Home

Improving the Accuracy and Usability of Iowa Falling Weight

1. IA Highway Division Iowa Department of Transportation Potter C J and Dirks K L 1989 Pavement Evaluation Using the Road Rater Deflection Dish Final Report for MLR 89 2 Ames IA Highway Division lowa Department of Transportation Ullidtz P 1987 Pavement Analysis The Technical University of Denmark 36
2. are expressed as follows SNeff 0 0045D lt Ep 3 keff SS Kaynamic 2 4 Where D total pavement thickness inch Ep effective modulus of pavement layers psi Kaynamic effective dynamic modulus of pavement layers calculated from deflection basin psi Following the AASHTO 1993 design procedure the pseudo codes and algorithms for implementing the computation of these two parameters SNerr and Kes in I BACK were developed First the deflection basin at the reference temperature is computed using the temperature normalized AC modulus and the ANN forward models Next the computed deflection at the center of the load plate do is related to the effective modulus of pavement layers Ep subgrade resilient modulus Mr and other parameters as shown in the following equation dy 1 5pa _ MR 1 Where do deflection measured at the center of the load plate inch at the reference temperature 68 F p load pressure psi a load plate radius inch pe 5 The unknown Ep value required in the SNeff computation See equation 3 is determined in I BACK by again employing the in built Excel Solver Tool Unlike SNeff computation for AC surfaced pavements the ket computation for rigid pavements and composite pavements is quite straightforward The backcalculated k values Kaynamic are inputted into equation 4 to obtain keff values Computation of Structural Rating SR and Soil Support k value
3. relative stiffness RRS were predicted In the case of composite pavements CPs where an HMA AC surface is overlaid on top of an existing PCC pavement Eac Epcc Ks opcc tensile stress at the bottom of the PCC and sac were predicted The developed methodology was successfully verified using results from long term pavement performance LTPP FWD test results as well as lowa DOT FWD field data All successfully developed ANN models were incorporated into a Microsoft Excel spreadsheet based backcalculation software toolbox with a user friendly interface The phase I study also concluded that the developed nondestructive pavement evaluation methodology for analyzing the FWD deflection data would be adopted by lowa DOT pavement and material engineers and technicians who do not employ any preferable FWD backcalculation analysis technique A follow up Phase II study of Iowa DOT Project Ceylan et al 2009 focused on the development of a fully automated software system for rapid processing of the FWD data The software system can automatically read the FWD raw data collected by the JILS 20 type FWD machine that lowa DOTowns process and analyze the collected data with the algorithms being developed during the phase I study This system smoothly integrates the FWD data analysis algorithms and the computer program being used to collect the pavement deflection data With the implementation of the developed software system the FWD data can be filtered
4. Menu button allows the user to go back to I BACK program main menu for selection of the pavement type The input requirements for conducting conventional flexible pavement analyses are FWD deflection data asphalt concrete thicknesses granular base thickness and FWD load The input requirements for conducting full depth asphalt pavement analyses are same as conventional flexible pavement analyses except not requiring granular base thickness Required input parameters for rigid pavement analysis are deflection data PCC layer thickness granular base thickness degree of bonding and estimated moduli ratio Ebase Ercc and FWD load To simplify the ANN based backcalculation methodology PCC layer and base layer thickness are adjusted into only one thickness value effective PCC thickness in the program 22 Ceylan et al 2007 During analysis the effective PCC thickness is automatically calculated from pavement layer information PCC layer thickness granular base thickness degree of bonding and estimated moduli ratio and used in backcalculation analysis The clicking Equation button in rigid pavement analysis tool menu provides the equations sheet as shown in Figure 19 This equations sheet summarized the equations used for calculation of effective PCC thickness for fully bonded PCC layers unbonded PCC layers and partially bonded PCC layers Effective thickness for fully bonded PCC layers as room por o 2 Ea 2 3 3 7
5. and Lopez A 1996 Statistical analysis of temperature and moisture effects on pavement structural properties based on seasonal monitoring data Transportation Research Record 1540 48 55 Washington DC Transportation Research Board National Research Council Baltzer S and Jansen J M 1994 Temperature correction of asphalt moduli for FWD measurements Proceedings of the 4th International Conference on Bearing Capacity of Roads and Airfields Volume 1 Minneapolis MN Ceylan H Guclu A Bayrak M B and Gopalakrishnan K 2007 Nondestructive Evaluation of Iowa Pavements Phase I Final Report CTRE Project 04 177 Ames IA Center for Transportation Research and Education CTRE Iowa State University Ceylan H Gopalakrishnan K Kim S Guclu A and Bayrak M B 2009 Nondestructive Evaluation of Iowa Pavements Phase II Development of a Fully Automated Software System for Rapid Analysis Processing of the Falling Weight Deflectometer Data Final Report CTRE Project 04 177Phase 2 Ames IA Center for Transportation Research and Education CTRE Iowa State University Chen D Bilyeu J Lin H H and Murphy M 2000 Temperature correction on falling weight deflectometer measurements Transportation Research Record 1716 30 39 Washington DC Transportation Research Board National Research Council Deacon J A Coplantz J S Tayebali A A and Monismith C L 1994 Temperature considerations in asphalt aggregate m
6. computation of Iowa DOT asphalt concrete AC overlay design related Structural Rating SR and k value k and 5 enhancement of user friendliness of input and output from the software tool A high quality easy to use backcalculation software package referred to as I BACK the Iowa Pavement Backcalculation Software was developed to achieve the project goals and requirements This report presents theoretical background behind the incorporated enhancements as well as guidance on the use of I BACK developed in this study The developed tool I BACK provides more fine tuned ANN pavement backcalculation results by implementation of deflection basin matching optimizer for conventional flexible full depth rigid and composite pavements Implementation of this tool within lowa DOT will facilitate accurate pavement structural evaluation and rehabilitation designs for pavement asset management purposes This research has also set the framework for the development of a simplified FWD deflection based HMA overlay design procedure which is one of the recommended areas for future research 17 Key Words 18 Distribution Statement analysis tool asset management FWD HMA pavement PCC No restrictions 19 Security Classification of this 20 Security Classification of this 21 No of Pages 22 Price report page Unclassified Unclassified 48 NA Form DOT F 1700 7 8 72 Reproduction of completed page authorized IMPROVING THE ACCURACY AND USABIL
7. equation 13 of SR Potter and Dirks 1989 for composite pavement expressed below has been incorporated into I BACK computation algorithm Temp Corrected SR Non Temp Corrected SR 70 F Pave Temp x 0 0145 9 Determination of k value for Flexible Rigid and Composite Pavements Figure 11 presents soil support k value relationship charts for flexible pavement and Figure 12 presents soil support k value relationship charts for rigid and composite pavements The base relationship charts for soil support k value used in lowa DOT AC overlay design procedure graphically relate soil support k values with average Road Rater do values and the ratio of Surface Curvature Index SCI values to Road Rater do values SCI do SCI is defined as the difference in mils between Road Rater Sensor No 1 do and Sensor No 2 d12 Similar to the development of SR determination equations described previously the correlation equations for subgrade support k values were developed using synthetic database derived from charts shown in Figure 11 and Figure 12 The developed equations were then incorporated into I BACK computation algorithms The developed correlation equations are expressed as follows k 456 52y 68 295xy 2 554x 421 04y 31 494x 80 628 10 Where k modulus of subgrade reaction for flexible pavement psi inch y SCI Road Rater do x Road Rater do mils k 2967 2y 561 988xy 26 61x 751 01y 71
