Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Developement of Reinforcement Learning based Optimisation Method for Side-Sill Design

Aditya Borse, Rutwik Gulakala, Marcus Stoffel

Abstract

Optimisation for crashworthiness is a critical part of the vehicle development process. Due to stringent regulations and increasing market demands, multiple factors must be considered within a limited timeframe. However, for optimal crashworthiness design, multiobjective optimisation is necessary, and for complex parts, multiple design parameters must be evaluated. This crashworthiness analysis requires computationally intensive finite element simulations. This challenge leads to the need for inverse multi-parameter multi-objective optimisation. This challenge leads to the need for multi-parameter, multi-objective inverse optimisation. This article investigates a machine learning-based method for this type of optimisation, focusing on the design optimisation of a multi-cell side sill to improve crashworthiness results. Furthermore, the optimiser is coupled with an FE solver to achieve improved results.

Developement of Reinforcement Learning based Optimisation Method for Side-Sill Design

Abstract

Optimisation for crashworthiness is a critical part of the vehicle development process. Due to stringent regulations and increasing market demands, multiple factors must be considered within a limited timeframe. However, for optimal crashworthiness design, multiobjective optimisation is necessary, and for complex parts, multiple design parameters must be evaluated. This crashworthiness analysis requires computationally intensive finite element simulations. This challenge leads to the need for inverse multi-parameter multi-objective optimisation. This challenge leads to the need for multi-parameter, multi-objective inverse optimisation. This article investigates a machine learning-based method for this type of optimisation, focusing on the design optimisation of a multi-cell side sill to improve crashworthiness results. Furthermore, the optimiser is coupled with an FE solver to achieve improved results.

Paper Structure

This paper contains 14 sections, 5 equations, 6 figures, 6 tables.

Table of Contents

  1. Introduction
  2. Side sill design and optimsiation
  3. Inverse optimisation function
  4. Method
  5. Finite Element impact simulations
  6. ML-based regression surrogate
  7. Optimisation using Genetic algorithm
  8. Optimisation using Network inversion
  9. Optimisation using Reinforcement Learning
  10. Results
  11. FE simulations results
  12. ML-based regression model outcomes
  13. Optimisation results
  14. Conclusion

Figures (6)

  • Figure 1: Cross-section of the side sill with the varied wall thickness
  • Figure 2: Assmebly view of one of the models of side sill and oblique pole impact
  • Figure 3: Data-correlation matrix of the FE simulation database
  • Figure 4: Residual distribution for the validation of the trained regression model
  • Figure 5: Training of SB3- A2C agent- Episodic rewards over episode
  • ...and 1 more figures