 102 NEES SAM T NEES_OpenSees NEES_ Zeus 03_Examples 6Beam5Col BUCKLE_InitLoading MOST MOST_LBCB SAC SDOF SDOF_Concrete SDOF_InitLoading ThreeSite UCB MOST PushoverTest SDOF 01_UCB Figure 14 Folder list of UI SIMCOR 7 4 Downloading and Updating UI SIMCOR Source Code The source code of UI SIMCOR is already installed in the 01 SIMCOR folder of UI SIMCOR installation directory However if you are interested in extending the software the latest version of UI SIMCOR source code is available for download using Subversion SV The Subversion is a version control system for software systems a major component of Source Configuration Management SCM of SIMCOR Project in 32 o
34. Simulation results continued 9 2 2 Hybrid test between UIUC and UCB by simulation After verification of the configuration of UI SIMCOR hybrid test between UIUC and UCB was conducted by simulation to verify the communication The results are shown in Figure 69 and the Expected and Actual mean the simulation results at UIUC only and at UIUC and UCB respectively As shown in figure the test results simulated at UIUC and UIUC and UCB is identical so the communication between UIUC and UCB was verified 133 Displacement Force o UI SIMC OR HSF UIUC site 40 30 Expected 20 7 10 0 10 20 30 40 0 0 5 1 1 5 2 2 5 3 3 5 4 4 5 5 Time HSF UIUC site 150000 Expected 400000 L iak Actual 50000 r 40 50000 0 10 20 30 40 100000 150000 200000 2 0000000 Displacement a Hybrid simulation framework Figure 69 Simulation results 134 o UI SIMC OR Remote UCB site 4 3 Expected 2 4 E D amp 0 1 a 2 3 4 0 0 5 1 1 5 2 2 5 3 3 5 4 4 5 5 Time Remote UCB site 3 Expected sse Actual 2 f e 4 3 4 3 Displacement a Hybrid simulation framework Figure 69 Simulation results continued 9 2 3 One site experiment at UCB The coupon will be used in this two site hybrid test is very similar to coupon used in the previous push
35. The PSD test for this example requires two processes one for UI SIMCOR and the other one for static analysis module Both processes can be run in a single computer or in multiple computers The procedure for running this example is given as below 1 Prepare configuration file The network setting e g port number in each static analysis module configuration file should be identical with the simulation coordinator file should be carefully set 2 Prepare static module a If FEDEA Lab or ABAQUS is used i Starta MATLAB ii For FEDEAS Lab Change current directory to the folder where FedeasCFG m and the model file SDOF m exist For ABAQUS Change current directory to the folder where AbaqusCFG m and the model file SDOF inp exist 53 ill o UI SIMC OR For FEDEAS Lab Run FEDEAS Lab by inputting following command NEES FL LabView2 enter for LabView2 protocol See Figure 21 For ABAQUS Run ABAQUS by inputting following command NEES Abaqus LabView2 enter for LabView2 protocol See Figure 22 b If ZEUS NL or OpenSees are used with NEES SAM 1 il iii Start NEES SAM by double clicking NEESSAM TCPIP exe tepip protocol or NEESSAM LabViewl exe LabVIEW1 protocol or NEESSAM LabView2 exe LabVIEW protocol Select the configuration file from the folder where configuration file and model file exist Check the open port and status See Figure 23 c If OpenFresco is used 1 il ill Start
36. UCB were used in this test and the experimental results show good agreement with the simulation results Therefore UI SIMCOR could be effectively used for hybrid test among geographically distributed equipments The three site example in this report has only one DOF at each site so it is required to use more DOFs during the hybrid test for more extensive verification of UI SIMCOR 157 o UI SIMC OR For this purpose the Loading and Boundary Condition Box LBCB which control 6 DOF will be considered for another three site test in the near future 9 5 Two site Experiment between UIUC and SDSC using NHCP To verify the functionality of UI SIMCOR with NHCP geographically distributed hybrid simulation between UIUC and SDSC was conducted at January 29 2007 The module 1 of MOST example structure explained Section 8 4 was replaced with MiniMOST 1 in UIUC using LabVIEW1 protocol module 3 is replaced with MiniMOST in SDSC using NHCP and module 2 is modeled by ZEUS NL with LabVIEW protocol However the rotational DOF of module 1 is removed because MiniMOST 1 in UIUC has only one DOF in translational direction The stiffness of Mini MOST 1 in UIUC and in SDSC can be found in the report of Installation of UI SIMCOR at SDSC and Implementation of NHCP in UI SIMCOR Spencer et al 2006 So the scale factor of each MiniMOST 1 can be obtained as follows Table 6 Scale factor of each MiniMOST 1 Site Displacement Rotation Force Moment
37. UI SIMC OR NEESforge website Subversion is commonly used in software development to record the history of sources files and documents At the top of the SIMCOR Project page of NEESforge click on SCM tab and you can see the Figure 15 E NEESforge UI SimCor Simulation Coordinator SCM Repository Microsoft Internet Explorer File Edit View Favorites Tools Help ev Q Back gt D P A gt Favorites ga Address e https neesforge nees org scm group id 21 Go z Ad d NEESforge Search the entire project Search ae Log In New Account UI SimCor Home My Page Project Tree Code Snippets Project Openings Simulation Coordinator Summary Forums Tracker Lists Tasks Docs Surveys News ETZN Files Documentation for Subversion sometimes referred to as SVN M Repositor SES Browse the Subversion Tree Anonymous Subversion Access Browsing the SVN tree gives you a great view into the current status of this project s code You may also view the complete histories of any file in the svn checkout repository https scm neesforge nees org svn simcor This project s SYN repository can be checked out through anonymous access with the following command s Developer Subversion Access via DAV browse Subversion Repository Only project developers can access the SVN tree via this method Substitute developername with the proper values Enter your site password whe
38. UI SIMCOR The communication protocol between MATLAB and OpenFresco is also provided by Schellenberg et al 2006 in MEX file format TCPSocket mexw32 The MEX file is dependent on the MATLAB version In the UI SIMCOR there are three different MEX files i e TCPSocket v65 mex TCPSocket R2006a mexw32 and TCPSocket R2006b mexw32 for MATLAB v6 5 R2006a and R2006b respectively The default MEX file for OpenFrescolD protocol sets to MATLAB version R2006b If user wants to MEX file for MATLAB v6 5 or MATLAB R2006a change file name from TCPSocket v65 mex or TCPSocket R2006a mexw32 into TCPSocket mex or TCPSocket mexw32 If a difference version of MATLAB is used the user needs to create a new MEX file based on the User Manual for OpenFresco Schellenberg et al 2006 The MEX file developed for the communication between MATLAB and OpenFresco is utilized in UI SIMCOR for communication with OpenFresco Currently only 1 DOF simulation case is supported even though OpenFresco provides two and three actuator cases and generic element which will be implemented in the next version of UI SIMCOR This section is based on the User Manual for OpenFresco so more details about OpenFresco can be found in the OpenFresco project page of NEESit https neesforge nees org projects openfresco 6 4 2 Usage of OpenFrescoID protocol setup connection Syntax socketID TCPSocket openConnection PORT OpenFresco IP OpenFresco where socket t
39. Unit mm N sec oo oe by Oh Sung Kwon okwon2Quiuc edu modified by Kyu Sik Park kspark uiuc edu Univ of Illinois at Urbana Champaign oe oe oe oe Last updated on 2007 01 26 11 05PM oe oo oo oe Common parameters oe Ground acceleration file name with extension The file should contains two 5 columns for time and acceleration The unit of acceleration should be consistent with the mass time and force i e mass acc force Sys GM Input elcentro dat Ground acceleration scale factor This factor will be multiplied to acceleration before starting simulation Sys GM SC 9810 Direction of ground acceleration x y or zZ Sys GM direction x Integration parameter related to the alpha OS method Alpha 0 1 3 In most cases SC Alph 0 05 worked Sys Alph 0 05 Sys Beta 1 4 1 Sys Alph 2 Sys Gamm 1 2 Sys Alph Evaluate Stiffness Yes 1 to run stiffness evaluation test No 0 to read stiffness matrix from file In this case there should exist oe stiffness matrices of individual module in the files MDLO1_K txt DL02 K txt etc ys Eval Stiffness 1 oe LD oe Number of initial static loading steps When ther xist static constant loading i e gravity forces apply then in Zeus NL or OpenSees as a incremental loading with n steps In this file SimConfig m specify the number of static steps in the following var
40. and K are matrix not scalar M K K K u M L 0 i K K K lu 0 1 A L 0 K Ky Kn 0 I M T K K K u MIA 0 K K K u 0 0 K K Ky ln 0 K K K u Mj A M ii F K K K u 0 0 Ke K Kaju 0 0 Condense out k DOfs K K K u F K K Kj u 0 K Ky Kallu 0 K K K el u K K u F K Kk k Kel I KJ i e ae u n ji Ji jk j ji Ji j J After rearranging terms i A 0 oja K K 4 0 ojl or M a Kd MIA o UI SIMC OR 4 3 MOST Experiment As described in the previous section the equation of motion of a structure can be established using only effective DOFs i e DOFs where lumped masses are defined or DOFs of our interest The MOST experiment example consists of two portal frames as given in Figure 4 a The stiffness matrix of the structure will be a size of 11 by 11 with DOFs in Figure 4 a Assuming that there are no axial deformations in columns and we are not interested in the rotation of hinge connections and rotation of mid column beam connection the number of DOFs can be reduced to 4 as shown in Figure 4 b The rotational DOF of the left column is of our interest since the column will be tested using two DOFs on top All the translational DOFs should be included since mass is defined in the DOFs and inertial forces should be applied Also it is assumed that the lumped masses have x directional compon
41. by one click run the all steps automatically by clicking the start button 54 o UI SIMC OR 5 There is a simulation monitor for each module This monitor shows the current status of hybrid simulation HSF option shows the results after scaling whereas Remote site shows the results before scaling Please note that the above procedure should be followed whenever new analysis starts Command Window Waiting for connection from client through port 11997 Figure 21 MATLAB command window after NEES FL LabView2 m command Command Window Waiting for connection from client through port 11999 Figure 22 MATLAB command window after NEES Abaqus LabView2 m command G NEES SAM TCPIP EN 16159 File Edit View Help static analysis module port number module name Ready Standby for connection SDOF 1188 Figure 23 TCP IP protocol window Zeus NL 7 55 WWINDOWSWsystem32 Wcmd exe OpenFresco Microsoft Windows XP Version 5 1 2666 lt C Copyright 1985 2001 Microsoft Corp D WHS FHExamplesWOpenFrescoiD_SDOF gt OpenFresco OpenFresco Open Framework for Experimental Setup and Control Version 2 8 Copyright lt c gt 2886 The Regents of the University of California fill Rights Reserved OpenFresco gt source SDOF tcl Socket successfully created Waiting for Simulation Application Client Figure 24 WINDOW DOS command window after running OpenFresco ex C WWINDOWS Wsyst
42. for CP 2 100000 Effective DOF for CP 3 Module 3 Right column MDL 3 node 3 Control point node number MDL 3 EFF DOF 1 0 00 0 0 Effective DOF for CP 3 Dismplacement for preliminary test for each module Del t Translation Del r Rotation in radian MDL 1 DEL t 0 005 MDL 2 DEL t 0 005 MDL 3 DEL t 0 005 MDL 1 DEL r 0 002 MDL 2 DEL r 0 002 MDL 3 DEL r 0 002 Enable GUI for each module GUI for each module can only display the data GUI for each module can not control the hybrid simulation Yes 1 enable the GUI for each module No 0 disable the GUI for each module MDL 1 EnableGUI 1 MDL 2 EnableGUI 1 MDL 3 EnableGUI 1 oe oe oe Advanced modular parameters oe oe These parameters need to be redefined for following situations 1 Different coordinate system between UI SIMCOR and static module 2 When scale factor needs to be applied either in experiment or simulation 3 To define force and displacement criteria for tolerance and safety 4 To trigger camera modules or DAQ system 5 5 When LBCB at UIUC is used for experiment 6 When NHCP protocol is used oe oe URL of remote site and NHCP mode for NHCP for i 1 Sys Num RF Module if strcmp lower MDL i protocol nhcp MDL i remote URL 127 0 0 1 99999 MDL i NHCPMode simld end 97 o UI SIMC OR end Stiffness
43. for UI SIMCOR Change current directory to the folder where SimConfig m and ground motion file is located Run UI SIMCOR by inputting following command UI SIMCOR enter 8 4 5 Result and verification MOST and SAC next section examples use artificial earthquake as shown in Figure 40 The results of PSD test using UI SIMCOR are very good correspondence with those of whole model run as shown in Figure 41 0 25 02 Fr 0 15 r 0 05 r 0 05 0 1 0 15 Fr 0 2 L 0 25 Acceleration g eo 0 0 5 1 1 5 2 2 5 3 3 5 4 4 5 5 Time sec Figure 40 Artificial earthquake 81 0 008 0 006 0 004 0 002 0 002 Displacement mm eo 0 004 0 006 0 008 0 008 0 006 0 004 0 002 0 002 Displacement m o 0 004 0 006 0 008 o UI SIMC OR Whole Model FEDEAS Lab dtdd PSD Test FEDEAS Lab 0 5 1 1 5 2 2 5 3 3 5 4 4 5 5 Time sec Whole Model ZEUS negat PSD Test ZEUS 0 5 1 1 5 2 2 5 3 3 5 4 4 5 5 Time sec Figure 40 Pseudo dynamic test result Horizontal displacement at control point 1 82 o UI SIMC OR 0 008 Whole Model OpenSees Leche PSD Test OpenSees 0 006 p 0 004 0 002 p 0 002 p Displacement m o L 0 004 0 006 0 008 0 0 5 1 1 5 2 2 5 3 3 5 4 4 5 5 Time sec 8 00E 03 Whole Model ABAQUS Sae PSD Test
44. for i 1 Sys Num RF Module MDL i TransM oe end oe Scale factor for displacement rotation force moment Experimental specimens are not always in full scale Use this factors to apply scale factors The displacement scale factors are multiplied before they ar sent to module Measured force and moments are divided with scale factors before used in the PSD algorithm for i 1 Sys Num RF Module MDL i ScaleF 1 1 1 1 Module i oe oe oe oe oe end oe Relaxation check If this parameter is 1 UI SimCor send commend to retrieve data and check relaxation just before the execution of proposed command If it s 1 the checking criteria needs to be provided for i 1 Sys Num RF Module MDL i CheckRelax 0 Module i if MDL i CheckLimit 1 define following variables Variable size should be number of control nodes 6 array oe oe oe oe oe oe Displacement variation ratio not increment MDL i MES D inco abcdef das IE Force variaiton ratio not increment MDL i MES_F_inc abcde f ds IE oe oe oe end oe Check displacement and force limit At every steps check if the displacement or force are approaching to the limitation of the equipments stroke or force capacity for i 1 Sys Num RF Module MDL i CheckLimit 0 Module i oe oe 99 o P oo end oo o UI SIMC OR if MDL i CheckLimit 1 define following vari
45. force 0 113 y rotation 0 0 z rotation 0 0 This is an often called method and should be asynchronous of the propose execute logic if possible In the LabVIEW implementation a separate loop in the control program handles these requests and returns instant DAQ readings enabling real time data as a move happens It uses a LabVIEW semaphore to mark the TCP send as a critical section so that the various threads do not collide in communicating with NTCP server get parameter Syntax get parameter TransactionID ParamName Return syntax OK 0 ParamName Parameter This is mainly used by the MATLAB code to pass simulation parameters around set parameter Syntax set parameter TransactionID ParamName Parameter This is the mirror of get parameter The plugin sends whatever it gets and LabVIEW duly ignores it with an OK response I5 o UI SIMC R 6 4 OpenFresco1D Protocol 6 4 1 Background and introduction OpenFrescolD protocol is used for OpenFresco module OpenFresco is a software framework intended to facilitate and help standardize the local or geographically distributed deployment of hybrid simulation and developed by UCB Schellenberg et al 2006 OpenFresco cannot perform hybrid simulations by itself it simply mediates in a modular and highly structured manner instructions between a host client numerical simulation computer and laboratory equipment It uses TCP connection to connect with simulation coordinator e g
46. in UIUC has 6 actuators so 6 displacements should be sent to LBCB for experiment For this purpose two variables i e MDL i LBCB and MDL i LBCB TransM are used in the UI SIMCOR If the LBCB is used for experiment the target displacements of all DOFs will be assigned as zero except effective DOFs Furthermore the transformation matrix for LBCB needs to be defined and this transformation matrix is dependent on the location of LBCB plates More details about LBCB can be found in User s Manual for MUST SIM Facility UIUC 2005 8 6 1 Structural configuration The MOST example used in Section 8 4 is used for this LBCB example The configuration of this example is exactly same with Section 8 4 except the left column as shown in Figure 46 and the left column represents the LBCB UI SIMCOR Communication Communication Communication protocol protocol protocol API API API All DOFs are free LBCB Module 2 Module 3 Figure 46 Simulation configuration for MOST_LBCB example 94 o UI SIMC GOR 8 6 2 Simulation configuration file The simulation configuration file of this example is exactly same with Section 8 4 except the LBCB variables CANSIMCORNOS Examples MOST LBCBNOO CoordinatorNSimConfig m function Sys MDL AUX SimConfig MDL MDL_RF AUX MDL_AUX Type definition Do not delete this line oe oe Configuration parameters fo
47. limitation of the equipments stroke or force capacity for i 1 Sys Num RF Module MDL i CheckLimit 0 Module i if MDL i CheckLimit 1 define following variables Variable size should be number of control nodes 6 array oe oe o o9 oe oe Displacement increment limit not ratio DL i TGT D inc abcdeef oe ae x oe Displacement limit DL i CAP D tot abcdeef oe oe oe Force limit MDL i CAP F tot a bc d e f EJ d A Displacement toleranc ratio MDL i TOL D inc abcdef end 98 o UI SIMC R Loading and Boundary Condition Box LBCB case If it s 1 the coordinate transformation matrix needs to be provided This can be also used for any other actuator which has diffrence number of DOF coordinate with those of UI SIMCOR oe oe oe oe MDL 1 LBCB 1 MDL 1 LBCB TransM 01 0 0 O O0 1 0 0 0 0 0 001000 00001 0 000 10 0 00000 1 oe oe oe Advanced modular parameters oe oe These parameters need to be redefined for following situations 1 Different coordinate system between UI SIMCOR and static module 2 When scale factor needs to be applied either in experiment or simulation 3 To define force and displacement criteria for tolerance and safety 4 To trigger camera modules or DAQ system oe oe Coordinate transformation If it needs the transformation matrix also needs to be provided
48. oe oe o oe oe o Connection port to controller MDL Port 11997 Module node number Order of node number should be identical to the corresponding node number in the simulation coordinator MDL MDL_Node 2 52 o UI SIMC OR Effective DOFs 1 specifiy the DOFs controlled by displacement MDL EFF_DOF 1 1 00 0 0 0 o Abaqus input file with extension MDL ABAQUS_MDL SDOF inp Dimension of the model 2 or 3 3 Dimensional model are not tested yet using Abaqus MDL MDL Dim 2 The number of static analysis steps for initial loading Added by GunJin Yun MDL Num Static Step 0 oo Number of Stiffness test Added by GunJin Yun If stiffness is evaluated through experiment th valuation need to be don several times and the average of the results are used as the initial stiffness This parameter is used when MDL Eval Stiffness 1 MDL Num Test Stiffness 1 oe oe oo Variable Description All variables except following two parameters are exactly same with FEDEASE Lab except two variables MDL Num Dynamic Step This variable defines the number of dynamic analysis steps It should be same with Sys Num Dynamic Step variable in simulation configuration file MDL Num Test Stiffness This variable defines the number of stiffness test It should be same with Sys Eval Stiffness variable in simulation configuration file 8 1 4 Running simulation
49. the IP address and port number of UCB site The default value of the displacement limit is 6 If the target displacement which will be sent to xPC Real time target is greater than the displacement limit then the simulation will be stopped with error message If you want to modify the displacement limit open the API Main and modify Disp Limit variable at line 17 UIUC Site 1 Run a MATLAB for UI SIMCOR and change the directory to C SIMCOR 03_ Examples UCB SDOF 00_ Coordinator 2 Type Ul SimCor and enter in MALAB command window to run UI SIMCOR New simulation xj e Please backup previous simulation results if necessary All previous files will be deleted cae Figure A 18 Warning message a Click Okay button then you will see the initialize message in a MATLAB command window All previous files will be deleted so if necessary backup previous results b Two popup windows will be displayed one window for control and other window for monitoring will popup In this example only one monitoring window will be displayed as this example has only one remote site 175 o UI SIMC R MUST 5I Feelin DS Ground Motion o0 f Stop by step contrat Al steps by ore cick Establish Connection for distributed pseudo dynamic ees FEDEASLab TEPIP network 08 Of Ds 02 0 02 Os 0 Of 1 x D 4 Mach E Sms Message Figure A 20 Monitoring window ee Site Run another MATLAB for
50. to send and receive information OpenFresco communicate with OpenFrescolD protocol NHCP Simulation Server MiniMOST 1 and 2 at UIUC or SDSC communicate with NHCP However OpenFrescolD and NHCP are not generalized yet The generalized version of OpenFrescolD and NHCP will be available in the next version of UI SIMCOR If the experiment site uses different protocol to communicate with actuator then the experiment site should provide a method of communication Any program language which can impose displacement and collect measured displacement and force from experiment can be used The Application Program Interface API is also need for communication between UI SIMCOR and experiment site The API will be further explained through the experiment example in Section 9 1 Figure 1 shows the framework of the simulation The complete segregation of integration scheme from static analysis module allows any combination of experiments and analyses Simulation Coordinator Object J of MDL RF class Component 7 Main Routine DOF Mapping Object n of MDL_RF class Static Equilibrium Z TCP IP Network sever API Dynamic Equilibrium Equipments E DAQ oo0o0nn i Pe e AUX gt 5 e ny S Simulation Control o o cp o lt Figure 1 Hybrid simulation framework o UI SIMC OR 3 Pseudo Dynamic Test Procedure In UI SIMCOR pseudo dynamic test consists of roughly four stages init
51. 1 5842359e 003 end end oe Coordinate transformation If it needs the transformation matrix also needs to be provided for i 1 Sys Num RF Module MDL i TransM end oe Scale factor for displacement rotation force moment Experimental specimens are not always in full scale Use this factors to apply scale factors The displacement scale factors are multiplied before they ar sent to module Measured force and moments are divided with scale factors before used in the PSD algorithm for i 1 Sys Num RF Module MDL i ScaleF 1 1 1 1 Module i end oe oe oe oe oe oe Relaxation check If this parameter is 1 UI SimCor send commend to retrieve data and check relaxation just before the execution of proposed command If it s 1 the checking criteria needs to be provided for i 1 Sys Num RF Module MDL i CheckRelax 0 Module i if MDL i CheckLimit 1 define following variables Variable size should be number of control nodes 6 array oe oe oe o o9 of o Displacement variation ratio not increment MDL i MES D inco abcde f m 1 Force variaiton ratio not increment MDL i MES F inco abcdef EJ 13 oe oe o end oe Check displacement and force limit At every steps check if the displacement or force are approaching to the limitation of the equipments stroke or force capacity for i 1 Sys Num RF Module MDL i CheckLimit 0 Module i if MDL
52. 1 SDSC Expected Actual 05 L 1 1 1 1 0 1 1 1 1 0 0012 0 001 0 0008 0 0006 0 0004 0 0002 0 0002 0 0004 0 0006 0 0008 0 001 Force 45 Displacement Figure 89 MiniMOST 1 at SDSC Hybrid test by experiment 500 steps Figures 90 92 show the three site hybrid test results It took 54 min and 12 sec for 500 steps 6 5 sec step to finish the pseudo dynamic stage in UI SIMCOR For MiniMOST 1 at UIUC and SDSC consume few seconds to stabilize the load cell to get the good measurements This is one of the reasons of the slow test During the test following websites were used to see the status of each equipment through webcamera For MiniMOST 1 at UIUC http cee axs3 cee uiuc edu view indexFrame shtml For uNEES at UCB http neespopv berkeley edu portal section local_video For MiniMOST 1 at SDSC http users sdsc edu ljmiller side tools RDVmm jnlp As shown in figures there are little difference between simulation and experiment especially at UCB site because of difference of coupon at UCB specimen The analytical model of uNEES specimen at UCB was obtained by using the pushover test results conducted during August 9 2006 August 16 2006 Furthermore the characteristic of specimen at UCB is highly dependent on the coupon installed in the specimen The coupon used in this test is little bit different from the coupon used for pushover test so there are little difference bet
53. 2 respectively These four files are combined into one MEX file i e NHCP dll for UI SIMCOR If file is compiled within MATLAB then the file will depend on the MATLAB version so we compiled the source code within Microsoft Visual Studio 2005 to make Dynamically Linked Library DLL file NCS server for NHCP and SimServer simulation server for NHCP are also modified to receive the port number for NCS and port number and stiffness for SimServer respectively 6 5 5 Usage of NHCP UI SIMCOR will call NHCP dll file if NHCP protocol is used for hybrid simulation Currently there are three modes in NHCP i e simulation mode with linear 1 DOF system MiniMOST 1 and 2 modes NEESit has not generalized the NHCP so NHCP for UI SIMCOR is restricted for a special case as mentioned in the above However when NEESit releases the generalized NHCP it will be implemented into UI SIMCOR There are three actions in each mode of NHCP i e open run and close During open action UI SIMCOR is connected with NHCP server 1 e NCS through OpenSSL connection and NCS is connected with simulation server 1 e SimServer for simulation mode or computer for Mini MOST 1 or 2 using TCP connection During run actions propose target displacement and query measurement are conducted simultaneously Processes for execute or query used in NTCP or LabVIEW protocol are not required Finally the UI SIMCOR and NCS are disconnec