8. equation is expressed as follows log10y 0 81779 x logy x 0 72032 8 Where y SR at testing temperature x Road Rater do at testing temperature mils Figure 10 compares SR determined from the rigid and composite pavement base relationship chart see Figure 9 and the developed equations using randomly selected data not used in equation development As seen in this figure the developed equation provides good SR estimations for rigid and composite pavements 12 ESTIMATED STRUCTURAL PATI AVERAGE ROAD RATER DEFLECTION Novenper 22 1982 ae E VERSUS Nge S ESTIMATED STRUCTURAL RATING ri 2 3 A 5 6 7 8 9 p 2 3 4 5 6 7 8 9 10 46 7003 EJ STIMATED STRULPURAL RATING Do ea ES TIT THTT KEK 7 Structural Rating Ae AASHO Structural Number Utes Correlation Coefficient R 0 895 o buen duns pep E pf Ti AT A E Aa i 2 3 EE EN 7J a Ty aa 2 3 4 7 7 10 AVERAGE ROAD RATER DEFLECTION HILS Figure 9 Rigid and composite pavement base relationship between SR and Road Rater Do adapted from Potter and Dirks 1989 R 0 9997 SR from equation O e N W FU DN CO WO SR from chart Figure 10 SR determination for rigid and composite pavements chart versus equation The Iowa DOT asphalt concrete overlay design procedure requires temperature correction of SR for composite pavement to 70 F but not for rigid pavement The temperature correction
9. h Bh B Ey n ER FER Effective thickness for unbonded PCC layers as zon ha 5 2 a a Effective thickness for partially bonded PCC layers as AL A o Geen 1 Fe Fy Effective thickness of the fully bonded PCC layers Effective thickness of the unbonded PCC layers Pr Effective thickness of the partially bonded PCC layers E orE Elastic modulus for layer 1or 2 hyorhz Thickness for layer 10r 2 eee Neutral axis distance from top of layer ee degree of bonding which ranges between 0 and 1 k coefficent of subgrade reaction and Epec equations in terms of radius of relative stiffness Harte 1287 R Pace PE noc 4 mh Equations Figure 19 Screen shot of effective PCC thickness equations sheet Required input parameters for composite pavement analysis are deflection data pavement layer information layer thicknesses estimated PCC modulus estimated coefficient of subgrade reaction and FWD load The default units used in the program are based on US customary units FWD deflection data Do till Deo should be entered in mils 10 inches layer thickness in inches and FWD load should be in kips The program will not run correctly 1f these input parameters are not in the desired ranges The user is requested to refer to the report for the appropriate ranges of these parameters Users can enter the FWD deflection database manually or obtain those directly from the JILS 2
10. results are modulus values strains and stresses Modulus and stress values are reported in psi and strains are reported in micro strains x10 The conventional flexible pavement analysis results are Eac modulus of AC Kp base modulus parameter Er subgrade resilient modulus ac tensile strain at the bottom of asphalt layer esc compressive strain at the top of subgrade and op subgrade deviator stress The full depth flexible pavement analysis results are Eac modulus of AC Eri subgrade resilient modulus ac tensile strain at the bottom of asphalt layer esc compressive strain at the top of subgrade and op subgrade deviator stress The rigid pavement analysis results include Epcc modulus of PCC ks coefficient of subgrade reaction opcc tensile stress at the bottom of the PCC layer and radius of relative stiffness RRS The composite pavement analysis results include Eac modulus of AC Epcc modulus of PCC ks coefficient of subgrade reaction ac tensile strain at the bottom of asphalt layer and opcc tensile stress at the bottom of the PCC layer Figure 22 a illustrates the sample analysis results of a conventional flexible pavement Failure to 25 supply all the input parameters will be reflected in the results column of that model The program will automatically write No Data For example if Dag is missing in the input data then output columns will display the error message of No Data At the end of each colum
11. right to see all results 31 Iowa DOT SR Figure 28 Road Rater Based SR and k for flexible pavement After completing computation of Road Rater Based SR and k value the final progress report window appears as shown in Figure 29 Its purpose is to inform the user about the completion of all analysis and to proceed with the generation of a summary result file which is a separate end result file Congratulations All calculations completed successfully Figure 29 Final Progress report window after completion of analysis Summary Result File After the successful completion of full backcalculation analysis both the inputs and outputs are written onto a separate output file in CSV format This eliminates the need to re save the Excel tool with all the macros for the sake of retaining the outputs from the analysis Through this enhancement in I BACK the Excel tool will always serve as a separate analysis tool and the output files less than 100 kbs can be saved for future reference 32 SUMMARY The objective of this study is to incorporate significant enhancements into the fully automated software system for rapid processing of the FWD data with the goal of improving the accuracy and usability of collected Iowa FWD data These enhancements included 1 deflection basin matching optimization 2 temperature normalization of HMA layer modulus 3 computation of 1993 AASHTO design guide based effective SN SNerr and effec
12. sheet ooooooconnccnnnccnoncconccconcnannnnnnnon 23 Figure 20 Screen shot of FWD data extraction through open FWD data file button 24 Figure 21 Filtering the FWD data a Filter options menu b Filtering results no filtered data IY ETAT ARNIS ED A E EEEE O 25 Figure 22 ANN based preliminary backcalculation analysis a Preliminary backcalculation analysis result outputs b Progress report window after completing ANN based preliminary backcalculation al SiS ie 26 Figure 23 Deflection basin matching optimization of backcalculation analysis a Optimized backcalculation analysis result outputs b Trial solution check message box 27 Figure 24 Progress report window after completing optimization of backcalculation analysis a flexible and composite pavements b rigid pavement ooooccnnnccnoconocnnonononnncnononannnona cono ncnnno 28 Figure 25 HMA modulus temperature correction a Temperature corrected HMA modulus result outputs b Progress report window after completing HMA modulus temperature CUIT A A A ia 29 Figure 26 Effective k value Kerr for rigid and composite pavements ooooccnoccccnoccninonacinnnccnanacnnns 30 Figure 27 Effective SN SNerr with effective modulus of pavement layer Ep for flexible PAVEMENE usina ari cisatii n 31 vil Figure 28 Road Rater Based SR and k for flexible pavement Figure 29 Final Progress report window after completion of analysis ooocc
13. use it to compute SR FLEXIBLE PAVEMENT BASE RELATIONSHIP gt eo m a r gt lt e mi XL gt H o 2 Ex H 12 Q 1 RE lt E Y u STRUCTURAL RATING AASHTO STRUCTURAL NUMBER CORRELATION COEFFICIENT R 0 874 8 9 1 0 2 3 4 5 AVERAGE ROAD RATER DEFLECTION MILS AT 80 F Figure 5 Flexible pavement base relationship between SR and Road Rater Do at 80 F adapted from Potter and Dirks 1989 STRUCTURAL RATING FROM ROAD RATER ECTI Corrected to 80 F pte A enone Figure 6 Flexible pavement SR determination nomograph adapted from Potter and Dirks 1989 10 To incorporate flexible pavement SR determination from nomograph and chart into I BACK computation algorithms a synthetic database was developed from chart in Figure 5 and nomograph from Figure 6 By using this synthetic database two correlation equations were developed These equations are Road Rater do temperature correction equation and SR determination equation expressed as follows y 1 3804 0 0046T x 1 2592 0 296 x In T 6 Where y Road Rater do corrected to 80 F mils x Road Rater do at testing temperature mils T Temperature F logioy 0 74171 x log px 0 7831 7 Where y SR at 80 F x Road Rater do corrected to 80 F mils Figure 7 and Figure 8 compare Road Rater Do at 80 F and SR determined from the nomograph see Figure 6 and the developed equations The data no
