Fault isolation of rail vehicle suspension systems by using similarity measure

In this paper, we consider fault isolation issue for Rail Vehicle Suspension Systems. The faults considered are the vertical damper fault and the vertical spring fault in suspension systems. A Kalman filter is applied to generate the residuals for fault isolation and a fault feature database in the frequency domain is built in which some typical suspension failures are included. When there is a fault detected, Fast Fourier Transform is applied to the residuals to obtain the amplitude-frequency fault feature. The amplitude-frequency information is compared with the fault feature in the database by using the so called Eros similarity measure method. The result with the largest similarity is accepted as the fault case. The effectiveness of the proposed approach is demonstrated by simulation results for several scenarios.