54. 2_NEES FL FedeasLab Section_Lib C ASIMCOR 02_NEES FL FedeasLab Solution_Lib C SIMCOR 02_NEES FL FedeasLab Utilities Path for ABAQUS CASIMCORY02 NEES ABAQUS For simple procedure 1 Runa MATLAB 2 click File Set Path Add with Subfolders in the menu bar Select CASIMCORY01 SIMCOR and click the OK button 3 Repeat step 2 for CASIMCORY02 NEES FL CASIMCORY02 NEES ABAQUS and C SIMCOR 04_API 01_UCB 4 Click Save button to save the paths 5 Click Close button Step 5 Run the preliminary test by simulation API Main sim and UI SimCor in UCB site only Purpose Verify the configuration of newly installed UI SIMCOR The default value of the displacement limit is 6 If the target displacement which will be sent to xPC Real time target is greater than the displacement limit then the simulation will be stopped with error message If you want to modify the displacement limit open the API Main sim and modify Disp Limit variable at line 17 1 Run a MATLAB for UI SIMCOR and change the directory to CASIMCORY0O3 Examples UCB SDOF 00_ Coordinator 2 Type UlL_SimCor and enter in MALAB command window to run UI SIMCOR New simulation xj Please backup previous simulation results if necessary All previous files will be deleted cr Figure A 1 Warning message a Click Okay button you will see the initialize message in a MATLAB command window All previous files will be deleted so if necessary backup previous results b Two pop
55. 4 0 0 5 1 1 5 2 2 5 3 3 5 4 4 5 5 Time sec Figure 28 El Centro earthquake 58 o UI SIMC OR 20 Whole Model FEDEAS Lab 415 JL gt 2638 PSD Test FEDEAS Lab Displacement mm eo 0 0 5 1 1 5 2 2 5 3 3 5 4 4 5 5 Time sec 20 Whole Model ZEUS jg L6 PSD Test ZEUS Displacement mm eo 0 0 5 1 1 5 2 2 5 3 3 5 4 4 5 5 Time sec 20 L Whole Model OpenSees dE J esee PSD Test OpenSees Displacement mm eo 0 0 5 1 1 5 2 2 5 3 3 5 4 4 5 5 Time sec Figure 29 Pseudo dynamic test result Horizontal displacement of top node 59 o UI SIMC OR 25 20 Whole Model ABAQUS 45 L sexes PSD Test ABAQUS 10 E L o E 7 S 5r Ey A 10 15 L 20 L 25 0 0 5 1 1 5 2 2 5 3 3 5 4 4 5 5 Time sec 25 20 Whole Mada ZEUS 454 sess PSD Test OpenFresco 0 E 10 5 L 0 amp 5 Q i O 10 15 20 25 0 0 5 1 1 5 2 25 3 3 5 4 45 5 Time sec 25 20 H Whole Model ZEUS b L p am PSD Test NHCPSim1D L E 10 L 5 LE 0 a 5 Q QO 10 15 20 Y 25 T T T T T T T T T 0 0 5 1 1 5 2 25 3 3 5 4 45 5 Time sec Figure 29 Pseudo dynamic test result Horizontal displacement of top node continued 60 o UI SIMC OR 8 2 Inelastic Cantilever Column with Initial Loading Ste
56. 70 0 0081 S Strain Figure 37 Steel and concrete response history o UI SIMC R Strain stress of concrete extreme fiber Retrieved from Zeus analysis 4 00E 6 4 5 00E 00 015 0 0010 1 50E 01 2 00E 01 2 50E 01 3 00E 01 3 50E 04 0 0005 Strain BR o o eo N e Displacement mm Whole Model ZEUS PSD Test ZEUS 0 0 5 1 1 5 60 Displacement mm 2 2 5 Time sec Whole Model OpenSees PSD Test OpenSees 3 5 4 4 5 0 0 5 1 1 5 Figure 38 Pseudo dynamic test result Horizontal displacement of top node 2 2 5 Time sec 72 3 5 4 5 o UI SIMC OR 8 4 MOST Example 8 4 1 Structural configuration The MOST experiment was briefly introduced in Section 4 3 of this document to explain the concept of control points and effective degree of freedoms The structural properties of the MOST experiment are well documented in the NEESgrid 3 2 documentation Nakata et al 2003 Thus this section focuses on the preparation of the configuration file for UI SIMCOR and static analysis modules The former experimental parts and computational parts can be replaced with FEDEAS Lab ZEUS NL OpenSees and ABAUQS UI SIMCOR Communication protocol Communication protocol Communication protocol API API
57. B To verify UI SIMCOR with other experimental site one site experiment with NEES facilities at UCB was conducted during August 9 2006 August 16 2006 This test was conducted as preliminary test for two site hybrid test between UIUC and UCB The two site hybrid test will be explained in the next section 9 1 1 Experiment setup NEES laboratory The uNEES facility at UCB was used for the hybrid test The overview of the small scale cantilever specimen setup is shown in Figure 56 This setup allows one or two columns to be tested They can be tested using the same control system or two different control systems There are two identical setups each consisting of a cantilevered beam column element positioned in a self equilibrating reaction frame Each beam column element is attached to an actuator on one end At the other end a special clevis attaches the cantilever to the supporting frame The clevis allows free rotation but no translation or axial movement To provide bending stiffness and strength at the clevis specially machined coupons are bolted to the clevises end plates By selecting the number of coupons their diameter and slenderness and the manner by which the coupons are bolted to the clevises a wide variety of hysteretic loop characteristics can be obtained 117 o UI SIMC OR Figure 56 Overview of uNEES specimen Schellenberg et al 2006 The experiment setup in the UNEES laboratory uses two actuators manufactur
58. Beta and Sys Gamm These variables define the integration parameters of OS scheme The recommend range of is 0 lt 0 1 3 and a 0 05 works in the most cases B 1 4 1 t a and y 1 2 a are used during the simulation Sys Eval Stiffness This variable defines the option of evaluation of stiffness If YES 1 the stiffness evaluation test will be run If NO 0 the stiffness matrix of structure is read from file In this case there should be exist stiffness matrices of individual module in the files MDLOI K txt MDLO02 K GL etc Sys Num Static Step 40 o UI SIMC OR In UI SIMCOR PSD test is composed of four stages as explained in Section 2 During static loading stage the gravity forces are applied incrementally and there are no inertial forces from ground acceleration The gravity forces are applied incrementally since those forces can cause inelastic deformation such as cracking of concrete elements and yielding During this stage inelastic static analysis is performed using initial stiffness Initial forces are applied in the static analysis module such as ZEUS NL and OpenSees For now we cannot include initial loading in FEDEAS Lab In this example there is no initial loading Thus the number of static step is zero Next example includes initial loading Further explanation will be given in next example Sys Num Dynamic Step This variable defines the number of dynamic analysis steps Sys dt This variable defines t
59. C Real time target You will see some messages indicating that the communication is disconnected in both MATLAB windows Following output files will be saved in the following folders C SIMCOR 03_ Examples UCB SDOF 00_ Coordinator Global K txt Stiffness matrix from initial stiffness evaluation MDLOI K txt Stiffness matrix of module 1 MDLOL recv txt Measured displacement and force of module 1 NetwkLog txt Communication log file NodeDisp txt Nodal displacements C ASIMCOR 04_API 01_UCB NetwkLog txt Communication log file UCB Results txt Target displacement measured displacement and force at UCB site Two tests with different peak ground accelerations of input earthquake were conducted and the results are shown in Figures 64 and 65 125 Force Displacement o UI SIMC OR Experiment Simulation Time a Horizontal displacement at top node in the HSF Experiment Simulation 15 100000 r 120000 Displacement b Force displacement relationship at top node in the HSF Figure 64 Pseudo dynamic test results SDOF test 1 126 o UI SIMC OR 20 15 Experiment 40 Simulation j A Displacement Time a Horizontal displacement at top node in the HSF Experiment Simulation Force Displacement b Force displacement relationship at top node in the HSF Fig
60. IMCOR 03_ Examples ThreeSite 00_Coordinator Cur Disp txt Cur Forc txt and Netlog txt located in C SIMCOR 03_Examples ThreeSite 01_UIUC Cur Disp txt Cur Forc txt and Netlog txt located in CASIMCORY03 Examples hreeSiteV02 Module2 b ForUCB site Cur Disp txt Cur Forc txt and Netlog txt located in C SIMCOR 03_Examples ThreeSite 03_ UCB c For SDSC site Cur Disp txt Cur Forc txt and Netlog txt located in C SIMCOR 03_ Examples ThreeSite 04_ SDSC B 2 Step 2 Hybrid test by experiment 500 steps This is experiment so it will send target displacement to each site The purpose of this step is to conduct the actual three site hybrid experiment among UIUC UCB and SDSC UI SIMCOR MiniMOST 1 ZEUS NL model UIUC pNEES UCB and MiniMOST 1 SDSC will be tested For this test following IP address and port number of each site will be used For MM1 at UIUC 130 126 240 138 44000 For uNEES at UCB 169 229 203 152 8090 For MM1 at SDSC 137 110 118 95 44000 The same displacement limit with step 1 is also used in this step The experimental procedure introduced in this section is very simple so if necessary more detailed information running the MiniMOST LabVIEW code can be found in http users sdsc edu hubbard neesgrid mmost and User s Manual of MUST SIM Facility UIUC 2005 Prepare the communication among UIUC UCB and SDSC UIUC Site For MiniMOST 1 UIUC 1 Turn on the MiniMOST 1 system 2 Log on to the Min
61. In this documentation the simulation framework brief theoretical background test procedure protocol specification installation guide etc are introduced from Section 2 to Section 7 In Section 8 various numerical examples are introduced Single degree of freedom SDOF system is used to show how to prepare configuration files and how to run the analysis Inelastic SDOF system with initial loading is introduced to explain a method to impose an initial static force In the later version of UI SimCor the time independent static forces will be applied in UI SimCor not in static modules SDOF system with concrete material is also given to demonstrate the capability of the ZEUS NL and OpenSees MOST example and three story three bay steel structure which is a part of three story SAC building are given Furthermore example for Loading and Boundary Condition Box LBCB in UIUC is explained This example can be used as a basis when experimental equipment has different number of actuators from the number of effective DOFs in UI SIMCOR Finally various experiments using UI SIMCOR at University of Illinois at Urbana Champaign UIUC University of California at Berkeley UCB and San Diego Supercomputer Center SDSC are given in Section 9 to demonstrate the efficacy of UI SIMCOR Furthermore the detailed procedure for two and three site experiments is explained in Appendix This experiment procedure can be used as a basis of other experiment with differ
62. Module 1 Module 2 Module 3 Figure 39 Simulation configuration for MOST example 73 o UI SIMC OR 8 4 2 Simulation configuration file As opposed to the SDOF example up to Section 8 1 the MOST experiment consists of multiple control points and multiple modules as shown in Figure 39 Thus when the configuration file is prepared for the distributed simulation the user should specify the address of server where each module communicates and node numbers and effective DOFs where each module is connected Explanation will be made through the example configuration file for the MOST experiment CANSIMCORNOS ExamplesN MOSTNOO CoordinatorNSimConfig m function Sys MDL AUX SimConfig MDL MDL_RF AUX MDL_AUX Type definition Do not delete this line oo oo Configuration parameters for SDOF experiment oe oe Unit mm N sec oe oo by Oh Sung Kwon okwon2Quiuc edu modified by Kyu Sik Park kspark uiuc edu Univ of Illinois at Urbana Champaign oe o9 oe oo Last updated on 2007 01 27 11 44AM oe oo oo oe Common parameters oe oo Ground acceleration file name with extension The file should contains two columns for time and acceleration The unit of acceleration should be consistent with the mass time and force i e mass acc force Sys GM Input acc475C dat oe oe Ground acceleration scale factor This factor will be multiplied to acceleration before s
63. NUTS RE o 1500 Measured Displacement in X mm 100000 L PSDTest ABAQUS ggoao t 4 Z L fegood N E J mx Ix DxOpx Dx E HE nob aee ae 2 T 07 g EA Ces a seed 2 o LL o S aera 5 2 bd o 5 Measured Displacement in Z mm Figure 53 PSD test results hysteresis loop 107 o UI SIMC OR 8 8 7 DOF Model using Five Protocols 8 8 1 Structural configuration The main objective of this example is to verify the recently enhanced UI SIMCOR including communication This example structure is a bridge like structure with five columns and six beams This example has 7 control points and each control point only has one horizontal DOF The material properties of this example are similar to those of MOST example in Section 8 4 The hybrid simulation configuration is shown in Figure 54 UI SIMCOR TCPIP LavView2 LavView1 OpenFresco1D NHCP Sim1D Module 1 Module 2 Module 3 Module 4 Module 5 OpenSees FedeasLab ZeusNL OpenFresco NHCPSim1D Figure 54 Hybrid simulation configuration 8 8 2 Simulation configuration file Simulation configuration file contains all the information that is necessary for UI SIMCOR for the multi site simulation Many parameters are already explained in the other examples CA SIMCOR 03_Examples 6Beam5CoN00_Coordinator SimConfi
64. Please do not attempt to modify the parameters unless you understand thoroughly system BandGeneral constraints Penalty 1E20 1E20 numberer Plain test NormDispIncr 1 0e 5 10 0 algorithm ModifiedNewton integrator LoadControl 1 analysis Static e Do not use recorder command It is used by NEES SAM to make a displacement output for control points and attached nodes The following gives explanation about the configuration file for OpenSees Module CANSIMCORNOS ExamplesNSSDOFNO3 OpenSeesWodule cfg Configuration parameters for Open Sees SDOF cantilever model Unit mm N sec Generated by Oh Sung Kwon okwon2Quiuc edu Univ of Illinois at Urbana Champaign Last updated on 2006 09 19 10 52PM Connection port Port 11997 Module application 1 for Zeus NL 2 for OpenSees MDL Type 2 Effective node numbers in Simulaiton coordinator The order of node number should be idntical to that specified in the simulaiton coordinator configuration file SC Node 1 Corresponding effective node numbers in this model DL Node 2 Effective DOFs in control point EFF DOF Use one line per each control point 00000 model file name without extension MODEL FILE SDOF Model dimension MDL Dim 3 Time history monitoring point Node number direction x y z rx ry rz 51 o UI SIMC OR D for disp or F for force This should be defined after SC Node and MDL Node
65. R o UI SIMC OR 4 Sub structuring of a Structure 4 1 Concept of Static Condensation and Sub structuring A portal frame with 16 nodes and 17 beam elements are used in this section for the purpose of illustration as shown in Figure 3 a Lumped masses are located at beam column joints and ground acceleration is applied in x direction In the equation of motion of the frame there will be mass and stiffness matrices with 48 by 48 elements If we model the same frame with 3 nodes and 4 elements as shown in Figure 3 b the mass and stiffness matrix size will be reduced to 9 by 9 but the analysis result will be identical to the model in Figure 3 a as long as the level of mesh refinement does not affect the analysis result In Figure 3 b one node is placed in the middle of right column just to show the displacement of the node is of our interest Even if the structure is modeled as in Figure 3 a the structure s mass and stiffness matrix can be also reduced using static condensation if the stiffness and mass matrices are known If the stiffness matrix cannot be retrieved the condensed stiffness matrix can be determined by applying a pre specified displacement to each DOFs of interest and measuring reaction forces as shown in Figure 3 c This method allows us that the stiffness of a certain DOF can be calculated by applying certain displacement to the whole structure as shown in Figure 3 c or by applying certain displacement to segmented str
66. SDOF cantilever model oe oe oo Unit mm N sec oo oe Generated by Oh Sung Kwon okwon2Quiuc edu Modified by Kyu Sik Park kspark uiuc edu Univ of Illinois at Urbana Champaign Last updated on 2006 11 29 10 03PM oe oe oo oe Connection port to controller MDL Port 11997 Module node number Order of node number should be identical to the corresponding node number in the simulation coordinator DL MDL Node 2 Z oe oe Effective DOFs 1 specifiy the DOFs controlled by displacement MDL EFF_DOF 1 1 0 00 0 0 Fedeas Lap input file name without extension MDL FEDEAS MDL SDOF Dimension of the model 2 or 3 3 Dimensional model are not tested yet using FEDEAS Lab MDL MDL Dim 2 Variable Description MDL Port This variable defines the port number for LabView2 protocol MDL MDL Node 46 o UI SIMC OR This variable defines the node number in the module In SimConfig m it was defined that this module has node number 1 That node number was for simulation coordinator MDL Node number is node number in FEDEAS Lab which corresponds to the node number in simulation coordinator See following figure gx SIMCOR FEDEAS Lab Figure 17 Module node number definition for FEDEAS Lab MDL EFF DOF i This variable defines the effective DOF for the i node This should be identical to that defined in SimConfig m In SimConfig m it is given t
67. SDOF system in Section 8 1 is modeled using rectangular concrete section as an example of concrete model ZEUS NL and OpenSees has a functionality to mesh the element s section into fibers and apply hysteretic concrete material model The material properties are given below FEDEAS Lab OpenFresco and NHCP Simulation Server model is not given for this example In ABAQUS concrete material model is very unstable for this example so ABAUQS model is not given Model property Material Steel E 200 000 N mm Fy 500 N mm Hardening ratio 3 Core Conc f 40 N mm f 4 N mm E 0 002 Confinement factor 1 2 Cover Conc f 40 N mm f 4 N mm amp 0 002 Stress Confinement factor 1 0 a Compressive Strain Section 400 mm jure 250 300 mm y Core concrete bn Cover concrete 350 mm Figure 36 Concrete section properties All configuration files and running procedures are identical to those of Section 8 1 thus further explanation is not given in this section Use this example as a reference for modeling concrete section in ZEUS NL and OpensSees Material behaviors and PSD test results are given in Figures 37 and 38 71 Stress Strain stress of steel extreme fiber 6 00E 02 Retrieved from Zeus analysis 4 00E 02 2 00E 02 9 90E 00 Stress 0 0020 0 0010 0 0 2 00E 02 4 00E 02 6 00E 02 00 0 0010 0 9620 0 00 0 0040 0 0050 0 0060 0 00
68. SD_MM1sdsc Z MM1uiuc l 0 06 4 l 0 02 4 i 0 00 TL NMN VT 0 02 4 l i Vl T Vl I h l 0 04 4 a B 0 06 10 I Displacement m 0 08 0 10 Time sec Figure 72 Horizontal displacement of left column 138 o UI SIMC R 9 4 Three site Experiment among UIUC UCB and SDSC One site and two site experiments explained in Section 9 1 and 9 3 were preliminary tests for three site hybrid test among UIUC UCB and SDSC These tests were very successful so UIUC UCB and UCB sites are ready for three site experiment This section will explain the model for three site hybrid test including calculation of scale factors Finally the results of three site hybrid test will be explained 9 4 1 Structural configuration The main objective of a three site hybrid test is to verify the recently enhanced UI SIMCOR including communication The example structure for a three site hybrid test among UIUC UCB and SDSC is a bridge like structure with four columns and five beams This example has 6 control points and each control point only has one horizontal DOF The material properties of this example are similar to those of MOST example Nakata et al 2003 The hybrid simulation configuration of this three site experiment is shown in Figure 73 Modules 1 3 and 4 will be replaced with MiniMOST 1 at UIUC uNEES at UCB and MiniMOST 1 at SDSC respectively and Module 2 will be m
69. SIMCOR Forc txt Other than that this version is identical to the Zeus NL Version 1 3 Last update on 5757 2005 Compatible with NEES SAM Ver 1 5 Reading analysis type Reading materials Reading sections Reading element classes Reading structrual nodes Reading non structural nodes Reading element connectivity Reading restraints Reading load curves Reading loads Reading equilibrium stages Reading iterative strategy Reading convergence criteria Reading output parameters ZeusNL Solver v1 3 build 2502 This version is recompiled for UI SIMCOR to redirect the reactions and displacements of control points to SIMCOR Disp txt and SIMCOR Forc txt Other than that this version is identical to the Zeus NL Version 1 3 Last update on 5 5 2005 Compatible with NEES SAM Ver 1 5 Variable Loading Time Disp for ID 1 v Ready Request processed MMT_UIU Connecte Zeus NL 7 Figure B 6 NEES SAM window for MM1_UIUC module after establish connection 8 Click Stiffness Evaluation button in UI SIMCOR control window UI SIMCOR will read the predefined stiffness matrix from files MDLOI K txt MDL_02 txt MDL 03 txt and MDL 04 txt located in C SIMCOR 03_ Examples ThreeSite 00_ Coordinator 9 Click Apply Static Loading button in UI SIMCOR control window In this stage the gravity force is applied however there is no plan to apply gravity as this exemplary test is to demonstrate the efficacy of hybr
70. Section 8 The distributed software is composed of following components e 01 SIMCOR Simulation coordinator a main control and integration module e 02 NEES ABAQUS Interface applications for ABAQUS e 02 NEES FL Interface applications for FEDEAS Lab e 02 NEES SAM Interface applications for ZEUS NL and OpenSees e 03 Examples Various examples including experimental test e 04 API API for specific experimental site 7 1 System Requirements A MATLAB v6 5 or higher version with Instrument Control Toolbox is required UI SIMCOR itself Various structural analysis programs and some additional programs for specific communication protocols are also required Installation instruction for each software will be explained 7 2 Installation of UI SIMCOR and Interface Application The executable file for UI SIMCOR v2 6 can be downloaded form the NEESforge website Start from the NEESforge UI SIMCOR page https neesforge nees org projects simcor as shown in Figure 12 27 o UI SIMC R E NEESforge UI SimCor Simulation Coordinator Project Info Microsoft Internet Explorer Back v x iz fo es Addre ss https neesForge nees ora projects simcor X Advanced Login NEESforge Search the entire project fd SAREN bod Account UI SimCor Home My Page Project Tree Code Snippets Project Openings Simulation Coordinator Forums Tracker lists Tasks Docs Surveys News SCM Files
71. Simulation Monitor PushOver 127 0 0 1 11998 Figure 60 Monitoring window for pushover test 122 o UI SIMC R UCBMon UCB Simulation Monitor x2 fane Nose 01 Comp 1 va Parone Node 0i Corp 1 T E 08 20 Q4 42 0 02 04 08 OD 1 Figure 61 Monitoring window for UCB site Results Following output files will be saved in the following folders CASMCORW03 Examples UCBWPushoverTestOO Coordinator MDL OO recv txt Measured displacement and force of specimen NetwkLog txt Communication log file C SIMCOR 04_API 01_UCB NetwkLog txt Communication log file UCB Results txt Target displacement measured displacement and force at UCB site The pushover test result is shown in Figure 62 and the analytical model was developed by using ZEUS NL based on these results Pushover Test of UCB specimen Experiment K_ini Keq K half dis Simulation Force Displacement Figure 62 Comparison between experiment and simulation pushover test 123 o UI SIMC OR 9 1 4 SDOF example A dynamic analysis of a cantilever column with a lumped mass explained in section 8 1 is used here to verify the UI SIMCOR The column is replaced with the specimen at UCB The simulation configuration file for this experiment is identical to section 8 1 except the scale factor Calculating scale factors for experimental site will be explained
72. TH MONITOR 2 x D Disp Force monitoring point DF MONITOR 2 x Variable Description All parameters are same for ZEUS NL Model ABAQUS Module Detailed description about modeling for ABAQUS is not given in this document However you can find very well documented information from following link http www abaqus com The current static analysis module for ABAQUS is tested under ABAQUS v6 4 3 and v6 5 1 If different version of ABQUAS uses different output format PostResult m in 02 NEES ABAQUS should be modified following new output format However this will be really minor modification Furthermore there is one restraint in ABAQUS input file It is option for imposing target displacement in the model The restraints is that the order of the DOFs in the command boundary should be the same as that of MDL i EFF_DOF in the ABAQUS configuration file This will be further explained in Section 7 5 The ABAQUS Module is developed based on the FEDEAS Lab Module so a description of configuration file for ABAQUS Module is very similar to that of FEDEAS Lab CA SIMCOR O03_Examples SDOF 04_Abaqus AbaqusCFG m Configuration parameters for AbaqusMDL SDOF cantilever model oe oe oe oe Unit mm N sec Modified by GunJin Yun gunjin73 hotmail com Generated by Oh Sung Kwon okwon2 uiuc edu Modified by Kyu Sik Park kspark uiuc edu Univ of Illinois at Urbana Champaign Last updated on 2006 11 29 10 09PM o oe oe
73. UI SIMCOR control window UI SIMCOR will read the predefined stiffness matrix from file MDLO1_K txt located in CASIMCORYO3 Examples UCB SDOF 00_ Coordinator Click Apply Static Loading button in UI SIMCOR control window In this stage the gravity force is applied however there is no plan to apply gravity as this exemplary test is to demonstrate the efficacy of hybrid simulation framework Click Start PSD Test to run test by experiment The target displacement and feedback displacement and force will be displayed in the monitoring window and UCB monitoring window Click Disconnect Modules after finishing hybrid simulation to disconnect UI SIMCOR from xPC Real time target You will see some messages indicating the communication is disconnected in both MALTAB command windows Backup the results files MDLOI recv txt Netwklog txt NodeDisp txt located in C SIMCOR 03_Examples UCB SDOF 00_Coordinator for UIUC site and another results files NetwkLog txt UCB Results txt located in CASIMCORY04 APIO1 UCB for UCB site Send results to UIUC 177 o UI SIMC OR B Three site Hybrid Test among UIUC UCB and SDSC The detailed experiment procedure for real three site hybrid test is introduced in this section All the three sites UIUC UCB and SDSC should understand the whole procedure for the successful completion of the experiment The experiment procedure consists of two steps Step 1 is to verify the communication among UIUC UCB
