Effectiveness of an interval computation approach to the dynamic simulation of a MacPherson Suspension system

In this paper, a new design approach based on methods by intervals adapted to the integration of a simulation step at the earliest stage of preliminary design for dynamic systems is proposed. The general idea consists in using the interval computation method to make a simulation by intervals in order to minimize the number of simulations. These intervals represent the domains of possible values for the design parameters of the system. So the parameterized model of system is solved by interval. This avoids launching n simulations with n values for each design parameter. The proposed method is evaluated by several tests on a scalable numerical example. It has been applied to solve parameterized differential equations of a Macpherson suspension system and study its dynamic behavior. The dynamic model of the MacPherson suspension is nonlinear but linearisable. It was transformed into a parameterized state equation by intervals. The solution to this state equation is given in the form of a matrix exponential. Three digital implementations of exponential have been tested to obtain convergent results. Simulations results are presented and discussed.