Research On The Diagnosis And Application Of Three-phase Asynchronous Motor Stator And Rotor Faults Based On Memetic Algorith

As a common precision equipment in daily life,motor is the most common equipment used in various power-driven operations.In the assembly line production of the workshop,due to the poor working environment,insufficient maintenance and long-term operation will full load,the motor is very prone to failure.Once it fails,the assembly line may need to be shut down,and it will cause huge economic losses to the enterprise.Meanwhile,in some highly sensitive occasion such as nuclear power plant and offshore drilling platform,any malfunction of the motor will put the entire facility in danger.Therefore,it is a very urgent topic for the research about the monitor of motor’s operation status,especially for the fault prediction of three-phase asynchronous motor.The common method of motor fault diagnosis is based on the analysis result of current signal,but this method will be limited by environmental constraints in practical application,which will result in frequent misjudgment,especially for the judgment of frequent malfunction as broken of rotor bars and short circuit between stator turns.Based on above,this paper takes the three-phase asynchronous motor as the research object,applied the parameter identification method and the meme algorithm in swarm intelligence optimization,designs the fault diagnosis research method of broken bar and short circuit between stators based on the meme algorithm,and verifies the effectiveness and feasibility of the above fault diagnosis method through experiments and applications.Firstly,this paper analyzes the disadvantages of traditional fault diagnosis methods,and created the method of using parameter identification to obtain the internal parameters of the motor,so as to master the feasibility of the motor running state.Through the coordinate transformation system of the original model from the three-phase asynchronous motor,the understanding coupling control is realized for the original strong coupling and time-varying motor model,which simplifies the calculation difficulty,so that the motor model can finally be applied by the meme algorithm.Secondly,this paper uses the meme algorithm as the actual method which based on the swarm intelligence optimization method,and solves the parameters to be identified in the parameter identification by transforming the parameter identification into the problem of solving the optimal value.Memetic algorithm is based on random search and emphasizes the ability to find feasible solutions on the premise of ensuring complex constraints.Therefore,when memetic algorithm is used,certain optimization is made for initial population generation,evolution selection operator and mutation operator,which ensures that the algorithm has the ability to consider both global search and local search.In order to verify the accuracy and effectiveness of the algorithm,the fault diagnosis experiment of the broken bar of three-phase asynchronous motor is carried out.The super parameters with the best identification effect for the experimental motor are determined.Through the comparison with other traditional methods,it is proved that the meme algorithm has good performance in accuracy and iteration times.Finally,aiming at the practical application of the meme algorithm,the fault diagnosis process of stator turn to turn short circuit is designed,and the parameter identification based on the meme algorithm is implemented.The method is used to identify the change trend of the internal parameters of the motor when the fault occurs,as well as the location and severity of the fault,which fully proves the practicability of the algorithm.