Research On Frequency Shift Cab Signal Detection And Fault Diagnosis Based On Generalized S Transform

Cab signal is one of the important components to ensure train’s efficient and safeoperation. It not only provides the ground control information for the driver but also providesthe required information for training running monitor and ATP (Automatic Train Protection)vehicle equipment. Since the outdated technology of cab signal detection, there is someimperfections in detecting the cab signal frequency information. Meanwhile the faultdiagnosis methods exist the problems of the low-level automation, low efficiency, and whichcan not diagnose whether occurred the fault, fault types and fault reasons thoroughly. Basedon the above problem, studies on cab signal detection and fault diagnosis are necessary.ZPW-2000A frequency shift signal is the main research target. After analyzing theworking principles, common faults and fault reasons of ZPW-2000A frequency shift signal,the detection algorithm of frequency shift cab signal is improved, the diagnosis model andknowledge base are proposed, which make the fault detection and diagnosis of cab signalwaveform data come true. The main contents are as follows:Firstly, the mathematical simulation model is established after the frequency shift cabsignal features are analyzed, and the common types of frequency shift cab signal faults aresummarized. Based on all kinds of time-frequency analysis techniques and generalized Stransform, a new generalized S transform is introduced to realize the accurate detection ofrailway frequency shift cab signal frequency. The railway frequency shift signal and diagnosethe fault types according to the formulate fault frequency criterion. The effective voltageamplitude of frequency shift cab signal is detected and the criterion of voltage fault is made tofind out whether the effective voltage signal of locomotive is normal. Through matching andsearching the normal code sequence and fault code sequence, the code sequence detection andcode sequence fault diagnosis of frequency shift cab signal are realized.Secondly, the frequency shift cab signal fault diagnosis model for Petri network isestablished with frequency, effective voltage and code sequence. The correlation matrix andstate equation of the model are used to realize reasoning and diagnosing of frequency shiftcab signal fault. A laboratory simulation environment is built and the correctness of the modelis verified. The results show that the model can judge out the occurrence of signal faults andthe corresponding fault types accurately.Finally, based on the discussion of the common no alarm fault and cab signal fault ofZPW-2000A system, the human-machine interface of frequency shift cab signal faultdiagnosis system is described by VC++6.0and Matlab joint programming techniques, thefield data is used to conduct detection and fault diagnosis of the frequency shift cab signal.The diagnostic results show that the system can not only diagnose the frequency shift cab signal fault accurately, but also locate the fault source to judge the types and analyze the faultreasons, which can deal with all kinds of fault information timely, thus the diagnosisefficiency of fault can be improved effectively.