74. UI for each module GUI for each module can only display the data GUI for each module can not control the hybrid simulation Yes 1 enable the GUI for each module No 0 disable the GUI for each module MDL 1 EnableGUI 1 oe oe oe oe oe oe oe Advanced modular parameters oe oe These parameters need to be redefined for following situations 1 Different coordinate system between UI SIMCOR and static module 2 When scale factor needs to be applied either in experiment or simulation 3 To define force and displacement criteria for tolerance and safety 4 To trigger camera modules or DAQ system E 5 When LBCB at UIUC is used for experiment 6 When NHCP protocol is used oe oe URL of remote site and NHCP mode for NHCP for i 1 Sys Num RF Module if strcmp lower MDL i protocol nhcp MDL i remote URL 127 0 0 1 99999 MDL i NHCPMode simld end end Stiffness for NHCP Only valid if NHCPMode SimlD for i 1 Sys Num RF Module if strcmp lower MDL i NHCPMode simld MDL i NHCPSimK 1000 end end oe Coordinate transformation If it needs the transformation matrix also needs to be provided for i 1 Sys Num RF Module MDL i TransM oe end oe Scale factor for displacement rotation force moment Experimental specimens are not always in full scale Use this factors to apply scale factors The displacem
75. UIUC 0 711 1 000 0 484 1 000 SDSC 0 711 1 000 0 243 1 000 Figures 93 and 94 show the test results and it shows good agreement between experiment and simulation results NEESit changed the signal conditioner of Mini MOSTI in SDSC to improve the load cell It took around 1 25 sec step average during 500 steps for MiniMOST1 in SDSC with new signal conditioner whereas it took around 2 89 sec step for MiniMOSTI in UIUC The signal conditioner of MiniMOST 1 in UIUC will be changed to improve the measuring time of load cell 158 o UI SIMC GOR Displacement 0 015 Ee 9911 SIM 0 005 MMHUIUC LabVIEWT 0 005 0 01 0 015 Tine sec MMHUIUC LabVIEWT1 Displacement rr Figure 93 MiniMOST 1 at UIUC 159 o UI SIMC OR MM1SDSC NHCP 0 015 Displacement i o 0 015 Tine sec MMHSDSC NHCP Exe SIM Force I 0 015 Displacement rr Figure 94 MiniMOST 1 in SDSC 160 o UI SIMC OR 10 References Combescure D and Pegon P 1997 a Operator Splitting Time Integration Technique for Pseudo dynamic Testing Error Propagation Analysis Soil Dynamics and Earthquake Engineering 16 427 443 Elnashai Amr S Papanikolaou V Lee D H 2006 User Manual for ZEUS NL Department of Civil and Environmental Engineer
76. W1 protocol has Propose Query Execute Query structure whereas LabVIEW2 has Propose Query structure 6 3 2 Basic sequence of commands The normal sequence of commands from LabVIEW protocol to the control system looks like this open session set parameter get parameter loop N times 1 propose 2 execute Pe 13 5 o UI SIMC R 3 get control point close session where steps 2 and 3 are semi optional and depend on the program running LabVIEW protocol This is sometimes called an client or simulation coordinator depending on the context 6 3 3 Detailed syntax for each verb open session Syntax open session TransactionID parameter parameter This can optionally initialize hardware or do other system specific work as required This should be the very first command of a connection The parameters are control system specific meaning you can use this for whatever you need they are optional The LabVIEW implementation ignores the transaction ID and any parameters sent close session This mirrors open session Currently performs no work but can be used as required Should be the last command received in an LabVIEW protocol session propose Syntax propose TransactionID ControlPoint GeomType ParameterType Parameter ControlPoint GeomType ParameterType Parameter where there can be zero to twelve parameters The spec also allows zero parameters for implied commands based on the contro
77. WINDOW DOS command window and change current directory to the folder where the model file SDOF tcl exists Run OpenFresco by inputting OpenFresco enter Load model file by inputting source SDOF tcl enter See Figure 24 d If NHCP Simulation Server is used i ii iii Start WINDOW DOS command window and change current directory to the folder where the ncs exe exists Run NCS by inputting ncs portnumber enter where portnumber is defined in the SimConfig m file MDL i URL see Figure 25 Start another WINDOW DOS command window and change current directory to the folder where the simserver exe exists Run SimServer by inputting simserver portnumber enter where portnumber is defined in the SimConfig m file MDL i remote_ URL see Figure 26 3 Run one MATLAB for UI SIMCOR Two MATLAB are required when FEDEAS Lab or ABAQUS is used as static module i e one is for communication and the other is for UI SIMCOR Change current directory to the folder where SimConfig m and ground motion file is located Run UI SIMCOR by inputting following command UI SIMCOR enter See Figure 27 4 There are two options in the UI SIMCOR control GUI i e Step by step control and All steps by one click a Step by step run the Establish Connection Stiffness Evaluation Apply Static Loading Start PSD Test and Disconnect Modules in order by clicking the each button b All steps
78. ZEUS NL In this example the stiffness of each module is obtained by stiffness evaluation step during PSD test and shown in Table 2 Table 2 Stiffness of each module Module Stiffness kN m Module 1 1 584x10 Module 3 1 584x10 Module 4 1 584x10 Because modules 1 3 and 4 will be replaced with specimen at each site the stiffness of those is needed to calculate scale factor of each site 9 4 3 Specimen at each site MiniMOST 1 UIUC The module 1 in the three site example will be replaced with MiniMOST 1 at UIUC as shown in Figure 76 Figure 76 MiniMOST 1 at UIUC Hybrid test between UIUC and SDSC with MiniMOST 1 at each site was conducted by Juan and Lawrence in September 12 2006 to test recently enhanced UI SIMCOR Based on these test results the stiffness of MiniMOST 1 at UIUC can be calculated as shown in Figure 77 141 o UI SIMC R Remote Site 40 y 2443 7x 30 R 0 9997 zZ 8 2 0 015 0 015 30 e T T 4 Displacement m Figure 77 Stiffness of Mini MOST 1 at UIUC uNEES UCB The specimen at UCB explained in section 9 1 is used for three site experiment MiniMOST 1 SDSC Figure 78 shows MiniMOST 1 at SDSC and Figure 79 shows the test results conducted between UIUC and SDCS in September 12 2006 The configuration of MiniMOST 1 at SDSC is very similar to MiniMOST 1 at UIUC
79. ables Variable size should be number of control nodes 6 array Displacement increment limit not ratio DL i TGT D inc abcdeef DL i CAP D tot abcdeef H Displacement limit Force limit DL i CAP F tot abcdeef ws i Displacement tolerance ratio DL i TOL D inc abe d e f oe oe oe Auxiliary module configuration e e Ds cc x xX AP oe oe oe oe MDL AUX URL 127 0 0 1 12000 1 protocol labviewl name Camera Module ID of this mdoule is 1 Command displacement z 3500 Variable Description Most variables in SimConfig m are explained in Section 8 4 for MOST example In this section variables that are related to LBCB are further explained MDL i LBCB This variable defines the option of Loading and Boundary Condition Boxes LBCB UI SIMCOR only calculates the target displacement of effective DOF However experimental equipment can have difference number of actuators with effective DOF The in UIUC has 6 actuators so 6 displacements should be sent to LBCB for experiment If YES 1 the target displacements of all DOFs will be assigned as zero except effective DOF in UI SIMCOR If MDL LBCB 1 MDL i LBCB TransM also needs to be defined MDL i LBCB TransM This variable defines the transformation matrix for LBCB The coordinate system of LBCB is dependent on the location of plate It is assumed that the x and y directions in UI SIMCOR is trans
80. ables pointed at are MATLAB arrays prhs is a length nrhs array of pointers to the right hand side inputs to the MEX file and pins is a length nins array that contains pointers to the left hand side outputs that your function generates For example if user invokes a MEX file from the MATLAB workspace with the command x fun y z the MATLAB interpreter calls nexrunction with the following arguments nlhs 1 nrhs 2 plha e prhs Y Z Figure 11 Relationship between MATLAB and C C variables plhs is a l element C C array where the single element is a null pointer prns is a 2 element C C array where the first element is a pointer to an mxarray named Y and the second element is a pointer to an mxArray named Z The parameter pihs points at nothing because the output x is not created until the subroutine executes It is the responsibility of the gateway routine to create an output array and to set a pointer to that array in pins 0 If p1nst0 is left unassigned MATLAB prints a warning message stating that no output has been assigned More details about MEX file can be found at MATHWORKS website www mathworks com 24 o UI SIMC R 6 5 4 NHCP DLL file The NHCP is implemented in UI SIMCOR by modifying the existing demo files of NHCP There are four demo files in NHCP Le simdemo cpp midriver cpp m2driver cpp baronDemo cpp simdemo cpp is for the simulation demo midriver cpp and m2driver cpp are for MiniMOST 1 and
81. acy of recently enhanced UI SIMCOR The NEES at UCB was used in this test and the functionality of UI SIMCOR with uNEES specimen was verified and UCB site is ready for three site hybrid test 131 o UI SIMC OR 9 2 Two site Experiment between UIUC and UCB The two site hybrid test between UIUC and UCB using recently enhanced UI SIMCOR were conducted at October 4 2006 using SDOF example in the Section 8 1 9 2 1 Hybrid test by simulation at UCB site only The purpose of this step is to verify the configuration of newly installed UI SIMCOR at UCB The results are shown in Figure 68 and the Expected and Actual mean the simulation results at UIUC and UCB respectively As shown in figure the test results simulated at UIUC and UCB is identical so the configuration of newly installed UI SIMCOR at UCB was verified HSF Expected op EE Actual Displacement eo 20 L 30 40 l 0 0 5 1 1 5 2 2 5 3 3 5 4 4 5 5 Time HSF 150000 Expected S Actual 100000 50000 Fr Force 0 50000 100000 150000 200000 Displacement a Hybrid simulation framework Figure 68 Simulation results 132 o UI SIMC OR Remote site 4 3 Expected ML Actual 1 E o a 2 3 4 0 0 5 1 1 5 2 2 5 3 3 5 4 4 5 5 Time Remote site Expected Force Displacement b Remote site Figure 68
82. ample in Section 8 1 except that there is static loading step and effective DOES are defined for x and y direction CNSIMCORNOS ExamplesNSDOF InitLoadingN00 CoordinatorNSimConfig m function Sys MDL AUX SimConfig MDL MDL_RF AUX MDL_AUX Type definition Do not delete this line oo oo Configuration parameters for SDOF experiment oe oe Unit mm N sec oo oo by Oh Sung Kwon okwon2Quiuc edu modified by Kyu Sik Park kspark uiuc edu Univ of Illinois at Urbana Champaign o o9 oe oo Last updated on 2007 01 26 11 55PM oe oo oo oe Common parameters oe oo Ground acceleration file name with extension The file should contains two columns for time and acceleration The unit of acceleration should be consistent with the mass time and force i e mass acc force Sys GM Input elcentro dat oe oe 5 Ground acceleration scale factor This factor will be multiplied to acceleration before starting simulation Sys GM SC 9810 Direction of ground acceleration x y or zZ Sys GM direction x Integration parameter related to the alpha OS method Alpha 0 1 3 In most cases SC Alph 0 05 worked Sys Alph 0 05 Sys Beta 1 4 1 Sys Alph 2 Sys Gamm 1 2 Sys Alph oe Evaluate Stiffness Yes 1 to run stiffness evaluation test O 0 to read stiffness matrix from file In this case there should exist stiffness matr
83. and FEDEAS Lab with specified open ports In the configuration of UI SIMCOR and FEDEAS Lab the URL and open port number of NTCP server should be given Both ends UI SIMCOR and FEDEAS Lab connect to NTCP server using same port number Figure 5 a shows the connection diagram The port number and URL in the figure are just example ZEUS NL OpenSees and test equipments use LabVIEW plugin to connect to the NTCP server As opposed to the MATLAB plugin NTPC server act as a server for the connection with UI SIMCOR and act as a client for the connection to the backend Figure 5 b Thus there is a URL and port number of NTCP server in UI SIMCOR and there is a URL and port number of backend when the NTCP server runs 10 o UI SIMC OR Client Client JI SIMCOR NTCP Server NTCP Server Server Server Matlab Plugin LabView Plugin Client URL http cee nsp3 cee uiuc edu Open Port 11151 FedeasMDL FEDEAS Lab URL http cee nsp3 cee uiuc edu Open Port 11151 NEES SAM ZEUS NL OpenSees a Connection through MATLAB plugin b Connection through LabVIEW plugin Figure 5 Connection diagram of NTCP server using MATLAB and LabVIEW plugin URL 130 126 243 165 Server Open Port 11150 Client 6 2 TCP IP Protocol 6 2 1 Background and introduction TCP connection with binary data format is used for TCP IP protocol For the TCP connection with remote site Instrument Control Toolbox
84. and a property value supported by PropertyName respectively For UI SIMCOR propertyName and Propertyvalue for initialization process are set to InputBufferSize and 1024x100 respectively For example Comm obj tcpip 127 0 0 1 11997 Set Comm obj InputBufferSize 1024x100 open Syntax fopen Comm obj to connect interface object to remote site fwrite Comm obj iniData to send initialize data to remote site fread Comm obj size to receive acknowledgement data from remote site For example fopen Comm obj fwrite Comm obj 500 set total number of simulation steps fread Comm obj 4 read 4 bytes of data propose Syntax fwrite Comm obj T DISP to write binary data to remote site where t_ptsp is target displacement calculated from UI SIMCOR query Syntax fread Comm obj size to read size bytes binary data from remote site close Syntax 12 o UI SIMC OR fwrite Comm obj closeData to send closing data to remote site fclose Comm obj to disconnect interface object from remote site 6 3 LabVIEW1 and LabVIEW2 Protocols for NTCP Server Many users of NTCP experienced slow communication and a little cumbersome setup process So LabVIEWI and 2 protocols use direct connection between coordinator and remote sites without NEESpop The original version of LabVIEW protocol was written for the communication between NEESPop and LabVIEW controlled hardware The communication is basically conducted in ASCII forma
85. ass 3 2 54628081981000 0 0 0 0 0 oe oe oe Restoring force module configuration oe e 5 Create objects of MDL RF DL 1 MDL RF DL 2 MDL RF DL 3 MDL RF oe Name of each module DL 1 name LeftCol LBCB Module ID of this module is 1 DL 2 name Middle Module ID of this module is 2 DL 3 name RightCol Module ID of this module is 3 oe URL of each module DL 1 URL 127 0 0 1 11997 DL 2 URL 127 0 0 1 11998 DL 3 URL 127 0 0 1 11999 oe Communication protocol for each module NTCP communicate through NEESPOP server TCPIP binary communication using TCPIP LabViewl ASCII communication with LabView plugin format 96 o UI SIMC GOR oe Propose Query Execute Query LabView2 same as LabViewl but Propose Query OpenFrescolD OpenFresco only 1 DOF is implemented now 5 NHCP NHCP linear 1 DOF simulation mode Mini MOST 1 and 2 at oe UIUC or SDSC LabView2 LabView2 LabView2 MDL 1 protocol MDL 2 protocol MDL 3 protocol Module 1 Left column MDL 1 node 1 MDL 1 EFF DOF 1 00 0 O 1 oe Control point node number Effective DOF for CP 1 oe Module 2 Middle column and beams MDL 2 node 12 3 Control point node number MDL 2 EFF_DOF 1 0000 1 Effective DOF for CP 1 100000 Effective DOF
86. ay button then you will see the Waiting for connection from client through port 11999 ina MATLAB command window b If you want modify the displacement limit click Reset button then the API Main sim for simulation or API Main for experiment file is opened automatically Modify the Disp Limit variable at line 17 Now ready to run the pushover test This test will run automatically 3 Runanother MATLAB for the pushover test and change the directory to CASIMCORY0O3 Examples UCBWPushoverTestOO Coordinator 4 Type PushOver and enter in MALAB command window to run the pushover test a This is running automatically b Another simulation monitor of GUI as shown in Figure A 12 for UCB site will be displayed in the MATLAB for API Main sim for simulation or API Main for experiment 171 o UI SIMC OR c The default value of maximum displacement for Pushover Test is 2 If you want to modify the maximum value please open the PushOver and modify the disp scale variable at line 66 LL iBpixi UCB Simulation Monitor LUCE 1 Coordinate system C U Ste Ais values Status Merzoge Figure A 12 Monitoring window for UCB site 5 Backup the results files MDLOO recv txt Netwklog txt located in C SIMCOR 03_ Examples UCBPushoverTest OO Coordinator and another results files NetwkLog txt UCB Results txt located in C SIMCOR 04_API 01_UCB 6 Send results to UIUC Step 8 Run the exper
87. binary communication using TCPIP LabViewl ASCII communication with LabView plugin format Propose Query Execute Query LabView2 same as LabViewl but Propose Query OpenFrescolD OpenFresco only 1 DOF is implemented now E NHCP NHCP linear 1 DOF simulation mode Mini MOST 1 and 2 at UIUC or SDSC MDL 1 protocol LabView2 MDL 2 protocol LabView2 MDL 3 protocol LabView2 Module 1 Left column MDL 1 node 1 Control point node number MDL 1 EFF DOF 10 O 0 0 1 Effective DOF for CP 1 Module 2 Middle column and beams MDL 2 node 1 2 3 Control point node number MDL 2 EFF DOF 100001 Effective DOF for CP 1 100000 Effective DOF for CP 2 100000 Effective DOF for CP 3 Module 3 Right column MDL 3 node 3 Control point node number MDL 3 EFF_DOF 1 00 0 0 0 Effective DOF for CP 3 Dismplacement for preliminary test for each module Del_t Translation Del_r Rotation in radian MDL 1 DEL t 0 005 MDL 2 DEL t 0 005 MDL 3 DEL t 0 005 MDL 1 DEL r 0 002 MDL 2 DEL r 0 002 MDL 3 DEL r 0 002 oe Enable GUI for each module GUI for each module can only display the data GUI for each module can not control the hybrid simulation Yes 1 enable the GUI for each module No 0 disable the GUI for each module oe oe oe oe MDL 1 EnableGUI 1 MDL 2 EnableGUI 1 MDL 3 EnableGUI 1 oe o
88. both of them if possible 120 o UI SIMC OR pe Object 1 of MDL_RF class Simulation co Monitor 1 1 UCB T i Ethernet TM Z 5 XPC Host PC OI eee gt o o o 8 MDL1 Pe B Stiffness Evaluation 5 passes E E API UCB e ri 2 xPC 5 Real time target o c zs Static Equilibrium C S x i MTS Hybrid Pacific Instruments Controller DAQ De Dynamic Equilibrium Simulation Controller Figure 58 Configuration of hybrid simulation framework at UCB with UI SIMCOR Two processes are required for this test one for UI SIMCOR and the other one for API module Both processes can be run in a single computer or in two computers A single computer was used in this test The procedure for running this test is given below The default value of the displacement limit is 6 If the target displacement which will be sent to xPC Real time target is greater than the displacement limit the test will be stopped with warning message If you want modify the displacement limit open the API Main and modify Disp Limit variable at line 17 1 Run a MATLAB for communication with xPC Real time target and change the directory to C SIMCOR 04_API 01_UCB 2 Type API Main then you will see the message box confirming the displacement limit as shown in Figure 59 Check Displacement Limit laj x
89. cally Modify the Disp Limit variable at line 17 c If displacement limit is confirmed click Okay button then you will see the Waiting for connection from client through port 11999 in a MATLAB command window The IP address and port number are dependent on experimental configuration d It is highly recommended to run the simulation by using API Main sim before the real experiment Run another MATLAB for UI SIMCOR and change the directory to C SIMCOR 03_ Examples UCB MOST 00_ Coordinator Type UI SimCor and enter in MATLAB command window to run UI SIMCOR You will see the Figure 63 again a Click Okay button you will see the initialize message in a MATLAB command window All previous files will be deleted so if necessary backup the previous test results b One control window and three monitoring windows which correspond to each module will be displayed 128 o UI SIMC OR Now ready to run experiment 7 10 11 Results Click Establish Connection button in UI SIMCOR control window You will see some messages indicating that the connection is established in both MATLAB command windows and in two APIs for simulation part Furthermore another simulation monitor of GUI for UCB site will be displayed in the MATLAB for API_Main as shown in Figure 61 a If the UI SIMCOR run in the other computer or the other site i e UIUC then this GUI helps UCB site seeing current status of experiment Click Stiffness Eval
90. cility for UI SIMCOR Following aspects were considered during the experiment at UCB to communicate with UI SIMCOR This is the most important and possible difficult step depending on the interface that people use for experiment Any program language which can impose displacement and collect measured displacement and force from experiment can be used If they use common protocol such as LabVIEW plugin via NTCP there is no need of customization Otherwise the experiment site should provide a method of communication In UCB Tony and Andreas provides us a MATLAB function which runs experiment and returns measured data It is recommended for user to keep in mind the communication procedure of the experiment control program with UI SIMCOR and followings e Figure out how to send the displacement command and get the feedback data from experiment UCB sites uses Shared Common RAM Network SCRAMNET and xPC Real time target to impose data and read feedback A MATLAB function is provided by Tony and Andreas for this purpose The functions are used in the scripts within the folder C ASIMCOR 04_API 01_UCB e Customize an interface program which can receive and send data through TCP IP network For UCB site MATLAB scripts were developed which can receive data from UI SIMCOR through network call the provided MATLAB function and send the feedback to UI SIMCOR through network e Set the IP address and port number Furthermore followings should be confir
91. col Saz 7j Absolute Position me me ill Description ERR Stiffness K Feedback channel name Feedback offset value Feedback value JSingle axis stepper motor d 1 000000 4 Be ome i o 00000E amp 0 p 00 Units p pons Move history Plot 0 E iP Ji 26876 64 Positions received 0 Range min 0 a 0 05010 r Range max Transaction ID Jo oso10 initial status pm M Ready i Position P Geometry 0 00000E 0 3 E J l i Cmd timestamp 5 254 Namejtype KE ER D m E L rare E displacement i jJ lisplacement 307 Initial position i Jio 35 Start time E x SOR 407 STOP E eres ooo 0 10 20 30 40 SO 60 70 80 90 100 110 120 130 140 155 Time eT pE Figure B 10 Control program in mini most 187 o UI SIMC OR a The control program commands MiniMOST 1 to execute the target displacements sent from UI SIMCOR by controlling the motor None of the variables need to be changed in this program To start receiving target displacements click on the white arrow located in the toolbar For ZEUS NL model UIUC 5 Log on to the Simulation Coordinator computer and run a NEES SAM API for ZEUS NL model by double clicking NEESSAM LabView2 exe which is located in C SIMCOR 03_Examples ThreeSite 02_Module2 a You will see the similar figure as shown in Figure B 2 details in figure are depends on the model b Check the status of NEES SAM module name and port number UCB Site For uNEES UCB 6
92. command This variable define the command for each auxiliary module 8 1 3 Static analysis module configuration In section 8 1 2 a configuration file for UI SIMCOR was explained In this section configuration file for static analysis module such as FEDEAS Lab ZEUS NL OpenSees and ABAQUS will be explained FEDEAS Lab Module Detailed description about modeling for FEDEAS Lab is not given in this document However the following should be kept in mind when a FEDEAS Lab model will be used as a part of simulation clear or CleanStart command should not be used Those commands wipe out all the configuration information during simulation 45 o UI SIMC OR Do not include Model Create Model command The command will be called in NEES FL LabView2 m The model file should have information about nodes elements material and section properties boundary conditions only All other parameters related to analysis will be given from NEES FL LabView2 m automatically The easiest way to start working on FEDEAS model for distributed simulation is starting from example file such as SDOF m rather than starting from scratch You can find very well documented manual from following link http www ce berkeley edu filippou FEDEASLab FEDEASLab htm The following is a description of configuration file for FEDEAS Lab Module C SIMCOR 03_Examples SDOF O1_FEDEAS FedeasCFG m oo oo Configuration parameters for FedeasMDL
93. communication with xPC Real time target and change the directory to C SIMCOR 04_API 01_UCB Type API Main then you will see the message box confirming the displacement limit as shown in Figure A 21 Check Displacement Limit Current displacement limit for UCB speicmen 6 00e 000 IF you want modify the displacement limit please open API_Main_sim m _Feset_ Figure A 21 Warning message for the displacement limit a If the displacement limit is confirmed click Okay button then you will see the Waiting for connection from client through port 11999 ina MATLAB command window If you want modify the displacement limit click Reset button then the API_Main file is opened automatically Modify the Disp_Limit variable at line 17 176 o UI SIMC OR Now ready to run the test by simulation UIUC Site 5 10 11 Click Establish Connection button in UI SIMCOR control window You will see some messages indicating that the connection is established in both MALTAB command windows and another simulation monitor of GUI for UCB site will be displayed in the MATLAB for API_Main as shown in Figure A 22 UCBMan alol xj UCB Simulation Monitor LUCE 1 Status Merzoge Figure A 22 Monitoring window for UCB site a If the UI SIMCOR run in the other computer or the other site 1 e UIUC then this GUI helps to UCB site seeing current status of experiment Click Stiffness Evaluation button in