14. 0 5 1 01715 1 022498 6 69E 01 6 69E 01 6 69E 01 6 6 56E 01 9 21E O1 4 13E 01 4 13E 01 4 13E 01 z 1 118753 2 40E 01 9 72E 01 9 72E 01 9 72E 01 8 1 503013 1 35E 02 8 73E 01 8 73E 01 8 73E 01 9 1 060284 9 88E 01 5 07E 02 5 07E 02 5 07E 02 10 8 58E 01 7 17E 01 3 87E 01 3 87 01 3 87 01 1 2 711796 5 81E 01 7 73E 01 7 73E 01 7 73E01 12 7 708158 60 88257 1 15 00 1 15 00 1 15 00 13 2 646723 28 36716 5 30E 01 5 30E 01 5 30E 01 14 1 50E 01 9 888139 4 42E 02 4 42E 02 4 42E 02 15 1 120319 3 309855 1 95E 01 1 95E 01 1 95 E 01 16 2 319184 3 983927 8 38E 01 8 38E 01 8 38E 01 17 1 77153 8 486038 6 29E 02 6 29E 02 6 29E 02 18 1 810938 16 54343 2 38E 01 2 38 01 2 38E 01 19 1 567331 21 72925 6 65E 01 6 65E 01 6 65E 01 M 4b M MAIN MENU FD ANN FD ANN 9 CFP ANN CFP ANN 9 RGD ANN_ COMP ANN lt 22 c Figure 16 ANN Information a ANN Information button in main menu b Screenshot of main menu on choosing ANN info show option c Sample Excel sheet for ANN model information Pavement Analysis Menus Pavement analysis menu consist of three main parts inputs analysis tool and outputs The user can provides the software with the information required for analysis in the inputs part of pavement analysis menu The analysis tool allows user processing data and executing analysis with several functions The results of analysis are provided in the output part of pavement analysis menu Figure 17 il
15. 0 type FWD raw data files clicking obtain FWD file data file The obtain FWD file data file allows the user load the FWD raw data files and extract the FWD deflections required into the program as shown in Figure 20 Based on FWD loads of deflections the program allows two types of flexible pavement analysis 9 kip constant FWD load analysis and variable FWD load analysis The 9 kip constant FWD load analysis uses the FWD deflection data normalized to 9 23 kip constant FWD load The variable FWD load analysis uses the raw FWD deflection data corresponding to the raw FWD loads Please select a file pa d n n _ oo ee py a Computer Local Disk C 1 My Post Doctor Reserch Jils FWD vi ty Search Jits FWD P Organize Newfolder vr 00 x Microsoft Excel Z ETE Date Z AC over PCC DAT 10 25 YA Favorites Conventional AC DAT 10 25 MZ Desktop Full depth AC DAT 10 25 BR Downloads Z PCC dat A E Recent Places grade A Libraries Documents P Music E Pictures Videos ME Computer amp Local Disk C DVD RW Drive D MEPDG_UNLOCK amp public orgfiles iastate edu engrS ccee P GP groups cceeresearch engineering iastate edu R GA Scanner S CA cunnhwan odilec isetate adhi AIIN bY Filename Conventional AC DAT X dils FWD Dat File dat Cancel Filtering Menu After entering the FWD data required there is a data preprocessing unit for
16. 0 892 x IR log d 1 25 0 448 x IR 0 621 1 day 1 83 x sin hr g 15 5 0 042 x IR x sin hryg 13 5 1 Where Ta Pavement temperature at depth d in C IR Infrared surface temperature measured at the time of FWD testing in C Log Base 10 logarithm d Depth at which mat temperature is to be predicted mm 1 day Average air temperature the day before testing sin Sine function on an 18 hr clock system with 2 radians equal to one 18 hr cycle hris Time of day in 24 hr clock system but calculated using an 18 hr asphalt concrete AC temperature rise and fall time cycle Temperature Correction for HMA Modulus Several equations Ullidtz 1987 Baltzer and Jansen 1994 Deacon et al 1994 Noureldin 1994 Kim et al 1995 Ali and Lopez 1996 Lukanen et al 2000 Chen et al 2000 have been proposed relating the HMA modulus to a standard reference temperature However different values of standard reference temperature are found in literature for HMA modulus characterization The commonly used standard reference temperatures are 68 F in AASHTO 1993 design procedure AASHTO 1993 70 F in AASHTO Mechanistic Empirical Pavement Design Guide MEPDG AASHTO 2008 and Pavement ME formerly DARWin ME AASHTO 2012 77 F in some previous studies Noureldin 1994 Among them the temperature correction equation developed by Chen et al 2000 using Texas Mobile Load Simulator MLS data is the only available model wit
17. 271x 176 76 11 Where k modulus of subgrade reaction for rigid and composite pavements psi inch y SCI Road Rater do x Road Rater do mils 14 0 00 0 104 al SCI SENS 1 RATIO Soil Support K And S Values For Flexible Pavements From Road Rater Deflection Dishes AVERAGE SENSOR 1 DEFLECTION MILS Figure 11 Flexible pavement base relationship for soil support k value determination adapted from Potter and Dirks 1989 0 00 SCI SENS 1 RATIO 0 30 Soil Support K Values For Rigid amp Composite Pavements 4 0 3 0 2 0 1 0 0 0 AVERAGE SENSOR 1 DEFLECTION MILS Figure 12 Rigid and composite pavement base relationship for soil support k value determination adapted from Potter and Dirks 1989 Figure 13 compares k values determined from the flexible pavement k value base relationship chart and the developed equations using randomly selected data not used in equation 15 development Figure 14 displays similar comparison for rigid and composite pavements As seen in these figures the developed equations provide good k value estimates for both the flexible and rigid composite pavements R 0 9721 c Oo 5 3 ej Lo Oo a gt x 100 150 200 k value from chart Figure 13 k value determination for flexible pavement chart versus equation R 0 995 k value from equation 100 150 k value from cha
18. 5 a illustrates the sample analysis results of a conventional flexible pavement The analysis results will be displayed on the right side of the deflection basin matching optimization based backcalculation analysis result columns in the screen The user should scroll right to see all results 28 Granular Base Modulus Param K psi AC Modulus Eac psi 2 302 909 3 723 641 3 880 613 3 880 613 3 880 613 3 880 613 3 000 11 260 12 000 3 167 11 999 3 232 083 3 880 614 3 077 587 3 880 615 3 223 963 3 545 989 3 711 295 3 880 619 3 880 613 3 880 619 3 880 613 11 998 11 999 12 000 1 818 12 000 a Microsoft Excel La gS cc Temperature Correction of Modulus Complete Proceeding to SNeff Computation O Figure 25 HMA modulus temperature correction a Temperature corrected HMA modulus result outputs b Progress report window after completing HMA modulus temperature correction After completing HMA modulus temperature normalization procedure progress report window appears as shown in Figure 25 b Its purpose is to give user the notice of completion of procedure and its intention to proceed to the computation of effective SN SNefrr for flexible pavement and effective k value Kerr for composite pavement 29 Output Results of the Effective SN SNeg for Flexible Pavement and the Effective k value keg of Rigid and Composite Pavements Clicking the OK button in progress rep
19. Catalog No InTrans Project 11 415 4 Title and Subtitle 5 Report Date Improving the Accuracy and Usability of Iowa Falling Weight Deflectometer Data 7 Author s 8 Performing Organization Report No Halil Ceylan Kasthurirangan Gopalakrishnan and Sunghwan Kim 9 Performing Organization Name and Address 10 Work Unit No TRAIS Institute for Transportation PE lowa State University 11 Contract or Grant No 2711 South Loop Drive Suite 4700 Ames IA 50010 8664 12 Sponsoring Organization Name and Address 13 Type of Report and Period Covered lowa Department of Transportation Federal Highway Administration Final Report 800 Lincoln Way U S Department of Transportation 14 Sponsoring Agency Code Ames IA 50010 1200 New Jersey Avenue SE SPR 90 00 RB04 012 Washington DC 20590 15 Supplementary Notes Visit www intrans iastate edu for color PDF files of this and other research reports 16 Abstract This study aims to improve the accuracy and usability of lowa Falling Weight Deflectometer FWD data by incorporating significant enhancements into the fully automated software system for rapid processing of the FWD data These enhancements include 1 refined prediction of backcalculated pavement layer modulus through deflection basin matching optimization 2 temperature correction of backcalculated Hot Mix Asphalt HMA layer modulus 3 computation of 1993 AASHTO design guide related effective SN SNerr and effective k value kerr 4
