Home

TEMPEST

1. Figure 2 4 New model with empty Sewer Lines form The model tabs can be used to switch between Sewer Lines Time Series and Results view tle HR Sim 2 Winter 1 tmo File Edit Model Compute View Help DSI HBSS s as 4 Sewer Lines Time Series 7 Results are 4943 Specifications 4943 2 4885 Sewer Node 4825 Inflow 4767 gt 4708 O Constant Value D 464 p 2 Time Series 25 02 27 02 08 Inflow 4535 A Su 4477 Inflow Temperature Twin C 4418 Constant Value o 4359 4300 2 Time Series Qwin m3 s phi TA p E TZ 1 25 02 27 02 08 Inlay deltas Ambient Temperature TA C 8 3 TS inf lambdaS Ambient Rel Humidity phia Ambient Air Pressure pA mbar Air Exchange Coeff b Sewer Pipe Type Kkst lambdaP f Type lambdas Length L m Penetration Depth deltaS m Nominal Diameter D m Soil Temperature TS inf C Wall Thickness s m Slope so COD Degradation Rate r mgCOD m3 s Figure 2 5 Main window displaying the Sewer Lines tab completed with values Figure 2 6 Dialog box for adding new sewer lines Add Line Name Copy Line q 8 Chapter 2 Program Handling Table 2 2 List of all parameters needed in order to perform simulations with TEMPEST Name Symbol Unit Constraints Sewer Node Inflow QWin m s QWin gt 0 m s for first and QWin gt 0 m s for successive sewer lines In
2. 25 02 2008 00 13 00 0 014717 13 08923 T im o N 22502 20098 NN LANN IN niaenz 12 114973 Figure 2 9 Time Series tab 2 2 Model Formulation 11 Time series columns and entire time series can be deleted if no references to sewer lines exist e To delete a column of a time series select the time series and click on in the toolbar or select Remove Column from the Model menu alternatively Then a dialog box shows up listing the columns of the current time series Select the column you want to be removed and confirm your choice e In order to delete an entire time series be sure that it is selected and click on in the toolbar or menu Model Delete Time Series 2 2 3 Model Settings The model settings dialog box includes data on material properties and numerical param eters It can be opened by choosing Model Settings from the Model menu This dialog box has three tabs the first two tabs allow the user to add and modify data for pipe Pipe Types Figure 2 10 and soil types Soil Types Figure 2 11 the third one lets the user change the numerical parameters The numerical parameters will be widely discussed in Section 2 3 4 A selection of parameter values of the most common pipe and soil types is given in Appendix B The forms to edit pipe and soil types have the same user interface but the two types have different parameters Types can be added and removed by clicking on
3. 2 2 2 Preferences 2 5 Model Alt M Add Sewer Line to 2 2 1 Remove Sewer Line te 2 2 1 Add Time Series Ed 2 2 2 Delete Time Series 2 2 2 Insert Column 2 2 2 Remove Column 2 2 2 Validate Entries and Data v Model Settings 2 2 3 Compute Alt C Steady State Solution 5 2 3 Dynamic Solution 5 2 3 View Alt V Sewer Lines 2 2 1 Time Series 2 2 2 Results 2 4 Box Zoom 4 2 4 Pan wi 2 4 Zoom In x 2 4 Zoom Out s 2 4 Restore View i 2 4 Help Alt H About 2 2 Model Formulation 5 2 1 2 File Handling TEMPEST models consisting of sewer lines time series and settings are stored in model files with the file extension TMO The File menu provides the basic functionalities to create TEMPEST models File New to save a model File Save or File Save As or to load a previously created model from the file system File Open 2 2 Model Formulation In TEMPEST the basic element used to build models of complex sewer systems is the Sewer Line which consists of a Node followed by a Conduit D rrenmatt and Wan ner 2008 Nodes are introduced in the sewer at each location where there are lateral inflows changes in the pipe geometry changes in the material properties of the sewer pipe or changes in the surrounding soil as shown in Figure 2 2 Lateral wastewater in flows to the system and inflow temperatures can either be constant values in time or time series Mat
4. Modeling with TEMPEST 29 15 T RS 4943 RS 3096 measured RS 3096 calculated Temperature C 8 11 Mar 2008 12 Mar 2008 13 Mar 2008 Figure 3 10 Result of the model validation 3 3 2 Model Validation The model parameter values identified during the model calibration phase now have to be validated by performing a simulation using time series data that have not already been used for the model calibration If the accuracy of the model validation more precisely the correspondence of the simulated and the measured downstream temperatures is satisfy ing the model can be accepted Otherwise the calibration must be revised In Figure 3 10 the results of a simulation based on the model parameters found by the calibration and based on inflow and inflow temperature values from March 8 to March 9 2008 are compared to the corresponding measurements The accordance of the two series is satisfying 3 3 3 Heat Recovery Scenarios By comparing two very simple heat recovery scenarios the potential of TEMPEST as a tool to investigate the effects of heat recovery on the downstream wastewater treatment plant is demonstrated In the first scenario a constant amount of heat of O 350kW shall be reclaimed at the upper end of the sewer at manhole RS 4943 and used for the fictitious building planned next to the manhole In the second scenario the amount of heat extracted is varied durin
5. added to the model and the parameter values of Table 3 1 have been entered 3 3 Modeling with TEMPEST 25 fe Model Calibration tmo File Edit Model Compute View Help Bea 8 tee v7 amp f Sewer Lines Time Series e Results Name Data Date Time Measuring Campaign 1 Figure 3 5 The newly added time series which is named Measuring Campaign 26 2 28 2 2008 is now available in the list to the left The time series data table currently only has a Date Time row name the series e g Measuring Campaign 26 2 28 2 2008 define the number of rows the time series table should have e g 1000 and choose between a relative time scale where the time series starts from second zero and an absolute time scale depending on the time format of the measured time series data Select Absolute Date Time The time series tab is now visible showing the Date Time column of the added time series Figure 3 5 Before adding inflow and temperature data to the data table a new column must be added for each of the two variables This is done by selecting Insert Column from the Model menu for both the Lateral Inflow m3 s and the Lat Inflow Temperature parameter The Default Value field can be left blank Note that Lateral Inflow also refers to the inflow at the upper end of the sewer to be modeled In order to fill the cells of the time series table for long time series it is recommended to copy past
6. compute a dynamic solution 2 4 Results View and Export 15 Progress SL Initialisation Done Steady State Solution Current step 195 curr tolerance e Dynamic Solution dt 14 90s Cr 0 79 maana Pause Abort Figure 2 15 Progress dialog showing typical information on the ongoing process of a dynamic solution Model Settings q _ Pipe Types soil Types Numerics Newton Raphson Iterator Tolerance for Convergence 0 0001 Pipe Soil Module Structure Number of Pipe Layers 5 PDE Solver Max Courant Number Max Rel Error Steady State Max Iterations Steady State Stepsize in Space m z0 g Stepsize in Time s autocompute 0 Apply Filter Enquist 2 1 OK Cancel Figure 2 16 Dialog box for editing numerical settings settings can be accessed in the Numerics tab of the Model Settings dialog box Figure 2 16 To open the dialog box click on Model Settings in the Model menu and select the appropriate tab or click on the Numerical Settings button in the Compute Solution dialog box Figure 2 13 and Figure 2 14 Three groups of parameters can be modified Parameters for the Newton Raphson iterator the soil module structure and the PDE solver cf Appendix A 2 The parameters together with a short description and the implemented default values are given in Table 2 3 The numerical settings are stored together with the model
