Overview
Optimizations using a genetic algorithm require time to run the FEA analyses of each generation. Any change to the design requirements inevitably results in extremely high computational costs (time/resources) for the FEA necessary to explore those design spaces.
Surrogate models are one solution that overcomes these computational costs. In particular, offline optimizations take advantage of surrogate models generated prior to an optimization. The offline approach explores design spaces via multiple optimizations free of any FEA computational costs to give designers the insight for fast decision-making at the initial design stage.
The JMAG Surrogate-Based Optimization (JMAG-SBO) tool capitalizes on JMAG surrogate models to run almost instantaneous multi-objective/multi-constraint optimizations. Surrogate models created using magnetic, thermal, structural, and electric field analysis results even facilitate multi-disciplinary optimizations.
Fig. 1 JMAG-SBO Window
JMAG-SBO simultaneously explores design spaces using multiple surrogate models from an extremely accessible interface. Users run optimizations by simply specifying
the surrogate model, objective functions, and constraint conditions.
The interactive result graphs streamline the data analysis process.
You need to sign in as a Regular JMAG Software User (paid user) or JMAG WEB MEMBER (free membership).
By registering as a JMAG WEB MEMBER, you can browse technical materials and other member-only contents for free.
If you are not registered, click the “Create an Account” button.
Create an Account Sign in
