Cross results

A cross result allows you to define an item that comprises the discrete variation of a specific data against result sets. In this case the results sets are combined into a single curve. You can create a cross result from a parameter, a variable, or a post-processing variable. Please refer to the Cross result section of the Simcenter Amesim manual for more information.

By default, the data are the final values of the result set temporal simulations. You can use others values, such as values at a specific simulation time, global maximum etc. through the post-processing variables.

When you create a plot of a cross result, the X-axis contains the IDs of the result sets and the Y-axis contains the sets of values. The ID of a result refers to its number and its studies (batch, design of experiments, experiment).

Note

From the result set selection of the plot, you can access the result ID of a cross result. It is then possible to select or deselect the result IDs that are included in the cross result.

Cross results are useful for comparing system designs, comparing system behavior at different operating conditions, or understanding the relationships between variables.

Compare physical quantity distribution

Since cross results deal with discrete values, all types of curve that support discrete data can be used.

When you create a plot, the cross result is displayed as a bar chart by default.

You can then compare the level reached by a physical quantity against the result sets.

Figure 136: Cross result bar chart

You can plot several cross result items on the same graph. This is useful when the items are related to the same quantity, for example energy. You can then use a stacked bar chart to analyze how the quantity is distributed over the system for each result set. Refer to the Stacked bar chart section to see how to set up your graph as a stacked bar chart.

Figure 137: Cross result stacked bar chart

If items are not related, you can use a scatter plot by setting the display of a 2D discrete line. Refer to the Scatter plot section for information on how to set up your graph with this kind of scatter plot.

Figure 138: Cross result scatter plot

Note

Even if parallel coordinate plots, spider charts, and pie charts are available, they are not really suitable for relevant analysis in most cases. It is recommended to use bar chart or 2D line displays.

Analyze relationships between one output and one input

You can create a graph from cross result items to check how a design output varies against a design input. A design output can be one variable of your system and a design input can be one study parameter. Add these two items as cross results and set the same result set IDs for both.

You can plot the cross results on the same graph (first the input and then the output) and convert it to an XY 2D curve. By default the curve is displayed as 2D line.

Figure 139: Connected scatter plot

Note

Save the plot data as a 1D table to use the curve as a component characteristic.

Analyze relationships between one output and two inputs

You can create a graph from cross result items to check how a design output varies against two design inputs. A design output can be one variable of your system and design inputs can be two study parameters. Add these three items as cross results and set the same result set IDs for all three.

You can plot the cross results on the same graph (first the two inputs and then the output). As a first step convert it to an XY curve. Then apply the XYZ curve conversion.

You can use the default 2D colormap display to visualize a component mapping for example.

Figure 140: 2D colormap

You can also switch to a 2D Bubble display to analyze the influence of the two inputs on the result sets. Refer to the Bubble plot section for information on how to set up your graph as a bubble chart.