CAGE User's Guide

Automated Tradeoff

Once you have set up an optimization you can use automated tradeoff. You can select cells to fill by applying an optimization. The cells you select in the tradeoff table define the operating point set for the optimization.

Set up a tradeoff as follows (also described in Creating a Tradeoff Calibration).

1. Select File -> New -> Tradeoff.

1. This takes you to the Tradeoff view. You need to add tables to the tradeoff.

2. Click . This opens the Add Table to Tradeoff dialog.
3. Select the Create a New Table radio button.
4. Enter Spark as the Table name.
5. Enter Speed as the X name and Load as the Y name.
6. Enter 10 as the size of the speed and load axes.
7. Fill the Speed axis with N and the Load axis with L.
8. Fill the table with SPK (spark angle).
9. Click OK.

A new SPK table appears in the Tradeoff tree. Set up the spark table as follows:

1. Expand the SPK node and select the normalizer Speed.
2. Click Initialize in the toolbar to space the normalizers over the ranges of N and L.

You need to select the cells where you want to apply automated tradeoff. Create a region within the table:

1. Highlight a rectangle of cells in the SPK table by clicking and dragging. Note that a large region can take a very long time to evaluate. Try four cells to start with.
2. Click  (or right-click and select Define Region) to define the region. The cells become blue.

To use automated tradeoff on the cells in a defined region,

1. Click any cell in a region.
2. To apply optimization, select Table -> Automated Tradeoff.

1. The Automated Tradeoff dialog appears, showing a list of available optimizations in your session that are set up and ready to run.

   

3. Select your NBI multiobjective optimization to apply to the tradeoff, and click OK in the following two dialogs to accept the default starting values for the optimization.

1. The automated tradeoff optimization runs, showing progress messages as each point in the region is evaluated. The results appear in the selected cells in the table, as shown in the example. You could optimize several regions and then use these results to extrapolate across the whole table.

   .

Sum Optimization

Worked Example Optimization