94. cture behaves in plane and if other DOFs are not of interest unnecessary DOFs can be condensed out By defining effective DOFs we can include DOFs of interest only and condense out unnecessary DOFs In this example inertial forces are applied in x direction and x direction is of our interest Thus only x directional DOFs are enabled and all other DOFs are disabled The order of the DOFs are x y z rx ry rz 42 o UI SIMC OR MDL i DEL t and MDL DEL Y These variables define the displacement level for the test for initial stiffness formulation These variables should be defined for each module The displacement for preliminary test may vary for each site and for translational and rotational displacement If the module is computational and it behaves in the elastic range the value is not important If the module is experimental and it can go to inelastic range these variables should be chosen with care MDL i EnableGUI This variable defines the option of GUI for each module It cannot control the hybrid simulation but only monitor the responses in real time e g target displacement and measured displacement and force This GUI is very similar to TH MONITOR and DF MONITOR of a static analysis module for ZEUS NL and OpenSees Besides above variables there are advanced modular parameters in MDL RF class Theses parameters need to be redefined for following situations Different coordinate system between UI SIMCOR and static modu
95. d simulation to disconnect UI SIMCOR from xPC Real time target You will see some messages indicating the communication is disconnected in both MALTAB command windows Backup the results files MDLOI recv txt Netwklog txt NodeDisp txt located in C SIMCOR 03_Examples UCB SDOF 00_ Coordinator and another results files NetwkLog txt UCB Results txt located in C SIMCOR 04_API 01_UCB Send results to UIUC 167 o UI SIMC OR Step 6 Run the preliminary test by simulation in UCB API Main sim and in UIUC UI SimCor Purpose Verify the communication between UIUC and UCB Most of procedure is same with Step 5 except the UI SIMCOR will run in UIUC site To do this test UIUC should know the IP address and port number of UCB site The default value of the displacement limit is 6 If the target displacement which will be sent to xPC Real time target is greater than the displacement limit then the simulation will be stopped with error message If you want to modify the displacement limit open the API Main sim and modify Disp Limit variable at line 17 UIUC Site 1 Runa MATLAB for UI SIMCOR and change the directory to C SIMCOR 03_ Examples UCBYSDOF 00 Coordinator 2 Type Ul SimCor and enter in MALAB command window to run UI SIMCOR New simulation R xj e Please backup previous simulation results if necessary All previous files will be deleted cr Figure A 6 Warning message a Click Okay button you will see the
96. ding should be 1 The control points should follow the order in the configuration file 2 The DOFs should be defined in the order of x y z rx ry and rz e n this example only one control point and one DOF is defined Thus order doesn t matter More examples will be given in the later section The following gives explanation about the configuration file for ZEUS NL Module CANSIMCORNO3 ExamplesNSSDOFNO2 ZEUSMModule cfg Configuratino parameters for Zeus NL Left column of MOST experiment Unit mm N sec Generated by Oh Sung Kwon okwon2 uiuc edu Univ of Illinois at Urbana Champaign Last updated on 2006 08 06 9 35PM Connection port to controller Port 11997 Module application 1 for Zeus NL 2 for OpenSees MDL_Type 1 Effective node numbers in Simulaiton coordinator The order of node number should be idntical to that specified in the simulaiton coordinator configuration file SC_Node 1 Corresponding effective node numbers in this model DL_Node n2 Effective DOFs in control point EFF_DOF Use one line per each control point 00000 Zeus model file name without extension MODEL SDOF Model dimension MDL_Dim 2 Time history monitoring point Node number direction x y Z rx ry rz D for disp or F for force This should be defined after SC Node and MDL Node H MONITOR 2 x D Disp Force monitoring point DF MONITOR 2 x
97. dule name IP address and port number in the monitoring windows Ut SimCor ke sigi xj Ul SimCor version2s MUST SIM Facility Control Ground Motion Figure B 4 Control window 182 o UI SIMC R lojxi Simulation Monitor Coordinate system MM1 UIUC 1270 0 1 11991 HSF Qa v1 e QQ Y2 C Remote site 05r Axis values mant gl al 1 EDITUXCUEN i module name IP address and port number x2 Step OS 1 oomen H E L L 4 L P n n A 08 06 04 02 0 02 04 06 0 5 1 Coordinate system C HSF C Remote ste 9s Axis values X1 T Disp Node 02 Comp 1 D Y la Fare Node 02 Comp 1 X2 M Disp Node 02 Comp 1 05 Y2 MForc Node 02 Comp 1 Qa vt QQ Y2 1 1 1 1 1 1 1 4 08 06 04 02 0 02 04 05 0 8 4 Status Message Figure B 5 Monitoring window Now ready to run the three site hybrid test Simulation Conduct the hybrid test UIUC Site 7 Click Establish Connection button in UI SIMCOR control window You will see some messages indicating that the connection is established in a MALTAB command window and NEES SAM API windows of each site 183 o UI SIMC OR E NEES SAM LabView File View Help ZeusNL Reader v1 3 build 2502 This version is recompiled for UI SIMCOR to redirect the reactions and displacements of control points to SIMCOR Disp txt and
98. e oe Advanced modular parameters oe oe These parameters need to be redefined for following situations 1 Different coordinate system between UI SIMCOR and static module 2 When scale factor needs to be applied either in experiment or simulation 3 To define force and displacement criteria for tolerance and safety 4 To trigger camera modules or DAQ system E 5 When LBCB at UIUC is used for experiment 6 When NHCP protocol is used oe 76 o UI SIMC OR URL of remote site and NHCP mode for NHCP for i 1 Sys Num RF Module if strcmp lower MDL i protocol nhcp MDL i remote URL 127 0 0 1 99999 MDL i NHCPMode simld end end Stiffness for NHCP Only valid if NHCPMode SimlD for i 1 Sys Num_RF_Module if strcmp lower MDL i NHCPMode simld MDL i NHCPSimK 1000 end end oe Coordinate transformation If it needs the transformation matrix also needs to be provided for i 1 Sys Num RF Module MDL i TransM oe end oe Scale factor for displacement rotation force moment Experimental specimens are not always in full scale Use this factors to apply scale factors The displacement scale factors are multiplied before they ar sent to module Measured force and moments are divided with scale factors before used in the PSD algorithm for i 1 Sys Num RF Module MDL i ScaleF 1 1 1 1 Module i oe oe oe oe
99. e Control daemon vi from this menu and click OK You will see the Figure B 9 a The control daemon program receives NTCP commands In order to start its ability to receive NTCP commands click on the white arrow located on the toolbar of this LabVIEW program 13 Now double click on the mini most icon located in the lv programs folder that was opened on the Desktop Choose Control program vi from the menu of LabVIEW programs and click OK You will see the Figure B 10 a The control program commands MiniMOST 1 to execute the target displacements sent from UI SIMCOR by controlling the motor None of the variables need to be changed in this program To start receiving target displacements click on the white arrow located in the toolbar Prepare the UI SIMCOR for hybrid simulation UIUC Site 14 Run a MATLAB in the Simulation Coordinator computer for UI SIMCOR and change the directory to CASIMCORV0O3 Examples WThreeSiteX0O0 Coordinator 15 Type UI SimCor and enter in MATLAB command window to run UI SIMCOR a Youwill see the warning message shown in Figure B 3 b Click Okay button then you will see the initialize message in a MATLAB command window All previous files will be deleted so if you want please backup previous results c Five popup windows will be displayed One window for control and other windows for monitoring of each module will popup d Check the module name IP address and port number in the monitoring windows 189
100. each module MDL 1 name STATIC Module ID of this module is 1 oe URL of each module Format IP address port number ex http c nsp4 cee uiuc edu 11997 for local machine 127 0 0 1 11997 MDL 1 URL 127 0 0 1 11997 oe Communication protocol for each module NTCP communicate through NEESPOP server TCPIP binary communication using TCPIP LabViewl ASCII communication with LabView plugin format Propose Query Execute Query LabView2 same as LabViewl but Propose Query OpenFrescolD OpenFresco only 1 DOF is implemented now 5 NHCP NHCP linear 1 DOF simulation mode Mini MOST 1 and 2 at UIUC or SDSC MDL 1 protocol nhcp Module 1 STATIC MDL 1 node 1 Control point node number 37 o UI SIMC OR MDL 1 EFF DOF 1 0 0 00 0 Effective DOF for CP 1 5 Dismplacement for preliminary test for each module Del t Translation Del r Rotation in radian MDL 1 DEL t 0 005 I MDL 1 DEL r 0 002 oe Enable GUI for each module GUI for each module can only display the data GUI for each module can not control the hybrid simulation Yes 1 enable the GUI for each module No 0 disable the GUI for each module MDL 1 EnableGUI 1 oe oe oe oe oe oe oe Advanced modular parameters oe oe These parameters need to be redefined for following situations 1 Differ
101. ed If ZEUS NL or OpenSees model is used then NEES SAM needs to be run The four processes can be run either in a single computer or in a multiple computers As long as network configuration is set correctly the four processes can run in four computers The procedure for running this example is as given below l Prepare configuration file The network setting e g port number in each static analysis module configuration file should be identical with the simulation coordinator file should be carefully set 2 Prepare static module a If FEDEA Lab or ABAQUS is used i Starta MATLAB for each module ii For FEDEAS Lab Change current directory to the folder where FedeasCFG m and the model file of each module exist For ABAQUS Change current directory to the folder where AbaqusCFG m and the model file of each module exist iii For FEDEAS Lab Run FEDEAS Lab by inputting following command NEES FL LabView2 enter for LabView2 protocol For ABAQUS Run ABAQUS by inputting following command NEES Abaqus LabView2 enter for LabView2 protocol b If ZEUS NL or OpenSees are used with NEES SAM i Start NEES SAM by double clicking NEESSAM TCPIP exe tcpip protocol or NEESSAM LabViewl exe LabVIEW1 protocol or NEESSAM LabView2 exe LabVIEW2 protocol for each module 80 o UI SIMC OR ii Select the configuration file from the folder where configuration file and model file exist iii Check the open port and status 3 Run one MATLAB
102. ed by Sheffer 2 5HHTF15 The brief description of 2 5 HHTF15 actuator is shown in Table 1 Table 1 2 5SHHTF15 actuator Load capacity 50 kN 12 kips Stroke 375 mm 15 in Velocity 625 mm sec 25 in sec The controllers provided for the NEES facility perform the real time or near real time digital control of the hydraulic system for the pseudo dynamic testing The controllers support a local testing with real specimens only a hybrid simulation with virtual specimens computer models and real specimens coupled in a complex system through geographically distributed network There are two separate digital control platforms provided by MTS The first is the MTS 493 Real Time Controller supporting software called Structural Test System STS that performs the real time or near real time pseudo dynamic structural testing capabilities The second digital control platform is the MTS FlexTest GT Controller that provides a highly flexible and user configurable control environment to support more conventional structural testing applications In addition ATS Control and DAQ cart with dSpace DS1104 real time environment are used for data acquisition system During this preliminary hybrid test only one actuator and one specimen were used More detailed information and documents about uNEES laboratory can be found in the following nees berkeley link http nees berkeley edu 118 o UI SIMC OR 9 1 2 Customization of NEES fa
103. eef o oe oe Displacement tolerance ratio MDL i TOL D inc abcdeef oe end oe Loading and Boundary Condition Box LBCB case If it s 1 the coordinate transformation matrix needs to be provided This can be also used for any other actuator which has diffrence number of DOF coordinate with those of UI SIMCOR oe oe oe for i 1 Sys Num RF Module MDL i LBCB 0 end for i 1 Sys Num RF Module MDL i LBCB TransM end oe oe Auxiliary module configuration AUX 1 MDL AUX AUX 1 URL 127 0 0 1 12000 AUX 1 protocol labviewl AUX 1 name Camera Module ID of this mdoule is 1 AUX 1 Command displacement z 3500 66 o UI SIMC R Variable Description Sys Num Static Step This variable defines the number of static step for initial loading such as gravity as explained in 8 1 2 In this example Sys Num_Static_Step is 10 to increases the initial loading gradually as explained in 8 2 1 8 2 3 Static analysis module configuration Static analysis modules for this example are identical to previous example except following points The effective DOFs are defined for x and y direction Time history static loading is defined in the model file for ZEUS NL OpenSees and ABAQUS The time history load factor is as shown in Figure 31 FEDEAS Lab OpenFresco NHCPSim1D models are not given for this example 8 2 4 Running simulat
104. ees version 1 6 2 and FEDEAS Lab version 2 6 Include example input files for OpenSees and ZEUS NL for MOST example Development of an adapter module NEES MW by Sung Jig Kim for experimental equipments LBCB in UIUC Version 2 0 release Reconstruct the architecture of UI SIMCOR using object oriented approach Modulize the main body of analysis algorithm a OS scheme Provide improved GUI to monitor the remote site Development of NEES ABAQUS for ABAQUS static analysis module by Gun Jin Yun Include example input files for ABAQUS for buckling example Remove NEES MW Update UI SIMCOR for experimental equipment LBCB in UIUC by Kyu Sik Park Include LBCB example Include API for uNEES facility in UCB Include one site experiment using NEES facility in UCB Include two site experiment between UIUC and UCB using NEES and MiniMOST 1 facilities Include two site experiment between UIUC and SDSC using MiniMOST 1 facilities Include three site experiment among UIUC UCB and SDSC using uNEES and MiniMOST 1 facilities Version 2 5 release Logo change iii February 2007 o UI SIMC OR Version 2 6 release Include OpenFrescolD and NHCP communication protocols Include SDOF example using OpenFrescolD and NHCP communication protocols Include 7 DOF example using TCP IP LabVIEW1 LabVIEW2 OpenFrescol1D and NHCP communication protocols Include two site experiment between UIUC and SDSC using MiniMOST 1 facilities through NHCP iv
105. el Material 8 2 1 Structural configuration In the previous example in Section 8 1 the cantilever column is subjected to ground acceleration only This example shows how to apply initial static forces in the structure The structural configuration is identical to the example in Section 8 1 except the column is subjected to initial gravity force and bilinear material is used as shown in Figure 30 The material is assumed as elasto perfectly plastic material The material s elastic modulus is 200000 N mm and yield strength is 50 N mm 200 000 N __ 200 000 N p n IE X X X m 10 N mm sec G m 10 N mm sec E E e S gt as Ax Agx Column dimension SIMCOR Static Analysis Module Figure 30 Initial load on cantilever column The initial loading is applied to the static analysis module As discussed in Section 3 the PSD test stage can be subdivide into initial stiffness formulation stage initial loading stage and dynamic loading stage Note that initial loading is defined as time history load factor as shown in Figure 31 The load factor should be zero during initial stiffness formulation Since the structure has two DOFs the load factor is zero up to the second step And then the factor increases up to 1 gradually This incremental application of static loading is to account for the inelastic behavior of structure In this example 10 steps are used to apply initial loading And du
106. em32Wemd exe ncs 11997 Microsoft Windows KP Version 5 1 2688 lt C gt Copyright 1985 2661 Microsoft Corp c gt WS IMCORWO3_Examples SDOFW 6_SimServerNHCPoncs 11997 Creating socket and binding to port 11997 Waiting for connection Figure 25 WINDOW DOS command window after running NCS for NHCP cA C WINDOWS Wsystem32Wemd exe simserver 99999 Microsoft Windows XP Version 5 1 2606 lt C gt Copyright 1985 2661 Microsoft Corp C gt WS IMCORWO3_ExamplesSDOFW 6_SimServerNHCP gt simserver 99999 Creating socket and binding to port 99999 Waiting for connection Figure 26 WINDOW DOS command window after running SimServer for NHCP 56 o UI SIMC OR DIETS T must sim recti GS Control simulation HT Sx mbines Zeus NL OpenSers FEDEASLab ADAQUS and experimetal sites vie TCPIP network zl Application for distributed pseudo dynamic SimMon B File x Simulation Monitor STATIC 127 0 0 1 11997 Coordinate system Qd Y1 G HSP X2 Y2 7 Remote site Os 4 Axis values xt Step z 5 4 Y4 T Disp Node 01 Comp 1 X2 Step E 05 Y2 M Disp Node 01 Comp 1 M M 4 08 06 04 02 0 02 04 06 08 1 Coordinate system Qa Y1 C HSF x2 Y2 T Remote site s h Axis values X1 T Disp Node 01 Comp 1 0 4 yt rere noae o1 comp E X2 M Disp Node 01 Comp 1 25 1 Y2 M Forc Node 01 C
107. ent coordinate system between UI SIMCOR and static module 2 When scale factor needs to be applied either in experiment or simulation To define force and displacement criteria for tolerance and safety To trigger camera modules or DAQ system When LBCB at UIUC is used for experiment When NHCP protocol is used oe oe oe oe oe oe o 3 4 5 6 oe oe oe URL of remote site and NHCP mode for NHCP for i 1 Sys Num RF Module if strcmp lower MDL i protocol nhcp MDL i remote URL 127 0 0 1 99999 MDL i NHCPMode simld end end Stiffness for NHCP Only valid if NHCPMode SimlD for i 1 Sys Num RF Module if strcmp lower MDL i NHCPMode simid MDL i NHCPSimK 6 2344023e 003 end end oe Coordinate transformation If it needs the transformation matrix also needs to be provided for i 1 Sys Num RF Module MDL i TransM oe end oe Scale factor for displacement rotation force moment Experimental specimens are not always in full scale Use this factors to apply scale factors The displacement scale factors are multiplied before they ar sent to module Measured force and moments are divided with scale factors before used in the PSD algorithm for i 1 Sys Num RF Module MDL i ScaleF 1 1 1 1 Module i oe oe oe oe oe end oe Relaxation check If this parameter is 1 UI SimCor send commend to retrie
108. ent facilities using UI SIMCOR 2 Simulation Framework In the developed framework the integration scheme a OS scheme Combescure and Pegon 1997 is modulized which enables users to easily implement new integration algorithm and all other modules perform static analysis or experiment For instance the MOST experiment example in NEESgrid 3 2 Nakata et al 2003 consists of simulation coordinator left column tested in UIUC right column tested in University of Colorado UCOL and remaining elements in National Center for Supercomputing Applications NCSA NCSA site had pseudo dynamic integration scheme in it In UI SIMCOR all sites UIUC UCOL and NCSA run static analysis or static experiments Integration scheme resides in simulation coordinator o UI SIMC OR UI SIMCOR can communicate through TCP IP LabVIEW1 LabVIEW2 OpenFrescolD and NHCP protocols with any experimental equipment or static analysis modules These protocols will be explained in detail in Chapter 6 Static analysis application which can take displacement inputs and make reaction force outputs can participate in the simulation as a static analysis module For this purpose an interface for FEDEAS Lab ZEUS NL OpenSees ABAQUS OpenFresco and NHCP Simulation Server are developed FEDEAS Lab and ABAQUS communicate with LabVIEW2 protocol ZEUS NL OpenSees and experimental modules communicate with TCP IP LabVIEW1 and LabVIEW protocols in which ASCII code is used
109. ent only If the lumped mass is significantly large in rotational direction then the rotational DOFs should be included in Figure 4 b Figure 4 c shows subdivided structure into three parts left column with two DOFs mid column and beam with four DOFs and right column with one DOF a MOST iri le m MOST E with reduced DOFs pet iu Fy c Subdivided structure Figure 4 Effective DOFs of the MOST example o UI SIMC OR 5 Static Analysis Module Interface In this distribution FEDEAS Lab ZEUS NL OpenSees ABAQUS OpenFresco and NHCP Simulation Server are included as static analysis modules The static analysis modules need to take displacements from simulation coordinator and give reaction forces to the simulation coordinator All these communications run through communication protocol Thus it was necessary to develop an interface module for these analysis softwares FEDEAS Lab FEDEAS Lab was not need to be modified to work as a static analysis module since input and out can be easily managed in MATLAB A few additional scripts need to be written to connect to UI SIMCOR and give displacement and take forces from FEDEAS Lab NEES FL LabView2 m plays this role using LabVIEW2 protocol In case when it is necessary to communicate through NEESpop using NTCP for Matlab NEES FL NTCP m can be used But NTCP for Matlab requires several prerequisite applications ZEUS NL Original version of ZEUS NL was modi
110. ent scale factors are multiplied before they ar sent to module Measured force and moments are divided with scale factors before used in the PSD algorithm for i 1 Sys Num RF Module MDL i ScaleF 1 1 1 1 Module i oe oe oe oe oe end 65 o UI SIMC OR oe Relaxation check If this parameter is 1 UI SimCor send commend to retrieve data and check relaxation just before the execution of proposed command If it s 1 the checking criteria needs to be provided for i 1 Sys Num RF Module MDL i CheckRelax 0 Module i if MDL i CheckLimit 1 define following variables Variable size should be number of control nodes 6 array oe oe oe oe oe o Displacement variation ratio not increment MDL i MES D inco abcdef cene 1 Force variaiton ratio not increment MDL i MES F inco abcdef PEE 15 end oe o o oe Check displacement and force limit At every steps check if the displacement or force are approaching to the limitation of the equipments stroke or force capacity for i 1 Sys Num RF Module MDL i CheckLimit 0 Module i if MDL i CheckLimit 1 define following variables Variable size should be number of control nodes 6 array oe oe o oe oe oe Displacement increment limit not ratio DL i TGT D inc abcdeef oe oe lt oe Displacement limit DL i CAP D tot abcdef S ae x oe Force limit DL i CAP F tot abcd
111. er of static steps in the following variable Sys Num Static Step 0 oe oe oo o Number of dynamic analysis steps Sys Num Dynamic Step 500 95 o UI SIMC OR o Dynamic analysis time steps Sys dt 0 01 Rayleigh damping xi_l and xi_2 Damping ratio Tn_1 Tn_2 Target period Sys xi 1 0 00 Sys Tn 1 0 00 Sys xi 2 0 00 Sys Tn 2 0 00 oe Number of Stiffness test If stiffness is evaluated through experiment th valuation need to be don several times and the average of the results are used as the initial stiffness This parameter is used when Sys Eval Stiffness 1 Sys Num Test Stiffness 1 oe oe oe oe Enable GUI for SimCor Yes 1 nable the GUI for SimCor o 0 disable the GUI for SimCor Hybrid simulation will be run automatically Not recommended for the experiment ys EnableGUI 1 Use GUI for SimCor oe oe oe oe LD Number of restoring force modules Sys Num RF Module 3 Number of auxilary modules Sys Num_AUX_Module 0 oe Total number of effective nodes Effective nodes are interface nodes between modules and nodes where lumped masses are defined ys Num Node 3 oe n oe Lumped mass assigned for each DOF for each node Node number x y Z rx ry rz directional mass Sys Node Mass 1 2 54628081981000 0 0 0 0 OJ Sys Node_Mass 2 5 49705587697000 0 0 0 0 OJ Sys Node_M
112. est are shown in following figures 178 o UI SIMC OR b ZEUS NL at UIUC Figure B 1 Responses of each site 179 UI SIMCOR Fare DOF 01 c KNEES at UCB f Remote i foeon E parr s 2 MDepDoFO F apr s d MMI at SDSC Figure B 2 Responses of each site continued 180 o UI SIMC OR Prepare the communication among UIUC UCB and SDSC UIUC Site 1 Log on to the MiniMOST 1 computer and run a NEES SAM API for MiniMOST 1 analytical model by double clicking NEESSAM LabViewl exe which is located in CASIMCORY03 Examples WThreeSite01 UIUC a You will see the similar figure as shown in Figure B 2 details in figure are dependent on the model b Check the status of NEES SAM module name and port number E NEES SAM LabView E ioj xl File View Help status of NEES SAM static analysis module port number module name Standby for connection IMMTLUIU 119 Ready IMMT UIU 11881 iZeus NL 7 Figure B 2 NEES SAM window for MM1 UIUC module standby for connection 2 Log on to the Simulation Coordinator computer and run a NEES SAM API for ZEUS NL model by double clicking NEESSAM LabView2 exe which is located in C SIMCOR 03_Examples ThreeSite 02_Module2 a You will see the similar figure as shown in Figure B 2 details in figure are dependent on the model b Check the status of NEES SAM module name and port number UCB Site 3 Log on to the NEES con
113. fied and recompiled The modified solver NEES Solver exe takes displacement input from console and makes reaction force output into a file in each step NEES SAM which stands for Static Analysis Module for NEESgrid gives displacement to the modified ZEUS NL and reads the force output file NEES SAM acts as an interface between UI SIMCOR and ZEUS NL There are three versions of NEES SAM for TCP IP LabVIEW1 and LabVIEW2 protocols The protocols will be explained in detail in Section 6 OpenSees OpenSees was also modified to generate an output file for each step Among the source code files NodeRecorder cpp was modified NEES SAM also works as an interface between UI SIMCOR and OpenSees ABAQUS It is not possible to modify the ABAQUS as it is commercial software So restart feature in ABAQUS is utilized to use ABAQUS as a static analysis module A MATLAB scripts are developed to connect to UI SIMCOR and give displacement to and take forces from ABAQUS NEES Abaqus LabView2 m plays this role using LabVIEW2 protocol This is very similar to FEDEAS Lab static analysis module interface NEES FL NTCP m OpenFresco Open Framework for Experimental Setup and Control OpenFresco is a software framework intended to facilitate and help standardize the local or geographically distributed deployment of hybrid simulation and developed by UCB Schellenberg et al 2006 OpenFresco can acts as an interface between experimental hardware and main simula
114. for NHCP Only valid if NHCPMode SimlD for i 1 Sys Num RF Module if strcmp lower MDL i NHCPMode simld MDL i NHCPSimK 1000 end end Coordinate transformation If it needs the transformation matrix also needs to be provided for i 1 Sys Num RF Module MDL i TransM end oe Scale factor for displacement rotation force moment Experimental specimens are not always in full scale Use this factors to apply scale factors The displacement scale factors are multiplied before they ar sent to module Measured force and moments are divided with scale factors before used in the PSD algorithm for i 1 Sys Num RF Module MDL i ScaleF 1 1 1 1 Module i oe oe oe oe oe end oe Relaxation check If this parameter is 1 UI SimCor send commend to retrieve data and check relaxation just before the execution of proposed command If it s 1 the checking criteria needs to be provided for i 1 Sys Num RF Module MDL i CheckRelax 0 Module i if MDL i CheckLimit 1 define following variables Variable size should be number of control nodes 6 array oe oe oe oe oe o Displacement variation ratio not increment MDL i MES D inco abcdef due 1 Force variaiton ratio not increment MDL i MES F inco abcdef deus J3 end oe oe o oe Check displacement and force limit At every steps check if the displacement or force are approaching to the
