Parametric Identification and Health Monitoring of Complex Ground Vehicle Models

Abstract

A systematic methodology is applied for performing parametric identification and health monitoring in the suspension substructures of complex vehicle models. The equations of motion are derived by applying a finite element method. As a consequence, they involve quite a large number of degrees-of-freedom (DOF). In addition, they include strongly nonlinear terms. In particular, the main nonlinearities arise due to the function of the suspension dampers and springs. Moreover, the action of the bushings connecting the suspension subsystems to the vehicle body is also strongly nonlinear. Since the resulting number of DOF is large, an appropriate coordinate condensation technique is applied first. This drastically reduces the dimension of the original system and allows the application of a statisticcal system identification methodology, which is effective for dynamical systems with relatively small dimension, in order to perform parametric identification and fault detection studies in the suspension subsystems of an example vehicle model. In the second part of this study, the methodology developed is applied and yields numerical results related to parametric identification and fault detection in the suspensions of the vehicle model examined. The results are found to be sufficiently accurate even in the presence of considerable measurement noise and model errors.