7. implemented to solve the system of PDEs for implementation de tails please refer to Appendix A 2 e TEMPEST applications range from simple steady state estimates of the changes of the wastewater temperature in a single sewer line to full scale simulations of the dynamics of the wastewater temperature in large systems of successive sewer lines with lateral inflows e Intuitive graphical user interfaces assist the user in managing data performing cal culations and plotting results Data can easily be imported from a text file or spread sheet and results can be plotted on the screen and exported for further use 1 3 System Prerequisites TEMPEST is written in the object oriented programming language C and uses the wxWidgets graphical user interface library It is compiled for the Microsoft Windows ope rating systems and consists of just one executable file has no dependencies on external libraries and does not require any software installation Therefore the only action to be taken in order to run the application is to copy or download the executable file and execute it The program has been extensively tested on Windows XP and Windows Vista but it should also run on previous versions of Windows Since it numerically solves partial differential equations and therefore performs a lot of calculations faster processor clocks yield shorter calculation times 1 4 File Format Models created with TEMPEST can be saved to disk and have th
8. 61 00m 2400 0 1055 0 0143 12 7494 9 9703 0 9098 12 3429 5 6255 11 9203 x 1220 00 m ba 1279 00 m 2 Figure 2 17 Main window displaying the calculation Results tab The wastewater and sewer air temperature as well as the temperature of the pipe layers are plotted 18 Chapter 2 Program Handling e Data selection top left and bottom left Either the time or the location for which the results are to be plotted and listed can be selected e Data table bottom right Lists the numerical values of the data series for the time value or location selected e Plot panel top center Visualizes the listed data e Data series panel top right The user can select the data series to be plotted Several tools are at your disposition to navigate and zoom within the plot panel The mouse cursor can either be in box zoom state or pan state In order to switch between these states use the and the icon in the toolbar If the box zoom state is activated a rectangular box can be drawn into the plot area which will then be zoomed in If the mouse cursor is in pan state the clipping can be shifted click and hold left mouse button while moving The and commands can be used to zoom in and zoom out respectively To restore the original view click on Cells rows columns or the entire data table can be selected and the selection may then be copied to the clipboard menu Edit Co
9. Case Study In this chapter it is illustrated step by step how TEMPEST can be used for the optimal planning of heat recovery from the sewer It is described which data must necessarily be collected and it is shown how a TEMPEST model is created calibrated and validated and then is used for the modeling of load scenarios 3 1 Introduction A fictitious heat recovery project was defined as follows It is planned to erect a new building close to manhole RS 4943 in R mlang see Figure 3 1 Since a lot of attention is given to ecological aspects during all phases of the building project the project team also wants to assess the potential of using heat that is reclaimed from the wastewater of a sewer system situated nearby for room heating and warm water production In this fictitious project TEMPEST is used to model the sewer system and to calculate different heat recovery scenarios For each scenario one must check that the legal con straints are fulfilled Those constraints assure that there will be no negative effects for the downstream wastewater treatment plant due to the heat recovery here we assume that the plant is located right after manhole RS 3096 1 845 km downstream of RS 4943 In the Canton of Zurich the inflow temperature to the plant must not fall below 8 C and the maximum temperature decrease resulting from temperature reclamation must not be higher than 0 5 C AWEL 2003 The preparatory work before being able to m
10. PEST 4 ep 4s ow us De 4 2 RE ES KS 3 3 1 Model Calibration sa sa as a u zu was AA e naeh 3 9 2 Model Validation 4 se s 2 2 8 0 oe a hk ner ee eg 3 3 3 Heat Recovery Scenarios 2 d nme 3 4 CONCUSSIONS 5 au a er Bibliography A Additional Material Al Analytical Mod l u OOS Ree under Ra A 1 1 Balance Equations and Transfer Processes 2 222222200 A124 Nodes sasmata eie Oak Ok ee ERE TR ae Ade PDE SOVET s o 4 ee Rk eRe E ADAN O ke SG B Material Properties B 1 Friction Coefficients ki o secs 222m nme B 2 Thermal Conductivity and Temperature Diffusivity of Different Pipe Types B 3 Thermal Conductivity and Temperature Diffusivity of Different Soil Types 20 20 21 23 27 29 29 30 33 35 36 37 37 40 41 42 43 ii Contents C TEMPEST Development 45 C 1 Program Versions and Changelog sema aaa a 46 1 Introduction The program TEMPEST was designed for the calculation of the dynamics and longitudinal spatial profiles of the wastewater temperature in the sewer This manual introduces the concepts behind TEMPEST and explains the program handling It is organized as follows It first discusses the program s fields of application its capabilities and the system prere quisites The second chapter focuses on the program handling and explains the conceptual model formulation Chapter 3 exemplifies the application of TEMPEST in a heat recovery project for the assessment of the recoverabl
11. al interval by which calculated results will be stored for further use 2 3 3 Computation Process Once the calculations have started a dialog box with a progress bar appears Figure 2 15 and informs the user about the ongoing process Usually TEMPEST occupies a lot of pro cessor time and the ability to work with other application while calculating can be limited this is defused if a multi core processor or multiple processors are at your disposition Therefore the calculations can be paused and resumed by clicking on the Pause button and ended by clicking on the Abort button If the steady state can not be found or the convergence is beyond the tolerance TEMPEST stops the calculation and gives advice for further optimization of the numerical parame ters When finished the dialog box with additional information such as total computation time stays on the screen until it is closed 2 3 4 Numerical Parameters TEMPEST tries to guess the best suiting numerical parameters though numerical settings can still be modified by the user to improve accuracy and optimize calculation time These Compute Solution O Compute steady state solution 25 02 2008 00 00 00 Compute dynamic solution Start Time Date Time 25 02 2008 00 00 00 Stop Time Date Time 27 02 2008 16 28 20 Output Timestep s N 600 Run Numerical Settings Cancel Figure 2 14 Dialog box for defining computation type and reference time to
12. ass transfer Soil type 2 gt ir 4 3 2 Sources sinks Conduit 1 Conduit 2 i Conduit 3 Figure 2 3 Compartments and processes considered in the sewer model Complex systems can be modeled by series of the elements node plus conduit 2 2 1 Sewer Lines Each model can contain an unlimited number of sewer lines The management of the sewer lines is done within the Sewer Lines tab see Figure 2 4 The Table to the left lists the successive sewer lines the form to the right has input fields for the various parameters A small image of a sewer cross section with the location where each model parameter is measured indicated helps the user In TEMPEST the input data needed for wastewater temperature simulations can be di vided into three groups sewer node data sewer pipe data and surrounding soil data The wastewater discharge and temperature ambient meteorological conditions and the Air Exchange Coefficient at the upstream end of a sewer conduit are specified in the Sewer Node section The Air Exchange Coefficient describes the capacity for air exchange between the sewer and the environment The Sewer Pipe and Soil sections are filled with data describing these two compartments By the select fields the user can choose appropriate values for thermal conductivities a biofouling factor and friction coefficients of many types of materials The COD Degradation Rate describes the