20. ITY OF IOWA FALLING WEIGHT DEFLECTOMETER DATA Final Report May 2013 Principal Investigator Halil Ceylan Associate Professor Institute for Transportation lowa State University Co Principal Investigators Kasthurirangan Gopalakrishnan Research Assistant Professor Institute for Transportation Iowa State University Sunghwan Kim Research Assistant Professor Institute for Transportation lowa State University Authors Halil Ceylan Kasthurirangan Gopalakrishnan and Sunghwan Kim Sponsored by Iowa Department of Transportation Federal Highway Administration SPR 90 00 RB04 012 Preparation of this report was financed in part through funds provided by the Iowa Department of Transportation through its Research Management Agreement with the Institute for Transportation nTrans Project 11 415 A report from Institute for Transportation Iowa State University 2711 South Loop Drive Suite 4700 Ames IA 50010 8664 Phone 515 294 8103 Fax 515 294 0467 www intrans iastate edu TABLE OF CONTENTS ACE NOW LEDOMEN De eo ties de danas de ix EXECUTIVE SUMMAR Y A A A ad xi INTRODUCTION re ae eee a REE E A ee ae E E E E 1 Background einne e ea e E OEE Gees uchdak vaste E TE ESAS 1 Object VE and SCONE ict SES 2 I BACK ENHANCEMENT FEATURES AND MODULES ooocococoncccoccnoncnoncnncnoncnnncnnacnncnnncnnnonnanns 2 Deflecti n Basin Matching iii iii dida ainda a ia 3 Temperature Normalization for HMA Modulus ooooccnoccnocnnononinoncnononon
21. Improving the Accuracy and Usability of lowa Falling Weight Deflectometer Data Final Report May 2013 Sponsored by IOWA STATE UNIVERSITY Iowa Department of Transportation Institute for Transportation Federal Highway Administration InTrans Project 11 419 About the Institute for Transportation The mission of the Institute for Transportation InTrans at lowa State University is to develop and implement innovative methods materials and technologies for improving transportation efficiency safety reliability and sustainability while improving the learning environment of students faculty and staff in transportation related fields Disclaimer Notice The contents of this report reflect the views of the authors who are responsible for the facts and the accuracy of the information presented herein The opinions findings and conclusions expressed in this publication are those of the authors and not necessarily those of the sponsors The sponsors assume no liability for the contents or use of the information contained in this document This report does not constitute a standard specification or regulation The sponsors do not endorse products or manufacturers Trademarks or manufacturers names appear in this report only because they are considered essential to the objective of the document Non Discrimination Statement lowa State University does not discriminate on the basis of race color age religion national origin sex
22. Iowa DOT AC Overlay Design Procedure Using Road Rater Deflection Measurements The Iowa DOT has developed an AC overlay design procedure for existing flexible rigid and composite pavements using the Road Rater measurements Heins 1979 Marks 1983 Potter and Dirks 1986 Potter and Dirks 1989 This overlay design procedure was patterned closely after the 1993 AASHTO design procedures For the lowa DOT AC overlay design procedure the concept of SR as the estimated AASHTO SN was developed to characterize present structural condition of existing pavement The SR was graphically related to average Road Rater sensor No 1 deflection do value to determine SR from Road Rater deflection measurement In addition to SR the Iowa DOT AC overlay design procedure also adapted soil support k value to estimate existing subgrade soil support condition The base relationship chart for soil support k value for flexible rigid and composite pavements was developed by relating soil support k values with Road Rater deflection measurements Correlation of Road Rater Deflection Measurements to FWD Deflection Measurements In the past the lowa DOT used the Road Rater based measurements for AC overlay design procedures pavement management system and research evaluations However with the acquisition of FWD by the Iowa DOT Office of Special Investigations the use of Road Rater was abandoned Consequently the FWD deflection measurements were correlated to Road Rater d
23. ackcalculation models Ceylan et al 2007 Ceylan eat al 2009 Then the pre final pavement layer moduli are adjusted through a deflection basin optimizer in I BACK to match the actual measured FWD deflection basin Note that the backcalculated moduli resulting from deflection basin optimization in I BACK are moduli at pavement temperature at the time of FWD testing In the case of flexible pavement analysis the temperature normalization routine is then invoked to correct the adjusted HMA moduli from optimizer to a standard reference temperature The SNerr values are computed for conventional flexible and full depth pavements in accordance with the AASHTO design procedure as follows The temperature normalized HMA moduli are inputted into ANN forward models to produce FWD deflection basin at a standard reference temperature The predicted FWD deflection basin at the standard reference temperature is then utilized to compute the overall section SNert for flexible pavements in accordance with the 1993 AASHTO pavement design procedure AASHTO 1993 For rigid pavement analysis the adjusted subgrade stiffness values from the deflection basin optimizer are utilized to compute effective subgrade support Kerr in accordance with the 1993 AASHTO design procedure AASHTO 1993 For composite pavement HMA overlaid PCC pavements analysis the adjusted HMA moduli from the deflection basin optimizer are corrected to a standard reference temperature throug
24. and variable FWD load and the composite and rigid pavement analysis module with 9 kip FWD loadings The software toolbox is programmed to give warning messages if the user the clicks anywhere else While working with the toolbox all other Excel features are accessible including open close copy paste save save as print and print settings The ANN information illustrated in Figure 16 provides the user the general information on ANN models employed By clicking ANN info show button as shown in Figure 16 a six Excel sheets as shown in Figure 16 b are created Each of Excel sheets as shown in Figure 16 c contains the ANN model information such as the ranges of the data used for ANN model development These Excel sheets can be hidden by choosing ANN info hide 17 Y fu gt F fe ON Sirancportaton Disclaimer This tool is for demonstration purposes only Copyright Dr Halil Ceylan hceylan iastate edu Mr Alper Guclu Dr M Birkan Bayrak Dr Kasthurirangan Gopalakrishnan Dr Sunghwan Kim FR ATTE XH h OWS yA TUN Ni IVER XS le Figure 15 I BACK program main menu 18 a gt MAIN MENU FD ANN lt FD ANN 9 CFP ANN lt CFP ANN 9 RGD ANN lt COMP ANN lt 2 y l b 1 9 38E 02 3 47E 01 1 65E 01 1 65E 01 1 65E 01 2 1 674708 4 21E 01 1 73E 01 1 73E 01 1 73E 01 3 1 461965 1 043324 7 69E 01 7 69E 01 7 69E 01 4 1 522861 1 307949 1 06E 00 1 06E 00 1 06E 0
25. cnnnncono nono ncconcnnnnnnnn non 5 Computation of Effective Structural Number SN and Subgrade Support k value AASHTO Pre ME Design APP ad 7 Computation of Structural Rating SR and Soil Support k value Iowa DOT AC Overlay Design Procedure Using Road Rater Deflection Measurements cesses 8 TFBACK USERMANU UA Doren eee an Soe cae EA E S e E E ae 17 Program MO MEUS a A acs faa aaO 17 Pavement Analysis Men usd ld ota 19 Summary Result Pile asada ito 32 SUMMARY craneo A A E E 33 REFERENCES AR Se A OS a ede AA R Se 35 LIST OF FIGURES Figure 1 I BACK analysis flow chart for flexible pavemMent oooonccnnncnnnnonnncnnonononncnnnnonnccnnncroncnnnnos 3 Figure 2 I BACK analysis flow chart for rigid pavement oooconnocnnocononnconnnnnnnnnoncnonccnnn conan cnn ncronccnnno 4 Figure 3 I BACK analysis flow chart for composite pavement HMA overlaid PCC pavement 5 Figure 4 Generalized linear correlation equation with coefficients between FWD and Road Rater deflection measurements Note senor 1 deflection do sensor 2 deflection d12 sensor 3 deflection d24 and sensor 4 deflection d36 adapted from Jones and Hanson 1991 9 Figure 5 Flexible pavement base relationship between SR and Road Rater Do at 80 F adapted from Potter and Dirks 188 Doa ead recs ck aces ede eh cba ei eee een 10 Figure 6 Flexible pavement SR determination nomograph adapted from Potter and Dirks 1989 10 Figure 7 Road Rater Do temperature c
