Multi-Objective and Robust Design of a Semi-Active Suspension System

Abstract

This paper presents a robust multi-objective optimal design (RMOP) of a passenger car with a semi-active suspension system. The mean-effective values of the root mean square of the passenger’s head acceleration, suspension travel, and tire deflection are considered as design objectives. The passive components of the suspension and the design details of the Linear Quadratic Regulator (LQR) algorithm are used as design parameters. During the design, global sensitivity analysis is carried out using the Fourier Amplitude Sensitivity Test (FAST) to specify the elements of the model that can highly alter the design objectives. The mass of the passenger’s head and upper body, the mass of the passenger’s lower body and cushion, passenger and cushion’s elastic properties, and the sprung mass of the vehicle are selected for the sensitivity analysis. Results show that the design criteria are very sensitive to the variations in the sprung mass of the vehicle as compared to the other parameters. As a result, the variations in this parameter and passive elements of the suspension system are considered. Constraints are applied on the objectives in compliance with the requirements of ISO 2631-1 on the design of car suspension systems.  The optimization problem is solved by the NSGA-II (non-dominated sorting genetic algorithm II) and robust Pareto front and set are obtained. The Pareto set includes multiple design options from which the decision-maker can choose to implement. Responses of the passenger’s head acceleration, suspension travel, and tire deflection show that the robust multi-objective design algorithm (RMOA) is very effective and guarantees less sensitivity to the suspension passive elements.