This tutorial demonstrates the use of HyperStudy to perform a DOE study with Madymo. The problem is a crash occupant safety application. The files needed for this tutorial are located in <install_directory>/tutorials/hst/Madymo/, and can be copied to your working directory.
Save both of the following files in the same directory:
Input file for Madymo.
Batch file for Madymo 61.
In this tutorial, you will:
|•||Use HyperStudy Editor to create the design variables.|
The design variables are the scale of force-penetration curve, the scale of recliner loading function and the position of the head rest. The objective is to study the effect of changing the dummy-seat contact stiffness, seat recliner loading function and the distance between the dummy head and the head rest on the dummy's neck injury criteria.
Step 1: Use HyperStudy Editor to create the design variables
|2.||Click Utilities from the top menu.|
|3.||Select HyperStudy Editor… to open the editor.|
|4.||Click File from the top menu of the editor.|
|5.||Select Open… to read in the Madymo input deck, a_hffoam_TEST.xml.|
|6.||Select Find… to search for FUNC_USAGE.|
|8.||Highlight 1.0 of Y_SCALE.|
|9.||Right click to activate the pop up menu and select Create Design Variable.|
|10.||In the Design Variable Properties menu, enter Recliner for the Name: field.|
|11.||Enter 1.0 for the Initial value:, 0.7 for the Lower bound:, 1.3 for the Upper bound:, %g for the Format: and click Ok.|
|12.||Use Find to search for BODY = "Hrest".|
|13.||Highlight 0.02 of POS = "0.02 0.0 0.22".|
|14.||Right click to activate the pop up menu and select Create Design Variables….|
|15.||In the Design Variable Properties menu, enter Fact1Level for the Name: field.|
|16.||Enter 0.02 for the Initial value:, 0.00 for the Lower bound:, 0.04 for the Upper bound:, %g for the Format: and click Ok.|
|17.||Use Find to search for FUNC_USAGE.|
|18.||Highlight 1.0 of Y_SCALE.|
|19.||Right click to activate the pop up menu and select Create Design Variables….|
|20.||In the Design Variable Properties menu, enter contactHR for the Name: field.|
|21.||Enter 1.0 for the Initial value:, 0.8 for the Lower bound:, 1.2 for the Upper bound:, %g for the Format: and click Ok.|
|22.||Use Find to search for FUNC_USAGE.|
|23.||Highlight 1.0 of Y_SCALE.|
|24.||Right click to activate the pop up menu, select Create Design Variables…. and select contactHR in the Define in field. This will apply the same design variable.|
|25.||Use Find… to search for FUNC_USAGE.|
|26.||Highlight 1.0 of Y_SCALE.|
|27.||Right click to activate the pop up menu, select Create Design Variables…. and select contactHR in the Define in field. This will apply the same design variable|
|28.||Click the Show/Generate Templex icon .|
|29.||A tpl file is created. Save it as hffoam_TEST.tpl.|
Step 2: Set up a study
|1.||In the HyperStudy window, click Add Study….|
|3.||In the HyperStudy – Add Study window, type a name for a study or select the default name and click OK.|
|4.||Under the Study Directory select the working directory.|
|7.||In the HyperStudy – Add Model window, type in a model name or select the default name.|
|8.||Verify the Model Type: as Template and click OK. |
|9.||Under the Template file select the saved parameterized deck, a_hffoam_TEST.tpl.|
|11.||In the Design variables page, the design variables defined in the a_hffoam_TEST.tpl will be read in.|
Step 3: Perform the base run
|1.||Click Tools from the top menu.|
|2.||Select Register Solver Script… to register the Madymo 61 script, Madymo61.bat.|
|3.||Select Add… in Register Solver Script window. |
|4.||In the HyperStudy-Register Script window, type in a name and variable for the Madymo script and click on Script to select Madymo61.bat.|
|5.||Click Open and click OK. |
|6.||Select Close in the Register Solver Script window.|
Note that if Save As… or Append is not used, HyperStudy only remembers the registered script for the current session. The next time HyperStudy is launched, users have to register the script again.
|7.||In the Solver Input file enter a_hffoam_TEST.xml.|
|8.||Select the registered script from the pull down menu under Solver execution script.|
|9.||Click Write and then Execute|
When the Madymo job is completed, a message appears saying:
End executing model(s) for approach ( Setup (nom_1) )
A nom_run/m_1 directory is created inside the study directory. The result files are created within the nom_run/m_1 directory.
The create responses panel is displayed.