13. ded is already at the user s disposition he or she can start to define sewer lines now Between manhole RS 4943 and RS 3096 there are actually around 30 additional man holes Ideally the sewer should be modeled by 31 sewer lines separated by the manholes However since there are no relevant parameter changes along the flow path and since the manhole covers only have small openings with very limited air exchange it is sufficient to set up the TEMPEST model with just one sewer line of 1845 m in length After clicking on Add Sewer Line in the Model menu a dialog box appears In this box a name e g R mlang for the new sewer line can by typed in and it can be chosen whether or not the parameters are to be copied from an already existing line since you are adding the first line now the dropdown list is still empty The new sewer line is added by clicking on OK and is now listed on the left side of the sewer lines tab see Figure 3 4 You may rename the sewer line at any time later by clicking its name The data of the R mlang sewer as given in Table 3 1 can now be entered into the Sewer Lines form For Inflow and Inflow Temperature the user can either enter a constant value or define a time series Because the objective of this project is to choose an heat extrac tion mode that does not negatively affect the performance of the downstream wastewater treatment plant but that reclaims as much heat as possible the amount of heat extract
14. e amount of heat 1 1 Background Raw wastewater contains a considerable amount of energy which can be recovered and used to produce warm water and for room heating This is done by means of a heat pump and a heat exchanger which is installed in the sewer The optimal location for the installation of a heat exchanger in the sewer depends on several criteria First of all it is important that the energy consumers are located close to the site where the heat is recovered More heat can be reclaimed from the wastewater if the discharge is high which is typically the case towards the end of the sewer On the other hand the wastewater temperature usually is highest in the initial part of the sewer system An additional aspect to be considered is that the efficiency of a nitrifying wastewater treatment plant is reduced if its influent wastewater temperature is lowered Wanner et al 2005 Therefore planning and design of facilities for heat recovery from raw wastewater require that the changes of the wastewater temperature along the flow path in the sewer and the effect of heat recovery on the influent temperature of wastewater treatment plants can be quantified The interactive simulation program TEMPEST temperature estimation has been devel oped to calculate the dynamics and longitudinal spatial profiles of the wastewater temper ature in the sewer The program is based on a new model of the heat balance in sewers D rrenmatt 2006 and for a s
15. e extension tmo Sewer line definitions material parameters and time series data as well as numerical settings will be saved Therefore TEMPEST model files can conveniently be exchanged between users and platforms 2 Program Handling The handling of TEMPEST and its dialog boxes are illustrated in this chapter The model formulation is explained and it is shown how sewer systems can be modeled in TEMPEST The order of the sections corresponds to a typical session in which a user first cre ates a model file then defines the model performs and optimizes computations and eventually analyzes the results 2 1 Overview 2 1 1 User Interface Figure 2 1 shows the main window of TEMPEST upon the start of the program The commands for file handling model creation and editing and computation can be found in the menus of the menu bar The commands most often used also are represented in the toolbar All commands are listed and described in Table 2 1 in TEMPEST IN Fie Edt model Compute View Hep M Menubar BBE 5 BeEBBe F RAR UuReER Toolbar Figure 2 1 Main window of TEMPEST Chapter 2 Program Handling Table 2 1 Menu and toolbar items Menu item Hot key Toolbar icon See also File Alt F New Ctrl N D 2 1 2 Open A 2 1 2 Close Ctrl W 2 1 2 Save Ctrl S LE 2 1 2 Save As Ctrl Shift S 2 1 2 Export Data 2 4 Recent Files Exit Alt F4 Edit Alt E Copy 2 2 2 Paste
16. e the data from a text file or a spreadsheet application like MS Excel Care has to be taken that the date time format in the data source is similar to the format specified in the TEMPEST preferences menu Edit Preferences Once the data of the source file is copied to the clipboard you can select the upper left cell of the time series table and select Paste from the Edit menu The time series named Measuring Campaign 26 2 28 2 2008 is now filled with data as shown in Figure 3 6 If you switch back to the Sewer Lines tab you can now select the new time series in the drop down lists for inflow and inflow temperature The last parameters to be set before you can start the model calibration are the sewer pipe and soil type Every TEMPEST model disposes of a pipe and soil type library which can be accessed via the menu Model Model Settings Add a new pipe type in the Pipe Types tab of the Model Settings dialog box and enter the estimates from Table 3 1 for the friction coefficient the thermal conductivity the thermal diffusivity and the fouling factor and assign a comprehensive label Figure 3 7 The same steps can be performed to add a new soil type to the library after switching to the Soil Types tab Save the changes and close the dialog by clicking on OK In the Sewer Pipe panel and the Soil panel of the Sewer Lines tab now select the appropriate Type for pipe and soil respectively from the list All the data needed to perform s
17. eawag aquatic research User Manual TEMPEST Computer Program for the Simulation of the Wastewater Temperature in Sewers Version 1 02 David J D rrenmatt and Oskar Wanner tempest eawag ch http www tempest eawag ch Atmosphere E Wastewater Sewer air E Sewer pipe E Soil type 1 E Soil type 2 Conduit 1 Contents 1 Introduction 1 1 Bach sroumd oc ERA AAA A AO AE ES 1 2 Program Capabilities sa cs u aa wi nr ae Is A dE 1 3 System Prerequisites mairaos 2444 sau Roo SG kranken ek ES 1 4 File Format se 4 05 00 REA ROSY La eee BER Ren RE 2 Program Handling Dell OVERVIEW e acid ra Beare he eb wk A a eG 21 ASST interface AA BIR eck AG COR Ae Ge eae 2 12 File Handling nao don Ae amp ao as am be na da 2 2 Model Formulation e 22 1 Sewer Lines ee he eee sin ass RR we 22 2 A ee Belg Sh ee wR ee es 22 Model Settings s a u asec ein aise aed ae ee 2 3 GOMPUutatiON au 4 oe rd ah oe ee Sree chee a n 2 3 1 Steady State Solubone or s 0er Keane 2 3 2 Dynamie Solutions s esas res AA a ES 2 3 3 Comp tation Process ara weeks 2 3 4 Numerical Parameters e e 2 4 Results View and Export x ro Bota a Bee a ae 257 System Preferentes yo e s de a e ede he RS A 3 Case Study 3 1 Introduction s oea ee ea en Se we Gee de ee ak He ee eS 3 2 Field Measurements and Data Acquisition o o o o 3 3 Modeling with TEM
18. ed is most certainly not constant over time Thus a dynamic simulation must be performed and therefore the latter be selected After clicking on the Time Series radio button for QWin in the Sewer Node panel of the Sewer Lines tab you can access a drop down list Initially there are no values in the list because no time series has been added to the TEMPEST model yet To add a new time series click on Add Time Series from the Model menu A dialog box appears that lets you 24 Chapter 3 Case Study Ile Model Calibration tmo File Edit Model Compute View Help BAR 4 BEBE Name Enea Sewer Node Inflow Inflow Temperature Ambient Air Pressure Air Exchange Coeff Sewer Pipe Type Length Nominal Diameter Wall Thickness Slope Ambient Temperature Ambient Rel Humidity COD Degradation Rate FE ag Sewer Lines Time Series si Results Line 1 Specifications QWin m3 s O Constant Value o Time Series 25 02 27 02 08 Infl Twin C Constant Value mu Time Series 25 02 27 02 08 fil TA ec 8 3 phia 0 75 p mbar 966 bE i 0 1 kst lambdaP f Concrete 2 stee x Lim 1845 D m 0 3 s m 0 1 sot 0 00091 r mgcoD m3 sy 2 8 phi TA p Soil Type Penetration Depth deltaS m Soil Temperature TS inf C Figure 3 4 The Sewer Lines tab A new empty sewer line has been
