Sequential probability ratio fault test method of diesel-electric locomotive system based on Kalman filter

This paper studies the system fault diagnosis of diesel-electric locomotives. First, a simplified model of locomotive speed control is established based on the diesel-electric locomotive system, and then use the delta operator to discretize the linear system and use the kalman filter to obtain the output value predicted by the system model, the predicted output value is compared with the actual value to generate a residual signal, and then the sequential probability ratio test method is introduced for analysis the statistical characteristics of the residual signal, the construction of the decision function and the decision rule, and the monitoring of the system operation status to detect system failures. The research results show that the fault detection method proposed in this paper can effectively realize the process monitoring, and the false alarm rate is low, and the detection speed is fast.