Using Analysis Sensors in an Optimization Problem

This section describes how you can use the Product Engineering Optimizer (PEO) workbench to solve for a desired result in a nonlinear structural or thermal analysis. Optimization requires the creation of local or global solution sensors and knowledgeware parameters to control the optimized solution. Once an optimization is created, it will run iteratively until the best solution is found for the current parameters.

The following topics are covered:

Basic Optimization Steps

The optimization process requires the following steps:

  1. Request field output or whole model history output of the desired optimization quantities.

  2. Run the initial analysis.

  3. Create local or global solution sensors (see Creating a Solution Sensor).

  4. Create new Knowledgeware parameters from the sensors (see Creating Optimization Parameters).

  5. Set up an optimization problem in the Product Engineering Optimizer (PEO) workbench.

  6. Run the optimization.

  7. Review the optimized parameter results in the Nonlinear Structural Analysis or Thermal Analysis workbench.

Creating Optimization Parameters

The solution sensor output for nonlinear and thermal analyses is a list of values for each sensor—one value for each increment in the required solution process. The parameters used to control the optimization input and output must each have a single value that the optimizer can compare over multiple analyses to reach a solution. The PEO workbench will run several analyses varying the input parameters and monitoring the output until the parameter value satisfies your optimization criteria.

You can use the Knowledgeware formula editor to create a single parameter for each list of values. For more information, see Creating a Solution Sensor. For general information on applying and using Knowledgeware, see Applying Knowledgeware and Using Knowledgeware Capabilities in the CATIA V5 Infrastructure User's Guide.

This task shows you how to create a Knowledgeware parameter for use in optimization.

  1. Click the Formula icon in the Knowledge toolbar.

  2. Use New Parameter of type to create a new parameter. Select an appropriate data type for the sensor you are using, such as Mass for a global mass sensor, and use the default Single Value selection.

  3. Select Add Formula to create the formula for the new parameter.

    CATIA opens the Formula Editor.

  4. Accept the default Parameters dictionary selection, and select List in the Members of Parameters column.

    Note:  You must use formula commands intended for list data, such as getItem(), to access the sensor values.

    The Members of List column should display a list for each of the solution sensors in the current model.

  5. Select the appropriate model output in the Members of List column.

Running an Optimization

Once you have created the solution sensors and parameters for use in an analysis, you can enter the Product Engineering Optimizer (PEO) workbench to create and run the optimization. The workbench is described in the Product Engineering Optimizer User's Guide.

Note:  The incremental solution techniques used to solve each analysis step produce solution sensors that are a list of values. This list format prevents the use of the Algorithm for Constraints and Derivatives Providers in the Optimization dialog box.

This task shows you how to create and run an optimization.

  1. Select the Optimization tool from the PEO toolbar.

  2. In the Problem tab, do the following:

    1. Select the optimization type.

    2. Select the parameter to optimize.

    3. Enter the “free” parameters that can be varied to optimize the desired result.

  3. In the Constraints tab, enter the parameters that will limit the analysis.

  4. Click Run optimization at the bottom of the dialog box.

    A progress bar indicates that the optimization process is running.

Once you define the optimization problem and select Run, the optimization process submits jobs, attaches the results, and updates the solution sensors until the termination criteria are met.

Updating a solution sensor or activating an image from the Nonlinear Structural Analysis or Thermal Analysis workbench will display results from the attached output database. If you are in the PEO workbench, updating a sensor or activating an image will trigger a new optimization run if any of the analysis parameters have changed.

Tip: To make parameters and optimizations available in the specification tree, select Tools>Options and toggle on Show parameters and Show relations in the General tabbed page for the Analysis & Simulation options.