115. for each DOF for each node Node number x y Z rx ry rz directional mass Sys Node Mass 1 2 54628 0 0 0 0 OJ Sys Node_Mass 2 5 49706 0 0 0 0 OJ Sys Node_Mass 3 5 49706 0 0 0 0 OJ Sys Node_Mass 4 5 49706 0 0 0 0 OJ Sys Node_Mass 5 5 49706 0 0 0 0 OJ Sys Node_Mass 6 5 49706 0 0 0 0 OJ Sys Node_Mass 7 2 54628 0 0 0 0 OJ oe oe oe Restoring force module configuration oe Create objects of MDL RF MDL 1 MDL RF MDL 2 MDL RF MDL 3 MDL RF MDL 4 MDL RF MDL 5 MDL RF Name of each module 5 Simulation witout scale factor DL 1 name OS_TCPIP Module ID of this module is 1 DL 2 name FL LV2 Module ID of this module is 2 DL 3 name ZEUS_LV1 Module ID of this module is 3 DL 4 name OF_OF1D Module ID of this module is 4 DL 5 name SimlD_NHCP Module ID of this module is 5 URL of each module DL 1 URL 127 0 0 1 11991 DL 2 URL 127 0 0 1 11992 DL 3 URL 127 0 0 1 11993 DL 4 URL 127 0 0 1 11994 DL 5 URL 127 0 0 1 11995 Communication protocol for each module TCP communicate through NEESPOP server TCPIP binary communication using TCPIP LabViewl ASCII communication with LabView plugin format Propose Query Execute Query LabView2 same as LabViewl but Propose Query oe OpenFrescolD OpenFresco only 1 DOF is i
116. formed to y and x directions in LBCB respectively So the transformation matrix for LBCB is 100 o UI SIMC OR 0 10 0 00 1 0 0 0 00 0 01 0 0 0 LBCB TransM T7 0 00 0 10 0 0 0 1 0 0 0 00 0 0 1 8 6 3 Static analysis module configuration The configuration files for static analysis module are similar to those for MOST example except the left column Thus further explanation is not given in this section In the folder CASIMCOR 03_Examples MOST_LBCB there are sub folders containing the structural model file and module configuration files for ZEUS NL and those of other static analysis modules are not given in this example There is one thing that needs attention When ZEUS NL is used as a static analysis module in which there are 6 DOFs for single control points the static time history load should be defined in the order of 6 DOFs number For instance the left column of this example has one control point in which two DOFs are used However the LBCB has 6 DOFs so time history load should be defined in the order of control points and then 6 DOFs in ZEUS NL input file time history loads nod name direction type crv name value n102 x displacement CV TE n102 y displacement CV ilr n102 Z displacement CV 1 n102 rx displacement CV Ls n102 ry displacement CV Ts n102 rz displacement CV 1 8 6 4 Running simulation The running procedure of this example is same with that of MOST example so further explanation is not given in t
117. g m function Sys MDL AUX SimConfig MDL MDL_RF AUX MDL_AUX Type definition Do not delete this line oe oe Configuration parameters for 6Beam5Col Examples to test five protocols i e TCPIP LabViewl LabView2 OpenFrescolD and NHCP oe oe oe Unit m kN sec oe oe by Kyu Sik Park kspark uiuc edu 108 o UI SIMC OR oe Univ of Illinois at Urbana Champaign oe oe Last updated on 2007 01 27 10 56AM oe oe oe oe Common parameters oe 5 Ground acceleration file name with extension The file should contains two 5 columns for time and acceleration The unit of acceleration should be consistent with the mass time and force i e mass acc force Sys GM Input acc475C dat Ground acceleration scale factor This factor will be multiplied to acceleration before starting simulation Sys GM SC 9 81 Direction of ground acceleration x y or z Sys GM direction x Integration parameter related to the alpha OS method Alpha 0 1 3 In most cases SC Alph 0 05 worked Sys Alph 0 2 Sys Beta 1 4 1 Sys Alph 2 Sys Gamm 1 2 Sys Alph oe Evaluate Stiffness Yes 1 to run stiffness evaluation test No 0 to read stiffness matrix from file In this case there should exist stiffness matrices of individual module in the files MDLO1_K txt MDL02 K txt etc ys Eval Stiffness 1 o oe oe oe n
118. hat only x directional translation will be participate in the simulation In this configuration file MDL REP DOF 1 should also be defined with same information MDL FEDEAS MDL This variable defines the FEDEAS Lab model file name without extension In the model file nodes elements boundary conditions section properties material properties etc are defined MDL MDL Dim This variable defines the dimension of the model i e 2 or 3 3 dimensional model are not tested yet using FEDEAS Lab ZEUS NL Module ZEUS NL Model and OpenSees Model can participate in the PSD test by using NEES SAM as an interface module Original version of ZEUS NL cannot take displacements from the other software or user Thus the Solver of the ZEUS NL was modified to take displacement input for each time step from console and make a force output file in each time step Following is key points for modeling of ZEUS NL e Analysis type should be Static Time History Analysis e Time history curve should be defined The value of the time history curve can be any value for the modified solver Uniform curve is used in the example 47 o UI SIMC OR The time history curve should be defined for a large enough duration for whole simulation e n the Applied Loading tab static time history load should be defined for each control points and for each DOF Any load factor can be used In this example load factor of 1 is used The order of the static time history loa
119. he dynamic analysis time steps Sys xi 1 Sys Tn 1 Sys xi 2 and Sys Tn 2 These variables define the Rayleigh damping parameters 1 e Damping ratio Sys xi 1 and Sys xi 2 and target period Sys Tn 1 and Sys Tn 2 Sys Num Test Stiffness This variable defines the number of stiffness test If stiffness is evaluation through simulation module one test is enough to evaluate the stiffness However if stiffness is evaluation through experiment the evaluation need to be done several times and the average of the results are used as the initial stiffness This parameter is used when Sys Eval Stiffness 1 Sys EnableGUI This variable defines the option of GUI for UI SIMCOR control window If NO 0 the hybrid simulation will be run automatically so it is not recommended for the experiment Sys Num RF Module This variable defines the number of static analysis or experimental module In this example there is only one static analysis module which represents column elements Thus it s 1 Sys Num AUX Module This variable defines the number of auxiliary modules Sys Num Node This variable defines total number of control points in the simulation coordinator As it was discussed in Section 4 nodes where masses are located or where the DOFs are of interest are used in the time integration 41 o UI SIMC R Sys Node Mass i This variable defines nodal mass for each node Lumped mass can be defined for each DOF for each control points
120. he shifting of this mass has little effect on the response of the structure since the inertial forces are acting on x direction only and the structure deformation is caused mainly by flexure not by the axial compression or elongation of beams The connection points between two modules are also defined as control points Thus in SimConfig m file there are total 14 control points UI SIMCOR I D Whole Structure Bine p ee speed gt Communication protocol Communication protocol API API Communication protocol API aa Module 1 Figure 42 Substructure PSD Test Modeling Module 3 Module 2 o UI SIMC GOR 8 5 2 Simulation configuration file CASIMCORNO3 ExamplesNSACNOO CoordinatorNSimConfig m function Sys MDL AUX SimConfig MDL MDL RF AUX MDL AUX Type definition Do not delete this line oo oo Configuration parameters for SDOF experiment oe oe Unit m kN sec oo oe by Oh Sung Kwon okwon2Quiuc edu modified by Kyu Sik Park kspark uiuc edu Univ of Illinois at Urbana Champaign o oe oe oe Last updated on 2007 01 27 12 03AM oe oo oo oe Common parameters oe oo Ground acceleration file name with extension The file should contains two columns for time and acceleration The unit of accele
121. his section 8 6 5 Result and verification The simulation results show in Figure 47 and the Normal and LBCB mean the results from normal system and LBCB system respectively As shown in figure the results of LBCB system are identical with those of normal system 101 Displacement m Rotation rad 0 008 0 006 0 004 0 002 0 002 0 004 0 006 0 008 0 0025 0 002 0 0015 0 001 0 0005 0 0005 0 001 0 0015 0 002 o UI SIMC OR Normal 1 5 2 2 5 3 3 5 4 4 5 5 Time sec 1 5 2 2 5 3 3 5 4 4 5 5 Time sec Figure 47 Pseudo dynamic test result Responses at control point 1 102 o UI SIMC OR 8 7 Buckling Example using ABAQUS In this section inelastic post buckling of cantilever column with unsymmetric open session and initial loading is briefly introduced using ABAQUS static analysis module Use this example as a reference for solving buckling problem in ABAQUS The section used in this example is shown in Figure 48 4000 N 20 mm E of Node E of Element 1682 Computing Time for PSDTest 1 hour 20 minutes 2000 mm 40 mm E 200 000 MPa f 400 MPa with 1 96 hardening Ground Motion Figure 48 Buckling example Figure 49 shows the effect of imperfection in dynamic responses As shown in figure effect of the geometric imperfection was negligible so the imperfection was not considered in this example 103 o UI SIMC R
122. i CheckLimit 1 define following variables Variable size should be number of control nodes 6 array oe oe o oe ox o Displacement increment limit not ratio MDL i TGT D inco abcdef o oe 1 Displacement limit 112 o UI SIMC OR MDL i CAP D tot abcdef s 1 Force limit MDL i CAP F tot abcdef eee E Displacement tolerance ratio MDL i TOL D inc abcdef caren us end Loading and Boundary Condition Box LBCB case If it s 1 the coordinate transformation matrix needs to be provided This can be also used for any other actuator which has diffrence number of DOF coordinate with those of UI SIMCOR oe for i 1 Sys Num RF Module MDL i LBCB 0 end for i 1 Sys Num RF Module MDL i LBCB TransM end oe oe oe Auxiliary module configuration oe AUX 1 MDL AUX AUX 1 URL 127 0 0 1 12000 AUX 1 protocol labviewl1 AUX 1 name Camera Module ID of this mdoule is 1 AUX 1 Command displacement z 3500 8 8 3 Running simulation 1 Module 1 OpenSees with TCP IP protocol a Start NEES SAM by double clicking NEESSAM_TCPIP exe b Select the configuration file from the folder where configuration file and model file exist c Check the open port and status 2 Module 2 FEDEAS Lab with LabVIEW protocol a Starta MATLAB b Change current directory to the folder where FedeasCFG m and the
123. iMOST 1 computer and open the lv programs folder located on the Desktop 186 o UI SIMC R 3 Double click on the daemon programs icon in this folder It should bring up a list of LabVIEW programs Choose Control daemon vi from this menu and click OK You will see the Figure B 9 lE Control daemon vi rev 64 DE x File Edit Operate Tools Browse Window Help d E a n 13pt Application Font Aes wde Running NTCP connected TCP error Dispatch error Logfile I E m h c temp ntcp log txt J Connected NTCP Commands received jo Incoming command Last command timestamp Dispatch error Dispatch timestamp Response Response timestamp Messages stop Not running j Figure B 9 Control daemon in daemon programs a The control daemon program receives NTCP commands In order to start its ability to receive NTCP commands click on the white arrow located on the toolbar of this LabVIEW program 4 Now double click on the mini most icon located in the lv programs folder that was opened on the Desktop Choose Control program vi from the menu of LabVIEW programs and click OK You will see the Figure B 10 I Control program vi rev 42 inf xi File Edit Operate Tools Browse Window Help ctr gt a uj 13pt Application Font tor 3s e5 sim s Actuator name Board ID dose Position Mode Absolute Moving Move complete Error Jucol t s ne e ju
124. iable Sys Num Static Step 0 oo oe oe o Number of dynamic analysis steps Sys Num Dynamic Step 500 o Dynamic analysis time steps Sys dt 0 01 36 o UI SIMC OR Rayleigh damping xi 1 and xi 2 Damping ratio Tn 1 Tn 2 Target period Sys xi 1 0 00 Sys Tn 1 0 00 Sys xi 2 0 00 Sys Tn 2 0 00 oe Number of Stiffness test If stiffness is evaluated through experiment th valuation need to be don several times and the average of the results are used as the initial stiffness This parameter is used when Sys Eval Stiffness 1 Sys Num Test Stiffness 1 oe oe oe oe Enable GUI for SimCor Yes 1 nable the GUI for SimCor o 0 disable the GUI for SimCor Hybrid simulation will be run automatically Not recommended for the experiment ys EnableGUI 1 Use GUI for SimCor oe oe oe oe n Number of restoring force modules Sys Num_RF_Module 1 5 Number of auxilary modules Sys Num AUX Module 0 Total number of effective nodes Effective nodes are interface nodes between modules and nodes where lumped masses are defined Sys Num Node 1 Lumped mass assigned for each DOF for each node Node number x y Z rx ry rz directional mass Sys Node_Mass 1 10 0 0 0 0 O oe oe oe Restoring force module configuration oe Create objects of MDL RF MDL 1 MDL RF Name of
125. ialization initial stiffness formulation static loading and dynamic loading stages During initialization stage the UI SIMCOR makes connections to experimental sites or simulation module and initializes each module Most variables that will be used in later stage are also initialized During Initial stiffness formulation stage UI SIMCOR evaluate stiffness matrix of whole structural system There are two options to establish stiffness matrix When the structural dimension is large stiffness matrix can be loaded from files each of which defines stiffness matrix of each module If the stiffness evaluation is affordable through network UI SIMCOR sends predefined displacement for each degree of freedom to each module and take measured forces to establish initial stiffness of the tested structure The established initial stiffness is used in the static loading stage and in the dynamic loading stage to determine target displacements Most structures are resisting gravity forces before they are hit by earthquake The effect of gravity forces cannot be ignored since the inelastic behavior of columns and beams with initial loading are significantly different from that of the elements without initial loading During the static loading stage displacements due to this gravity forces are imposed When there are no gravity forces this step can be skipped During dynamic loading stage PSD test using a OS scheme Combescure and Pegon 1997 is performed Each ex
126. ices of individual module in the files MDLO1_K txt MDLO2_ K txt etc ys Eval_Stiffness 1 oe oe oe oe LD oe Number of initial static loading steps When ther xist static constant loading i e gravity forces apply then in Zeus NL or OpenSees as a incremental loading with n steps In this file SimConfig m specify the number of static steps in the following variable Sys Num Static Step 10 oe oe oo o Number of dynamic analysis steps Sys Num Dynamic Step 500 63 o UI SIMC OR o Dynamic analysis time steps Sys dt 0 01 Rayleigh damping xi_l and xi_2 Damping ratio Tn_1 Tn_2 Target period Sys xi 1 0 00 Sys Tn 1 0 00 Sys xi 2 0 00 Sys Tn 2 0 00 oe Number of Stiffness test If stiffness is evaluated through experiment th valuation need to be don several times and the average of the results are used as the initial stiffness This parameter is used when Sys Eval Stiffness 1 Sys Num Test Stiffness 1 oe oe oe oe Enable GUI for SimCor Yes 1 nable the GUI for SimCor o 0 disable the GUI for SimCor Hybrid simulation will be run automatically Not recommended for the experiment ys EnableGUI 1 Use GUI for SimCor oe oe oe oe LD Number of restoring force modules Sys Num RF Module 1 5 Number of auxilary modules Sys Num AUX Module 0 oe Total number of effective nodes Effective
127. id simulation framework 10 Click Start PSD Test button in UI SIMCOR control window to run test by simulation The target displacement and feedback displacement and force will be displayed in the monitoring windows in UIUC site Furthermore current status of test will be displayed in NEES SAM API of each site as shown in Figure B 7 184 o UI SIMC OR E NEES SAM LabView File View Help Disp for ID The received displacements are applied to structure Reaction forces and displacements at converged state are saved amp Disp for ID 1 The received displacements are applied to structure Reaction forces and displacements at converged state are saved amp Disp for ID The received displacements are applied to structure Reaction forces and displacements at converged state are saved amp Disp for ID 1 The received displacements are applied to structure Reaction forces and displacements at converged state are saved amp Disp for ID 1 The received displacements are applied to structure Reaction forces and displacements at converged state are saved amp Disp for ID 4 1 The received displacements are applied to structure Reaction forces and displacements at converged state are saved amp Disp for ID 2 1 The received displacements are applied to structure Reaction forces and displacements at converged state are saved amp Disp for ID 1 The received displacements are applied to structure
128. iment in UCB only API Main and UI SimCor Purpose Verify the one site hybrid simulation Most of procedure is same with Step 5 except the running API_Main instead of API Main sim The default value of the displacement limit is 6 If the target displacement which will be sent to xPC Real time target is greater than the displacement limit then the simulation will be stopped with error message If you want to modify the displacement limit open the API Main and modify Disp Limit variable at line 17 1 Runa MATLAB for UI SIMCOR and change the directory to C SIMCOR 03_ Examples UCB SDOF 00_ Coordinator 2 Type Ul SimCor and enter in MALAB command window to run UI SIMCOR 172 o UI SIMC OR New simulation S xi Please backup previous simulation results if necessary All previous files will be deleted Figure A 13 Warning message a Click Okay button you will see the initialize message in a MATLAB command window All previous files will be deleted so if necessary backup previous results b Two popup windows will be displayed one window for control and other window for monitoring will popup In this example only one monitoring window will be displayed as this example has only one remote site Ul SimCor sigi xj Ul SimCor Version25 pista Facing Application for distributed pseudo dynamic simulation n UI SimCor combines Zeus NL OpenSees FEDFASLab ABAQUS and experimetal sites via TCPIP network Yt Mo
129. in section 9 4 Running Procedure 1 Run a MATLAB for UI SIMCOR and change the directory to C SIMCOR 03_ Examples UCB SDOF 00_ Coordinator 2 Type Ul SimCor and enter in MATLAB command window to run UI SIMCOR New simulation z xj Please backup previous simulation results if necessary All previous files will be deleted coe Figure 63 Warning message Click Okay button you will see the initialize message in a MATLAB command window All previous files will be deleted so if necessary backup the previous test results Two popup windows will be displayed one window for control and other window for monitoring will popup In this example only one monitoring window will be displayed as this example has only one remote site 3 Run another MATLAB for communication with xPC Real time target and change the directory to C SIMCOR 04_API 01_UCB 4 Type API_Main you will see the message box confirming the displacement limit as shown in Figure 59 a b Similar to the pushover test the default value of displacement limit of the hybrid simulation is 6 If you want to modify the displacement limit click Reset button then the API_Main file is opened automatically Modify the Disp_Limit variable at line 17 If displacement limit is confirmed click Okay button then you will see the Waiting for connection from client through port 11999 ina MATLAB command window IP address and port number are dependent on ex
130. ing University of Illinois at Urbana Champaign Nakata N Spencer B Johoson E A Elnashi A S Kuchma D A 2003 Finite Element Modeling in MOST Simulation Code Department of Civil and Environmental Engineering University of Illinois at Urbana Champaign Schellenberg A Kim H Takahashi Y Fenves G Mahin S 2006 OpenFresco Framework for Hybrid Simulation Installation and Example Guide Department of Civil and Environmental Engineering University of California at Berkeley Schellenberg A Stojadinovic B Mahin S 2006 The NEES Hybrid Simulation Setup Department of Civil and Environmental Engineering University of California at Berkeley Spencer B F Jr Elnashai Amr K S Park O S Kwon 2006 Installation of UI SIMCOR at SDSC and Implementation of NHCP in UI SIMCOR Department of Civil and Environmental Engineering University of Illinois at Urbana Champaign University of Illinois at Urbana Champaign 2005 User s Manual for MUST SIM Facility 161 o UI SIMC OR Appendix Experimental Procedure A Two site Hybrid Test between UIUC and UCB A 1 Step by step procedure for UIUC site 162 Stage _ Ste To do Purpose 1 Answer to questions e Understand the experiment procedure 2 oneal Backup the previous version of UI Preliminary SIMCOR API UCB and example files 5 e Uninstall previous version o Stage 3 Conti SIMCOR and delete all of previous fi
131. initialize message in a MATLAB command window All previous files will be deleted so if necessary backup previous results b Two popup windows will be displayed one window for control and other window for monitoring will popup In this example only one monitoring window will be displayed as this example has only one remote site Ul SimCor Version26 MUST SIM Facility Figure A 7 Control window 168 o UI SIMC OR STATIC 12700 111998 nerra 74398 358 34 33 o 03 04 ob 08 1 E 4 08 026 04 02 o 02 04 0 os 1 Figure A 8 Monitoring window UCB Site 3 Run another MATLAB for communication with xPC Real time target and change the directory to C SIMCOR 04_API 01_UCB 4 Type API Main sim you will see the message box confirming the displacement limit as shown in Figure A 9 Check Displacement Limit 5 xj Current displacement limit for UCB speicmen 5 00e 000 IF you want modify the displacement limit please open API_Main_sim m e Figure A 9 Warning message for the displacement limit If the displacement limit is confirmed click Okay button then you will see the Waiting for connection from client through port 11999 ina MATLAB command window If you want modify the displacement limit click Reset button then the API Main sim file is opened automatically Modify the Disp Limit variable at line 17 Now ready to run the test by simulation 169
132. ion Simulation procedure is identical to Section 8 1 4 Again please note that the procedure should be followed whenever new analysis starts 8 2 5 Result and verification When the column is subjected to gravity force the column shortens 0 058333 mm Figures 33 and 34 shows the PSD test result that the vertical displacement reaches 0 058333 mm after 10 static steps and hold the displacement constant during dynamic stage During static stage the translational displacement remains zero Figures 33 and 34 show the vertical and horizontal displacement history from PSD test using OpenSees and ABAQUS as a static module When ZEUS NL is used for the initial loading problem there could be a fluctuation in translational displacement during static loading stage It s because ZEUS NL always accounts for geometric nonlinearity Thus the estimated stiffness matrix is not diagonal for this problem and result in inaccurate displacement estimation The effect of geometric nonlinearity however can be controlled by choosing small displacement during initial stiffness formulation stage Figure 35 shows very good correspondence between whole model run and PSD test using UI SIMCOR 67 Vertical displacement mm Horizontal displacement mm 0 01 0 02 L 0 03 p 0 04 L 0 05 0 06 0 07 0 05 0 1 0 15 L 02 L 0 25 o UI SIMC OR Static Stage Dynamic Stage 6 8 4 8 9 e e a e e e e e a
133. ist of all files of the project The release you have chosen is highlighted Before downloading you may want to read Release Notes and ChangeLog accessible by clicking on release version F 2 E a Date Packa Release amp Notes Filename dii ee R Size D L Arch Type UI SIMCOR UI SIMCORV2 5 2006 12 01 10 39 simCar msi 10 37 MB 29 Other Other 2006 01 20 2006 01 20 15 39 Release zip 5 02 MB 60 i386 zip users manual pdf 4 93MB 156Any pdf Project totals 2 3 21 28 MB 245 LZ el T TL D B ep vene Z Figure 13 UI SIMCOR download webpage on NEESit Download UI SIMCOR v2 6 by clicking on SimCor msi Currently UI SIMCOR v2 5 and v2 6 are available in the NEESforge website By double clicking the downloaded file you can install UI SIMCOR interface applications for structural analysis programs and example files When you install software DO NOT change the default folder location The software should be installed in C SIMCOR At the end of the installation process you will see a message that some folders should be set in the search path of MATLAB To do that open MATLAB and click File Set Path Following folders should be in the search path of MATLAB Path for UI SIMCOR CASIMCORV1 SIMCOR C SIMCOR 01_SIMCOR parmatlab 29 o UI SIMC R Path for FEDEAS Lab and NEES FL C SIMCOR 02_NEES FL C SIMCOR 02_NEES FL FedeasLab C A SIMCOR 02_NEES FL FedeasLab Element_Lib C SIMCOR 02_NEES FL FedeasLab General C SIMCOR 02_NEES FL Fedea
134. ity to transfer data between MEX files and MATLAB and the ability to call MATLAB functions from C C or Fortran code The main reasons to write a MEX file are 1 The ability to call large existing C C or Fortran routines directly from MATLAB without having to rewrite them as M files 2 Speed user can rewrite bottleneck computations like for loops as a MEX file for efficiency The source code for a MEX file consists of two distinct parts 1 A computational routine that contains the code for performing the computations that you want implemented in the MEX file Computations can be numerical computations as well as inputting and outputting data 2 A gateway routine that interfaces the computational routine with MATLAB by the entry point mexFunction and its parameters prhs nrhs plhs nlhs Where prhs is an array of right hand input arguments nrhs is the number of right hand input arguments pins is an array of left hand output arguments and nins is the number of left hand output arguments The gateway calls the computational routine as a subroutine In the gateway routine user can access the data in the mxarray structure and then manipulate this data in your C C computational subroutine For example the expression mxGetPr prhs 0 returns a pointer of type double to the real data in the mxArray pointed to by prns o User can then use this pointer like any other pointer of type double in C C After calling user s C C com
