Difference between revisions of "SMS:Error vs. Simulation Plot"
Revision as of 22:10, 16 September 2009
An Error vs. Simulation plot is generally used with constant simulations and measurement types, although it may be used in transient simulations. This plot can display the mean error, mean absolute error, and root mean squared error between successive solutions and a set of observed data. Various simulations would be run after changing model parameters, such as material roughness values and/or eddy viscosities. The plot will show trends in the solution to see if model parameter changes are causing better calibration with measured field data. Error vs. Simulation plots are created in the Plot Wizard by setting the plot type to Error vs. Simulation. A sample plot is shown in the figure.
Error vs. Simulation Plot
After the plot type is set in the first step of the Plot Wizard, the second step of the Plot Wizard shows the options for the Error vs. Simulation plot.
Solutions. This box lists all available solutions for the simulation. Select the desired solution to use its data sets in the plot.
Move Up/Move Down. SMS initially shows the solutions in the order they were opened. However, this is not necessarily the order in which they were run. To change the order, highlight a solution and move it up or down to rearrange their order.
Check Box Options. There are three options that can be turned on or off. The three options determine whether the mean error, mean absolute error, and root mean squared error plots should be shown. Because these values are an average of all observation points, their line and symbol styles are not linked to any one observation point, but can be defined by clicking on the appropriate canvas window in the dialog.