Step 4: Create a response
|2.||In the HyperStudy – Add Response window, type in the name of the response, NTF, in Label: field.|
The HyperStudy - Response Expression Builder window appears.
|6.||On the Vectors tab, click Add.|
A result vector called Vector 1 with varname v_1 is added. The name of Vector 1 can be changed by right clicking on the vector and editing the name. This vector now needs to be defined.
|7.||Click the browser button under the Vector resource file field and select the a_hffoam_TEST.injury file. |
|8.||Under Type select MADYMO Results.|
|9.||Under Request, select /3/2047 ( /Hybrid_III_50th/NeckUp_Fx_lce ) LOAD_CELL Force (N) FILTER=CFC1000 (on input link).|
|10.||Under Component select -. |
|11.||On the Vectors tab, click Add to add another vector.|
A result vector with varname v_2 is added.
|12.||Click the browser button under the Vector resource file field and select the a_hffoam_TEST.injury file. |
|13.||Under Type select MADYMO Results.|
|14.||Under Request, select /3/2049 ( /Hybrid_III_50th/NeckUp_Fz_lce ) LOAD_CELL Force (N) FILTER=CFC1000 (on input link).|
|15.||Under Component select -. |
|16.||On the Vectors tab, click Add to add another vector.|
A result vector with varname v_3 is added.
|17.||Click the browser button under the Vector resource file field and select the a_hffoam_TEST.injury file. |
|18.||Under Type, select MADYMO Results.|
|19.||Under Request, select /3/2052 ( /Hybrid_III_50th/NeckUp_My_lce ) LOAD_CELL Torque (Nm) FILTER=CFC600 (on input link).|
|20.||Under Component select -.|
|21.||Click Function… in the HyperStudy - Response Expression Builder window.|
|22.||In the Functions window, select the function nij and click OK.|
|23.||Click Apply to use all three vectors in the expression field as shown below.|
|24.||In the Response expression field, edit to obtain nij(v_1,v_2,v_3,5,5,1,0).|
|25.||Check the Evaluate response expression box. |
The expression should change to 0.022117.
|27.||Click Add response to add second neck injury criteria, NTE.|
|28.||In the Response expression field, edit the index to obtain nij(v_1,v_2,v_3,5,5,1,0).|
|29.||Check the Evaluate response expression box. |
The expression should change to 0.029185.
|30.||Click Add response to add third neck injury criteria, NCF.|
|31.||In the Response expression field, edit the index to obtain nij(v_1,v_2,v_3,5,5,1,0).|
|32.||Check the Evaluate response expression box. |
|33.||The expression should change to 0.015077.|
|34.||Click Add response to add fourth neck injury criteria, NCE.|
|35.||In the Response expression field, edit the index to obtain nij(v_1,v_2,v_3,5,5,1,0).|
|36.||Check the Evaluate response expression box. |
The expression should change to 0.010352.
This completes the study setup. You can now proceed to the desired study type (DOE, Optimization, or Stochastic study).
Step 5: Run a DOE study
|1.||Click the blank box for Create DOE study in the navigation tree.|
|4.||Under Controlled factors, set DOE Class as Full Factorial. |
|5.||Keep the default DOE Class as None for Uncontrolled factors.|
|7.||Select the Fact1Level design variable and increase the level from 2 to 3 in No. of Levels:.|
|8.||Repeat this process to change the level of Recliner and contactHR design variables to 3.|
|9.||Click Next to continue to the Interactions dialog. |
By default all interactions that are selected for Full Factorial design.
This page shows you the design matrix that is automatically selected and contains 27 design variable combinations or designs.
|11.||Click Next to skip the Uncontrolled variables that are not defined in this study. |
This brings you to the DOE responses dialog.
This gets you to the dialog to actually run the study.
|15.||Answer the question that pops up with Yes. |
This executes all 27 runs.
|16.||Click Next to get to the Extract responses page.|
This extracts the responses defined from the four runs.
|18.||After the results are extracted, click Next to continue to the Post-processing window.|
Step 6: Post-process the results of the DOE study
|1.||Check the Curve legend box in the Main Effect tab.|
|2.||Only check the NTF box to see the main effect of NTF.|
|3.||Repeat step 2 for different responses.|
|4.||Select the Controlled Design Variables Interactions tab to see the interaction between two different variables.|
|5.||Check the C0-C1 interaction to see the interaction between the first and second variable (Fact1Level and Recliner).|
|6.||Repeat step 5 for C0-C2 (Recliner and contactHR) and C1-C2 (Fact1Level and contactHR) interaction to select response NTF.|
|7.||Repeat steps 4 and 5 for other responses.|
Step 7: Create approximations
|1.||Click on the blank box for Create approximation in the navigation tree.|
An approximation can be created for each response in the list on the left that we want to use in an optimization study.
|3.||Select Approximation Type: Least Squares Regression. Make sure that the Add to all tick box is selected. |
|6.||Click Import Matrix and select doe_1. |
This brings you to the Build Approximation page. A second order approximation is built by default.
|9.||Click Next to go to the Residuals page to assess the accuracy of the regression equations.|
|10.||Review the other responses NTE, NCF, NCE.|
|11.||Select Anova in the navigation tree to assess the contributing percentage of each design variables to the selected response.|