135. l ZEUS PSD Test ZEUS E E t 5 E D o S Qa E a 0 0 5 1 1 5 2 2 5 3 3 5 4 4 5 5 Time sec 0 1 Whole Model ABAQUS 0 05 PSD Test ABAQUS E E t 0 d E o 0 05 a E a 0 1 0 15 0 0 5 1 1 5 2 2 5 3 3 5 4 4 5 5 Time sec Figure 44 Pseudo dynamic test result Horizontal displacement at top node of left column 92 o UI SIMC GOR 0 08 0 06 L Whole Model FEDEAS duae PSD Test FEDEAS 0 04 0 02 0 02 Displacement mm 0 04 0 06 0 08 0 1 1 L 1 L 1 L 1 L 1 0 0 5 1 1 5 2 2 5 3 3 5 4 4 5 5 Time sec 0 08 0 06 Whole Model ZEUS PSD Test ZEUS 0 04 0 02 0 02 Displacement mm 0 04 0 06 0 08 0 1 1 1 L 1 L 1 L 1 L 0 0 5 1 1 5 2 2 5 3 3 5 4 4 5 5 Time sec 0 08 0 06 Whole Model ABAQUS I PSD Test ABAQUS 0 04 0 02 0 02 Displacement mm 0 04 0 06 0 08 0 1 L L n L n L L L L 0 0 5 1 1 5 2 2 5 3 3 5 4 4 5 5 Time sec Figure 45 Pseudo dynamic test result Horizontal displacement at the second floor of the third column 93 o UI SIMC OR 8 6 LBCB Example The UI SIMCOR calculates the target displacements related to effective DOFs and send these target displacements to actuators However experimental equipment can have different number of actuators with effective DOFs For example the LBCB
136. l point name ceonrype mirros the NTCP specification and is a single character x y Or z ParamterType also mirrors the specification and can be displacement force rotation OT moment Parameter is simple a scalar floating point number For example propose TransactionO0 ANCO table x force 0 3 x displacement 3 1 y rotation 180 1 execute cancel Syntax execute TransactionID This simply triggers execution of a previously accepted proposal Obviously the transaction ID must have been both proposal and accepted for this to be valid The control system should not return until the execution is complete Syntax 14 o UI SIMC R cancel TransactionID This command cancels a pending command usually due to a proposal being rejected by another control system somewhere else and the plugin throw an exception This should return OK if the transaction was canceled successfully get control point Syntax get control point TransactionID ControlPoint where the transaction ID is usually ignored but required for consistent parsing The return format has a variable number of parameters depending on what the given control point supports Mirroring the propose syntax it returns to up to twelve triplets of Geomtype ParamType Paramter With the same types as propose above e g x displacement 1 414 For example get control point DummyTransactionID003 ANCO replies OK 0 DummyTransactionID003 x displacement 0 0 x
137. lacement will be sent to each site 146 o UI SIMC OR UI SIMCOR Protocol CP2 CP3 CP 4 CP 5 gt gt gt UIUC ZEUS NL UCB SDSC Wisconsin Miwaukee Chicago Kansas Missouri San Francisc o Wichita o Springteld California Bakersfieldo Las Vegas Albuquerque m rt Oklahoma Arkansas Arizona Los Angeleso New Mexico dLbbock Mississippi A Dallas Abilene o b San H El Paso Shreveport Alab 5n Texas Mobile Tn Louisiana o Corpus p Christi oMami SDSC MiniMOST1 Figure 81 Hybrid simulation configuration of three site experiment 147 Displacement 0 015 Displacement 1 5 L 2 5 Displacement 0 5 0 015 o UI SIMC OR UIUC 0 01 0 005 L 0 005 p 0 01 Displacement Limit Displacement Limit 2 5 Time a MiniMOST 1 at UIUC UCB 15 k 0 5 Displacement Limit Dat DTA pps pa ee ey ay E V Displacement Limit 0 015 0 01 0 005 p 0 005 p 0 01 0 015 Time b uNEES at UCB SDSC Displacement Limit tan preyed AaYVMUVUWV Y V Displacement Limit Time c MiniMOST 1 at UIUC Figure 82 Hybrid test results input case 1 148 Force Force Force o UI SIMC OR UIUC 15 0 015 0101 0 005 0 015 Displacement Limit Displacement Limit Displacement a MiniMOST 1 at UIUC UCB
138. le When scale factor needs to be applied either in experiment or simulation To define force and displacement criteria for tolerance and safety To trigger camera modules or DAQ system When LBCB at UIUC is used for experiment When NHCP protocol is used MDL i remote_ URL MDL i remote URL directs the address of remote site for each module when NHCP protocol is used For local machine MDL i remote URL 127 0 0 1 port number MDL i NHCPMode This variable defines the mode for NHCP protocol This variable one of Sim1D for simulation case MM1 and MM2 for Mini MOST 1 and 2 at UIUC or SDSC respectively If NHCPMode is Sim1D then MDL 1 NHCPSimK is required MDL i NHCPSimK This variable defines the stiffness of Simulation Server for NHCP Currently only linear 1 DOF simulation case is available MDL i TransM This variable defines the transformation matrix If the coordinate system between UI SIMCOR and static module is different transformation matrix need to be given 43 o UI SIMC OR MDL i ScaleF This variable defines the scale factor for displacement rotation force and moment Experimental specimens are not always in full scale These factors can be used to apply scale factors The displacement and rotation scale factors are multiplied before they are sent to module or experimental specimen Measured forces and moments are divided with scale factors before used in the PSD algori
139. les S Install new version of UI SIMCOR and 4 is copy required files for UCB site 5 e Receive the results from Verify the configuration of newly UCB installed UI SIMCOR e Check IP address and port number e Run the preliminary test by simulation in UCB API Main sim and imulation API Mam ag UIUC UL SimCor Stage Verify the communication between 9 6 Does UIUC site UIUC and UCB communicate with UCB site e ITs it running Receive the results from UCB e Backup the results o UI SIMC R A 2 Step by step procedure for UCB site Stage Step To do Preliminary Stage 1 Print this document UCB_Report doc Read document very carefully especially 5 Hybrid test between UIUC and UCB Is there any questions N Backup C SIMCOR folder Uninstall UI SIMCOR and delete C SIMCOR folder Install new version of UI SIMCOR Copy 05 API UCB and 06 Examples UCB folder to C SIMCOR Add paths in MATLAB Simulation Stage Run the preliminary test by simulation API Main sim and UI SimCor in UCB site only Is it running Backup the results and send to UIUC Give IP address and port number to UIUC Run the preliminary test by simulation in UCB API Main sim and UIUC UI SimCor Does UCB site communicate with UIUC site Is it running Backup the results and send to UIUC 163 o UI SIMC OR A 3 Detailed procedure of each step Installation procedu
140. mber RH v xX au ea Figure 19 Time history monitoring window DF MONITOR Deformation Force monitor By defining this variable real time monitoring window pops up to show displacement force relation ship The chart is updated at each time step This monitor is also available in UI SIMCOR Real time Monitoring 1 x Monitoring Window 30 000 20 000 l 10 000 n2 e 10 000 4 S 20 000 30000 40 000 4 C 50 000 L 60 000 10 8 6 4 2 0 2 4 6 X disp n2 lsz a0 v Xx 3u Figure 20 Displacement Force monitoring window OpenSees Module OpenSees Model can participate in the PSD test by using NEES SAM as an interface module Original version of OpenSees was modified to make a displacement output file for each time step The file is read by NEES SAM and convert displacement to reaction forces Detailed explanation about the modeling of OpenSees is not given in this document You can find very well documented manual from following link http opensees berkeley edu To use OpenSees as a static analysis module in PSD test simulation using UI SIMCOR and NEES SAM following points should be accounted for 50 o UI SIMC OR e In modeling file only node element material section properties and boundary conditions are given Other parameters as shown below should be given in a separate file named as StaticAnalysisEnv tcl
141. med for xPC Real time target e The application name for xPC Real time target of the experiment is HybridControllerPoly3 1Act e The function which uses for initialize the xPC Real time target execute experiment propose the target displacement and acquire the measured data is HybridControlxPCtarget e The directory for initialization of xPC Real time target is D OpenSees PredictorCorrector RTActualT estModels c amp mCode xPCTarget STS 9 1 3 Pushover test Before the hybrid test the pushover test was conducted to calculate scale factor in UI SIMCOR and to make an analytical model for comparison after test The specimen provided by UCB is shown in Figure 57 119 o UI SIMC OR Figure 57 Specimen at UCB Running procedure The procedure in this section is for the UCB site The procedure may vary with different experimental configurations Figure 58 shows the configuration of hybrid simulation framework in UCB The most important customization was writing API for UCB site which acts key role for communication between UI SIMCOR and experimental site Before starting the experiment make sure followings Restart the PC for xPC Real time target for every new hybrid test After upgrading MATLAB from 7 0 to 7 1 the PC needed to be restarted It seems like a version compatibility problem of MATLAB Set the displacement limit of the specimen for safety Zero the initial displacement and initial force of controller
142. model file MDLO2 m exist c Run FEDEAS Lab by inputting following command NEES FL LabView2 enter 3 Module 3 ZEUS NL with LabVIEW protocol a Start NEES SAM by double clicking NEESSAM TCPIP exe b Select the configuration file from the folder where configuration file and model file exist c Check the open port and status 4 Module 4 OpenFresco with OpenFresco1D protocol a Start WINDOW DOS command window and change current directory to the folder where the model file MDL04 tcl exists 113 o UI SIMC OR b Run OpenFresco by inputting OpenFresco enter c Load model file by inputting source MDL04 tcl enter 5 Module 5 NHCPSimID with NHCP protocol a Start WINDOW DOS command window and change current directory to the folder where the ncs exe exists b Run NCS by inputting ncs portnumber enter where portnumber is defined in the SimConfig m file MDL 1 URL c Start another WINDOW DOS command window and change current directory to the folder where the simserver exe exists d Run SimServer by inputting simserver portnumber enter where portnumber is defined in the SimConfig m file MDL i remote URL 8 8 4 Result and verification The results of PSD test using UI SIMCOR are very good correspondence with those of whole model run as shown in Figure 55 114 Displacement Displacement o 5 0 025 0 02 0 015 0 01 0 01 0 015 4 0 02 4 o UI SIMC OR Whole Model
143. mote site 6 5 NEES Hybrid Communication Protocol NHCP 6 5 1 Background and introduction Previous communication protocol of NEESit i e NTCP is designed for slow tests over unreliable networks spanning one or more laboratories Indeed it has been used successfully in several of these and researchers continue to use it NEESit has been reworking NTCP such that it can be used for fast distributed tests In particular one proposed experiment they are using as a goal is 75 Hz over a 500 km link NEESit are not going toe try and address fast hybrid tests that run in hard real time instead they will focus on distributed tests both slow and fast attempting to run as fast as possible One area where NEESit can contribute is streaming data NEESit have an existing system based on the Create Date Turbine that works well for data video and audio NEESit plans 17 o UI SIMC OR on implementing an interface to the turbine for streaming commands responses events and other experimental data for observers and participants Security is another area where NEESit can contribute The framework will have authentication authorization and hooks or tools to administer permissions GridAuth http www gridauth com Kerberos http web mit edu kerberos and JAAS http java sun com products jaas are all intriguing here Based on initial results and feedback from the Denver meeting http it nees org weblog hybridsim p 3
144. motion recorded at a site in El Centro California during the Imperial Valley earthquake of May 18 1940 is used for all cantilever column examples 8 1 1 Structural configuration The column height is 3 500 mm with 10 N mm sec of mass at the top Figure 16 Frame element is used to model the column Only x directional DOF is used for dynamic analysis El Centro ground motion was applied along the x direction at the base Lumped mass is defined in simulation coordinator Stiffness evaluation and static analysis is performed in static analysis module FEDEAS Lab ZEUS NL OpenSees ABAQUS OpenFresco and NHCP Simulation Server X X X gt m 10 N mm sec B m 10 N mm sec o S EE lt gt lt gt A A Column dimension SIMCOR Static Analysis Module Figure 16 Column configuration 8 1 2 Simulation configuration file Simulation configuration file contains all the information that is necessary for UI SIMCOR for the multi site simulation Many parameters are trivial and self explained 35 o UI SIMC R Note that most of the parameters should be defined for all analysis and variable names should not be modified The MATLAB script is case sensitive C SIMCOR 03_Examples SDOF 00_Coordinator SimConfig m function Sys MDL AUX SimConfig MDL MDL_RF AUX MDL_AUX Type definition Do not delete this line oo oo Configuration parameters for SDOF experiment oe oe
145. mparison between experimental and analytical results of each site 144 Force Force o UI SIMC OR UIUC 40 Experiment 30 L Simulation 0 015 0 015 Displacement a MiniMOST 1 at UIUC UCB e Experiment Simulation Displacement b uNEES at UCB SDSC 20 Experiment Simulation 15 F 10 Force 0 015 0 01 0 015 Displacement c MiniMOST 1 at SDSC Figure 80 Comparison between experimental and analytical model of each site 145 o UI SIMC OR As shown in Figure 80 the results of analytical model are well matched with those of experiment test so theses analytical models could be used to estimate the target displacement during the hybrid test 9 4 6 Simulation results The modules in the three site example shown in Figure 73 are replaced with the specimen of each site as shown in Figure 81 The simulation results are shown in Figures 82 and 83 and the target displacement for each site is well within the displacement limit Figures 84 and 85 show the simulation results under magnified excitation The input excitation is increased 1 8 times to increase the target displacement In this case same scale factors obtained in Section 9 4 4 are used The target displacement for each site is also well within the displacement limit and these estimated target disp
146. mplemented now oe oe UIUC or SDSC MDL 1 protocol TCPIP MDL 2 protocol LabView2 MDL 3 protocol LabViewl MDL 4 protocol OpenFrescolD 110 NHCP NHCP linear 1 DOF simulation mode Mini MOST 1 and 2 at o UI SIMC OR MDL 5 protocol NHCP o 6 o 6 o 6 2 6 Module 1 OpenSees with TCPIP DL 1 node 2 Control point node number DL 1 EFF_DOF 1 0 0 00 0 Effective DOF for CP 2 Module 2 Fedeas Lab with LabView2 DL 2 node 3 Control point node number DL 2 EFF DOF 00000 Effective DOF for CP 3 Module 3 ZeusNL with LabView2 DL 3 node 23 4 5 6 7 Control point node number DL 3 EFF_DOF 1 00000 Effective DOF for CP 1 100000 Effective DOF for CP 2 100000 Effective DOF for CP 3 00000 Effective DOF for CP 4 100000 Effective DOF for CP 5 100000 Effective DOF for CP 6 i 0 00 0 0 Effective DOF for CP 7 Module 4 OpenFresco with OpenFrescolD MDL 4 node MDL 4 EFF_DOF 2 6 Module 5 MDL 5 node MDL 5 EFF_DOF oe oe o oe oe oe ANP oe oe oe oe oe NHCPSim1D with NHCP oe 5 1 000 0 0 Control point node number Effective DOF for CP 5 oe oe 6 1 000 0 0 Control point node number Effective DOF for CP 6 oe Dismplacement for preliminary test for each module Del t
147. n ZEUS NL is used as a static analysis module in which there are multiple control points or multiple DOFs for single control points the static time history load should be defined in the order of control point number and DOFs number For instance the left column of the MOST experiment consists of one control point in which two DOFs are used In ZEUS NL input file time history load should be defined in the order of control points and then DOFs 1 time history loads 21 nod name direction type crv name value 3 n102 x displacement CV T 79 o UI SIMC OR 4 n102 rz displacement CV 1 In other words the third and fourth line in the above script cannot be swapped Furthermore the type of protocol and port number should be identical between simulation coordinator and each module configuration file This and following SAC examples need to run three ABAQUS simultaneously However it is impossible to run three ABAUQS under limited license Instead of running three ABAQUS in one computer multiple computers can be used for example which has modules more than three 8 4 4 Running simulation MOST experiment requires four processes one for UI SIMCOR and the other three for static analysis modules UI SIMCOR requires a MATLAB If FEDEAS Lab or ABAQUS models are going to be used as a static analysis module another MATLAB should be started If all three modules are FEDEAS Lab or ABAQUS models then three another MATLAB processes should be start
148. n is ID for connection openconnection for open action and PORT OpenFresco and rP OpenFresco are port number and IP address for OpenFresco respectively set data sizes for remote site Syntax TCPSocket sendData socketID dataSizes 11 16 o UI SIMC OR where sendData for sending data to OpenFresco datasizes is size of data which will be sent to OpenFresco For 1 DOF system datasizes int32 1 0 0 0 0 0 0010 dataSize where dataSize 2 send target responses to remote site Syntax TCPSocket sendData socketID sData dataSize where spata is for target displacement of size 1 x 2 spata 1 3 for sendData command and sData 2 is target displacement which was sent from UI SIMCOR dataSize is size of sDate i e 2 get measured restoring force Syntax TCPSocket sendData socketID sData dataSize rData TCPSocket recvData socketID dataSize where sData is scalar value of 10 for query action aatasize is 2 which was defined during set data sizes for remote site rData is measured restoring force received from OpenFresco UI SIMCOR assumes that the measured displacement is same with the target displacement disconnect from remote site Syntax TCPSocket sendData SocketID sData dataSize TCPSocket closeConnection socketID where sData is a scalar value of 99 for close action datasize is 2 which was defined during set data sizes for re
149. n prompted svn checkout username jemeJopernmnawe https scm nessforge ness org svn simcor Download The Nightly SYN Tree Snapshot i https Insesforge nees org IBI ap internet Z Figure 15 UI SIMCOR SCM webpage on NEESit To browse the source code click Brows Subversion Repository on the right side of the page However it is recommended to use Subversion software for downloading and updating the source code Only authorized user can update the source code The freely available Subversion software including documentation can be found in the following link http tortoisesvn net 33 o UI SIMC OR The files in the SCM tab of NEESforge are the latest version of UI SIMCOR source code so it may have some bug or may not working If user wants stable UI SIMCOR source code user needs install the latest version of UI SIMCOR 34 o UI SIMC OR 8 Numerical Examples In this section configuration and modeling guidelines for the UI SIMCOR and NEES SAM are introduced through various numerical examples 8 1 Cantilever Column A dynamic analysis of a cantilever column with a lumped mass is introduced here to demonstrate and verify each analysis module The column is tested with elastic material without initial loading using three static analysis applications FEDEAS Lab ZEUS NL OpenSees and ABAQUS a OS scheme with 20 05 f 1 4 1 a and y 1 2 a is used for all the simulation N S component of the ground
150. nalytical model was simulated at SDSC It took 1 min and 29 sec for 100 steps 0 9 sec per to finish the pseudo dynamic analysis stage in UI SIMCOR As shown in figures the results simulated at each site were exactly 152 o UI SIMC GOR matched with those simulated at one computer Therefore the communication among UIUC UCB and SDSC was verified through this test MM1 UIUC 0 001 0 0008 L Expected 0 0006 Actual 0 0004 L E 0 0002 fs 8 H i i Z 0 0002 9 0 2 0 4 0 6 0 8 1 Z 0 0004 v 0 0006 L 0 0008 L 0 001 0 0012 Time MM1 UIUC 25 Expected 2r Actual 15r 1L 05r 8 C 0 0012 0 001 0 0008 0 0006 0 0004 20 5 0 0002 0 0004 0 0006 0 0008 0 001 A F 1 5 F 2 L 2 5 L 3 Displacement Figure 87 MiniMOST 1 at UIUC mNEES UCB 0 2 0 15 L Expected Actual 0 1 L 0 05 L e B o i 8 S 0 2 0 4 0 6 0 8 1 a 0 05 L E a 0 1 L 0 15 L 0 2 L 0 25 Time mNEES UCB 0 5 Expected 04 r Actual 0 3 L co g 025 0 15 0 2 Displacement Figure 88 uNEES at UCB 153 o UI SIMC R MM1 SDSC 0 001 0 0008 Expected Actual 0 0006 0 0004 Fr 0 0002 f 0 1 1 1 1 0 0002 02 04 0 6 0 8 1 0 0004 Fr 0 0006 Displacement 0 0008 0 001 L 0 0012 Time MM
151. nassigned The unc c gateway routine uses the mxcreate function to create the MATLAB arrays for your output arguments It sets pins 0 1 to the pointers to the newly created MATLAB arrays It uses the mxcet functions to extract user s data from prhs 0j 1 Finally it calls C C subroutine passing the input and output data pointers as function parameters On return to MATLAB pins 0 is assigned to c and pins 1 is assigned to D The two components of the MEX file may be separate or combined In either case the files must contain the include mex n header so that the entry point and interface routines are declared properly The name of the gateway routine must always be mexFunction and must contain these parameters 23 o UI SIMC OR void mexFunction int nlhs mxArray plhs int nrhs const mxArray prhs more C C code The parameters nins and nrhs contain the number of left and right hand arguments with which the MEX file is invoked In the syntax of the MATLAB language functions have the general form of a Dy ss fun d e Tf where the ellipsis denotes additional terms of the same format The a b c are left hand arguments and the a e are right hand arguments The parameters pins and prns are vectors that contain pointers to the left and right hand arguments of the MEX file Note that both are declared as containing type mxarray which means that the vari
152. nator is lost Normally NHCP would just terminate the plugins and reset but it needs to consider error handling Did the coordinator crash Was it killed Or was the network temporarily damaged Since NHCP can not tell from within the server its generates a critical error callback and lets the plugins do whatever error handling is required The coordinator can reconnect and resume when able If the coordinator logs out and unloads the plugins then they are terminated in the normal fashion The server has a simple interface MSG QUERY LOAD and UNLOAD for dynamic discovery and plugin management There are a number of benefits to this design l Server configuration is greatly reduced In most uses there needs be only one server running on a single well known TCP port Coordinators can simply connect login and then load the plugins required for their test This compares well to the effort required to configure NTCP All plugins see all messages making possible things like the observer plugin that can be used for monitoring logging database back ends and eventually RBNB all transparently Users can load multiple instances of a single plugin into the server and configure them on the fly to do different tasks For example SDSC will use the NTCP ASCII plugin twice to connect to MiniMOST 1 and 2 This is accomplished by loading the plugin then using the new MSG SET INSTANCE NAME and MSG SET to configure the backend and in
153. nodes are interface nodes between modules and nodes where lumped masses are defined ys Num Node 1 oe n Lumped mass assigned for each DOF for each node Node number x y Z rx ry rz directional mass Sys Node_Mass 1 10 0 0 0 0 O oe oe oe Restoring force module configuration oe Create objects of MDL RF MDL 1 MDL RF Name of each module MDL 1 name STATIC Module ID of this module is 1 oe URL of each module Format IP address port number ex http c nsp4 cee uiuc edu 11997 for local machine 127 0 0 1 11997 MDL 1 URL 127 0 0 1 11997 oe Communication protocol for each module NTCP communicate through NEESPOP server TCPIP binary communication using TCPIP LabViewl ASCII communication with LabView plugin format Propose Query Execute Query LabView2 same as LabViewl but Propose Query OpenFrescolD OpenFresco only 1 DOF is implemented now E NHCP NHCP linear 1 DOF simulation mode Mini MOST 1 and 2 at oe UIUC or SDSC 64 o UI SIMC R MDL 1 protocol LabView2 Module 1 STATIC MDL 1 node 1 Control point node number MDL 1 EFF DOF 1 1 O 0 O 0 Effective DOF for CP 1 Dismplacement for preliminary test for each module Del t Translation Del r Rotation in radian MDL 1 DEL t 0 005 I MDL 1 DEL r 0 002 oe Enable G
154. odeled by ZEUS NL Elnashai et al 2006 UI SIMCOR Communication Communication Communication Communication Protocol Protocol Protocol Protocol API API API API Module 1 Module 2 Module 3 Module 4 Figure 73 Hybrid simulation configuration of three site experiment 139 o UI SIMC OR 9 4 2 Simulation results Before the real experiment hybrid test is conducted only by simulation using the model of each module at one computer The scale factor of each site for UI SIMCOR will be calculated using the stiffness of each module and responses The input excitation used in this example is shown in Figure 74 Acceleration g 0 25 0 2 0 15 0 1 0 05 0 05 0 1 0 15 0 2 0 25 Time sec Figure 74 Input ground motion Figure 75 shows the pseudo dynamic PSD test results and compared with whole model Whole model and each module of PSD test are modeled by ZEUS NL The PSD results show good agreement with those of whole model Horizontal Displacement m 0 03 Whole ZEUS PSDTest ZEUS 0 03 0 0 5 1 1 5 2 2 5 3 3 5 4 4 5 5 Time sec Figure 75 Horizontal displacement at CP2 140 o UI SIMC OR The stiffness of each module can be calculated by structural parameters or stiffness evaluation step during PSD test by imposing single step displacement in UI SIMCOR or pushover test of each module in
155. odule i 1 1 i I Open TCP IP or NTCP connettion i B L l T i Acknowledge 3 2 Hardware system 1 Open TCP IP or NTCP connection i i ly T i 1 H Acknowledge 1 1 p l 4 Stiffness evaluation l i 1 for each module 1 1 1 i 1 i i i H 4 1 Ask for stifness 1 i i Yes i i l Tetum K 1 1 1 i TNo evaluate K D i i i for each DOF SetTargetDisp H gt H 1 j Acknowledge l Execute 1 i Acknowledge Execute test fe cg Trigger DAQ 1 L for each hardware T i H Trigger i 1 1 Acknowledge Query i 1 i l SS E R H i Measured data i 4 Felumk Assemble k l i 5 Simulation i levery time step i i i i H 5 1 Get displacement increment H Retu irn displacement increment i 5 2 Run experiment for each module i 1 1 H H Displacement SetTargetDisp H H i __ ee o O IM i i i Acknowledge 1 Execute 1 1 H l n I g Acknowledge Execute test i i i la Trigger DAQ 1 i Tor each hardware H 1 T gt Trigger 1 i Acknowledge Query 1 i i i n I i Measured Force 1 1 Measured data i e a i 6 Close connection i 1 i i H rl each module i i i 6 1 Close remote iig Close TCP IP or NTCP connection h fi Acknowledge 6 2 Close hardware system 1 i 1 Close TCP IP or NTCP connection 1 i La Acknowledge Figure 2 Flow chart of the PSD test using UI SIMCO