26. e pavement layer stiffnesses and further estimate pavement remaining life Although the Office of Special Investigations at lowa Department of Transportation DOT has collected the FWD data on regular basis the pavement layer moduli backcalculation techniques used so far have been cumbersome and time consuming Thus there was a need for more efficient and faster methods In a previous Iowa DOT project entitled Nondestructive Evaluation of Iowa Pavements Phase I advanced backcalculation models were developed using the Artificial Neural Networks ANN methodology Ceylan et al 2007 These ANN models are capable of predicting pavement layer stiffnesses as well as pavement critical responses forward modeling fully based on FWD test results and pavement layer thickness information For the three pavement types over 300 models in total were developed for varying input parameters The primary pavement types considered were flexible conventional and full depth rigid and composite Predicted flexible pavement parameters were Eac modulus of hot mix asphalt HMA or asphalt concrete AC Kp base modulus parameter Eri subgrade resilient modulus gac tensile strain at the bottom of asphalt layer gsc compressive strain at the top of subgrade and op subgrade deviator stress For rigid pavements Epcc modulus of portland cement concrete PCC ks coefficient of subgrade reaction opcc tensile stress at the bottom of the PCC layer and radius of
27. eflection measurements to continue the use of AC overlay design procedure Jones and Hanson 1991 Figure 4 displays the developed linear correlation equations between FWD and Road Rater deflection measurements along with the model coefficients These correlation equations Figure 4 were incorporated into the I BACK computation algorithms to convert the measured FWD deflection measurements into Road Rater deflection measurements FWD x x R R C PCC Sections Composite Sections AC Sections A Std Std Std X c r Error x Cc r Error x C r Error Sensor 1 3 745 0 83 0 92 0 62 4 890 0 83 0 96 1 11 11 830 3 89 0 92 4 23 Sensor 2 3 822 0 67 0 91 0 60 4 034 0 64 0 99 0 35 8 918 1 30 0 96 1 07 Sensor 3 3 850 0 63 0 90 0 53 3 803 0 86 0 98 0 41 7 622 0 51 0 96 0 52 Sensor 4 4 056 0 48 0 91 0 39 3 816 0 86 0 96 0 41 6 116 0 189 0 92 0 35 Figure 4 Generalized linear correlation equation with coefficients between FWD and Road Rater deflection measurements Note sensor 1 deflection do sensor 2 deflection d12 sensor 3 deflection d24 and sensor 4 deflection d36 adapted from Jones and Hanson 1991 Determination of SR for Flexible Pavement Figure 5 presents flexible pavement base relationship chart to determine SR from Road Rater do corrected to 80 F The Road Rater do at testing temperature were corrected to 80 F A nomograph as shown in Figure 6 was developed to correct the Road Rater do at testing temperature to 80 F and then
28. filtering the data It is optional to use the filtering window Figure 21 shows the available options for filtering The two options are e Range Check Deflection basin should form a bowl shape and therefore deflections should be in decreasing order Data that falls outside this range are red colored e Model Check ANN models are normalized according to the model ranges and therefore any input outside the range used in ANN training will form a poor quality input As a result the model check will determine the outliers and color them in red The filtering is applied by changing the color of the input parameter to red Therefore results for these parameters are also calculated With this approach engineers will have a better understanding of the sources of errors 24 Filter Options Filtering Completed Total Filtered data is gt 0 42 b Figure 21 Filtering the FWD data a Filter options menu b Filtering results no filtered data in this example ANN Based Preliminary Backcalculation Result Outputs After preprocessing the data clicking the Run button will activate a neural network based analysis of pavements The program will analyze the employed ANN model for the pavement properties For each model the analysis results will be displayed on the right side of the screen The user should scroll right to see all results The default units used in the program are based on US customary units Reported
29. h temperature normalization routine similar to flexible pavement analysis Similar to rigid pavement analysis the adjusted subgrade stiffness values from the deflection basin optimizer are utilized to compute ker values in accordance with the 1993 AASHTO design procedure AASHTO 1993 The SR and k values are also computed for all three types of pavement in accordance with Iowa DOT Road Rater based AC overlay design procedure AC Pavement SN Computation Temperature Predictions Calculated Deflection ANN FWD Analysis Basin at Reference Tool Forward Temp calculation Temperature Correction eee tenes Calculated Deflection ANN FWD Analysis differences betwee E a Basin at Measured Tool Forward Measured and F Calculated Deflections Temp calculation Terminate Measured FWD Deflection Basin ANN FWD Analysis Backcalculated E Pavement Layer Tool Backcalculation values prefinal Thicknesses FWD Deflection Inputs Correlation Conversionto Road Equations Rater Measurements Figure 1 I BACK analysis flow chart for flexible pavement Deflection Basin Matching Deflection basin matching provides a fool proof method to validate the ANN based backcalculation results since the predicted pavement layer moduli are optimized or fine tuned to obtain a very close match between the actual FWD and predicted deflections The pre final pavement layer moduli from ANN backcalculation models are inputted into the ANN forward mode
30. h the flexibility to normalize to any reference temperature with good accuracy Considering this advantage it was incorporated into the temperature normalization routine in I BACK Erw Erc 18Ty 32 446 1 8T 32 724467 2 Where Erw the adjusted modulus of elasticity at Ty MPa Erc the adjusted modulus of elasticity at Te MPa w the temperature to which the modulus of elasticity is adjusted C Te the mid depth temperature at the time of FWD data collection C Although a standard reference temperature of 68 F is utilized in the current implementation of I BACK the temperature correction equation proposed by Chen et al 2000 could be easily adapted later for any reference temperature should a need arise Computation of Effective Structural Number SN and Subgrade Support k value AASHTO Pre ME Design Approach The concept of SNerr is typically used for evaluating the overall structural condition of flexible pavements Similarly the ket is used for determining the subgrade support for PCC rigid pavement and composite pavement analysis Typically when the ratio of SNerr to as built SN based on in place pavement structure falls below 90 the evaluated section is recommended for structural improvement The AASHTO 1993 design procedure AASHTO 1993 outlines a method for calculation of SNerr and the ke using the measured deflection data The SNerr and the ker equations in the AASHTO 1993 design procedure AASHTO 1993
