摘 要：為準確評價車內加速噪聲聲品質，利用成對比較法對14款汽車從30 km/h加速到80 km/h行駛時的車內噪聲進行主觀評價試驗，同時計算5個主要心理聲學客觀參量，并通過相關分析得出對聲品質有重要影響的客觀參量。采用支持向量機建立車內加速噪聲聲品質的預測模型，經驗證其預測相對誤差都在9.5%以內，表明該模型可以準確地對車內噪聲聲品質進行預測。
Study on the prediction of car acceleration noise sound quality based on
support vector machine
ZHOU Minggang， LIU Yang， CHEN Yuan， WEN Yao
（Engineering Design and Research Institute of Agricultural Machinery，
Hubei University of Technology，Wuhan 430068，China）
Abstract: To accurately evaluate the sound quality of car interior accelerating noise， the noise samples from 14 different brands of cars during acceleration from 30 km/h to 80 km/h were selected， and subjective evaluation test was carried out with the paired comparison method， meanwhile， five main psychoacoustic parameters were calculated， and the objective parameters having significant effect on the sound quality were extracted by the correlation analysis. A car interior accelerating sound quality prediction model was built based on support vector machine （SVM）， and the relative predict error had been proved less than 9.5% these shows that the prediction model can accurately predict the sound quality of the car interior noise.
Keywords: sound quality； accelerating condition； support vector machine； prediction model