CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   CFX (https://www.cfd-online.com/Forums/cfx/)
-   -   Stop Run by user point convergence (https://www.cfd-online.com/Forums/cfx/138029-stop-run-user-point-convergence.html)

Daniel Frederik June 27, 2014 02:39

Stop Run by user point convergence
 
Hi

Does anyone know if there's a way to stop a solver run by a user point convergence criterion. For instance, I'm plotting the user point of isentropic efficiency. Is it possible to stop the solver run saying that it should stop when the efficiency does not change any more than 1% within 50 timesteps?

Opaque June 27, 2014 08:51

From the ANSYS CFX documentation.

Quote:

Interrupt control conditions are used to specify the interrupt criteria for a solver run. These conditions are specified using logical expressions that are evaluated by CFX-Solver and reported in the CFX output file. After executing each coefficient iteration and time step (or outer iteration), the solver evaluates all internal termination conditions and user defined interrupt control conditions. If any of these conditions are true, then solver execution stops and the outcome is written to the CFX output file.

Typically, interrupt control conditions are defined by single-valued logical expressions. However, single-valued mathematical expressions can also be used. In this case, a single-valued mathematical expression is considered to be true if, and only if, the result of the expression is greater than or equal to 0.5. Otherwise it is deemed to have a value of false. For a discussion of logical expressions, see CFX Expression Language Statements in the CFX Reference Guide.

You may need to evaluate the running average of the quantity of interest, and able to provide the return value to the logical expression described in the documentation. You could ask ANSYS CFX support if there is any existing UserFortran for your needs, or similar that you could use as a starting point.

Daniel Frederik June 27, 2014 09:01

Thanks Opaque that's a good hint right there! I will try that out


All times are GMT -4. The time now is 20:47.