Changing the Response Surface

Changing response surface types or changing the available options for the current response surface can improve the Goodness of Fit. To change your response surface type:

  1. On the Project Schematic, right-click the Response Surface cell and select Edit.

  2. In the Outline view, select the Response Surface node.

  3. In the Properties view, under Meta Model, select a different choice for Response Surface Type.

  4. Update the response surface either by either clicking the Update button in the toolbar or right-clicking the Response Surface node and selecting Update.

  5. Review the Goodness of Fit information for each output parameter to see if the new response surface provides a better fitting.

Once you've achieved the desired level of Goodness of Fit, add verification points and then recheck the Goodness of Fit. If needed, you can refine the response surface further manually. For more information, see Performing a Manual Refinement.

Guidelines for Changing the Response Surface Type

Standard Response Surface 2nd-Order Polynomial

If the default Genetic Aggregation takes too long to generate response surfaces, you can switch to Standard Response Surface 2nd-Order Polynomial.

The standard response surface is particularly effective when the variation of the output is smooth with regard to the input parameters.

Kriging

If Standard Response Surface 2nd-Order Polynomial does not produce a response surface with the desired level of Goodness of Fit, try the Kriging response surface.

After updating the response surface with the Kriging method selected, recheck the Goodness of Fit.

For Kriging, Coefficient of Determination must be set to 1. If it is not, the model is over-constrained and not suitable for refinement via Kriging.

Kriging fits the response surface through all the design points, which means that many of the other metrics are always perfect. Therefore, it is particularly important to run verification points with Kriging.

Non-Parametric Regression

If the model is over-constrained and not suitable for refinement via Kriging, try switching to Non-Parametric Regression.

Other Response Surface Types

If you decide to use one of the other response surface types, consider your selection carefully to ensure that the response surface suits your specific purpose. For more information, see Response Surface Characteristics.

Response Surface Characteristics

Different response surface types have varying characteristics and so lend themselves to different types of responses. The response surface used impacts the quality and Goodness of Fit of the resulting response surface. Use the following characteristics to guide your selection of the response surface most suited to your design scenario.

Genetic Aggregation
  • Default response surface type.

  • Efficient and reliable.

  • Suited to all types of responses.

  • Can be slow to compute.

Standard Response Surface - Full 2nd-Order Polynomial
  • Creates a standard response surface.

  • Effective when the variation of the output is smooth with regard to the input parameters.

Kriging
  • Efficient in a large number of cases.

  • Suited to highly nonlinear responses.

  • Do not use when results are noisy. Kriging is an interpolation that matches the points exactly.

  • Always use verification points to check the Goodness of Fit.

Non-Parametric Regression
  • Suited to nonlinear responses.

  • Use when results are noisy.

Neural Network
  • Suited to highly nonlinear responses.

  • Use when results are noisy.

  • Control over the algorithm is very limited.

Sparse Grid
  • Suited for studies containing discontinuities.

  • Use when the solve is fast.


Release 18.2 - © ANSYS, Inc. All rights reserved.