Date and time: 22 February 2021, 15.00 - 16.30
Track: Structural Engineering
Topic: High-dimensional numerical optimization of fiber-reinforced polymers with variational autoencoders and Bayesian optimization
Location: Microsoft Teams
Description: In the design process of any structure, many parameters have a significant influence on the resulting structure. To evaluate a single configuration of parameters, often a Finite Element Analysis is required, which can take a substantial computational effort. Traditional optimization techniques could therefore take days or weeks to optimize complex designs. In this Thesis, recent developments in the field of machine learning are combined to substantially decrease the required computational time during optimization.