【摘要】 基于采集的汽车匀速与加速工况下的车内噪声信号,提取响度和A计权声压级两个主要心理声学客观参量作为声品质评价模型的输入特征,以参考语义细分法得到的综合烦躁度主观评分作为模型的输出量,基于SVM建立非平稳工况车内声品质客观评价模型。预测检验结果表明,与运用多元线性回归方法建立的评价模型相比,该模型预测误差均值、标准差及平均相对误差更小,车内声品质评价的预测精度、稳定性均有提高,所建的SVM客观评价模型具有较好的泛化能力,可用于非平稳工况车内噪声品质的预测。更多还原

【Abstract】 Based on the collected vehicle interior noise under uniform speed and acceleration conditions,an objective model for vehicle interior sound quality evaluation under non-stationary conditions is established based on SVM,with two main psychoacoustics objective parameters of loudness and A Weighted Sound Pressure Level as the input characteristics and with the subjective score of comprehensive irritability obtained by reference semantics subdivision method used as the output of the model.The test results show that the mean,standard deviation and average relative error of the model are smaller than those of the evaluation model established by the multivariate linear regression method,and the prediction accuracy and stability of sound quality evaluation are improved.The SVM model has good generalization ability and can be used to predict the vehicle interior noise quality of uniform speed and accelerated working conditions.更多还原