Research On Residual Life Prediction Method Of Permanent Magnet Synchronous Motor

Permanent magnet synchronous motors(PMSM)are widely used in numerical control equipment(CNC),electric vehicles and other products,because of it's small size,high efficiency,low moment of inertia and fast dynamic response etc.The PMSM is a key component of CNC,and can lead to CNC downtime when a fault occurs.Therefore,during the operation of PMSM,if the remaining life can be predicted based on its current health status,historical operation data,expected tasks and working environment conditions,it can ensure safe and effective operation under the prescribed conditions and avoid disasters accident,prolonging service life and reducing property damage.In this paper,the PMSM is taken as the research object,by using experimental analysis and theoretical analysis,the methods to predict the remaining life of PMSM was studied based on ARIMA.The main research contents and conclusions of this paper are listed as following:Chapter 1,presents the background,purpose and significance of this research.Give the overview of remaining life prediction methods,reliability testing of PMSM s and their remaining life prediction and development trend in domestic and abroad.Chapter 2,combining with PMSM working principle and structure,the typical failure causes of stator windings,sensors,inverters,permanent magnets,bearings and their influence on the performance of PMSM are determined through theoretical analysis and experiment.These lays the foundation of determining degradation performance parameters of PMSM and designing degradation tests.Chapter 3,according to the fault mechanism of PMSM,it is revealed that the negative sequence current can be used as a parameter to characterize the performance degradation of permanent magnet synchronous motor.Based on the fundamental test and analysis of the test data,a method for determining the failure threshold of performance degradation parameters is established.A PMSM performance degradation test device is designed,and reliability test is carried out,which provides reliable data for establishing the remaining life model of PMSM.Chapter 4,based on the ARIMA analysis model,combined with the experimental data of performance degradation parameters,combined with the test data of performance degradation test,firstly,the stationary time series is obtained through the differential operation of the time series of the nonstationary performance parameters.The model is preliminarily identified by the truncation and trailing of the character of partial correlation and autocorrelation coefficient,and the parameters of the models are estimated by the least square method.The model is discriminated and judged according to the AIC and SC information criterion to determine the best prediction model,and then the suitability of the prediction model is further determined by the chi square distribution test;Finally,predicting their remaining life.In order to improve the accuracy of ARIMA prediction results,the remaining life of each sample PMSM is been predicted several times,and the life distribution of the PMSM of each sample is obtained by fitting the multiple predictive values.Then a certain sample number is extracted from the life distribution function of each sample by the Monte Carlo model sampling method.And the lifetime distribution of the PMSM of this model is obtained by using the data extracted.Chapter 5,summarizes the research content of this article,and prospects for further research.