Fault Detection for Passive Suspension Systems Using Wavelet Transform 2007-01-2368

In this paper, the analysis of vibration signals for detecting the faults of suspension system, using wavelet transform is developed. Based on the continuous wavelet transform (CWT), two methods have been presented, in which Morlet wavelet is used to fault detection of the suspension systems. In both methods, natural frequencies have been estimated by the cross-wavelet analysis (Cr-WA). The changes of the signal energy in natural frequencies and also the damping ratio due to the faults occurred in dampers and bushings are investigated in the first and second method, respectively. A full vehicle model with seven degree-of freedom with upper strut mount effect has been simulated. Damper and strut mount fault is supposed as damping decreasing and gap, which may lead to model non-linearity, respectively. The vehicle considered to be excited under the frequency sweep and bump inputs. Final results show the inability of spectral analysis using power spectral density (PSD) function for analysing the mentioned systems and signals including transient specifications is shown and then the ability and accuracy of CWT for these purpose is investigated. In practice, the results can be used to detect the faults of suspension components such as dampers and bushings.