Research On The Fault Diagnosis And Tolerant Control Of Three-phrase Induction Motor

Induction motors are widely used in the industry, agriculture, transportation, and national defense projects, as well as everyday life on a wide range of applications because of its simple structure, low price, high reliability and convenient maintenance. However, as a result of the motor assembly, operational characteristics, structure, work methods and the complex load conditions change, it will have a variety of electrical failure, the electrical fault will cause a chain reaction, resulting in equipment and even the whole process can not operate normally. Therefore, the study of intelligent on-line monitoring, diagnostic systems, and failure to prevent the occurrence of motor and timely detection and eradication of failure to ensure reliable operation of motors, reducing excess of maintenance, reduce maintenance costs, improve production efficiency, is of great economic and social significance This has become a domestic and foreign scholars and attaches great importance to the forefront of the key issues.In the research of the inter-turn faults on the stator of the induction motor, this paper presents a new model of three phase induction motors having inter-turn faults on the stator. The new model is obtained by transforming the original phase axes model of three phase induction motor. It can be used to diagnose the fault when the parameters of the motor's stator are unbalance. The fault induction model is established by MATLAB/SIMULINK to analyze the stator current. The result shows that when the number of shorted turns increases above a definite degree, the corresponding stator current's virtual value is increasing in evidence.In the research of the rotor broken bar faults of the induction motor, this paper presents a model of three phase induction motors having rotor broken bar faults. The fault induction model is established by MATLAB/SIMULINK. The frequency variety law of the rotor broken bar fault feature component in the startup electromagnetic torque signal can be extracted based on the wavelet ridge algorithm.The purpose of fault-tolerant control is for the fault in the different sources and fault characteristics, to take corresponding measures to deal with the fault-tolerance of fault compensation, eliminate or automatically repair, in order to ensure continued safe and reliable operation of equipment, or at the cost of sacrificing performance losses and ensure equipment within the specified time to achieve its basic functions. A RBF neural network adaptive parallel control combining model reference adaptive control (MRAC) and RBF adaptive neural network is proposed for induction motor. This method combined the MRAC with neural network control, MRAC is used to track the reference input, and the effect of the uncertainties and disturbances is overcome by online adaptive tuning network parameters of RBFNN which can implement the mapping between speed error and control current.