Fault isolation for light rail vehicle suspension system based on multi-sensor information fusion

In this paper, fault isolation issue for Light Rail Vehicle suspension system based on multi-sensor information fusion is discussed. The faults considered are the vertical damper fault and the vertical spring fault in suspension systems. The Eros is applied for similarity measurement in the fault feature database, Fast Fourier Transformation for nine sensor outputs equipped on the vehicle. After the belief function assignments are obtained by using Eros method, the belief function assignments are fused by using the rules of D-S evidence combination and the faulty component is determined and isolated. The effectiveness of the proposed method is demonstrated by a case study.