156. of MATLAB is used This toolbox is a collection of M file functions build on the MATLAB technical computing environment The toolbox provides a framework for communicating with instruments that support the GPIB interface the VISA standard and the TCP IP and UDP protocols The transferring data can be binary numerical or text The transfer can be synchronous and block the MATLAB command line or asynchronous and allow access to the MATLAB command line More details about Instrument Control Toolbox can be found in the manual of MATLAB or MATHWORKS website www mathworks com 6 2 2 Basic sequence of commands A typical sequence of commands from TCP IP protocol to the control system looks like this 1 initialize 1 create TCP IP communication object 2 set parameters 2 open 1 connect interface to remote site 2 send initialize data to remote site 3 receive acknowledgement from remote site 3 loop N times 11 o UI SIMC OR 1 propose 2 query 4 close 1 send closing data to remote site 2 disconnect interface object from remote site 6 2 3 Detailed syntax of each step initialize Syntax Comm obj tcpip IP PORT to create TCP IP object set Comm obj PropertyName PropertyValue to set parameters where comm obj is TCP IP communication object returned from TCP connection tp and port are IP address and port number of remote site respectively PropertyName and PropertyValue area property name for Comm obj
157. omp 1 7 4 1 1 1 08 06 04 02 0 02 04 06 08 1 Status Message Figure 27 UI SIMCOR control window and simulation monitor window 8 1 5 Result and verification Following files will be generated after analysis C SIMCOR 03_ Examples SDOF00 Coordinator Global K txt Stiffness matrix from initial stiffness evaluation MDLO1 K txt Stiffness matrix of module 1 In this example only one module exists MDLOI recv txt This file contains received data from module 1 The displacements and forces are after coordinate transformation and scaling if necessary NodeDisp txt Nodal displacements Node numbers corresponds to the node number for UI SIMCOR NetwkLog txt Communication log file C SIMCOR 03_Examples SDOF 01_Fedeas 57 o UI SIMC OR NetwkLog txt Communication log file CASIMCOR 03 Examples SDOF 02_ZEUS Cur Disp txt Current displacement after each step analysis Cur Forc txt Current force after each step analysis NetLog txt Communication log file CASIMCOR 03 Examples SDOF 03_ OpenSees Same as 02 ZEUS C SIMCOR 03_Examples SDOF 04_ Abaqus Same as 01 Fedeas In the folder CASIMCOR 03 Examples SDOF WholeModel there is a whole model for each static analysis application Figures 28 and 29 show the input earthquake and the PSD test results for each static analysis module 0 4 0 3 L 02 L 0 1 L Acceleration g oO 0 1 F 0 2 0 3 F 0
158. or any other actuator which has diffrence number of DOF coordinate with those of UI SIMCOR oe oe oo for i 1 Sys Num RF Module MDL i LBCB 0 end for i 1 Sys Num RF Module MDL i LBCB TransM end oe oe Auxiliary module configuration AUX 1 MDL AUX AUX 1 URL 127 0 0 1 12000 AUX 1 protocol labviewl AUX 1 name Camera Module ID of this mdoule is 1 AUX 1 Command displacement z 3500 Variable Description Most variables in SimConfig m are explained in Section 8 1 for SDOF system In this section variables that are related to multiple control points and multiple modules are further explained Sys Num RF Module In this example the MOST experiment will be simulated using three analysis modules Thus Sys Num RF Module is 3 Each module has its own module number The one representing left column is assigned number 1 beams and middle column is number 2 and the right column are designated as number 3 Sys Num Node The MOST experiment consists of three control points As discussed in Section 4 control points are nodes with lumped masses or with DOFs of interest Sys Node_Mass i Lumped masses are assigned for each node In this example it is assumed that only x directional masses exist MDL i name 78 o UI SIMC OR Since there are three modules three names i e LeftCol Middle and RightCol for each module are defined in the configura
159. other computer or the other site i e UIUC then this GUI helps to UCB site seeing current status of experiment Click Stiffness Evaluation button in UI SIMCOR control window UI SIMCOR will read the predefined stiffness matrix from file MDLO1_K txt located in C SIMCOR 03_ Examples UCB SDOF 00_ Coordinator Click Apply Static Loading button in UI SIMCOR control window In this stage the gravity force is applied however there is no plan to apply gravity as this exemplary test is to demonstrate the efficacy of hybrid simulation framework Click Start PSD Test to run test by experiment The target displacement and feedback displacement and force will be displayed in the monitoring window and 174 o UI SIMC OR UCB monitoring window 9 Click Disconnect Modules after finishing hybrid simulation to disconnect UI SIMCOR from xPC Real time target You will see some messages indicating the communication is disconnected in both MALTAB command windows 10 Backup the results files MDLOI recv txt Netwklog txt NodeDisp txt located in C SIMCOR 03_Examples UCB SDOF 00 Coordinator and another results files NetwkLog txt UCB_Results txt located in C A SIMCOR 04_API 01_UCB 11 Send results to UIUC Step 9 Run the experiment in UCB API Main and in UIUC UI SimCor Purpose Verify the hybrid simulation between UIUC and UCB Most of procedure is same with Step 8 except the UI SIMCOR will run in UIUC site To do this test UIUC should know
160. over test Before the hybrid test the one site experiment at UCB was conducted to verify the configuration of xPC Real time target and the test results are shown in Figure 70 It took 3 min and 20 sec for 500 steps 0 4 sec step to finish the pseudo dynamic stage in UI SIMCOR There are little difference between simulation and experiment because of difference of coupon at UCB specimen The analytical model of uUNEES specimen at UCB was obtained by using the pushover test results conducted during August 9 2006 August 16 2006 Furthermore the characteristic of specimen at UCB is highly dependent on the coupon installed in the specimen The coupon used in this test is little bit different from the coupon used for pushover test so there are little difference between simulation and experiment 135 0 8 0 6 o UI SIMC OR old coupon sim L new coupon exp O04 F T I 2 0 2 L E LV l N l Bap UI 19 N I 02 VY V Y P nU 0 4 0 6 L JIL i 0 8 y 1 0 0 5 1 1 5 2 5 3 3 5 4 4 5 5 Time sec 5 old coupon sim new coupon exp S C 0 8 1 t45 Displacement Figure 70 Experiment results 9 2 4 Two site experiment between UIUC and UCB Finally two site experiment between UIUC and UCB was conducted and the results are shown in Figure 71 It took 3 min and 16 sec for 500 steps 0 4 sec step to finish
161. perimental configuration It is highly recommended to run the simulation by using API Main sim before the real experiment 124 o UI SIMC OR Now ready to run experiment 5 Results Click Establish Connection button in UI SIMCOR control window You will see some messages indicating that the connection is established in both MATLAB command windows and another simulation monitor of GUI for UCB site will be displayed in the MATLAB for API Main as shown in Figure 61 a If the UI SIMCOR run in the other computer or the other site 1 e UIUC then this GUI helps UCB site seeing current status of experiment Click Stiffness Evaluation button in UI SIMCOR UI SIMCOR will either read stiffness of specimen from file or run stiffness test depending on the parameter in configuration file When the stiffness is manually provided the stiffness matrix each module should be provided in the files located in the O0 Coordinator folder In this test the predefined stiffness matrix is used Click Apply Static Loading button In this stage the gravity force is applied There is no plan to apply gravity as this exemplary test is to demonstrate the efficacy of hybrid simulation framework Click Start PSD Test to run experiment The target displacement and feedback displacement and force will be displayed in the monitoring window and UCB monitoring window Click Disconnect Modules after finishing hybrid test to disconnect UI SIMCOR from xP
162. perimental stage i e initial stiffness formulation static and dynamic stages consists of similar functions such as evaluating target displacements sending target displacements to each module executing the commands checking relaxation and receiving the measured responses Figure 2 describes data sequence of data flow of a pseudo dynamic simulation in UI SIMCOR User Interface Hybrid Simulation Framework UI SIMCOR Experiment or Analysis L c Remote Site N Module N famera os Module 2 Static Krypton Remote Site 2 MOST Module 1 Transient DAQ Hardware other GUI or User ibd SupElement Analysis HdwControl Bemote Site 1 2 Initialization 1 Start simulation i 1 i Server Server i i i standby standby 2 1 Add modules i i i i to System 2 1 i 1 i 2 2 Add analysis 1 objects to System i i H 1 Monitor 2 3 Add hardware contro T i i experiment objects to System H H 8 Open connection 1 1 1 1 foreach m
163. putational routine from the gateway user can set a pointer of type mxArray to the data it returns MATLAB is then able to recognize the output from your computational routine as the output from the MEX file The following C C MEX cycle figure shows how inputs enter a MEX file what functions the gateway routine performs and how outputs return to MATLAB 22 o UI SIMC OR MATLAB INPUTS const mxArray B A call to B prhs 1 MEX file func C D 2func A B const mxArray A A tells MATLAB to ee pass variables A and B to your MEX file Tanasa C and D are left i l unassigned void mexFunction int nlhs mxArray plhs int nrhs const mxArray prhs In the gateway routine Use the mxCreate functions to create the MATLAB arrays for your output arguments Set plhs 0 1 tothe pointers to the newly created MATLAB arrays Use the mxGet functions to extract your data from prhs 0 1 Call your C subroutine passing the input and output data pointers as function parameters MATLAB On return from MEX file func C D func A B mxArray D D plhs 1 plhs 0 is assigned to C and plhs 1 is mxArray C assigned to D C plhs 0 OUTPUTS Figure 10 C C MEX cycle In this figure a call to the MEX file named func of the form c D func a B tells MATLAB to pass variables a and s to user s MEX file c and n are left u
164. r MOST LBCB example oe oo Unit mm N sec oe oe by Oh Sung Kwon okwon2Quiuc edu modified by Kyu Sik Park kspark uiuc edu Univ of Illinois at Urbana Champaign oe oe oe oe Last updated on 2007 01 27 11 46AM oe oe oe oe Common parameters oo 5 Ground acceleration file name with extension The file should contains two 5 columns for time and acceleration The unit of acceleration should be consistent with the mass time and force i e mass acc force Sys GM Input acc475C dat Ground acceleration scale factor This factor will be multiplied to acceleration before starting simulation Sys GM SC 9 81 Direction of ground acceleration x y or z Sys GM direction x Integration parameter related to the alpha OS method Alpha 0 1 3 In most cases SC Alph 0 05 worked Sys Alph 0 05 Sys Beta 1 4 1 Sys Alph 2 Sys Gamm 1 2 Sys Alph oe Evaluate Stiffness Yes 1 to run stiffness evaluation test O 0 to read stiffness matrix from file In this case there should exist stiffness matrices of individual module in the files MDLO1_K txt MDLO2_K txt etc ys Eval_Stiffness 0 oe oe oe oe n oe Number of initial static loading steps When ther xist static constant loading i e gravity forces apply then in Zeus NL or OpenSees as a incremental loading with n steps In this file SimConfig m specify the numb
165. ration should be consistent with the mass time and force i e mass acc force Sys GM Input acc475C dat oe oe oe Ground acceleration scale factor This factor will be multiplied to acceleration before starting simulation Sys GM SC 32 2 Direction of ground acceleration x y or zZ Sys GM direction x Integration parameter related to the alpha OS method Alpha 0 1 3 In most cases SC Alph 0 05 worked Sys Alph 0 05 Sys Beta 1 4 1 Sys Alph 2 Sys Gamm 1 2 Sys Alph oe Evaluate Stiffness Yes 1 to run stiffness evaluation test Oo 0 to read stiffness matrix from file In this case there should exist stiffness matrices of individual module in the files MDLO1_K txt MDL02 K txt etc ys Eval Stiffness 1 oe oe oe oe LD oo Number of initial static loading steps When ther xist static constant loading i e gravity forces apply then in Zeus NL or OpenSees as a incremental loading with n steps In this file SimConfig m specify the number of static steps in the following variable Sys Num Static Step 0 oe oe oo o Number of dynamic analysis steps Sys Num Dynamic Step 500 o Dynamic analysis time steps Sys dt 0 01 Rayleigh damping xi_l and xi_2 Damping ratio Tn_1 Tn_2 Target period 85 o UI SIMC OR Sys xi 1 0 03 Sys Tn 1 1 064 Sys xi 2 0 03 Sys Tn 2 0 338 oe Number of Stiffnes
166. re Installation of UI SIMCOR Before install the UI SIMCOR backup C SIMCOR folder and remove previous version of UI SIMCOR Start Set Program Access and Defaults Change or Remove Programs then select UI SIMCOR and Remove and delete C SIMCOR folder By double clicking the distributed installation file SimCor msi you can install UI SIMCOR NEES SAM and example files When install software DO NOT change the default folder location The software should be installed in C SIMCOR After install the UI SIMCOR copy separately distributed files 05 API UCB and 06 Examples UCB to CASIMCORY folder These folders include the API and example files for UCB site Finally add the following paths in MATLAB To do that open MATLAB and click File Set Path Following folders should be in the path of MATLAB Paths for UI SIMCOR C SIMCOR 01_SIMCOR C ASIMCOR 01_SIMCOR parmatlab Paths for UCB site This is necessary only for UCB site Scripts in the folder 04 API 01_UCB are written to operate xPC Real time trarget in UCB site C A SIMCOR 04_API 01_UCB Paths for FEDEAS Lab This is necessary only for when FEDEAS Lab is used to model part of a structure CASIMCOR 02 NEES FL C SIMCOR 02_NEES FL FedeasLab C SIMCOR 02_NEES FL FedeasLab Element_Lib C SIMCOR 02_NEES FL FedeasLab General C SIMCOR 02_NEES FL FedeasLab Geometry C A SIMCOR 02_NEES FL FedeasLab Material_Lib 164 o UI SIMC OR C SIMCOR 02_NEES FL FedeasLab Output C ASIMCOR 0
167. re Nose 01 Comp 1 B M lu per Node 01 Como Z va urec no OF Come T 74 38 45 44 23 0 oz 04 06 99 1 Me ssge Figure A 15 Monitoring window Run another MATLAB for communication with xPC Real time target and change the directory to C SIMCOR 04_API 01_UCB Type API Main you will see the message box confirming the displacement limit as shown in Figure A 16 173 o UI SIMC OR Check Displacement Limit ia xl e Current displacement limit for UCB speicmen 6 00e 000 If you want modify the displacement limit please open API_Main m i Okay Reset Figure A 16 Warning message a Ifthe displacement limit is confirmed click Okay button then you will see the Waiting for connection from client through port 11999 in a MATLAB command window b If you want modify the displacement limit click Reset button then the API_Main file is opened automatically Modify the Disp_Limit variable at line 17 Now ready to run experiment real experiment 5 Click Establish Connection button in UI SIMCOR control window You will see some messages indicating that the connection is established in both MALTAB command windows and another simulation monitor of GUI for UCB site will be displayed in the MATLAB for API_Main as shown in Figure A 17 gt UCBMon 0905 UCB Simulation Monitor LUCE 1 Status Merzoge Figure A 17 Monitoring window for UCB site a If the UI SIMCOR run in the
168. ring this stage UI SIMCOR performs inelastic static analysis not pseudo dynamic test using a OS scheme Right after the initial loading stage ground acceleration is applied The inertial forces are applied from UI SIMCOR The static analysis module is controlled by displacement In the previous example we were interested in x directional displacement only In this example however we should include y directional displacement since the initial loading will shorten the column Thus in the configuration file the effective DOFs are enabled for x and y direction From the basics of structural analysis we cannot control a DOF both for displacement and force 61 o UI SIMC OR Since the control point will be displacement controlled in x and y direction the initial loading need to be applied somewhere else To resolve this problem a very short frame element is attached on the control point and the initial loading is applied on that element Figure 32 12 1 0 0 8 0 6 0 4 0 2 Load Factor for Initial Loading 0 0 0 5 10 15 20 25 30 Initial Loading Stage Dynamic Loading Stage Step for Initial Stiffness Formulation Figure 31 Time history of initial load factor 000 N 200 000 N x Loading E i J DON Dummy Element y on e Control Point Figure 32 Application of initial a 62 o UI SIMC OR 8 2 2 Simulation configuration file The simulation configuration file for this model is identical to previous ex
169. rrent displacement at each DOF is larger than the displacement defined in MDL i CAP D tot UI SIMCOR generates error message and cancel the proposed step MDL CAP F tot Absolute force constraint This variable does same role with MDL i CAP D tot except that it check for force MDL i TOL D inc Displacement tolerance criteria In the experimental module the hydraulic system cannot exactly impose the proposed target 44 o UI SIMC OR displacement There can be difference between proposed and actually imposed displacement If this displacement difference ratio is larger than MDL i TOL D inc UI SIMCOR cancels the current step MDL i LBCB This variable defines the option of Loading and Boundary Condition Box LBCB UI SIMCOR only calculates the target displacements of effective DOFs However experimental equipment can have difference number of actuators with effective DOFs The LBCB at UIUC has 6 actuators so 6 displacements should be sent to LBCB for experiment If YES 1 the target displacements of all DOFs will be assigned as zero except effective DOFs in UI SIMCOR If MDL i LBCB 1 MDL i LBCB TransM also needs to be defined MDL i LBCB_TransM This variable defines the transformation matrix for LBCB The coordinate system of LBCB is dependent on the location of plate MDL_AUX Class AUX i URL AUX i protocol and AUX i name Same as MDL i URL MDL protocol and MDL i name for each auxiliary module AUX i
170. s number of control nodes 6 array Displacement increment limit not ratio a amp b cde f LBCB case If it s 1 the oe VEZ T2004 0 1 12000 5 labviewl Camera Module ID of this mdoule is 1 Auxiliary module configuration AUX 1 MDL AUX AUX 1 URL AUX 1 protocol AUX 1 name AUX 1 Command Variable Description displacement 2 3500 39 o UI SIMC OR Three classes are used in the simulation configuration file i e Sys MDL RF and MDL AUX Sys Class A structure that includes all data about PSD simulation MDL RF Class A class for restoring force module All remote experimental specimens will communicate with objects of this class MDL AUX Class A class for sending signals to auxiliary hardware such as DAQ and camera This class also has similar member functions with MDL RF The detailed descriptions of variables in each class are as follows Sys Class Sys GM Input This variable defines the ground acceleration file name with extension The file should contain two columns for time and acceleration The unit of acceleration should be consistent with the mass time and force i e massxacc force Sys GM SC This variable defines the ground acceleration scale factor This factor will be multiplied to acceleration before starting simulation Sys GM direction This variable defines the direction of ground acceleration x y or z Sys Alph Sys
171. s 1 oe oe oe oe Enable GUI for SimCor Yes 1 nable the GUI for SimCor o 0 disable the GUI for SimCor Hybrid simulation will be run automatically Not recommended for the experiment ys EnableGUI 1 Use GUI for SimCor oe oe oe oe n o Number of restoring force modules Sys Num RF Module 23 5 Number of auxilary modules Sys Num AUX Module 0 Total number of effective nodes Effective nodes are interface nodes between modules and nodes where lumped masses are defined Sys Num Node 3 Lumped mass assigned for each DOF for each node Node number x y Z rx ry rz directional mass Sys Node_Mass 1 2 54628081981000 0 0 0 O 0 Sys Node_Mass 2 5 49705587697000 0 0 0 0 0 Sys Node Mass 3 2 54628081981000 0 0 0 0 0 oe oe oe Restoring force module configuration oe oe Create objects of MDL RF MDL 1 MDL RF MDL 2 MDL RF MDL 3 MDL RF Name of each module MDL 1 name LeftCol Module ID of this module is 1 MDL 2 name Middle Module ID of this module is 2 MDL 3 name RightCol Module ID of this module is 3 URL of each module MDL 1 URL 127 0 0 1 11997 MDL 2 URL 127 0 0 1 11998 73 o UI SIMC OR MDL 3 URL 127 0 0 1 11999 oe Communication protocol for each module NTCP communicate through NEESPOP server TOPTEP
172. s ThreeSite 00_Coordinator Click Apply Static Loading button in UI SIMCOR control window In this stage the gravity force is applied however there is no plan to apply gravity as this exemplary test is to demonstrate the efficacy of hybrid simulation framework Click Start PSD Test button in UI SIMCOR control window to conduct the experiment The target displacement and feedback displacement and force will be displayed in the monitoring windows in UIUC site For UCB site the target displacement and measured displacement and force in remote site not HSF will be displayed in GUI 18 19 190 o UI SIMC OR 20 Click Disconnect Modules button in UI SIMCOR control window after finishing hybrid simulation to disconnect UI SIMCOR from each site You will see some messages indicating the communication is disconnected in a MALTAB command window and API of each site 2 Backup the results files and send to UIUC a ForUIUC site e MDLOI recv txt MDLO2 recv txt MDL 03 recv txt MDLO04 recv txt Netwklog txt and NodeDisp txt located in CASIMCORY03 ExamplesYThreeSiteX00 Coordinator Cur Disp txt Cur Forc txt and Netlog txt located in C SIMCOR 03_ Examples ThreeSite 02_Module2 b For UCB site NetwkLog txt UCB Results txt located in C SIMCOR 04_API 01_UCB 191
173. s test If stiffness is evaluated through experiment th valuation need to be don several times and the average of the results are used as the initial stiffness This parameter is used when Sys Eval Stiffness 1 Sys Num Test Stiffness 1 oe oe oe oe Enable GUI for SimCor Yes 1 nable the GUI for SimCor o 0 disable the GUI for SimCor Hybrid simulation will be run automatically Not recommended for the experiment ys EnableGUI 1 Use GUI for SimCor oe oe oe oe n Number of restoring force modules Sys Num RF Module 3 Number of auxilary modules Sys Num_AUX_Module 0 Total number of effective nodes Effective nodes are interface nodes between modules and nodes where lumped masses are defined Sys Num Node 14 Lumped mass assigned for each DOF for each node Node number x y Z rx ry rz directional mass Sys Node Mass 1 8 875 0 0 0 0 OJ Sys Node Mass 2 8 875 0 0 0 0 0 Sys Node Mass 3 8 875 0 0 0 O OJ Sys Node Mass 4 8 875 0 0 O O OJ Sys Node Mass 5 8 1875 0 0 0 0 0 Sys Node Mass 6 8 1875 0 0 0 0 0 Sys Node Mass 7 8 1875 0 0 0 0 0 Sys Node Mass 8 8 1875 0 0 0 O OJ Sys Node_Mass 9 0 0 0 0 0 0 Sys Node Mass 10 8 1875 0 0 0 0 0 Sys Node Mass 11 8 1875 0 0 0 0 0 Sys Node_Mass 12 8 1875 0 0 0 0 OJ Sys Node_Mass 13
174. s used oe oe URL of remote site and NHCP mode for NHCP for i 1 Sys Num RF Module if strcmp lower MDL i protocol nhcp MDL i remote URL 127 0 0 1 99999 MDL i NHCPMode simld end end Stiffness for NHCP Only valid if NHCPMode SimlD for i 1 Sys Num RF Module if strcmp lower MDL i NHCPMode simld MDL i NHCPSimK 1000 end end oe Coordinate transformation If it needs the transformation matrix also needs to be provided for i 1 Sys Num RF Module MDL i TransM end oe Scale factor for displacement rotation force moment Experimental specimens are not always in full scale Use this factors to apply scale factors The displacement scale factors are multiplied before they ar sent to module Measured force and moments are divided with scale factors before used in the PSD algorithm for i 1 Sys Num RF Module MDL i ScaleF 1 1 1 1 Module i oe oe oe oe oe end oe Relaxation check If this parameter is 1 UI SimCor send commend to retrieve data and check relaxation just before the execution of proposed command If it s 1 the checking criteria needs to be provided for i 1 Sys Num RF Module MDL i CheckRelax 0 Module i if MDL i CheckLimit 1 define following variables Variable size should be number of control nodes 6 array oe oe oe oe oe o Displacement variation ratio not increment MDL i MES D inco abcde f
175. sLab Geometry C SIMCOR 02_NEES FL FedeasLab Material_ Lib C SIMCOR 02_NEES FL FedeasLab Output C SIMCOR 02_NEES FL FedeasLab Section_Lib C SIMCOR 02_NEES FL FedeasLab Solution_Lib C A SIMCOR 02_NEES FL FedeasLab Utilities Path for NEES ABAQUS C SIMCOR 02_NEES ABAQUS Path for API of UCB If uNEES at UCB is used for experiment C SIMCOR 04_API 01_UCB For simple procedure 1 Runa MATLAB 2 Click File Set Path Add with Subfolders in the menu bar and select C A SIMCOR 01_SIMCOR and click the OK button 3 Repeat step 2 for C SIMCOR O2_NEES FL C SIMCOR 02_NEES ABAQUS and C SIMCOR 04_API 01_UCB 4 Click Save button to save the paths and click Close button 7 3 Installation of Structural Analysis Software The modified analysis softwares for the multi site simulation are included in the distribution Also FEDEAS Lab is included in the distribution To edit and verify structural model for ZEUS NL and OpenSees each software should be installed separately The latest version of ZEUS NL can be downloaded from the following link By double clicking installation file you can install the latest version of ZEUS NL If you encounter difficulties in modeling or running ZEUS NL contact Oh Sung Kwon okwon2 a uiuc edu or Kyu Sik Park kspark uiuc edu http mae ce uiuc edu OpenSees can be downloaded from the following link http opensees berkeley edu 30 o UI SIMC OR In the above link there is detailed information about