31. hted benefits e Provides more fine tuned ANN pavement backcalculation results by implementation of deflection matching optimizer for conventional flexible full depth rigid and composite pavements Provides temperature normalized corrected hot mix asphalt HMA layer modulus at a standard reference temperature for conventional flexible full depth and composite pavements Provides effective SNerr and the effective k value ke as final outputs for pavement asset management purposes Provides SR and k value k as final outputs to make FWD deflection measurements suitable for use in the existing lowa DOT AC overlay design procedure Produces separate smaller sized output files from backcalculation analysis Produce separate smaller sized output files from backcalculation analysis INTRODUCTION Background Evaluating the structural condition of existing in service pavements is part of the routine maintenance and rehabilitation activities undertaken by the most Departments of Transportation DOTs In the field the pavement deflection profiles or basins gathered from the nondestructive Falling Weight Deflectometer FWD test data are typically used to evaluate pavement structural condition FWD testing is often preferred over destructive testing methods because it is faster than destructive tests and does not entail the removal of pavement materials This kind of evaluation requires the use of backcalculation type structural analysis to determin
32. ing optimization based backcalculation analysis Figure 23 a illustrates the sample analysis results of a conventional flexible pavement The analysis results will be displayed on the right side of ANN based preliminary backcalculation analysis result columns in the screen The user should scroll right to see all results If user have a trial solution checking screen shown in Figure 23 b during deflection basin matching analysis user can hit Stop button See Figure 23 b to proceed with analysis since the solutions do not generally improve beyond this point Subgrade s Mean Abs Error bet Meas and Comp FWD Deflections AC Modulus i Eac psi i Computed FWD Deflections after Deflection Basin Matching mils Actual Temp Show Trial Eolutionl The maximum number of feasible solutions was reached continue anyway Figure 23 Deflection basin matching optimization of backcalculation analysis a Optimized backcalculation analysis result outputs b Trial solution check message box The deflection basin matching optimization based backcalculation analysis refines ANN based backcalculation results for each pavement type to obtain a very close match between the actual FWD and predicted deflections Reported results are refined optimized modulus for each 27 pavement type FWD deflection predictions and Mean Absolute Error MAE between measured and predicted FWD deflections for assessment of the accuracy of refined
33. is flow chart for composite pavement HMA overlaid PCC pavement Temperature Normalization for HMA Modulus The stiffness or modulus of HMA is very temperature sensitive The temperature normalization routine was incorporated into I BACK in close consultation with the TAC in order to correct the backcalculated HMA moduli to a standard reference temperature for the section being analyzed The incorporated temperature normalization routine consists of 1 HMA pavement temperature estimation and 2 temperature correction algorithm for HMA modulus Prediction of HMA Pavement Temperature Before correction of the backcalculated HMA moduli to a standard reference temperature the mid depth pavement temperature at which FWD deflections were taken should be identified The direct measurement of this temperature requires the time consuming process of installation of temperature probe in depth of pavement Alternatively this temperature may be estimated from approximate methods based on air and surface temperatures measured at the time of FWD testing Lukanen et al 2000 developed a set of equations called as BELLS models for predicting in depth pavement temperatures in LTPP testing based on empirical data Among these equations the BELLS3 model accounts for shaded condition of pavement surface under routine testing conducted by most highway agencies The BELLS3 model employed in temperature normalization routine is expressed as follows Ta 0 95
34. ixture analysis and design Presented at the Annual 73thMeeting of the Transportation Research Board Washington DC Transportation Research Board Heins D 1979 Road Rater Dynamic Deflections for Determine Structural Rating of Flexible Pavements Final Report for IHRB 178 Ames IA Office of Materials lowa Department of Transportation Jones K and Hanson T 1991 Correlation of the Road Rater and the Dynatest Falling Weight Deflectometer Final Report for MLR 91 4 Ames IA Office of Materials lowa Department of Transportation Kim Y R Hibbs B O and Lee Y C 1995 Temperature correction of deflections and backcalculated moduli Transportation Research Record 1473 55 62 Washington DC Transportation Research Board National Research Council Lukanen E O Stubstad R and Briggs R 2000 Temperature Predictions and Adjustment Factors for Asphalt Pavement FHWA RD 98 085 Mclean VA Federal Highway Administration Marks V J 1983 Dynamic Deflections to Determine Roadway Support Ratings Final Report for IHRB 245 Ames IA Office of Materials lowa Department of Transportation Noureldin A S 1994 Temperature gradient in a full depth asphalt and its effect on modulus and shear Gradients The Proceedings of 6th Conference on Asphalt Pavements for Southern Africa CAPSA 35 Potter C J and Dirks K L 1986 Structural Analysis and Overlay Design of Pavements Using the Road Rater Final Report for MLR 84 3 Ames
35. ls to predict the deflection basin and compare how closely the predicted deflection basin matches the measured deflection basin The differences between the field measured and predicted deflection basins are minimized by adapting the in built Excel Solver tool employing evolutionary optimization or GRG Nonlinear optimization in I BACK Thus the robustness of I BACK predictions has been greatly improved by incorporating the deflection basin matching optimization routine ka Computation Optimizer Minimize Calculated Deflection ANN FWD Analysis differences between au aa Basin at Measured Tool Forward Measured and Calculated Deflections Temp calculation Terminate Measured FWD Deflection Basin ANN FWD Analysis Backcalculated E Tool Backcalculation values prefinal Pavement Layer Thicknesses FWD Deflection Correlation l Conversionto Road Inputs Equations Rater Measurements Figure 2 I BACK analysis flow chart for rigid pavement AC Pavement Temperature Predictions Temperature Correction pa ES Calculated Deflection ANN FWD Analysis asians apa Basin at Measured Tool Forward Measured and sian Calculated Deflections Temp calculation Terminate Measured FWD k Computation Deflection Basin ANN FWD Analysis Backcalculated E Pavement Layer Tool Backcalculation values prefinal Thicknesses eres FWD Deflection Conversionto Road Rater Measurements Correlation Equations Figure 3 I BACK analys
36. lustrates the conventional flexible pavement analysis menu of I BACK as a typical layout example The analysis tool for all pavement types has four button functions Run Filter Open FWD data file and Main Menu The analysis tool for rigid pavements has additional submenus of Equation The detail descriptions of these button functions are presented in the following subsection Table 1 summarizes key inputs required and outputs produced through I BACK oo oh Inputs Analysi Tool ITT Anil Al e Figure 17 I BACK pavement analyses menu for conventional flexible pavement analysis 20 Table 1 Key required inputs and outputs produced by I BACK Type Inputs Outputs Flexible e FWD loads applied and e Eac modulus of HMA or AC Ko conventional deflection measurements base modulus parameter and Eri full depth e Pavement layer thicknesses subgrade resilient modulus at actual FWD testing temperature e Eac at reference temperature e SNerr effective SN e Road Rater based SR and k Rigid e FWD loads applied and e Epcc modulus of PCC and ks deflection measurements coefficient of subgrade reaction e Pavement layer thicknesses e ker effective k value e Road Rater based SR and k Composite e FWD loads applied and e Eac Ercc and ks at actual FWD deflection measurements testing temperature e Pavement layer thicknesses e Eac at reference temperature o kerr e Road Rate
