Multi-objective robust optimization of chassis system with polynomial chaos expansion method

Abstract

Chassis technical specifications such as durability and ride comfort are the focuses during vehicle development projects. Additionally, robustness is also important as it is related to manufacturing costs as well as chassis performance in the total life cycle. This article proposes a robust multi-objective optimization approach to improve chassis specifications while taking the physical random factors into consideration. The adaptive–sparse polynomial chaos expansion (PCE) method with Chebyshev polynomials of the second kind has been applied to predict responses' uncertainty moments according to the uncertain-but-bounded parameters. Industrial cases related to chassis development have been tested, including an analytical formula, a quarter-car model and a finite element twist beam. For each problem a Pareto front that represents the best compromises between objectives and their robustness is obtained and the comparison tests prove a good robustness prediction by the Chebyshev PCE method. These examples demonstrate the effectiveness and reliability of the approach, in particular its ability to save computational costs for a complex system.