Research On Electrical Fault Diagnosis Method Of In-wheel Motor Based On Multi-feature Parameters Fusion

The emergence of in-wheel motor makes the driving form of new energy vehicles change subversively.It integrates the functions of driving,carrying and braking,and is flexible in control,high in transmission efficiency and compact in structure.However,the immature technology of in-wheel motor makes the safety of the vehicle can't be guaranteed.Therefore,it is necessary to carry out the research on the fault diagnosis of the in-wheel motor.To realize the electrical fault diagnosis of the motor in its actual operating condition,the power source of the independent driving new energy vehicle is combined.The common feature parameters are calculated on the premise of analyzing the failure mechanism of in-wheel motor and the fusion feature parameters are selected based on the statistical hypothesis testing principle.Finally,the parameters fusion weight is allocated to fuse multiple feature parameters based on the wolf group algorithm.First,the practical application value of in-wheel motor fault diagnosis and the research status of three important links in fault diagnosis are expounded.And to improve the limitations of existing research methods are analyzed,the electrical fault diagnosis method of in-wheel motor based on multi feature parameters fusion is proposed.Secondly,based on the working principle of in-wheel motor,the leakage fault mechanism is analyzed,and the simulation model of in-wheel motor is built by using MATLAB/Simulink to verify the correctness of the fault mechanism analysis,which lays the foundation for the initial selection of the characteristic parameters.Thirdly,the in-wheel motor fault simulation test bench is designed to analyze and select the feature parameters of the phase current signal obtained by the test.The feature parameters of high sensitivity of fault discrimination based on the probability statistics and the Weibull distribution fitting function include the effective value p₁,the extreme value p₂and the waveform rate p₅.Finally,the fusion function of the feature parameters and the fitness function of the target are defined.Based on the wolf pack algorithm,the equation parameters of the fusion function are solved.The Weibull distribution fitting function of the fusion characteristic parameters in different fault states is compared and analyzed,and the feasibility of the multi feature parameter fusion diagnosis method in the in-wheel motor fault diagnosis is verified.The electrical fault diagnosis of in-wheel motor based on multi-feature parameters fusion shows that the Weibull distribution of the fusion feature parameters is distinctly differentiated between the normal and the fault states.The fusion characteristic parameters can fully characterize the fault state of the in-wheel motor.And by fitting the scale range parameter a_i of the Weibull distribution,the failure degree of the wheel motor is reflected.