37. n a statistical summary 1 e mean standard deviation and coefficient of variation of the results is presented After completing ANN based preliminary backcalculation a progress report window appears as shown in Figure 22 a Its purpose is to give user the notice of completion of ANN based preliminary backcalculation and its intention to proceed to deflection basin matching analysis to determine final backcalculation results Subgrade Granular Base strain ue Deviator Stress Modulus Eri Modulus Strain uz Top of psi Top of psi Actual Param K psi Bottom of AC Subgrade Subgrade Temp Actual Temp Actual Temp Actual Temp AC Modulus Eac psi Actual Temp 5 261 155 6 000 000 6 000 000 6 000 000 6 000 000 6 000 000 ls D 8 8 5 113 697 8 6 000 000 6 000 000 6 000 000 6 000 000 6 000 000 MEN esoo MET MOTO MOTO MIT smes MOTO MOTO 8 8 s Microsoft Excel Preliminary Backcalculation Analysis Complete Proceeding to Deflection Basin Matching b Figure 22 ANN based preliminary backcalculation analysis a Preliminary backcalculation analysis result outputs b Progress report window after completing ANN based preliminary backcalculation analysis 26 Deflection Basin Matching Optimization of Backcalculation Analysis Result Outputs Clicking the OK button in progress report window see Figure 22 b will activate deflection basin match
38. nnnccnnncnnoccnonnnacnnnnnnon LIST OF TABLES Table 1 Key required inputs and outputs produced by I BACK 0 eeeeeeeeeeneecneeenseeeeeeeeneees viii ACKNOWLEDGMENTS The authors would like to thank the Iowa Department of Transportation DOT for sponsoring this research and the Federal Highway Administration for state planning and research SPR funds used for this project The project technical advisory committee TAC members from the Iowa DOT including Jason Omundson Chris Brakke Fereidoon Ben Behnami Kevin Jones Scott Schram Todd Hanson and Tom Muhlenbruch are gratefully acknowledged for their guidance support and direction throughout the research 1X EXECUTIVE SUMMARY Highway agencies periodically evaluate the structural condition of roads as part of their routine maintenance and rehabilitation activities The falling weight deflectometer FWD test measures road surface deflections resulting from an applied impulse loading simulative of a truck passing on the highway The measured surface deflections are utilized to determine pavement layer stiffnesses through a backcalculation type structural analysis Although the Iowa Department of Transportation DOT has been collecting the FWD data on a regular basis the pavement layer moduli backcalculation techniques used so far have been cumbersome and time consuming More efficient and faster methods in FWD test data analysis were deemed necessary for routine analysis P
39. optimized modulus After completing the deflection basin matching optimization based backcalculation analysis progress report window appears as shown in Figure 24 a for flexible and composite pavements and Figure 24 b for rigid pavement Its purpose is to give user the notice of completion of analysis and its intention to proceed to HMA modulus temperature correction for flexible and composite pavements or computation of effective k value Kerr for rigid pavement E a Deflection Basin Matching Optimization Complete Proceeding to Temperature Correction of HMA Modulus Deflection Basin Matching Optimization Complete Now Proceeding to Computation of Effective k value b Figure 24 Progress report window after completing optimization of backcalculation analysis a flexible and composite pavements b rigid pavement HMA Modulus Normalization Result Outputs Clicking the OK button in progress report window see Figure 24 a will activate HMA modulus temperature normalization routine for flexible and composite pavements The HMA modulus temperature normalization routine in I BACK corrects the backcalculated HMA moduli to a standard reference temperature 68 F by using HMA pavement temperature estimation and HMA modulus temperature correction algorithm Reported results are temperature corrected modulus at a standard reference temperature 68 F for flexible and composite pavements Figure 2
40. orrection for flexible pavement nomograph versus EU A eich et wg way ae gst vege oad a ae Vas eave avcege 11 Figure 8 SR determination for flexible pavement nomograph versus equation ccceeeeees 12 Figure 9 Rigid and composite pavement base relationship between SR and Road Rater Do adapted from Potter and Dirks 1989 000000 Gaaelana ee 13 Figure 10 SR determination for rigid and composite pavements chart versus equation 13 Figure 11 Flexible pavement base relationship for soil support k value determination adapted from Potter a d Dirks 1989 coin nia ias been a heat eee ede in ede 15 Figure 12 Rigid and composite pavement base relationship for soil support k value determination adapted from Potter and Dirks 1989 oo eee eeceeeneeeseeceeeeeeeeeeaeesaeenes 15 Figure 13 k value determination for flexible pavement chart versus equation cecceeeeeeees 16 Figure 14 k value determination for rigid and composite pavements chart versus equation 16 Figure 15 I BACK program main A a See 18 Figure 16 ANN Information a ANN Information button in main menu b Screenshot of main menu on choosing ANN info show option c Sample Excel sheet for ANN model informan epen ie A AE EA R A A E AAA 19 Figure 17 I BACK pavement analyses menu for conventional flexible pavement analysis 20 Figure 18 General information WII W 00 A A A A A 22 Figure 19 Screen shot of effective PCC thickness equations
41. ort windows shown in Figure 24 b will activate computation of effective k value Kerr for rigid pavement Similar to this clicking the OK button in progress report windows shown in Figure 25 b will activate computation of effective SN SNefrr for flexible pavement and effective k value Ker for composite pavement Figure 26 illustrates the sample analysis results of rigid pavement Reported results for rigid and composite pavements is effective k value Kerr Figure 27 illustrates the sample analysis results of a conventional flexible pavement Reported results for flexible pavement are effective modulus of pavement layer Ep and effective SN SNefr The user should scroll right to see all results keff Effective k value pci 87 90 88 89 57 58 76 77 73 77 64 66 71 72 62 62 73 74 49 Figure 26 Effective k value Kerr for rigid and composite pavements 30 Effective Structural Number SNeff Effective Pavement Modulus Ep psi 5 4340 Figure 27 Effective SN SNerr with effective modulus of pavement layer Ep for flexible pavement Output Results of the Road Rater Based SR and k Figure 28 illustrates the sample analysis results of Road Rater Based SR and k value computation for a conventional flexible pavement The analysis results will be displayed on the farthest to the right side on the screen The user should scroll
42. processed and analyzed on the fly Objective and Scope The objective of this study is to incorporate significant enhancements into the developed fully automated software system for rapid processing of the FWD data Based on the requirements by the Technical Advisory Committee TAC members and engineers who will be using the program on a routine basis the following enhancements were incorporated into the fully automated software system referred to as I BACK the Iowa Pavement Backcalculation Software e Deflection basin matching After predicting the pavement layer modulus based on ANN models adjust them to match the input deflection basin e Temperature normalization correction of HMA layer modulus e Computation of overall pavement section effective Structural Number SNefrr and subgrade support Kerr following 1993 AASHTO design guide based procedures e Computation of Structural Rating SR and soil support k following lowa DOT Road Rater based AC Overlay Design Procedure e Enhancement of user friendliness of input and output from the software I BACK ENHANCEMENT FEATURES AND MODULES The I BACK analysis flow charts are depicted in Figure 1 for flexible pavements conventional HMA and full depth HMA pavements Figure 2 for rigid pavements PCC surface pavements and Figure 3 for composite pavements HMA overlaid PCC pavements As a first step in the I BACK analysis the pre final pavement layer moduli are calculated from ANN b