19. ed by clicking the OK button Its name appears in the list on the left of the Time Series tab and is automatically selected The time series now consists of a column for time values only 4 In order to add a column for inflow or inflow temperature click on in the toolbar or open the Model menu and select Insert Column and specify the parameter you want to add to column for and a default value not required Figure 2 8 5 Repeat step 4 to add another column and go to step 2 to add additional time series If new time series are added the appropriate select fields in the Sewer Lines tab will be extended To input and edit values two main methods can be applied The user can use the arrow keys or the mouse cursor to navigate in the table and enter values directly or more easily paste values previously copied from a text file or a spreadsheet application select the upper left cell were the pasted area should start and select Paste from the Edit menu Please consider that TEMPEST expects the date time values to have the same format as specified in the system preferences menu Edit Preferences and that date and time values must be provided If this is not the case TEMPEST tries its best to guess the right format Furthermore consider that data gaps are not allowed and that the date time value must increase with increasing row number The program ignores empty cell at the end of the table The columns for time series parameters do no
20. eentnahme aus Abwasser Institut fiir Bauingenieurwesen V M nchen Lehrstuhl und Pr famt f r Wasserg tewirtschaft und Gesundheitsingenieurwesen Institut f r Siedlungswasserwirtschaft und Abfalltechnik Hannover Cunge J Holly F Verwey A 1980 Practical Aspects of Computational River Hydraulics Pitman D rrenmatt D J 2006 Berechnung des Verlaufs der Abwassertemperatur im Kanalisa tionsrohr Diploma Thesis ETH Zurich D rrenmatt D J Wanner O 2008 Simulation of the wastewater temperature in sewers with TEMPEST Water Science and Technology 57 11 1809 1815 Hager W H 1994 Abwasserhydraulik Theorie und Praxis Springer Berlin Hohmann R Setzer M J Wehling M 2004 Bauphysikalische Formeln und Tabellen Werner Verlag Holman J P 2002 Heat transfer 9 Edition McGraw Hill New York Incropera F P DeWitt D P 2002 Fundamentals of heat and mass transfer 5 Edition Wiley New York Press W H 2005 Numerical recipes in C the art of scientific computing 2 Edition Cambridge University Press Cambridge Scheffer F Schachtschabel P Blume H P 2002 Lehrbuch der Bodenkunde 15 Edi tion Spektrum Akademischer Verlag Heidelberg Unsworth M H Monteith J L 1990 Principles of environmental physics 2 Edition Edward Arnold London VDI 1963 VDI W rmeatlas Berechnungsbl tter f r den W rme bergang VDI Verlag D s seldorf 33 34 Bibliograp
21. flow Temperature TWin Ambient Temperature TA C Ambient Relative Humidity phiA O lt phiA lt 1 Ambient Air Pressure pA mbar pA gt 0 mbar Air Exchange Coefficient b O lt b lt 1l Sewer Pipe Type kst lambdaP aP f Value from library Length i m L gt 0m Nominal Diameter D m DS Om Wall Thickness s m s gt 0m Slope so s0 gt 0 COD Degradation Rate r mg m s Soil Type lambdaS aS Value from library Penetration Depth deltaS m deltaS gt 0 m Soil Temperature TS inf E 2 2 Model Formulation 9 2 2 2 Time Series The inflow of a sewer line can either be described by a constant value or a time series To define a time series the Time Series tab which has a spreadsheet like interface can be used If the user wants to add a new time series for wastewater discharge or wastewater temperature he or she has to perform the following steps 1 Switch to the time series tab click on the Time Series tab or select Time Series in the View menu 2 Select Add Time Series in the Model menu or click on L in the toolbar to open the dialog box showed in Figure 2 7 3 Name the time series specify the initial number of rows corresponds to the number of data point you want to enter and select whether your time series should have a relative or an absolute time scale The former has seconds as unit starting from second zero whereas the latter uses absolute time values e g 25 02 2004 12 00 The time series is then add
22. g the day Oye 500kW from 7 am to 10 pm and O 100kW from 10 pm to 7 am The total amount of heat recovered is the same for both scenarios When heat is recovered from wastewater the wastewater temperature decreases Given the amount of heat extracted O the wastewater temperature and discharge upstream 30 Chapter 3 Case Study of the heat exchanger Ty in and Qw respectively the temperature after the heat exchanger Twou can be calculated using the equation Orec _ 3 1 Cp w Pw Qw Tw out Tw in where cpw stands for the specific heat capacity of water and pw for the density of water In order to investigate the effect of the decrease of Tw ou on the wastewater temperature at the downstream end of the sewer the new time series Twou stored e g in an Excel spreadsheet can be used as input time series for the model developed above To do so add a new time series in the Time Series tab add the column for the wastewater temperature and copy paste the values of Ty ou with the corresponding date time values into the data table proceed as described in the model calibration section Now switch to the Sewer Lines tab and select the new time series as input for the inflow temperature In Figure 3 11 the wastewater temperatures at the downstream end of the sewer manhole RS 3096 calculated for the two scenarios are compared One can see that the first scenario which reclaims a constant amount of heat causes cold d
23. geometrical and material properties were made available by the engi neering company that has built the sewer section Based on the type of concrete that has been used the parameters of the thermal properties and the friction coefficient of the pipe were estimated The thermal properties of the soil were estimated based on the soil dis covered on site Meteorological data were taken from the closest meteorological station The values of these parameters are listed in Table 3 1 RS 4943 RS 4943 a RS 3096 a g 02 g 02 wu v D Do 5 5 01 501 a a 0 0 26 Feb 2008 27 Feb 2008 28 Feb 2008 11 Mar 2008 12 Mar 2008 13 Mar 2008 14 14 Y Y g g 5 12 5 12 t t Qa o E 10 E 10 2 RS 4943 Ao RS 4943 RS 3096 RS 3096 8 8 26 Feb 2008 27 Feb 2008 28 Feb 2008 11 Mar 2008 12 Mar 2008 13 Mar 2008 Figure 3 2 Discharge and temperature data measured from February 26 to February 28 left column and from March 11 to March 13 right column at the manholes RS 4943 and RS 3096 22 Chapter 3 Case Study Table 3 1 Geometrical and meteorological parameters and material properties that describe the sewer sec tion RS 4943 RS 3096 in Riimlang Estimates were made where no data was available Key to Origin column IP implementation plan E estimate AS ANETZ meteorological station L literature cf Appendix B Parameter Lengt
24. h Nominal diameter Wall thickness Sewer slope Friction coeff Fouling factor COD degradation rate Soil penetration depth Air exchange coeff Ambient temperature Ambient air pressure Relative humidity Soil temperature Thermal conductivity pipe Thermal conductivity soil Thermal diffusivity pipe Thermal diffusivity soil Reinforced concrete Symbol L D S So kst f s b Ta Value 1845 0 9 0 1 0 0091 70 200 0 6 2 8 0 1 0 1 8 3 966 0 75 5 5 0 3 2 5 0 25 2 5 0 4 0 64 0 3 0 8 Gravel with a porosity of 50 and 50 saturated Unit Origin m IP m IP m IP IP miga E W m K L mgCOD m s E m E E C AS mbar AS AS C M E W m K IP L W m K IP L m s IP L m s IP L 3 3 Modeling with TEMPEST 23 fa TEMPEST File Edit Model Compute View Help BEE 4 BESE Figure 3 3 Main window of TEMPEST as it shows up when the TEMPEST executable file is run 3 3 Modeling with TEMPEST In order to simulate the effect of heat recovery at the upstream site on the downstream wastewater temperature a new model has to be setup and the parameters of the sewer section and the time series data must be entered into the model To do so run the TEM PEST executable file in Windows The main window as shown in Figure 3 3 appears After selecting New from the File menu a model is created but does not yet contain any sewer line or time series Since the relevant data nee
25. hematically Conduits represent sets of mass balance equations and Nodes repre sent sets of continuity conditions of the variables of successive conduits The variables considered are discharge and temperature in the wastewater compartment airflow tem perature and humidity in the sewer headspace and temperature in the sewer pipe and the surrounding soil The hydraulics is modeled with the St Venant equations Cunge et al 1980 and the airflow with a recently developed model for circular pipes The heat and mass transfer processes considered in the model are shown in Figure 2 3 The rate expressions of the transfer processes were taken from the literature e g Holman 2002 Incropera and DeWitt 2002 Wanner et al 2004 The complete mathematical model im plemented in TEMPEST is described and discussed in two publications D rrenmatt 2006 and Wanner and D rrenmatt in preparation an overview is given in Appendix A 1 The following sections explain how to input sewer lines and read in time series data Unsaturated soil Sewer line O Node mw Conduit Flow direction A i Air exchange Figure 2 2 Discontinuities require that nodes are introduced which separate sewer conduits having different properties 6 Chapter 2 Program Handling Atmosphere Wastewater Airflow Sewer air Wastewater discharge Se ul E Sewer pipe gt Heat transfer ile er Soil type 1 gt M
