Abstract

The ride quality, handling, and stability are three main issues in vehicle suspension design. Different suspension systems have been designed in the past to fulfil these conflicting requirements. One of the popular suspension systems integrated in small and midsize passenger cars is MacPherson suspension system. A suspension system is either passive if a conventional damper is incorporated or is semi-active with a variable damper. A new control oriented dynamic model of the MacPherson suspension system is developed in this thesis to consider the effects of the suspension structure on the dynamic response and a new kinematic model is proposed to investigate those suspension kinematic parameters affecting both handling performance and stability of the vehicle. The performance of MacPherson suspension system under alternative hybrid semi-active controls is evaluated. It is shown that the contribution of different control strategies on the ride quality enhancement of the vehicle could be similar whereas their effectiveness on the performance of suspension kinematc parameters is completely different. Using the H {592} robust control theory, a full state feedback controller is designed to improve MacPherson suspension specifications. The gain of the controller is optimized so that the trade-off between the requirements is achieved. To be more practical and to reduce the design cost, H, output feedback control theory is employed to design a controller with the minimal cost design. To optimize the controller gain, the LMI and Genetic Algorithm optimization tools are used. It is shown that the output controller can improve the suspension performance close to that of a full state feedback controller. A magnetorheological damper with continuously variable damping is considered as the actuator to the system. In order to tune the current signal of the damper so as to track the desired force calculated from the controller unit, a mathematical dynamic model of the damper is required. For modelling the damper, the MR damper is characterized by a piece-wise polynomial model which is identified by using the data acquired from various tests in the laboratory. The dynamic behaviour of the MR damper on control performance is investigated. The Hardware-in-the-Loop Simulation is made and the effectiveness of the controllers is evaluated through experiments.