Application of genetic algorithms to optimum design of elasto-damping elements of a half-car model under random road excitations

Abstract

Vehicle ride comfort could be improved by controlling passenger vertical accelerations induced by random road excitations. In the current paper, to provide a certain level for the passenger's sensation of comfort, optimum values for all elasto-damping elements of the seat and passive suspension system are selected numerically by a genetic algorithm (GA). The root-meansquare (RMS) values of vertical acceleration are first calculated at a series of centre frequencies by power spectral density of an actual random road excitation and then the optimized parameters are found by minimization of difference between the obtained RMS values of acceleration and the boundary values specified by the ISO2631. For determining the applicability of the GA in optimization of the passive suspension system, this method is initially applied to a simple twodegrees-of-freedom (2DOF) half-car model and the results are compared with those obtained by non-linear programming as a gradient-based method. Since the GA proves to be an efficient optimization method, it is easily extended to a more complete 5DOF model. Simulation results demonstrate improvement of vehicle ride comfort and capability of the suggested method in optimizing multi-DOF (MDOF) mechanical systems.