Non-intrusive surrogate modelling using sparse random features with applications in crashworthiness analysis

TL;DR

This paper introduces a novel surrogate modelling approach using Sparse Random Features combined with self-supervised dimensionality reduction, demonstrating superior performance over traditional methods in crashworthiness analysis.

Contribution

The paper presents a new surrogate modelling method that outperforms Polynomial Chaos Expansions and Neural Networks in crashworthiness applications.

Findings

Outperforms state-of-the-art surrogate models

Effective in synthetic and real crashworthiness data

Demonstrates improved accuracy and efficiency

Abstract

Efficient surrogate modelling is a key requirement for uncertainty quantification in data-driven scenarios. In this work, a novel approach of using Sparse Random Features for surrogate modelling in combination with self-supervised dimensionality reduction is described. The method is compared to other methods on synthetic and real data obtained from crashworthiness analyses. The results show a superiority of the here described approach over state of the art surrogate modelling techniques, Polynomial Chaos Expansions and Neural Networks.