基于频带声能量的关门声声品质评价模型  

Sound quality evaluation model of door closing based on frequency band sound energy

作　　者：金龙娥 黄泽好 陈华语[1] 邹艾宏 陈宝 JIN Long’e;HUANG Zehao;CHEN Huayu;ZOU Aihong;CHEN Bao(Key Laboratory of Advanced Manufacturing Technology for Automobile Parts,Ministry of Education,Chongqing University of Technology,Chongqing 400054,China;Venicle Engineering Institute,Chongqing University of Technology,Chongqing 400054,China;Chongqing Jinkang Seres New Energy Vehicle Design Institute Co.,Ltd.,Chongqing 401120,China)

机构地区：[1]重庆理工大学车辆工程学院,重庆400054 [2]重庆理工大学汽车零部件先进制造技术教育部重点实验室,重庆400054 [3]重庆金康赛力斯新能源汽车设计院有限公司,重庆401120

出　　处：《重庆理工大学学报（自然科学）》2022年第10期111-118,共8页Journal of Chongqing University of Technology：Natural Science

基　　金：重庆市教委科研项目(KJQN20181101);重庆市研究生导师团队项目(渝教研发[2018]6号);重庆理工大学研究生创新项目(clgycx 20202024)。

摘　　要：针对汽车关门声信号的非稳态特性,采用小波包时频分析方法,建立以频带声能量为输入、主观偏好值为输出的汽车关门声声品质客观量化模型。采集汽车关门声声样本并进行声品质主观偏好性评价试验,建立基于24Bark域的小波包滤波器组,提取各听觉频带内的声能量,利用遗传算法优化的BP(back propagation)神经网络建立频带声能量与主观偏好值之间的客观量化模型。结果表明:与基于心理声学客观参数的关门声声品质评价模型相比,基于频带声能量的关门声声品质评价模型在误差绝对值均值、均方误差、拟合系数方面均有明显提高,说明基于频带声能量的关门声声品质评价模型的预测精度更高。Aiming at the unsteady characteristics of automobile door closing sound signal,an objective quantitative model of automobile door closing sound quality is established by using wavelet packet time-frequency analysis method,which takes frequency band sound energy as input and subjective preference value as output.The sound samples of car door closing are collected and the subjective preference evaluation test of sound quality is carried out.The wavelet packet filter banks based on 24 Bark domain are established to extract the sound energy in each auditory frequency band.The objective quantitative model between the sound energy and subjective preference value is established by BP(Back Propagation) neural network optimized by genetic algorithm.The results show that compared with the door-closing sound quality evaluation model based on psychoacoustic objective parameters,the door-closing sound quality evaluation model based on frequency band sound energy has significant improvement in the absolute value of the error,the mean square error,and the fitting coefficient.The prediction accuracy of the door-closing sound quality evaluation model based on the frequency band sound energy is higher.

关 键 词：汽车关门声声品质 小波包 遗传算法 神经网络 频带声能量 

分 类 号：U467.493[机械工程—车辆工程]