Fault Diagnosis Of Traction Motor Based On Independent Component Analysis In High Speed Train

With the development and breakthrough of high-speed train,China Railway High-speed(CRH)has been rapidly popularized in China and has gone abroad to the world.CRH has also been rated as one of four new inventions in China and has become into the business card of China in the new era.High-speed railway is of great importance for the safety of the passengers.Traction motor provides power for the high-speed railway traction system and is the core unit which ensures the operation performance and reliability of the whole train.Fault diagnosis of traction motor is the foundation of the safety and stability for the high-speed railway traction system.This paper focuses on fault diagnosis of the traction motor in CRH2 high-speed train,and the research work of fault diagnosis based on independent component analysis is carried out in this paper.Firstly,the significance of the research is introduced according to the development background of high-speed train.The research status of traction motors for high-speed trains at home and abroad is summarized.The main methods of fault diagnosis and the characteristics of independent component analysis technology are introduced.The mechanism of traction motors in high-speed trains is studied and the data characteristics of traction motors are analyzed.Secondly,the current,voltage and speed in traction motor are collected,and the independent component analysis model and detection statistics are established to monitor the health status of the traction motor.When the speed sensor fault in the traction motor occurs,the detection statistics will exceed the threshold and give an alarm.The multi-model structure can realize fault isolation the fault location.The simulation results of the four-variable system prove the effectiveness and superiority of the proposed method.The method is further validated by the experimental data from the semi-physical simulation platform of high-speed railway traction system in CRRC Zhuzhou Institute.Then,considering the incipient faults of the speed sensor in the traction motor of high-speed train,an incipient fault detection method based on wavelet transform and ensemble independent component analysis is proposed and is applied on the speed sensor in the traction motor.The wavelet transform technology is used for noise reduction and the fault features are extracted by integrated independent component analysis(ICA).The detection statistics are established to realize incipient fault detection.The simulation results and experimental platform data verify the effectiveness of the method.Finally,in order to verify the applicability of the proposed method in practical engineering,the experimental verification is carried out on the traction motor control experimental platform.This paper uses YS7124 three-phase asynchronous motor as the research object,which is developed by Nanjing University of Aeronautics and Astronautics Laboratory.The experimental equipment is equipped with Quanser QPID control board and the speed sensor fault is injected into traction motor.The experimental results show that the proposed method has good performance and good applicability.