HS-3020: Prediction of Vector Data

In this tutorial you will build a vector fit model using data source data and complete post-processing using the Trade-off tool.

Before you begin, copy the model files used in this tutorial from <hst.zip>/HS-3020/ to your working directory.

Data source of interest is the internal energy of a RADIOSS simulation. Later, the fit model will be used to predict internal energy curve for any given design variable combination. Training data for the fit model has already been prepared through a DOE approach in the provided archive study, Vector_Fit.hstx.

Build a Vector Model

  1. Start HyperStudy.
  2. From the menu bar, click File > Import to open an existing archive.
  3. Browse and select the Vector_Fit.hstx archive.
    The study contains a DOE approach with training data needed for building a fit model.
  4. Go to the Define Models step of the DOE and review design variables, responses, and data sources.
  5. Add a Fit.
    1. In the Explorer, right-click and select Add from the context menu.
    2. In the Add dialog, select Fit Existing Data.
    3. For Definition from, select DOE.
  6. Define Specification and import matrix.
    1. Go to the Specifications step of the Fit.
    2. Activate Internal Energy – MAG and deactivate IE_peak as shown in Figure 3.
      Scalar response, IE_peak is not needed as this tutorial focuses on vector fit.


      Figure 1.
    3. Import training data of DOE 1 as shown in Figure 2.


      Figure 2.
    4. Click Apply.
  7. Go to the Evaluate step and click Evaluate Tasks.
  8. Go to Post-Processing step, Trade-off Tab, and move the sliders to a combination of your choice to plot the predicted curve as shown in Figure 3.


    Figure 3.