Modelling chassis flexibility in vehicle dynamics simulation

This thesis deals with the development of advanced mathematical models for the assessment of the influence of chassis flexibility on vehicle handling qualities. A review of the literature relevant to the subject is presented and discussed in the first part of the thesis. A preliminary model that includes chassis flexibility is then developed and employed for a first assessment of the significance of chassis flexibility. In the second part of the thesis a symbolic multibody library for vehicle dynamics simulations is introduced. This library constitutes the basis for the development of an advanced 14-degrees-of-freedom vehicle model that includes chassis flexibility. The model is then demonstrated using a set of data relative to a real vehicle. Finally, simulation results are discussed and conclusions are presented. The advanced model fully exploits a novel multibody formulation which represent the kinematics and the dynamics of the system with a level of accuracy which is typical of numeric multibody models while retaining the benefits of purpose-developed handderived models. More specifically, a semi-recursive formulation, a velocity projection technique and a symbolic development are, for the first time, coupled with flexible body modelling. The effect of chassis flexibility on vehicle handling is observed through the analysis of openand closed-loop manoeuvres. Results show that chassis flexibility induces variations of lateral load transfer distribution and suspension kinematics that sensibly affect the steady-state behaviour of the vehicle. Further effects on dynamic response and high-speed stability are demonstrated. Also, optimal control theory is employed to demonstrate the existence of a strict correlation between chassis flexibility and driver behaviour. The research yields new insights into the dynamics of vehicles with a flexible chassis and highlights critical aspects of chassis design. Although the focus is on sports and race cars, both the modelling approach and the results can be extended to other vehicles.