Stator And Rotor Winding Of Synchronous Motor For Mine Hoist And Excitation Device Fault Diagnosis

Synchronous motor has high power factor and good speed regulation performance,and is widely used in deep well hoisting systems.It is widely used in the system.Synchronous motor is a key component in elevator systems,If it fails,it will bring serious economic,losses and even cause casualties.Therefore,it is of great significance for the fault diagnosis of the synchronous motor of the elevator.In this paper,the synchronous motor of TBPS710-8 mine hoist is taken as the research object,and the stator winding fault,rotor winding fault and the rectifier element fault of the excitation device are studied in detail.The failure mechanism of each fault is analyzed in detail and summarized.Phenomenon of the phenomenon when a fault occurs.The finite element model of the stator and rotor winding failure simulation of the 4000 k W synchronous motor and the Simulink simulation model of the excitation device failure are established respectively.Independent simulation analysis of the winding failure and the excitation device component failure are conducted.Time,and summarized the characteristics change law,and selected the characteristic parameters of fault diagnosis.Based on the advantages of a stack autoencoder with high degree of automation and strong learning ability,a fault diagnosis model of stator and rotor windings and excitation devices for synchronous motors based on SAE-Softmax is proposed.First,the basic composition structure of the model is designed,and the stack autoencoder is combined with the Softmax classifier to solve the problem that the stack autoencoder does not have classification capabilities.Secondly,the training method of the model is determined.The stack is automatically encoded from the bottom up,and the input features are learned layer by layer,which improves the training efficiency of the model.Finally,the proposed model is implemented and trained and tested under the framework of Tensor Flow.This model greatly improves the degree of automation of fault diagnosis by inputting the original simulation data,but the accuracy is relatively low.For SAE-Softmax model,due to the slow convergence speed and easy to fall into local optimum,it leads to the problem of low diagnostic accuracy.,a fault diagnosis model of stator and rotor winding and excitation device of synchronous motor based on SAPO improved SAE-Softmax was proposed.Adding a simulated annealing mechanism to the particle swarm algorithm increases the diversity of the particle swarm and improves the global search capability of the algorithm when optimizing the SAE-Softmax network.Simulation results show that the proposed fault diagnosis method of stator and rotor windings and excitation devices of the synchronous motor based on improved SAE-Softmax has strong self-adaptability and high diagnostic accuracy.A fault diagnosis experimental platform for synchronous motors was established,and the feasibility of the fault diagnosis method proposed in this paper was verified.