Research on fault isolation of rail vehicle suspension system

As a significant component of rail vehicle, suspension system is really important to the safety of train, so the real-time condition monitoring and fault diagnosis on it is very necessary. It can not only improve the safety and reliability of vehicles, but also reduce the cost of preventive maintenance. Fault diagnosis and isolation methods of rail train suspension system will be discussed in this paper, using the SIMPACK and MATLAB co-simulation environment, the experiment platform of fault simulation was built and used to generate data for fault diagnosis. Furthermore, Support Vector Machine (SVM) and Fuzzy Min-Max Neural Network (FMMNN) were applied to the issue of fault isolation. The simulation results demonstrated that both of them could achieve fairly good accuracy and the method of SVM could obtain a higher one than FMMNN. Besides, the approach proposed in this paper provides a new solution for fault isolation application of suspension system.