Multi-Objective Optimization — Lesson 6

Once again, a wizard is used to easily set up a multi-objective optimization.
Due to its very good prediction quality, the Adaptive Metamodel of Optimal Prognosis (AMOP) can be used to approximate the responses with sufficient accuracy. Thus, the optimization process can be performed in a much shorter time compared to time-consuming numerical simulations. Nevertheless, it will be shown how best designs along the Pareto-Front can be validated.
Furthermore, it will be shown how to easily evaluate and comprehensibly visualize the results in the post-processing.

Lecture

Alternate video link.

Video Highlights

00:09 – Start using the optimization wizard

00:36 – Step 2: Define optimization criteria

00:57 – Step 3: Choose the optimization method

01:08 – Step 4: Additional optimization options

01:31 – Reducing the number of pareto designs to be validated

01:51 – Multi-objective optimization postprocessing

02:08 – Control validated design

02:35 – Cluster analysis to divide pareto front by color

02:52 – Summary (audio only)