[L-OP-03] Multi-Objective Optimization with JMAG

Sign in to download the data

Download PDF

Overview

The multi-objective optimization function using genetic algorithms (GA) is a new addition to the optimization calculation option in JMAG. Design exploration can be carried out without error even for design spaces with multimodality, because GA retains several resultsfrom which the narrowing down of results can occur. for the optimization.
JMAG provides a set of functions necessary in evaluating designs presented as a workflow. Since the 2D and 3D geometry modeling and mesh-based modeling are both supported, a geometry can be changed immediately while preserving the analysis conditions, even when a geometry parametric study or optimization calculations are run. Optimization calculations combined with multiphysics analysis or multi-stage analysis can also be run by utilizing analysis groups. In addition to the response surface method and GA embedded in JMAG, an engine used for optimization calculations can call a user-defined algorithm directly.

Search Filter