Multi-objective stochastic optimisation of the suspension system of road vehicles

This paper addresses a new approach for designing automotive suspension systems, based on the theory of multi-objective programming together with the theory of robust design. A two-degrees-of-freedom (2 dof) linear model is used to describe the dynamic behaviour of vehicles running on randomly profiled roads. The road irregularity is considered a Gaussian random process and modelled by means of a simple exponential PSD. The performance indices considered are discomfort, road holding and working space. The design variables to be optimised are the suspension stiffness and damping (passively suspended vehicle) and the controller gains (actively suspended vehicle). The mass of the vehicle's body and the tyre radial stiffness are considered as stochastic parameters, together with the design variables (stochastic design variables). The optimal trade-off solutions (Pareto-optimal solutions) are derived in a stochastic framework and, whenever possible, in a non-dimensional analytical form. The analytical expressions are derived by means of a new method based on the Fritz John necessary condition.