Active variable geometry suspension for cars

Abstract

This thesis investigates the characteristics and performance of a new type of active suspension for cars through modelling, simulation, control design and experimental testing. The Series Active Variable Geometry Suspension (SAVGS) concept is first put in context by reviewing the history and current trends in automotive suspensions. Its potential is then critically evaluated and work is carried out to maximise its performance for various suspension functions. A multi-model multi-software modelling and simulation approach is followed throughout the thesis in order to cross-check and substantiate simulation results in the absence of experimental data. The simpler linear models are used to inform the selection of suitable parameter sets for the case studies, to synthesise control systems and to qualitatively validate the more complex, nonlinear multi-body models. The latter are developed as a platform to virtually test the system and its control algorithms. When possible, these tests are based on standard open-loop test manoeuvres and on standardised external disturbances. The SAVGS-retrofitted suspension displays a very nonlinear behaviour, which is at the same time a liability and an opportunity from the point of view of control. Nevertheless, different linear control techniques are effectively applied to improve various suspension functions: PIDs are applied to the lower frequency suspension functions such as mitigation of chassis attitude motions, and the H∞ framework is applied to the higher frequency suspension functions such as comfort and road holding enhancement. In all cases, a cascade control approach is employed, and mechanisms are implemented to ensure that physical and design actuator constraints are always respected. This thesis also covers the design and construction of a quarter-car experimental test rig facility. Step-by-step recommendations for its refinement as well as a testing plan are also outlined.