26. hy Wanner O D rrenmatt D J in preparation Heat recovery from sewers and its effect on the influent temperature of wastewater treatment plants Wanner O Panagiotidis V Clavadetscher P Siegrist H 2005 Effect of heat recovery from raw wastewater on nitrification and nitrogen removal in activated sludge plants Water Research 39 19 4725 4734 Wanner O Panagiotidis V Siegrist H 2004 W rmeentnahme aus der Kanalisation Einfluss auf die Abwassertemperatur Korrespondenz Abwasser 51 5 489 495 A Appendix Additional Material This appendix contains additional material mainly on the analytical model model equations and transfer processes implemented in TEMPEST Additionally the algorithm used to numerically solve the partial differential equations is introduced 35 36 Appendix A Additional Material A 1 Analytical Model The sewer system is modeled based on two basic elements conduits and nodes Con duits in which the wastewater discharge airflow water vapor and temperature are contin uous functions in time and space and are modeled by one dimensional balance equations The element conduit is assumed to represent a prismatic pipe with circular cross section and without any discontinuity Nodes describe discontinuities caused by lateral inflows head space openings sudden changes of the sewer geometry or of material properties and are modeled by continuity conditio
27. ifferent pipe types is listed in Table B 2 Table B 2 Values of the thermal conductivity A and the temperature diffusivity a for different pipe ma terials Thermal conductivity Temperature diffusivity 2 10 4 By Concrete Medium density 1800 kg m 1 15 0 64 Medium density 2000 kg m 1 35 0 68 Medium density 2200 kg m 1 65 0 75 High density 2400 kg m 1 65 0 75 Reinforced 1 steel 2 30 1 00 Reinforced 2 steel 2 50 1 04 Brick Clay 1 00 0 63 Concrete 1 50 0 71 Brick Masonry outside wall 0 64 Masonry inside wall 0 52 Brick 0 38 0 52 Masonry saturated with water 0 60 Cement hardened 0 69 Concrete Reinforced concrete 1 12 Gravel concrete 0 95 Slag concrete masonry 0 52 Concrete saturated with water 1 Hohmann et al 2004 VDI 1963 Baehr and Stephan 2006 dBischofsberger and Seyfried 1984 44 Appendix B Material Properties B 3 Thermal Conductivity and Temperature Diffusivity of Different Soil Types In Table B 3 the thermal conductivity A and temperature diffusivity a of different soils are listed Table B 3 Values of the thermal conductivity A and the temperature diffusivity a for different soil types Water content Heat conductivity Temperature diffusivity 21 10 6 Gravel coarse 0 52 Gravel crushed rock 0 37 Sandy soil 0 0 0 30 0 24 40 pore space 0 2 1 80 0 85 0 4 2 20 0 74 Sandy soil dr
28. imulations has now been entered If you want to verify whether the model is complete select Validate Entries and Data from the Model menu Before starting the model calibration it is recommended to save the model file to disk File Save 26 Chapter 3 Case Study Ma Model Calibration tmo File Edit Model Compute View Help DEG 46 03458 vs a Lt Sewer Lines El Time Series pe Results Name Date Time Inflow m3 s Inflow Temperature C Messin Ganpson ts Zr OZ 200e 0837700 UUSII TZ 8Z491 27 02 2008 08 38 00 0 032701 12 78423 27 02 2008 08 39 00 0 031961 12 79247 27 02 2008 08 40 00 0 032182 12 76375 27 02 2008 08 41 00 0 032851 12 84034 27 02 2008 08 42 00 0 034719 12 83555 27 02 2008 08 43 00 0 035258 12 87385 27 02 2008 08 44 00 0 034699 12 87385 27 02 2008 08 45 00 0 03449 12 84512 27 02 2008 08 46 00 0 034576 12 91533 27 02 2008 08 47 00 0 034892 12 84513 27 02 2008 08 48 00 0 034086 12 94565 27 02 2008 08 49 00 0 032915 12 89779 27 02 2008 08 50 00 0 032501 12 91693 27 02 2008 08 51 00 0 031985 12 89779 Figure 3 6 The data table of the series named Measuring Campaign 26 2 28 2 2008 is filled with data copied from a spreadsheet application Model Settings Pipe Types Soil Types Numerics Label Friction Coeff Heat Cond Thermal D Concrete 2 st 70 2 c K Label Friction Coeff m 0 33 s 70 Heat Conductivity w mkK 2 Therma
29. in the TEMPEST file Model name tmo 2 4 Results View and Export The results of a computation are displayed in the Results tab Figure 2 17 It consists of five sub panels 16 Chapter 2 Program Handling Table 2 3 List of the numerical parameters needed by simulations performed with TEMPEST Parameter Description Default value Newton Raphson Iterator Tolerance for Convergence Pipe Soil Module Structure Number of Pipe Layers PDE Solver Max Courant Number Rel Tolerance Steady State Max Iterations Steady State Stepsize in Space Stepsize in Time Apply Filter This is a relative value which spec ifies the accuracy that must be ob tained in order to stop the iteration Defines the discretization of the pipe in radial direction the num ber of radial layers the pipe will be divided in If the step size in time is auto computed the solver tries to keep the Courant number lower than the given value Defines the relative tolerance which must be reached for the solver to abort the relaxation for steady state If steady state cannot be reached the solver will stop after this num ber of steps Defines the spatial discretization along the flow path in the sewer The discretization in time can either be automatically computed or spec ified by the user To improve stability and avoid nu merical artifacts a non linear fil ter can be applied to suppress high frequency oscillations
30. l Diffusivity 1E 6 m2 s 0 4 Fouling Factor W m2K 200 La OK Cancel Figure 3 7 Pipe Types library with an entry for the R mlang sewer pipe type Since no exact values were available rough estimates have been typed in 3 3 Modeling with TEMPEST 27 3 3 1 Model Calibration In the model calibration phase one tries to change to model parameters within certain limits so that the output of the model the simulated wastewater temperature at the lower end of the sewer section matches the measured temperature as good as possible The time series of the measuring period from February 26 to February 28 2008 Figure 3 2 page 21 will be used for the calibration You can first try to calculate the downstream wastewater temperature using the parameters given in Table 3 1 Click on Compute Dy namic Solution to open the Compute Solution dialog The option Compute dynamic solution is preselected and start and stop times are set to the first and the last time series value defined in the time series used in the model Figure 3 8 The Output Timestep s can be changed in order that more or less calculated data are saved as results Usually there is no need to change the numerical settings see Section 2 3 4 If your sewer system is rather long but not very dynamic and complex you can try to increase the Stepsize in Space m in order to speed up the calculation time click on Numerical Settings If you want to fi
31. midity must be between O and 1 Value 5 3823 Penetration depth must be greater than O m Value O m 4P Numerics Stepsize in space must be positive Value O m Figure 2 12 Dialog box that displays errors Compute Solution Compute steady state solution Reference Time Date Time 25 02 2008 00 00 00 O Compute dynamic solution 25 02 2008 00 00 00 27 02 2008 16 28 20 se Numerical Settings Cancel Figure 2 13 Dialog box for defining computation type and reference time to compute a steady state solution 14 Chapter 2 Program Handling TEMPEST uses linear interpolation algorithms to calculate values between known data The first value of the time series is taken for calculation times earlier than the first data value and vice versa for computation times beyond the last data value of a time series 2 3 2 Dynamic Solutions In Figure 2 14 the dialog box of Figure 2 13 is set up for a dynamic calculation Here the same rules for time series apply as discussed in the previous paragraph It has to be noted that TEMPEST automatically calculates a steady state solution first in order to generate initial conditions for the dynamic simulation The reference time for the calculation of the steady state solution is set equal to the start time of the dynamic simulation In addition to a Start Time and an End Time one can also choose the Output Timestep It specifies the tempor
