Study On Evaluation Method Of Sound Quality Of The Vehicle Exhaust Noise

With the rapid development of Automobile industry and the popularization of ordinary passenger cars, Noise、Vehicle and Harshness(NVH) of vehicles are increasingly concerned by people, this also caused the attention of the scholars at home and abroad and car manufacturers. People’s attitude toward the car has become the psychological and physiological to enjoy together. While the evaluation about the vehicle exhaust noise more tend to psychological feelings. Therefore consequence, the objective evaluation index such as the traditional sound pressure level meter right has been far cannot meet the requirements. Not only can the sound quality of exhaust noise response noise levels, also can reflect the ear from people to noise of subjective feeling. So, it is great significant to study in the sound quality of vehicle exhaust noise.In this paper, we adopted the method of field measurement to collect exhaust noise under vehicle start conditions, and the noise of collected in the sample completed the preprocessing, and in accordance with the requirements of the subjective evaluation reference level evaluation method, to deal with the sample and establish the database. According the standard of ISO and the calculating of Zwicker,the traditional psychoacoustic parameters values(loudness, sharpness, roughness and dither strength)was calculated according to the relevant standard and calculation method; and at the same time using the theory of analytic wavelet decomposition of noise samples in each frequency band, and calculate the energy characteristic value of each frequency band. Through multiple linear regression and least squares support vector machine to establish sound quality evaluation model of exhaust noise respectively. The traditional psychoacoustic parameters and the energy characteristic as the input of the model, the exhaust noise subjective evaluation results for the output of the model.The analysis results of the model shows that: both traditional psychoacoustic parameters and the frequency band energy characteristic value are has good correlation with subjective assessment results, which are able to be used as the objective parameter of analysis the sound quality of exhaust noises. Analytic wavelet decomposition are used to get the energy characteristics value of each band as objective parameters, based on least squares support vector machine has the highest sound quality model prediction accuracy, can control the error within 10% very well. It was obviously seen that, it is effective that the method is applied to analysis and study the exhaust noise sound quality, can be used to predict the sound quality of exhaust noise. It has certain guiding significance to the auto harshness NVH research and control the environmental noise.