Design of Active Tyre-Suspension-Seat System Through Multibody Model and Genetic Algorithms

Abstract

The tyre-suspension-seat dynamic system, driveline and engine vibrations are generally considered in the vibrational field as the main factors that influence the particular feeling of comfort perceived by passengers on a vehicle. Hence, the development of several criteria and models for the optimal estimation of the design parameters of such systems. Among these parameters, the most detrimental impacting on the passenger comfort are undoubtedly acceleration and its variation. The two types of suspension systems (conventional passive suspension system and active suspension system) differ as the first foresees the spring-damper characteristics to be adjusted so that only one of several conflicting objectives (such as passenger comfort, road holding, and suspension deflection) is followed. In active suspension systems, instead, these objectives are balanced by the designer in a more efficient manner thanks to the feedback-controller actuator assembly. However, this approach presents some limitations linked to the extremely wide spectrum of magnitude and frequency of external forces that the tyre-suspension-seat system has to efficiently control and mitigate. It remains that in the existing optimisation models and systems time exposure limits established by unification agencies and road authorities are not generally considered. This paper illustrates the development of an active tyre-suspension-seat system control for passenger cars, using both a non-linear multibody model and Genetic Algorithm (GA) controls. A benefit of the proposed active tyre-suspension-seat system control is also to consider various time exposure limits and an active damping element. The main innovative element introduced by this work consists in having coupled an active control to passive mechanical parameters in order to minimize the seat acceleration. The 3 DoF multibody model, applied to a quarter body for symmetry reasons, treated road roughness as an input variable in the GA control so as to determine the vertical component of acceleration. The numerical and experimental applications of the proposed model to a specific case study allowed to validate the effectiveness of the active system towards the vibrations transmitted to the passenger.