32. nce equations Table A 1 Process wL Qwz Tw Tz GsNb KsNb Tin f a A Zur 1 9 7 dew kpw T h Pb w Gsn Ksn Ti f a if a grt kp 7 T Pt L Jo csB FCSB dvp Op Psat Tw PL JyP hrg dvp 2 1 dkP Opt p Psat zh Je h que Gut hfe PL x Xsat JE PL x Xsat Description Convective heat transfer from the compartment wastew ater to the compartment sewer head space Heat flux from the surrounding soil with temperature Ts inf to the outermost lower pipe layer segment N Heat flux from the lower pipe layer segment 1 to the lower pipe layer segment j Heat flux from the lower pipe layer segment 1 the com partment wastewater Heat flux from the surrounding soil with temperature Ts inf to the outermost upper pipe layer segment N Heat flux from the upper pipe layer segment 1 to the upper pipe layer segment j Convective heat transfer from the innermost lower pipe layer segment the the compartment sewer head space Heat produced by biochemical activity in the compart ment wastewater Heat transfer due to evaporation condensation at the water air interface Mass flux between the compartments wastewater and sewer head space caused by evaporation condensa tion Mass flux due to condensation at the innermost upper pipe layer Loss of water vapor caused by condensation at the i
33. nd the optimal stepsize start with a rather long stepsize and perform the calculation Now decrease to stepsize gradually until the results of the actual and the last calculation do not differ significantly By clicking on Run the calculation process starts A progress bar keeps you informed about the ongoing process First the steady state solution is calculated using the time series values at the specified start time The steady state solution is then used as initial condition for the calculation of the dynamic solution Since the heat transfer processes in the sewer pipe and the soil are rather slow the relaxation time of the steady state calculation might be shorter than the time needed A possible way to cope with this phenomenon is to duplicate the input time series and perform a simulation e g for 6 days instead of the original 3 days of available input data and use the results of the last 3 days of the simulation only When the calculation process has finished you can close the progress dialog TEMPEST will now show the Results tab To compare the values of the calculated and measured variables at the downstream end of the sewer the calculated variables must be transferred to a data analysis or spreadsheet application You can either export the data by using the Data Export dialog File Export Data cf Section 2 4 or first choose the row x 1845 00 m in the Parameter vs Time list lower left then select the variable
34. nner most upper pipe layer Condensation in the compartment sewer head space because of oversaturation reduction of the latent heat Condensation in the compartment sewer head space because of oversaturation reduction of the content of water vapor For a mathematical formulation of the parameters OL ksyb ksn kp Kew krpi kpr Op and Qxp please consult D rrenmatt 2006 gt For a more accurate calculation of the pipe temperature the pipe is discretized in N radial layers where the innermost layer is layer j 1 the outermost j N Each layer is then further divided in two segments the lower segment Pb interfaces the compartment wastewater the upper segment Pr the compartment sewer head space A 1 Analytical Model 39 Table A 3 Nomenclature used in Table A 1 and A 2 Symbol Description Geometrical Variables An cross section area n W for water L for air and P for a pipe layer P Water level width Un Wetted perimeter n W for water L for air and P for a pipe layer Material Properties Com Specific heat capacity n W for water L for air P for the pipe and S for soil Ts inf Temperature of the undisturbed soil An Thermal conductivity n W for water L for air P for the pipe and S for soil Pn Density n W for water L for air P for the pipe and S for soil Transfer Processes j Mass transfer gq Heat transfer Heat and Mass Transfer Coefficients k Thermal tra
35. ns Complex sewer systems can be modeled by series of basic elements node plus conduit which is called sewer line The compartments considered in the model are wastewater sewer head space sewer pipe and surrounding soil The compartments and the transport heat and mass transfer pro cesses considered in the model are indicated in Figure A 1 Below a short overview of the analytical model including the balance equations the equa tions for the process rates and details on the modeling of the nodes is given For more detailed explanations the reader is referred to D rrenmatt 2006 and Wanner and D r renmatt in preparation Wastewater Sewer air Sewer pipe Surrounding soil gt Heat transfer gt Mass transfer NA a Heat sources sinks Figure A 1 Cross section of a sewer line with transfer processes by which the humidity in the sewer headspace and the temperatures in the wastewater sewer headspace sewer pipe and soil are affected A 1 Analytical Model 37 Table A 1 Balance equations which are solved by TEMPEST The mathematical formulation of the heat and mass transfer processes j and q can be found in Table A 2 The underlying assumptions are discussed in depth in D rrenmatt 2006 and Wanner and D rrenmatt in preparation The mass and the momentum balance equations of water are known as the St Venant equations Mass Balances Water discharge Ow 2er Qu j
36. nsmission coefficient rcoD Biological degradation rate a heat transfer coefficient Miscellaneous Variables ecoD Reaction enthalpy of COD degradation g Gravitational force hfe Evaporation enthalpy PL Partial pressure of water Psat Partial pressure of water of saturated air So Sewer slope Sr Friction slope Ta Ambient temperature Kg Water vapor loading of saturated air 40 Appendix A Additional Material A 2 PDE Solver To solve the system of one dimensional partial differential equations the two step Lax Wendroff scheme an explicit finite volume scheme which is second order in both space and time O Ax Ar is used It omits excessive numerical dispersion has no amplitude dissipation and no instabilities due to mesh drifting A one dimensional initial value problem can be written in a flux conservative form as du F u dt 0x S u A 6 where u state variables F flux terms and S source terms are vectors The two step Lax Wendroff method calculates interim values u at half time steps 1 1 1 2 17 j At j j o 7 a 1 FR Fi Fi A 7 The fluxes P can be calculated using e similar for o and FI a and finally the values u l at the full time step can be calculated by H y ME pY pia py gi A 8 MA le aha j After evaluating ul 1 the interim values T and AS can be discarded To assure stability the Courant Friedrichs Lewy CFL criterion must be met Press 2005 Since the St Venan
37. odel the sewer system with TEMPEST is dis cussed in Section 3 2 Then it is illustrated how the considered section of the R mlang Figure 3 1 Top view of a section of the sewer between the villages R mlang and Oberglatt in the Canton of Zurich Switzerland 20 3 2 Field Measurements and Data Acquisition 21 sewer system can be modeled in TEMPEST Section 3 3 and how calibration Section 3 3 2 and validation Section 3 3 2 are performed using the available data Two exem plary heat recovery scenarios are simulated in Section 3 3 3 Finally some conclusions drawn from the project are addressed in Section 3 4 3 2 Field Measurements and Data Acquisition In order to model the considered sewer section with TEMPEST geometrical as well as material properties must be known Further discharge and temperature measurements and meteorological data must be measured gathered or estimated for the period in time which will be used for model calibration and validation For the R mlang sewer data were measured by an ultrasonic flow meter and a tempera ture logger mounted at RS 4943 manhole and a temperature logger in manhole RS 3096 TEMPEST automatically calculates the hydraulics The measured discharge and temper ature data are plotted in Figure 3 2 To get a good estimate for the ground temperature a temperature logger was buried in 1 2 m depth and in 2 m distance from the sewer at manhole RS 4943 Information on the
