Life And Reliability Analysis Of Related Parts Of Pure Electric Vehicle

With the increasing attention to energy and environmental protection,all countries in the world are speeding up the pace of research and development of electric vehicles.However,the safety and reliability of electric vehicle is an important problem in the development of electric vehicle because of the lack of research on the reliability of electric vehicle parts and components due to the rush of automobile manufacturers to launch their products.In this context,the main parts of electric vehicles are studied in detail.First of all,by consulting relevant materials,the domestic and foreign related studies were analyzed and summarized.Secondly,determine the research program and technical route of this paper,research methods.Reference parts reliability evaluation is presented in this paper,on the basis of using the artificial neural network and least squares support vector machine(SVM)and the reliability theory and method,in view of the pure electric car motor service life evaluation for the following research:First of all,introduce the weibull distribution reliability analysis,correlation and regression analysis,the least squares support vector machine and neural network relevant theories and research status at home and abroad about auto components reliability analysis.Secondly,select the same configuration of 18 Electric pure electric vehicle motors,collect relevant fault data in the same test environment,and make a statistical description and analysis of their life,and give the related results.On this basis,the mathematical expectation of the empirical distribution function to calculate the probability analysis method was established,and the neural network model of the service life of the motor the least squares support vector machine,and two kinds of estimation method,neural network standard mean square error is less than the minimum two by support vector machine.Finally,according to the actual fault data of the motor,the shape parameter of Weibull distribution is determined to be 4.5003,and the scale parameter is 3.7626.The shape parameter of Weibull distribution predicted by neural network algorithm is 4.4995,and the scale parameter is 3.7619.The mean square deviation is NRMS=0.0300.Therefore,the prediction results of neural network algorithm are better.