How to Enable Complex Simulations: The Power of Multiphysics & Digital Thread Seminar

The Vital Role of Machine Learning in Enhancing Simulations

Author: Peter Chien - University of Wisconsin-Madison

Abstract

Simulations and digital twins stand as pivotal components within the realm of digital thread and engineering. However, simulations and digital twins pose several challenges:

  • Resource-Intensive Nature: The execution of simulations and digital twins is often time- consuming. This presents a significant obstacle when attempting to undertake tasks heavily reliant on sampling.
  • Discrepancies with Test Data: Disparities frequently arise between simulation outcomes and test or physical data.
  • Ubiquitous Uncertainties: Simulations and digital twins are subject to a multitude of uncertainties, stemming from sources like boundary and initial conditions and geometric variations.

Addressing these challenges necessitates the integration of modern machine learning techniques. In this talk, I will delve into the application of contemporary machine learning methodologies to solve these problems. The presentation will be structured around the following focal points:

  • Advanced design of experiments, including Sudoku.
  • Rapid predictive modeling
  • Calibration techniques to narrow the disparities between simulation results and test data.

This site uses cookies that enable us to make improvements, provide relevant content, and for analytics purposes. For more details, see our Cookie Policy. By clicking Accept, you consent to our use of cookies.