38. of calculated variables 0 001 0 95 5x10 10 000 10 metres autocompute Enquist 2 1 2 4 Results View and Export 17 HR Sim 2 Winter V1 tmo File Edit Model Compute View Help ta to E FE Sewer Lines L Time Series gt Resuts l Parameter vs Space wre lol Data Series t 25 02 2008 00 00 tfc i J t 25 02 2008 00 10 Tswi c i t 25 02 2008 00 20 JSwZ2 ec oe t 25 02 2008 00 30 rw t 25 02 2008 00 40 t 25 02 2008 00 50 pero t 25 02 2008 01 00 phi t 25 02 2008 01 10 aL t 25 02 2008 01 20 3 TS bottom t 25 02 2008 01 30 N j i _ TS top t 25 02 2008 01 40 s i Fe t 25 02 2008 01 50 b 25 02 2008 NNN Parameter vs Time x 0 00m x 295 00 m x 354 00 m x 409 00 m all x 468 00 m 0 0e 000 5 0e 004 1 0e 005 1 5e 005 2 0e 005 x 526 00 m Time s e x 585 00 m x 644 00 m gt 201 x 700 00 m B c D E FE x 760 00 m ine water Level Water Discharge Water Temperature Air Temperature Rel Humidity Pipe Layer 1W Pipe Layer 14 Pipe Layer an Im mals PC PC El Pc PC PC x 916 00 m 0 1078 0 0151 13 0279 10 1323 0 9091 12 4597 5 6196 11 9617 x Ka 600 0 1054 0 0143 13 0328 10 1125 0 9083 12 4605 5 6219 11 9618 se 1200 0 1048 0 0142 12 9730 10 0841 0 9085 12 4447 5 6232 11 9590 x 1121 00 m 1800 0 1084 0 0156 12 8414 10 0368 0 9106 12 4064 5 6247 11 9472 x 11
39. ons for down stream wastewater treatment plants 3 4 Conclusions 31 600 600 z z 3 400 400 U E a E E E 2 200 200 s E H o H o 11 Mar 2008 12 Mar 2008 13 Mar 2008 11 Mar 2008 12 Mar 2008 13 Mar 2008 04 04 03 03 eB cB oD Dd 0 2 0 2 lt lt Y Y a 0 1 a 0 1 0 11 Mar 2008 12 Mar 2008 13 Mar 2008 11 Mar 2008 12 Mar 2008 13 Mar 2008 T T E E oO w _ Der gt _ _ U 0 E E w 0 E 5 Rs 11 Mar 2008 12 Mar 2008 13 Mar 2008 11 Mar 2008 12 Mar 2008 13 Mar 2008 Figure 3 11 Simulation results of two alternative heat recovery scenarios In the first scenario left column 350 kW heat were constantly reclaimed from the wastewater In the second scenario right column 350 kW of were reclaimed from 7am to 10pm and 100 kW in the remaining time The discharge measured at the heat exchanger is plotted in the middle row In the third row the temperature before the heat exchanger dashed black line the temperature after the heat exchanger gray line and the downstream temperature bold black line are indicated Bibliography AWEL 2003 W rmenutzung aus Abwasser Kanalisation ungereichnigt und ARA Abl ufe gereinigt AWEL Standard Baehr H D Stephan K 2006 W rme und Stoff bertragung 5 Edition Springer Berlin Bischofsberger W Seyfried C 1984 W rm
40. ownstream temperature peaks at times where the discharge in the sewer is low though there is even some warming up during the flow time in the sewer In contrast the diurnal variation of the downstream temperature calculated for the second scenario is much smaller Additional heat recovery scenarios could start from the second scenario and try to further increase the amount of heat extracted or to optimize the pattern of the diurnal extraction profile 3 4 Conclusions By means of a simple and fictitious example a procedure to use TEMPEST for planning of heat recovery projects is proposed To achieve this the data needed in order to model the sewer section has to be acquired Model parameters can be measured on site or found in the literature When data is miss ing even a priori estimates may be used Because TEMPEST yields model predictions whose accuracy depend on the accuracy of the input data available sensitivity analyses can be performed in order to assess the accuracy of the predicted results Model calibration is used to adjust the model parameters to the specific conditions of the sewer considered and model validation serves to test the reliability of the model predic tions Once the model is set up and tested it can be used e g to estimate the effect of heat recovery on the downstream wastewater temperature to analyze alternative patterns of heat recovery with diurnal variation or to identify potentially critical conditi
41. p P Sewer airflow Qz Jn _ 224 Water vapor loading X Ar X A a GiveP jke Ur ju AL Heat Balances Water temp Ty At Fw HQw Tw o zy dw Uw dw P dyp P 4a Aw Sewer air temp Tz nn aam gt 5 pr UL Gwe P 6 der AL Pipe layer temp TY aa 19 GC us q a Momentum Balance d 2 d Water Ou 2 Q g Aw 8 Aw So Sp A 1 1 Balance Equations and Transfer Processes The balance equations for mass heat and momentum are given in Table A 1 The transfer processes used in the balance equations are described in detail in Table A 2 p 38 Please notice that the nomenclature used in this section is explained in Table A 3 p 39 A 1 2 Nodes If the sewer is modeled as a series of conduits additional continuity conditions must be fulfilled at the nodes between them Continuity of Ty T and X requires that at the nodes i 2toN Ow Twi Owi 1 Twi Owin i Twin A 1 and el Dr A 2 Or XLi Oria XLi Q i Xj A 3 if air is inhaled at the node 9 gt 0 or Ti TL i 1 4 4 XLi XLi 1 A 5 if air is exhaled at the node O lt 0 In these equations Qwin and Twin are the dis charge and water temperature of a lateral inflow respectively and T and X4 are the temperature and water vapor loading of the ambient air respectively 38 Appendix A Additional Material Table A 2 Mathematical description of the transfer processes used in the bala
42. production of heat by biological processes and the Penetration Depth describes the depth to which the soil temperature TS inf is assumed to be affected by the sewer Table 2 2 lists all parameters needed to model a sewer system with TEMPEST and Figure 2 5 shows a screenshot of the Sewer Lines form with typical parameter values To add a new sewer line click on fo in the toolbar or alternatively select Add Sewer Line from the Model menu A dialog box as shown in Figure 2 6 appears You can now name the series and if desired choose a sewer line whose parameter values should be copied and used for the new sewer line Confirm by clicking the OK button Sewer lines can be deleted by using the Remove Sewer Line command from the Model menu or by using the a button in the toolbar 2 2 Model Formulation 7 Mz unnamed1 File Edit Model Compute View Help es Time Series Results pecifications Sewer Node Inflow Inflow Temperature Ambient Temperature Ambient Rel Humidity Ambient Air Pressure Air Exchange Coeff Model tabs Quin m3 5 mstant Twin C TA C phia pA mbar bE phid TA p N deltas ITSint lambdaS Sewer Pipe Soil Type Kst lambdaP f Type lambdas Length L m Penetration Depth deltas m Nominal Diameter D m Soil Temperature TS inf C Wall Thickness s m Slope 50 COD Degradation Rate r mgCOD m3 s
43. py The calculated values can also be exported for external use Figure 2 18 shows the Data Export dialog box File menu Export Data in which the variables time and location as well as output format can be specified One can export spatial data for a given time step time series data at a fixed coordinate or all data together 2 5 System Preferences System preferences such as plot pen colors and the default date time format can be changed in the Preferences dialog box Figure 2 19 To open it select Preferences in the Edit menu Data Export Data Columns Y h Water Level QW Discharge Y Tw water Temperature TL Sewer Air Temperature phi Sewer Rel Humidity QL Sewer Airflow Data Selection all O Fixed Time t 0 005 O Fixed Distance x 0 00m File Format Comma Separated Tab Delimited Export Cancel Figure 2 18 Dialog box for selecting data to export 2 5 System Preferences Preferences General Display Date Time Format 5 DD MM YYYY hh mm ss 23 12 2007 12 26 03 O YYYY MM DD hh mm ss 2007 12 23 12 26 03 Ommyopjyvvy hh mm ss 12 23 2007 12 26 03 Plot Pen Colors h Water Level 1 TW Water Temp h Profile Elevation Ol TL Air Temp Qu Discharge phi Rel Humidity QL Airflow O TSW Pipe Layer Temp bottom TSA Pipe Layer Temp top Restore Defaults Figure 2 19 Dialog box for editing the system preferences 3
44. row by clicking on its header in the data table lower right and select Copy from the Edit menu If the parameters Ts in Ap ap As as and ds are systematically changed within the ranges indicated in Table 3 1 on page 22 a good correspondence between the temperatures sim ulated and measured in R mlang is achieved using the values listed in Table 3 2 The measured and calculated time series are plotted in Figure 3 9 28 Chapter 3 Case Study Compute Solution O Compute steady state solution 25 02 2008 00 00 00 Start Time Date Time 25 02 2008 00 00 00 Stop Time Date Time 27 02 2008 16 28 20 Output Timestep s 600 Run Numerical Settings Cancel Figure 3 8 The Compute Solution Dialog Table 3 2 Parameters changed during the model calibration and values yielding good correspondence of measured and calculated data Parameter Symbol Value Unit Soil penetration depth Os 0 11 m Soil temperature TS inf 5 5 C Thermal conductivity pipe Ap 2 3 W m K Thermal diffusivity pipe ap 0 5 m s Thermal conductivity soil As 0 7 W m K Thermal diffusivity soil as 0 6 m s 15 T 14 A ech po M i Wih A LAN I mn MM i LV Mama 13 VA Ww py TAWA Ki ify 9 1 a 12 5 w a E 11 7 oO ER 10 gL RS 4943 RS 3096 measured RS 3096 calculated 8 I 26 Feb 2008 27 Feb 2008 28 Feb 2008 Figure 3 9 Result of the model calibration 3 3
