Variability (Stochastics) Study

Study the effects of parameter variations on squeak and rattle issues.

The manufacturing of parts is often related to variations of tolerances and quality of materials. Also, the assembly process adds a set of variations on dimensions and connection of stiffnesses resulting in a unique product at the end of each build. These two major processes contribute to the quality and performance of the final product.

squeak and rattle are phenomena that are tightly related to the quality of the built parts, and performance can vary through the population of produced products.

The Manufacturing Variations Study workflow will take the you through the steps of setting up a variability study to analyze the risk and effects of the manufacturing problems. Below is an illustration for Manufacturing Variations Study.
Note: SnRD requires a runnable solver model, which should either contain a Dynamic or Static Loadcase. This is required since HyperStudy works on the principle of comparing results of a nominal run with the consecutive run results.
In this tutorial, you will:
  • Prepare the FE model for analyzing squeak and rattle issues.
  • Setup variability study for the following-
    • Material Parameters
    • Attachment Stiffness
    • Thickness, and
    • Damping
  • Run analysis and post-process the results.
For this tutorial, use a new model and prepare the model analysis setup. For this workflow, refer the following sections from Detailed Risk and Root Cause Analysis: Import a model with Dynamic Event loadcase. For this workflow, you can use the model with solver deck created in the Detailed Risk and Root Cause Analysis usecase along with the Dynamic Loadcase. Choose the workflow according to your need and refer to sections mentioned above for the procedures.

For this tutorial, you will use the solver deck exported from the Detailed Risk and Root Cause Analysis usecase. Once you import the Dynamic Loadcase solver deck, you can proceed with Variability (Stochastics) Study setup.

Variability Study Setup

In this step, you will setup variability study(s).

  1. From Setup group, select Variability Study.
    Figure 1.
    A guide bar opens.
  2. From the guide bar, click to open the advanced selection option.
  3. Select all the materials from the list.
    The five materials in the model are selected. A microdialog opens.
    Figure 2.
  4. Select E from the list, and enter 20 in the and value fields.
    The parameter variation percentage for the study is defined.
    Figure 3.
    Tip: Click to expand the microdialog. You can define a desired name for the study or retain the default.
  5. Click .
    The material Young's Modulus (E) variation study is created.
  6. From the guide bar, click to open advanced selection option. Select all the materials from the list.
  7. Select all the materials from the list.
    The five materials in the model are selected. A microdialog opens.
  8. Select RHO from the list, and enter 20 in the and value fields.
    The parameter variation percentage for the study is defined.
  9. Click .
    The material Density (Rho) variation study is created.
  10. From the guide bar, select Attachment Stiffness.
  11. From the guide bar, click to open the advanced selection option.
  12. In the Advanced Selection dialog, select the following properties.
    • ControlPanel_IPSub_Snaps
    • GloveBox_RubberBumper
    • ClusterBezel_IPSub_Clips
    • DriveSidePanel_IPSub_Snaps
  13. From the microdialog, select K1, K2, and K3 stiffnesses.
  14. Click .
    The Attachment Stiffness variation study is created.
  15. From the guide bar, select Thickness.
  16. From the graphics area, select the following components.
    • IPSubstrate
    • Glove Compartment
    • Driver Side Panel
    • Lower Control Panel
  17. Retain the default values in the microdialog and click .
    The material Thickness variation study is created.
  18. From the guide bar, select Damping.
  19. From the guide bar, click to open the advanced selection option.
  20. From the Advanced Selection dialog, select all the damping curves.
  21. Click .
    The material Damping variation study is created.
  22. Select Review Variations.
    Figure 4.
    A table opens containing all the variations and the related parameters.
    Figure 5.
  23. Click Close.

Export Solver Deck

  1. From Analyze group, click Export OptiStruct Solver File.
    Figure 6.
    The Model Export dialog opens.
    Figure 7.
  2. In the Model Export dialog, enable the Export Variability Study checkbox.
    The Variability Study properties are displayed.
    Figure 8.
  3. Accept the default options and click Export.
    A folder selection dialog.
    Tip: You can modify any properties in the dialog.
  4. Browse and select the required folder.
    The OptiStruct solver deck, along with the XML and TPL files are exported to the selected folder.
  5. Click Close to close the model export option.
Use the exported TPL template file to solve in HyperStudy.

Conduct Stochastics Runs

In this step, you will generate results database and summary files. These files will be later used in SnRD Post-processing.

SnRD exports the variability study parameters into the following files, which are required by HyperStudy for results database generation:
  • TPL
  • XML

The template (TPL) file us created with the required parameters that are readable by HyperStudy.

  1. Launch HyperStudy2023.1.
  2. Click Open.
    A file browser option will be displayed. Browse and select tutorial XML file. Once the XML file is loaded, a session tree will populate in HyperStudy and the process will start from the Sto 1 node.
    Figure 9.
    Note: Green checkboxes signify that the respective data is available. Any Yellow, Red, or empty boxes mean inputs are required.
  3. Click Next to proceed to the Sto 1 > Definition > Test Models task.
    Note: You can skip Define Models and Define Input Variables tasks since they are pre-defined.
    Figure 10.
  4. In the Test column, click All.
  5. Click to proceed to Define Output Responses and click Evaluate.
  6. Click to proceed to Specifications and click Apply.
    Figure 11.
  7. Click to proceed to Evaluate and click Evaluate to run the stochastics variations.
    Figure 12.
    Note: Evaluation tasks will take considerable amount time to complete. Do not interfere with HyperStudy when the process is running.
    Figure 13.
    Once the evaluation is completed, remaining tasks, Post-processing and Report, turn green.
  8. Click twice to skip Post-processing and navigate to the Report task.
  9. Click Create Report.
    A .data file is generated, which will be the results database that will be utilized later in SnRD Post.
    Tip: You can navigate to Tutorial working model folder and verify the files. The respective files will be created in the export folder. Learn more about Altair HyperStudy.

Stochastics Post Processing

Note: Copy the summary.data file created in reports/sto_1 after the Create Report steps in HyperStudy, to approaches/sto_1.
  1. Load the model and CSV file in the Post Processing tab and open the DoE/Stochastics tab.
  2. Click the HyperStudyResults Database .
  3. Click against HyperStudy Results Database.
    A file browser window opens.
  4. Browse and select the XML file generated in Export Solver Deck.
  5. Click Calculate.
    Note: Since the number of runs is high and the process will take a considerable amount time, do not interfere with process.
    Once the calculation is complete, a confirmation message will be displayed.
    Figure 14.
  6. Click OK.
    Note: To run the results processing for the second time, you can use the SnR Database XML that was created during the calculation can be utilized.
  7. Select the following options from the Selection lists.
    1. For E-Lines, select All.
    2. For Line ID, select All.
    3. For Results, select SnR Index.
  8. For Show Runs, select All.
  9. For Vary Variable by, select Value.
  10. For Show Variables, enable the All checkbox.
  11. Click Apply Filters.
    All the runs in the Run Selection table are filtered.
    Figure 15.
  12. Click Display Results.
    The results for the all the filtered runs are plotted.
    Tip: You can select the required runs from the Run Selection table.
  13. Click Display Selection.
    The results for the selected runs are plotted.
    Figure 16.