Abstract
This paper presents an innovative approach for the fault isolation of Light Rail Vehicle (LRV) suspension system based on the Dempster-Shafer (D-S) evidence theory and its improvement application case. The considered LRV has three rolling stocks and each one equips three sensors for monitoring the suspension system. A Kalman filter is applied to generate the residuals for fault diagnosis. For the purpose of fault isolation, a fault feature database is built in advance. The Eros and the norm distance between the fault feature of the new occurred fault and the one in the feature database are applied to measure the similarity of the feature which is the basis for the basic belief assignment to the fault, respectively. After the basic belief assignments are obtained, they are fused by using the D-S evidence theory. The fusion of the basic belief assignments increases the isolation accuracy significantly. The efficiency of the proposed method is demonstrated by two case studies.
Highlights
With the rapid development of economy, the efficiency and safety of transportation are paid much more attention than ever before
The possible occurred fault in the fusion isolation results is the actual fault among these inconsistent faults. This method, multi-sensor information fusion based on the D-S evidence theory and Eros, is effective and accurate on fault isolation that reduces the uncertainty of fault isolation substantially, and effectively improves the accuracy of recognition to the fault model
This paper proposes a new method to isolate faults of Light Rail Vehicle (LRV) suspension system and its further case
Summary
With the rapid development of economy, the efficiency and safety of transportation are paid much more attention than ever before. Isolate issue of fault of suspension system in LRVs based on multi-sensor information fusion is investigated. An innovative approach for the fault isolation of Light Rail Vehicle (LRV) suspension system based on the Dempster-Shafer (D-S) evidence theory is studied. This approach mainly uses the Eros, which is applied for similarity measurement in the fault feature database. By using the fault detection method in our precious work [6], 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 system failures are included. The obtained seven fault features (BBAs) are combined by D-S evidence theory for enhancing the isolation accuracy
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
