| In recent years,consumers have higher and higher requirements for car interior comfort,and the interior noise quality is the most intuitive indicator of car comfort.For electric vehicles,the driving motor is the main noise source in the vehicle,which has high frequency and high intensity,and brings a very irritable experience to people’s subjective perception.In order to reduce the interior noise of electric vehicles and optimize the interior noise quality,this paper takes new energy vehicle permanent magnet synchronous motor as the research object,analyzes the noise characteristics of the motor,and analyzes and predicts its sound quality.The main work is as follows.1)By analyzing the objective parameters of the motor,the noise characteristics of the motor are studied.Firstly,the permanent magnet synchronous motor test bench is built to test and collect the noise signal of the motor,and monitor and screen it.According to the sound quality theory,the objective noise parameters of the motor under different working conditions are calculated,and the correlation analysis,factor analysis and principal component analysis of the objective parameters are carried out to calculate the principal component score.Then the objective parameters corresponding to the electromagnetic noise and mechanical noise characteristics of the motor are determined,and the noise characteristics of the motor are obtained.Furthermore,the dimension of motor noise is reduced by the result of principal component score,which reduces the calculation workload.(2)An improved component pair comparison method based on Bradley Kullback selection weights to select the associated samples is established,and the improved subjective evaluation results are obtained.The subjective evaluation experiment is used to verify the method,and the results show that the method can solve the problem of unstable correlation effect in the sample group.(3)The MPGA-RBF prediction model for sound quality evaluation of permanent magnet synchronous motor is established.Based on the RBF neural network,the multi population genetic algorithm(MPGA)is used to optimize the RBF neural network,and the MPGA-RBF neural network based subjective evaluation and prediction model for the sound quality of permanent magnet synchronous motor is established.The model is tested by randomly reserved test samples,and the error between the predicted results and the actual evaluation value is less than 2%.It has good practicability,and can provide a certain degree of reference for the design and research of permanent magnet synchronous motor sound quality... |
