*Please prepare a license ID and password for the license administrator.
*It is different from the service for JMAG WEB MEMBER (free membership). Please be careful.
Overview
Optimizations that use the genetic algorithm (GA) require time because each generation runs finite element analyses (FEA). JMAG runs offline optimizations that take advantage of surrogate models to run as the FEA and check the CAD geometry of each individual in each generation. These surrogate models significantly reduce the time required to run an optimization.
This tutorial describes the procedures to create the samples (training data) and surrogate models to run an offline optimization using a response value surrogate model and geometry check surrogate model.
Keywords
surrogate model, response value surrogate model, geometry check surrogate model, AI, machine learning, optimization, parametric