176. sing NHCP Security There are several things to consider here l 2 W Message integrity was the data corrupted either accidentally or deliberately Authenticity are you talking to the process you expected or the man in the middle Authorization is user J allowed to control hardware X Encryption is you have proprietary e g commercial data you might want to prevent against interception For solution NEESit proposes using a tiered solution 1 2 OpenSSL for communications a Plain socket mode default fastest simple b Message digests plaintext this addresses 1 c Encrypted with digests External username password mechanism TBD There are many possibilities here kerberos PAM JAAS Gridauth Globus etc ete The use of OpenSSL does complicate the development and compilation a bit but there are many offsetting factors It s a well established standard cross platform stable and can be accelerated with crypto cards if desired If user employ server and client certificates they can ensure secure communications with minimal overhead both runtime and administrative 21 o UI SIMC OR 6 5 3 MEX file MEX stands for MATLAB Executable MEX files are dynamically linked subroutines produced from C C or Fortran source code that when compiled can be run from within MATLAB in the same way as MATLAB M files or built in functions The external interface functions of MATLAB manual provide functional
177. stance human readable label name That way the Observer can log commands in an intelligible format e g Coordinator 1 tells M1 to move to 0 01m The server handles addressing All plugin instances get a unique numeric ID as do all coordinators These are returned to the coordinator after loading a plugin so that the coordinator can send messages to specific places For example user can send different displacements to the two mini mosts so that they both do the move correctly and when they generate reaction force messages the coordinator will know their origin The server actually has a simpler job now It manages plugin loading loads the security plugin first and then mainly spends its time checking messages queues and moving them around as required All the heavy lifting is done in plugins running in their own thread The plugin authors have a simpler job There are less than five callbacks to implement and then they can focus on handling and generating messages They can use whatever method of communication is easiest including our new OpenSSL wrapper classes if they like Looking at a more detailed picture let s add the data turbine and authorization authentication bits 20 o UI SIMC OR Simulation Coordinator i NEES comms library NEES comms NEES comms library plugin gaz NEES comms library Data turbine RDV RDV RDV Figure 9 Detailed functional view of hybrid simulation u
178. t data In UI SimCor the specific ASCII based communication is referred as LabVIEW or LabVIEW2 But it does not necessarily involve LabVIEW software These protocols are a simple ASCII format designed for easy parsing and communications occur over a single TCP connection This section is based on the document about LabVIEW NTCP plugin which can be found in NEESit website http it nees org document pdf TR 2004 58 pdf 6 3 1 Background and introduction Any system wishing to use this protocol must be capable of having a process open and listen on a TCP port i e have threading or some equivalent form of multitasking and interprocess communications All messages are tab delimited ASCII on a single line variable length Newline is the message delimiter and should be human readable The protocol should as much as possible simply act as a serialization layer with all the work being done on the client side This means less assumption about client functionality and higher implementation flexibility As a side benefit simper protocol code reduces complexity and the number of bugs Any language that can host a TCP connection and parse string can be a full fledged NTCP controller This opens several doors to lightweight control of small devices such as pan tilt zoom camera bases Transaction ID is included in every message so that the client side can be asynchronous and multithreaded This is how the LabVIEW control code was implemented LabVIE
179. tarting simulation Sys GM SC 9 81 Direction of ground acceleration x y or zZ Sys GM direction x Integration parameter related to the alpha OS method Alpha 0 1 3 In most cases SC Alph 0 05 worked Sys Alph 0 05 Sys Beta 1 4 1 Sys Alph 2 Sys Gamm 1 2 Sys Alph oe Evaluate Stiffness Yes 1 to run stiffness evaluation test O 0 to read stiffness matrix from file In this case there should exist stiffness matrices of individual module in the files MDLO1_K txt MDL02 K txt etc ys Eval_Stiffness 1 oe oe oe oe LD oo Number of initial static loading steps When there exist static constant loading i e gravity forces apply then in Zeus NL or OpenSees as a incremental loading with n steps In this file SimConfig m specify the oe oe 74 o UI SIMC OR o number of static steps in the following variable Sys Num Static Step 0 Number of dynamic analysis steps Sys Num Dynamic Step 500 Dynamic analysis time steps Sys dt 0 01 Rayleigh damping xi 1 and xi 2 Damping ratio Tn 1 Tn 2 Target period Sys xi 1 0 00 Sys Tn 1 0 00 Sys xi 2 0 00 Sys Tn 2 0 00 oe Number of Stiffness test If stiffness is evaluated through experiment th valuation need to be don several times and the average of the results are used as the initial stiffness This parameter is used when Sys Eval Stiffness 1 Sys Num Test Stiffnes
180. ted from NCS and SimServer or computer for MiniMOST 1 or 2 respectively during close action open Syntax NHCP NHCPMode open IP NCS PORT NCS IP SERVER PORT SERVER STIFFNESS SIM nd where NHcPMode is one of simip for simulation case with linear 1 DOF system m MM2 for MiniMOST 1 and 2 respectively open command for open action and 1P NCs PORT NCS IP SERVER PORT SERVER are IP address and port number for NCS and simulation server for simulation case with linear 1 DOF system computer for MiniMOST 1 and computer for Mini MOST 2 respectively Finally srirrwEss sTM is the stiffness of simulation case and required only when NHCP is used for simulation case i e NHCPMode iS Sim1D It does not include units and units are depends on the simulation coordinator i e UI SIMCOR All of input variable for NHCP should be bracketed by For open action there is no return value from NHCP protocol Instead if there are errors during connection to NCS and SimServer or MiniMOST 1 and 2 there are errors messages in the NCS command window 25 run close o UI SIMC OR For example NHCP Siml1D open 127 0 0 1 11997 127 0 0 1 11998 1000 NHCP MM1 open 127 0 0 1 11997 127 0 0 1 11998 Syntax For simip and mm modes M FORC NHCP NHCPMode run T DISP For mm mode M DISP M ROT M
181. the pseudo dynamic stage in UI SIMCOR and the two site test was successfully accomplished 136 Displacement Force A Y Low AMA 0 LA Ys VA 1 f J J f A i 7 T TN n 0 6 1 0 8 r o UI SIMC R 1 0 8 r old coupon sim 0 6 r new coupon exp 04 r N f 0 0 5 1 1 5 2 2 5 3 3 5 4 4 5 Time sec old coupon sim new coupon exp Displacement Figure 71 Experiment results 137 o UI SIMC OR 9 3 Two site Experiment between UIUC and SDSC Two site hybrid test between UIUC and SDSC was conducted at September 12 2006 and these experiments were very successful The structural configuration was similar to the MOST example with the only difference that the left column was assumed pinned at the top to be able to conduct the test using two MiniMOST 1 systems The modules 1 and 3 in MOST example are replaced with MiniMOST 1 at SDSC and UIUC respectively and module 2 is simulated at UIUC using ZEUS NL This test was conducted as a preliminary test for a three site experiment among UIUC UCB and SDSC so further explanation including running procedure is not given in this section Instead all of related explanation will be given in the next section Figure 72 shows the comparison between experiment and simulation 0 10 4 WholeModel Zeus 0 08 4 PSD ZZz Small scale P
182. thm This scale factor can also be used to adjust the target displacement within the limit of experiment specimen MDL i CheckRelax If the module is going to be experimented the CheckRelax variable should be enabled Then UI SIMCOR checks relaxation of displacement and force before execution of proposed displacement in each time step If MDL i CheckRelax 1 following variables also need to be defined MDL i MES D inc Displacement variation ratio not increment MDL i MES F inc Force variation ratio not increment MDL CheckLimit If the module is going to be experimented the CheckLimit variable should be enabled Then UI SIMCOR checks the target displacement target displacement increment measured force and the difference between target and measured displacement If MDL i CheckLimit 1 following variables also need to be defined All of following values are defined in the hybrid simulation framework not in the remote site MDL i TGT D inc Target displacement increment constraint Target displacement increment is defined as below Disp incre abs TGTD i Disp TGTD_O i Disp If the target displacement increment is larger than MDL i TGT D incvalues UI SIMCOR generates an error message and cancel the proposed step MDL i CAP D tot Absolute displacement constraint Theses values should be given for each control points and for each DOFs If the DOFs are not used in the UI SIMCOR you can input any number except zero If any cu
183. tion coordinator In addition it also has a functionality to simulate experimental o UI SIMC OR hardware so that the PSD simulation can be verified without experimental setup The OpenFresco 2 0 was distributed with Matlab OpenFresco examples and OpenSees OpenFresco examples As UI SimCor is written in Matlab the Matlab OpenFresco example was utilized so that OpenFresco can work with UI SimCor OpenFresco uses TCP IP connection to communicate with simulation coordinator written in any platform such as Matlab Ul SimCor or OpenSees The communication protocol between MATLAB and OpenFresco is provided by Schellenberg et al 2006 in MEX file format TCPSocket mexw32 This file is combined with UI SIMCOR for hybrid simulation Currently only 1 DOF simulation case is supported in UI SimCor The protocol is referred as OpenFrescolD protocol in UI SimCor The original version of OpenFresco provides two and three actuator cases and also generic element for generic FEM application This extension will be combined with the next version of UI SIMCOR NHCP Simulation Server The NEES Hybrid Simulation Protocol NHCP is at the early stage of development As a pilot example NEESit provided a simple example for which OpenSSL is used to encrypt communication data and data for very simple system 1 DOF can be transferred Currently NHCP can be used for numerical simulation and MiniMOST 1 and 2 in UIUC or SDSC For numerical simulation only linear 1
184. tion file MDL i URL Since there are three modules three URL for communication are defined in the configuration file MDL Z protocol Since there are three modules three protocols for each module are defined in the configuration file MDL Z node This variable defines the control points that are connected to a certain module In Figure 22 the left column is connected to control point 1 only Thus in the configuration file MDL 1 node 1 The second module where beams and middle column are defined is connected to three control points Thus in the configuration file MDL 2 node 1 2 3 Third module has only one control point 3 MDL i EFF DOF Effective DOFs also need to be defined for each control point for each module In module 1 x translational and rotational DOFs are effective as shown in Figure 23 Thus Effective DOFs are 1 00 00 1 MDL DEL t and MDL i DEL Y There are three control points in this example so displacement level for the test for initial stiffness formulation should defined for each module 8 4 3 Static analysis module configuration The configuration files for static analysis module are similar to those for SDOF system Thus further explanation is not given in this section In the folder CA SIMCOR 03_Examples MOST there are sub folders containing the structural model file and module configuration files for FEDEAS Lab ZEUS NL OpenSees and ABAQUS There is one thing that needs attention Whe
185. trol computer and run a NEES SAM API for NEES analytical model by double clicking NEESSAM LabView2 exe which is located in C SIMCOR 03_Examples ThreeSite 03_UCB a You will see the similar figure as shown in Figure B 2 details in figure are dependent on the model b Check the status of NEES SAM module name and port number 181 o UI SIMC GOR SDSC Site 4 Logon to the MiniMOST 1 computer and run a NEES SAM API for Mini MOST 1 analytical model by double clicking NEESSAM LabViewl exe which is located in C SIMCOR 03_Examples ThreeSite 04_SDSC a You will see the similar figure as shown in Figure B 2 details in figure are dependent on the model b Check the status of NEES SAM module name and port number Prepare the UI SIMCOR for hybrid test UIUC Site 5 Runa MATLAB in the Simulation Coordinator computer for UI SIMCOR and change the directory to C SIMCOR 03_ Examples ThreeSite 00_Coordinator 6 Type UI SimCor and enter in MATLAB command window to run UI SIMCOR New simulation E xj es Please backup previous simulation results if necessary All previous files will be deleted cn Figure B 3 Warning message a Click Okay button then you will see the initialize message in a MATLAB command window All previous files will be deleted so if necessary backup previous results b Five popup windows will be displayed One window for control and other windows for monitoring of each module will popup c Check the mo
186. uation button in UI SIMCOR UI SIMCOR will either read stiffness of specimen from file or run stiffness test depending on the parameter in configuration file When the stiffness is manually provided the stiffness matrix each module should be provided in the files located in the 00 Coordinator folder In this test the predefined stiffness matrix is used Click Apply Static Loading button In this stage the gravity force is applied There is no plan to apply gravity as this exemplary test is to demonstrate the efficacy of hybrid simulation framework Click Start PSD Test to run experiment The target displacement and feedback displacement and force will be displayed in the monitoring window and UCB monitoring window Click Disconnect Modules after finishing hybrid simulation to disconnect UI SIMCOR from xPC Real time target You will see some messages indicating that the communication is disconnected in both MATLAB windows and in two APIs for simulation parts Following output files will be saved in the following folders CASIMCORY03 Examples UCBWMOST 00 Coordinator Global K txt Stiffness matrix from initial stiffness evaluation MDLOI K txt Stiffness matrix of module 1 MDLO2 K txt Stiffness matrix of module 2 MDLO3 K txt Stiffness matrix of module 3 MDLOL recv txt Measured displacement and force of module 1 MDLO2 recv txt Measured displacement and force of module 2 MDLOS3 recv txt Measured displacement and force of module 3 NetwkLog txt Comm
187. ucture and take summation of reaction forces from each segment of structure as shown in Figure 3 d ead ch Ah a Portal frame with 48 DOFs b Portal frame with 9 DOFs A A T TT d Measurement of stiffness from segmented structure Figure 3 Concept of static condensation and sub structuring c Measurement of stiffness o UI SIMC OR 4 2 Effective DOFs for Dynamic Analysis A dynamic analysis can be performed for a structure using reduced DOFs The stiffness matrix of the structure with reduced DOFs can be established by applying known amount of displacement to the whole structure or segmented part of the structure Following simple matrix manipulation shows that certain DOFs can be condensed out from the equation of motion through static condensation As shown in the following derivation remaining DOFs or effective DOFs are the DOFs where lumped masses are defined or the DOFs of our interest This is very important concept for the preparation of input file for PSD testing using UI SIMCOR DOF designation i DOFs where mass is defined j interface DOFs that are of our interest k internal DOFs where mass is not defined and we are not interested in For the PSD test we want to know displacement at i and j DOFs to impose displacement corresponding to inertial forces to experimental or static module DOFs k can be condensed out since we are not interested in them In the equations below alphabet letters M 0 u F
188. unication log file NodeDisp txt Nodal displacements C ASIMCOR 04_API 01_UCB NetwkLog txt Communication log file UCB Results txt Target displacement measured displacement and force at UCB site 129 o UI SIMC OR Two tests with different peak ground accelerations of input earthquake were conducted and the results are shown in Figures 66 and 67 0 008 Experiment 0 006 Simulation 0 004 0 002 0 002 Displacement eo 0 004 0 006 0 008 Time a Horizontal displacement at CP3 in the HSF Experiment Simulation Force 0 008 0 006 004 Jo 9 0 004 0 006 0 008 Displacement b Force displacement relationship at CP3 HSF Figure 66 Pseudo dynamic test results MOST test 1 130 o UI SIMC OR 0 015 Experiment 0 01 L Simulation x 0 005 o 5 S 0 Q 2 a 0 005 0 01 0 015 Time a Horizontal displacement at CP3 in the HSF Experiment Simulation Force Displacement b Force displacement relationship at CP3 in the HSF Figure 67 Pseudo dynamic test results MOST test 2 The test results are better than those of SDOF example because most part of the structures is analytical model except one component of horizontal DOF 9 1 6 Summary The two site hybrid test between UIUC and UCB was conducted to verify and demonstrate the effic
189. up windows will be displayed one window for control and other 165 o UI SIMC OR window for monitoring will popup In this example only one monitoring window will be displayed as this example has only one remote site UI SimCor Ul SimCor version Z ff Stop by step control 7 A steps by one click Application for distributed paeudo dynamic nl UI SimCor combines Zeus NL OpenSees FEDEASLab ABAQUS and experimetal sites via TCPIP network NE T pap Node 01 Comp 1 B X laen KA T Dep nese 0t comp 1 vt urere Meke 01 Comp T lure c 01 comm T 2 V2 M Fore Node 0i Come 1 a 4 08 026 04 02 0 02 04 05 os U Figure A 3 Monitoring window Run another MATLAB for communication with xPC Real time target and change the directory to C SIMCOR 04_API 01_ UCB Type API Main sim then you will see the message box confirming the displacement limit as shown in Figure A 4 Check Displacement Limit Current displacement limit for UCB speicmen 6 00e 000 If you want modify the displacement limit please open API_Main_sim m Figure A 4 Warning message for the displacement limit a Ifthe displacement limit is confirmed click Okay button then you will see the Waiting for connection from client through port 11999 in a MATLAB command window The IP address and port number are dependent on experimental configuration 166 o UI SIMC OR b If you want modify the displacement limit
190. ure 65 Pseudo dynamic test results SDOF test 2 There are some differences between experiment and simulation because of the initial value in displacement and force during the experiment 127 o UI SIMC R 9 1 5 MOST example MOST example used in section 8 4 is used in this section The structural configuration is except scale factors also same as section 8 4 The modules 1 and 2 were simulated using the ZEUS NL and module 3 was replaced by the specimen at UCB Running procedure The running procedure of the MOST example is very similar to that of the SDOF system except there are two addition simulation parts 1 Go to C SIMCOR 03_Examples UCB MOST 01_Left_ZEUS folder and double click the NEESSAM LabVIEW to run API for simulation part module 1 You will see the API similar with Figure 23 Similar to step 1 go to C SIMCOR 03_ Examples UCB MOST 02_ Middle ZEUS folder and double click the NEESSAM_LabVIEW2 to run API for another simulation part module 2 You will see the API similar with Figure 23 Run a MATLAB for communication with xPC Real time target and change the directory to C SIMCOR 04_API 01_UCB Type API Main you will see the message box confirming the displacement limit as shown in Figure 59 a Similar to the pushover test the default value of displacement limit of the hybrid simulation is 6 b If you want to modify the displacement limit click Reset button then the API Main file is opened automati
191. ve data and check relaxation just before the execution of proposed command If it s 1 the checking criteria needs to be provided oe oe oe 38 o UI SIMC R for i 1 Sys Num_RF_Module MDL i CheckRelax 0 Module i if MDL i CheckLimit 1 define following variables Variable size should be number of control nodes 6 array 5 Displacement variation ratio not increment MDL i MES D inco abcde f A 1 Force variaiton ratio not increment 5 MDL i MES F inco abcde f l end Check displacement and force limit oe At every steps check if the limitation of the equipments for i 1 Sys Num RF Module MDL i CheckLimit 0 if MDL i CheckLimit 1 Variable size should be oe o oe oe oe oe z DL i TGT D inc oe oe Displacement limit DL i CAP D tot oe oe oe Force limit E DL i CAP F tot oe Displacement toleranc DL i TOL D inc oe z 9 oe end oe oe ge oe DOF coordinate with those for i 1 Sys Num RF Module MDL i LBCB 0 end for i 1 Sys Num RF Module MDL i LBCB TransM end oe abcdef abcdef ratio abcdef Loading and Boundary Condition Box coordinate transformation matrix needs to be provided This can be also used for any other actuator which has diffrence number of of UI SIMCOR displacement or force are approaching to the stroke or force capacity Module i define following variable
192. vers one per site probably but each server will load and run multiple plugins This is worth discussing a bit both for how it came about and also for the consequences of the design First off the server is now more event driven with messages arriving and departing all the time from given displacement This is sent to all plugins starting with the security plugin Each plugin can look at the command source and optional destination and decide if they need to act or not It has an Observer plugin that as pictured generates no messages but logs all to disk In the example the control plugin would take the command move the hardware or simulation compute reaction forces and then generate a new MSG REACTION which the server would send to the coordinator and also all the plugins This enables a number of nice things and makes the plugins interface simpler but requires a bit more of the server to keep it from slowing as the number of communicating processes increases To wit 1 Each plugin runs in its own thread 2 Each message queue to from plugin to from coordinator is a thread safe queue of fixed maximum length If a plugin gets too far behind the server will drop messages rather than wait generating a critical error callback to the plugin upon doing so This keeps a slow plugin from rate limiting the server 3 The server has to decide whether a plugin running or not if the connection to the 19 o UI SIMC OR coordi
193. ween simulation and experiment This difference also affects 154 o UI SIMC GOR the other results at UIUC and SDSC because the example structure consists of four modules as shown in Figure 73 Displacement Force MM1 UIUC 0 008 0 006 Simulation 0 004 Experiment 0 002 0 0 002 0 004 0 006 0 008 0 01 0 1 2 3 4 5 Time MM1 UIUC 20 Simulation 15 F Experiment 0 01 0 008 0 006 0 004 0 002 0 004 0 006 0 008 Displacement Figure 90 MiniMOST 1 at UIUC 155 Displacement Force Simulation Experiment o UI SIMC OR mNEES UCB Time MM1 UIUC Simulation Experiment Displacement Figure 91 uNEES at UCB 156 o UI SIMC OR 0 008 MM1 SDSC 0 008 0 006 Simulation A A 0 004 Experiment A g 0002 X 0 g 0 002 A 0 004 V 0 006 0 008 k V 0 01 l l 0 5 1 1 5 2 2 5 3 5 4 4 5 5 Time MM1 UIUC m ees Simulation 8 Experiment 6 4 2 o g C 0101 0 008 0 006 0 004 Q 0 002 0 004 0 006 4 L 6 L 8 10 42 Displacement Figure 92 MiniMOST 1 at SDSC 9 4 8 Summary The three site hybrid test among UIUC UCB and SDSC was conducted to verify and demonstrate the efficacy of recently enhanced UI SIMCOR The MiniMOST 1 at UIUC and SDSC and uNEES at
194. y conducted at UIUC UCB and SDSC following properties of each site can be obtained 143 o UI SIMC OR Table 4 Properties of each site Site Stiffness Diimit E UIUC 2444X10 0 01 24 44 UCB 1 230 2 0 2 46 SDSC 1 172 lt 10 0 01 11 72 The displacement limit Diii for UIUC UCB and SDSC is assumed as 0 01 2 and 0 01 respectively However the actual displacement limit of each site is lager than theses values For this test the displacement limit of each site is assumed little smaller than actual one for safety issue The MiniMOST 1 at UIUC and SDSC is linear specimen whereas the UNEES at UCB is nonlinear specimen Therefore the equivalent stiffness obtained when the displacement is 1 is used for UCB specimen Khalf dis in Figure 62 However this stiffness may or may not appropriate for calculation of scale factor depend on how much target displacement will be sent to the specimen Finally following scale factors of each site can be obtained using above results Table 5 Scale factor Site Displacement Rotation Force Moment UIUC 0 439 1 0 677 1 UCB 87 719 1 0 068 1 SDSC 0 439 1 0 324 1 9 4 5 Analytical model of each site To predict and check the target displacement which will be sent to each site during the hybrid test analytical model of specimen of each site is developed based on the experimental results by using ZEUS NL Figure 80 shows the co
195. ys Tn 1 Sys xi 2 and Sys Tn 2 Rayleigh damping with critical damping ratio of 3 in the 1 and 2 modes is considered in this example Sys Num Node There are 14 control points in this example as explained in 8 5 1 so Sys Num Node 14 89 o UI SIMC OR Sys Node Mass There are 14 control points thus there are also 14 masses for each node only in x direction However there are no mass in the one of control points of module 1 and control point of module 2 as shown in Figure 42 See the static module configuration file of each module for reference 8 5 3 Static analysis module configuration The configuration files for static analysis module are similar to those for MOST example Thus further explanation is not given in this section In the folder CASIMCOR 03_Examples SAC there are sub folders containing the structural model file and module configuration files for FEDEAS Lab ZEUS NL and ABAQUS OpenSees model is not given for this example There is one restraint in ABAQUS input file It is option for imposing targeted displacement in the model The restraint is that the order of the DOFs in the command boundary should be the same as that of MDL EFF DOF i For example in SAC building From ABAQUS input file Frame inp BOUNDARY 4 1 1 1 0 42 2 1 0 8 1 1 1 0 8 2 2 1 0 12 1 1 1 0 12 2 2 1 0 16 1 1 1 0 16 2 2 1 0 3 1 1 1 0 3 2 2 1 0 7 1 1 1 0 7 2 2 1 0 11 1 1 1 0 11 2 2 1 0 15 1 1 1 0 15 2 2 1 0 2 1 1 1 0 22214

UI-SimCor Manual - No website was found at this address

Contents

Download Pdf Manuals

Related Search

Related Contents