43. r based SR and k General Information Inputs After selecting one of the pavement types from the main menu a general information window appears Its purpose is to get information that represents a project site at the beginning of each analysis see Figure 18 The information required in this window are project name project location FWD testing date time and temperature conditions of air and pavement surface The user is required to fill in the information to continue with pavement analysis Note that the Time Infrared IR temperature at the time of FWD testing and previous day air temperature are required inputs for carrying out temperature correction of AC modulus and subsequent SNerr computations 21 Project Name Project Location Date Comments Time HH MM Time ex 9 50 14 30 15 26 IR Temperature F Pavement Surface Temperature Prev Day Air Temperature F Prev Day Air Temperature Figure 18 General information window FWD Deflection Database and Inputs At the next step users are expected to enter the FWD deflection database and inputs for the I BACK program Required analysis parameters are deflection data pavement layer information layer thicknesses and FWD load for variable FWD load analysis Depending on pavement type the number of layers can be changed If any of the required parameter is missing the program will display an error message No Data in the results section Clicking the Main
44. revious Iowa DOT research projects focused on developing advanced pavement layer moduli backcalculation models using the artificial neural networks ANN methodology The developed models were successfully validated using field data and incorporated into a Microsoft Excel spreadsheet based backcalculation software toolbox with a user friendly interface This study was undertaken with the objectives of improving the accuracy and usability of lowa FWD data and the pavement inverse analysis tools Based on the requirements by the technical advisory committee TAC members representing potential users of the developed backcalculation software system at the lowa DOT significant enhancements were incorporated into the fully automated software system for rapid processing of the FWD data These enhancements include the following e Refined prediction of pavement layer modulus through deflection basin matching optimization e Temperature correction of HMA layer modulus e Computation of 1993 AASHTO design guide based effective structural number SNefrr and effective k value Kerr e Computation of Iowa DOT asphalt concrete AC overlay design based structural rating SR and k value k e Enhancement of user friendliness of input and output from the software tool A high quality easy to use backcalculation software package called Iowa Pavement Backcalculation I BACK software was developed to achieve the project goals and yielded the following highlig
45. rt Figure 14 k value determination for rigid and composite pavements chart versus equation 16 I BACK USER MANUAL The password protected Excel Spreadsheet based I BACK was developed by writing Macros using Microsoft s Visual Basic for Applications VBA programming language In case of troubleshooting the user is requested to change the macro security Tools gt Macro gt Security to the medium or low level to allow the macros to run I BACK provides user interaction for data editing and pasting and all other functionalities available in Excel The Excel sheets in I BACK include a main menu and analysis menu for each pavement type for recording inputs and displaying the results of analysis The input and output space is divided by a control button with command buttons in each of the analysis spreadsheets I BACK also generates an output file in CSV format at the completion of analysis in the same folder where the software tool is saved Program Main Menus I BACK starts by displaying the main menu Figure 15 As a first step users are expected to select the pavement type conventional full depth flexible rigid or composite pavements by clicking on it to activate the selected pavement analysis Excel sheet interface There are six Excel pavement analysis sheets including the conventional flexible pavement analysis module with 9 kip and variable FWD load the full depth flexible pavements analysis module with 9 kip
46. t used in equation development were randomly selected and used in these comparisons As seen in these figures the developed equations provide good estimations of both Road Rater Do at 80 F and SR determined from nomograph 10 9 v 3 E R 0 9997 7 o S 6 o 5 E 2 4 L 3 2 sa 2 oO a 1 0 0 1 2 3 4 5 6 7 8 9 10 D at 80 F from nomograph mils Figure 7 Road Rater Do temperature correction for flexible pavement nomograph versus equation 11 j o R 0 9997 ec 2 3 o te e k 4 aL un OrR NW kU DN CO WO SR from nomograph Figure 8 SR determination for flexible pavement nomograph versus equation Determination of SR for Rigid and Composite Pavements Figure 9 presents the rigid and composite base relationship chart to determine SR from Road Rater do in Iowa DOT AC overlay design procedure No temperature correction is applied to do for both of rigid and composite pavements However for composite pavement temperature correction to 70 F is applied to SR at testing temperature Similar to the development of SR prediction equation for flexible pavement a correlation equation between SR and Road Rater do was developed using synthetic database derived from chart in Figure 9 The chart based rigid and composite pavement SR determination procedures were then incorporated into the I BACK computation algorithms The developed correlation
47. tive k value Kerr 4 computation of Iowa DOT AC overlay design based Structural Rating SR and k value k and 5 enhancement of user friendliness of input and output from the software A high quality easy to use backcalculation software package referred to as I BACK the Iowa Pavement Backcalculation Software was developed to achieve goals with the following highlighted benefits e Provide more fine tuned ANN pavement backcalculation results by implementation of deflection matching optimizer for conventional flexible full depth rigid and composite pavements e Provide temperature normalized corrected HMA layer modulus at a standard reference temperature for conventional flexible full depth and composite pavements e Provide effective Structural Number SNegr and the effective k value ker as final outputs for pavement asset management purposes e Provide Structural Rating SR and k value k as final outputs to make FWD deflection measurements suitable for use in existing Iowa DOT AC overlay design procedure e Produce separate smaller sized output files from backcalculation analysis 33 REFERENCES AASHTO 1993 AASHTO Guide for Design of Pavement Structures Washington DC AASHTO 2008 Mechanistic Empirical Pavement Design Guide Interim Edition A Manual of Practice Washington DC AASHTO 2012 AASHTOWare Pavement ME Design Washington DC lt http www darwinme org MEDesign Index html gt February 22 2013 Ali H
48. ual orientation gender identity genetic information sex marital status disability or status as a U S veteran Inquiries can be directed to the Director of Equal Opportunity and Compliance 3280 Beardshear Hall 515 294 7612 Iowa Department of Transportation Statements Federal and state laws prohibit employment and or public accommodation discrimination on the basis of age color creed disability gender identity national origin pregnancy race religion sex sexual orientation or veteran s status If you believe you have been discriminated against please contact the lowa Civil Rights Commission at 800 457 4416 or the lowa Department of Transportation affirmative action officer If you need accommodations because of a disability to access the Iowa Department of Transportation s services contact the agency s affirmative action officer at 800 262 0003 The preparation of this report was financed in part through funds provided by the lowa Department of Transportation through its Second Revised Agreement for the Management of Research Conducted by lowa State University for the lowa Department of Transportation and its amendments The opinions findings and conclusions expressed in this publication are those of the authors and not necessarily those of the lowa Department of Transportation or the U S Department of Transportation Technical Report Documentation Page 1 Report No 2 Government Accession No 3 Recipient s

Improving the Accuracy and Usability of Iowa Falling Weight

Contents

Download Pdf Manuals

Related Search

Related Contents