Fault Diagnosis of Vehicle Driveline System Using Modal Response Optimization

This paper describes an optimization methodology for damage detection in an automobile driveline system. When a constant nature of excitation such as harmonic or pulse is transmitted from the engines, the torsional response at various points along the driveline should remain same under various trails. But in actual practice, due to the deterioration and minute cracks developed along the driveline, a slight variation from the baseline response data is often observed. In other words the changes in the response history is an indication of the changes in the system parameters. In this paper driveline system is modeled using finite element method with consideration of reductions in stiffness of each element. If there be no damage in any element, the overall response when substituted in original equation of motion should not be a non zero quantity. Otherwise, the difference termed as residue has to be accounted and minimized. The unknown damage factors in terms of stiffness reduction coefficients corresponding to a current state of response is predicted using the metaheurestic method known as Genetic Algorithms (GA). The methodology is illustrated with help of a truck driveline system. The results are shown in the form of figures and tables.