45. s dialog box Figure 2 12 appears Each error is listed its category is illustrated with an icon and the sewer line parameter the time series data point or the setting where it occurred is given Once all the errors are corrected computations can be performed 2 3 1 Steady State Solutions The same dialog box is used to compute both steady state and dynamic solutions A radio button allows changing the type of calculation within the dialog box A configuration ready to perform a steady state simulation is shown in Figure 2 13 If time series are used in the model care has to be taken when filling in the Reference Time The reference time defines the link between the simulation time in the program and the data for inflow inflow temperature or both If all time series used have a relative time scale the reference time has the unit seconds If all time series used have an absolute time scale the reference time must have a value as illustrated in Figure 2 13 If relative and absolute time scales are mixed an absolute date and time must be specified and the time series with relative time scales are shifted to start with the reference time specified 2 3 Computation 13 Verification Errors amp The following errors have been encountered while verifying the model RS5060 Time scale values must increase with increasing row number Row 78 Column 3 _ RS5060 Data gap detected Row 1728 Column 3 ra 3922 Relative hu
46. t equations contain a non linear term it is advisable to apply a filter that suppresses oscillations caused by waves of short wavelength in order to improve sta bility and avoid numerical artifacts after each time step In TEMPEST the Enquist 2 1 filter is implemented for this purpose B Appendix Material Properties In this appendix a collection of different parameter values to de scribe the thermal properties of the most common soils and build ing materials and friction coefficient for different pipe materials is provided The values are collected from the literature 41 42 Appendix B Material Properties B 1 Friction Coefficients k In order to describe the wastewater discharge a friction coefficient is needed A selection for several pipe materials is given in Table B 1 Table B 1 Friction coefficients ks Manning Strickler for different pipe materials by Hager 1994 Condition ka m P4s7 Pipes Asbestos cement pipe 67 91 Brick 58 77 Cast iron pipe new cemented 67 91 Concrete monolithic smooth 70 83 rough 58 67 Concrete pipe 67 91 Plastic pipe smooth 70 90 Fire clay pipe 70 90 Canals Coated with Asphalt 60 77 Brick 55 83 Concrete 50 90 B 2 Thermal Conductivity and Temperature Diffusivity of Different Pipe Types 43 B 2 Thermal Conductivity and Temperature Diffusivity of Different Pipe Types A selection of the thermal conductivity and temperature diffusivity values for d
47. t need to have the same number of rows but the number of rows of the time scale column must be at least as high as the length of the longest data column In Figure 2 9 the screenshot of the Time Series tab of a model which contains one time series 25 02 27 02 08 is shown The time series has an absolute time scale and values for both inflow discharge and inflow temperature are specified 10 Chapter 2 Program Handling Add Time Series Name Series A Initial Number of Rows 100 Time Scale Relative v Relative Absolute Date Time Figure 2 7 Dialog box for adding Time Series Add Time Series Column Parameter Lateral Inflow m3 s iv Default Value OK Cancel Figure 2 8 Dialog box for adding new columns ln HR Sim 2 Winter 1 tmo File Edit Model Compute View Help BARB 5 BEBZ ys a Sewer Lines Time Series Im Results Name Data Date Time Inflow m3 s Inflow Temperature C 25 02 27 02 08 1 0 015147 13 16581 25 02 2008 00 01 00 0 015077 13 21365 25 02 2008 00 02 00 0 014989 13 18973 25 02 2008 00 03 00 0 014307 13 16581 25 02 2008 00 04 00 0 014284 13 14668 25 02 2008 00 05 00 0 014294 13 1706 25 02 2008 00 06 00 0 014497 13 13232 25 02 2008 00 07 00 0 013789 13 13232 25 02 2008 00 08 00 0 014403 13 12274 25 02 2008 00 09 00 0 014127 13 12754 25 02 2008 00 10 00 0 01465 13 11796 25 02 2008 00 11 00 0 014685 13 12327 25 02 2008 00 12 00 0 014796 13 12275
48. the respective button indicated in Figure 2 10 In order to edit one of the types select it in the list on top of the dialog box and change the parameter values in the text fields below The changes are applied if the user clicks on the OK button Model Settings p Pipe Types Soil Types Numerics Label Friction Coeff Heat Cond Thermal Diff Concrete 2 70 2 3 D 5 Label Concrete 2 reinfi Friction Coeff m 0 33 s Remove type Heat Conductivity wim Add type Thermal Diffusivity 1E 6 m2 s 0 5 Fouling Factor W m2K 200 La La OK Cancel Figure 2 10 Pipe Types tab of the Model Settings dialog box 12 Chapter 2 Program Handling Model Settings ez Pipe Types Soil Types Numerics Label Heat Cond Thermal Diff Gravel 0 7 0 6 Label Heat Conductivity w mk Thermal Diffusivity 1E 6 m2 s OK Cancel Figure 2 11 Soil Types tab of the Model Settings dialog box 2 3 Computation If the model formulation is complete calculations can be performed To open the Compute Solution dialog box either select Steady State Solution or Dynamic Solution in the Compute menu or click on 2 in the toolbar Before the Compute Solution dialog box is displayed TEMPEST verifies the model the verification can also be manually started by clicking on Y in the toolbar or by selecting Validate Entries and Data in the Model menu If errors are found the Verification Error
49. ummary of the model equations see Appendix A 1 Appli cations range from simple steady state estimates of the changes of the wastewater tem perature in a single sewer line to full scale simulations of the dynamics of the wastewater temperature in successive sewer lines with lateral inflows 1 2 Program Capabilities TEMPEST targets high applicability in practice The implemented model requires as few input data and a priori estimates as possible TEMPEST applies default values wherever 2 Chapter 1 Introduction possible and disposes of a library that already contains parameter values for the most com mon soil and pipe types Generally it prefers parameters that are either easy to measure or available from the literature and internally converts them into the appropriate format However the model includes all processes which are of primary importance Its clear and intuitive graphical user interfaces make it easy to handle Some among the major capabilities are e The sewer hydraulics is modeled with the dynamic form of the St Venant equations Cunge et al 1980 Therefore the user has to provide upstream discharge data only To model parameters such as temperature humidity and airflow balance equa tions which form a system of one dimensional partial differential equations PDEs are used for an overview see Appendix A 1 e To yield high accuracy and numerical stability the two step Lax Wendroff algorithm Press 2005 is
50. y 0 27 Sandy soil moist 0 58 Clay soil 0 0 0 25 0 18 40 pore space 0 2 1 18 0 53 0 4 1 58 0 51 Clay soil 1 28 Peat soil 0 0 0 06 0 10 80 pore space 0 4 0 29 0 13 0 8 0 50 0 12 Humus 0 25 Soil 2 00 Baehr and Stephan 2006 PVDI 1963 Unsworth and Monteith 1990 dScheffer et al 2002 Bischofsberger and Seyfried 1984 C Appendix TEMPEST Development 46 Appendix C TEMPEST Development C 1 Program Versions and Changelog The TEMPEST version history together with a documentation of the major changes is given in Table C 1 Table C 1 TEMPEST changelog Version Release Date Major Changes 1 01 29th October 2008 Initial public release 1 02 8th December 2012 Fix Condensation process in air compartment wrong under certain hydraulic conditions low discharge to gether with small pipe diameter

TEMPEST

Contents

Download Pdf Manuals

Related Search

Related Contents