About Virtual Execution

A virtual execution allows you to execute a model during the conception phase, to perform early validation. Behaviors must be defined to perform a virtual execution on a model.

Within the Dymola Behavior Modeling app, the virtual execution tools provide the following capabilities:

This page discusses:

Experiment Setup Definition

Various preferences can be applied for the execution.

For more information, see Setting up a Simulation.

Model Check and Translation

Before execution, the model is checked and translated. Model equations are manipulated to a form suitable for efficient virtual execution.

Errors and warnings from the check are displayed in the Message Reporting window, under the Authoring tab. (Messages during the virtual execution are displayed under the Simulation tab.) See Messages Reporting Window For more information about this window.

Important: The check is a recursive process that follows the Modelica structure of the checked class. Therefore, a parent class can be checked several times. Note that the Message Reporting window does not reorder the messages but displays them as soon as the check is performed.

Model Execution

The virtual execution is performed.

For more information, see Simulating a Behavior.

Important: To be able to execute, a valid compiler must be installed. The existence of such compiler can be verified. See Compiler.
Notes:

Execution Result Analysis

The systems virtual execution tools let you analyze the execution result in a number of ways.

Some examples are:

  • Plotting and monitoring variables
  • Sweeping parameters
  • Analyzing variable dependencies
  • Analyzing numeric integration and events
  • Saving and exporting simulation results

See Analyzing Virtual Execution Results.

Variable Selections

You can use variable selections to reduce the amount of saved and displayed variables, to simplify post processing and reduce memory demands.

See Managing Variable Selections.

Saving Snapshots

To improve the possibility to restart a lengthy simulation from multiple time points, periodic snapshots of results can be saved during simulation. The reason for such a restart can be some problems in the simulation that needs debugging, or to perform parameter studies from a certain time point. You can also save shapshots at events.

See Saving Snapshots of Results During Simulation.

Comparing Simulated Models

You can compare simulations of evolutions of the same model or simulations of different models. This can help you to analyze the effect of model alternatives on systems performance.

See Comparing Simulated Models.

Saving Initial Values in the Model

You can save changed initial values of parameters and variables into the model code, once the model has been validated by a simulation with those values.

See Saving Initial Values in the Model.

Scripting

You can automate the simulation by scripting. The script facility makes it possible to for example load libraries, set parameters, set start values, simulate, and plot variables. You can use functions or more traditional script files.

See Scripting.