变速器啸叫声品质的RBF神经网络预测与权重分析

PDF(1352 KB)

振动与冲击 ›› 2017, Vol. 36 ›› Issue (6) : 175-180.

论文

施全1，柳培海2，郭栋2，易鹏3

作者信息 +

Transmission whine noise quality prediction and weight analysis based on RBF neural network

SHI Quan1，LIU Pei-hai1 ，GUO Dong1 ， YI Peng2

Author information +

文章历史 +

摘要

以变速器啸叫为研究对象，提出使用径向基(RBF)神经网络的方法来确定变速器声品质评价中客观评价参数对主观评价结果的影响权重。采集变速器不同位置的声音信号作为试验样本，用等级评分法对111个样本信号进行了主观评价试验，同时计算11个声品质客观评价参数。以客观评价参数计算结果为神经网络输入，声品质主观评价结果为输出，引入径向基神经网络建立了变速器声品质预测模型。以预测模型为基础，利用各网络层间连接权值，计算变速器声品质客观评价参数对主观评价结果的影响权重。研究结果表明：变速器啸叫声品质主要受SIL-4、总响度和随时间响度三个客观参量的影响。

Abstract

The method of calculating the impact weight of objective psychoacoustic metrics on subjective evaluation results is proposed using RBF neural network. The gear whine signals were collected in different locations using as the evaluating samples. Subjective sound quality evaluation testing for the 111 noise samples were conducted. Meanwhile, eleven sound quality objective parameters were calculated. By using objective parameters as inputs and subjective values as outputs, a RBF neural network was adopted to establish gear whine sound quality prediction model. The network connection coefficients of the prediction model were used to calculate the impact weight of objective parameters on the results of subjective evaluation. The calculation results show that the SIL-4, the sharpness and loudness over time are the key psychoacoustic parameters to conduct gear whine sound quality.

导出引用

施全1，柳培海2，郭栋2，易鹏3. 变速器啸叫声品质的RBF神经网络预测与权重分析[J]. 振动与冲击, 2017, 36(6): 175-180

SHI Quan1，LIU Pei-hai1,GUO Dong1,YI Peng2 . Transmission whine noise quality prediction and weight analysis based on RBF neural network[J]. Journal of Vibration and Shock, 2017, 36(6): 175-180

参考文献

[1] Norm otto, Scott Amman, Chris Eatonm, et al. Guidelines for Jury Evaluations of Automotive Sounds [J]. SAE International 1999-01-1820-2.
[2] 王登峰,刘宗巍,梁杰，等. 车内噪声品质的主观评价试验与客观量化描述[J]. 吉林大学学报 (工学版), 2006,S2:41-45.
    WANG Deng-feng, LIU Zong-wei, LIANG Jie, et al. Subjec-
tive evaluation test and objective quantificational description of vehicle interior noise quality [J], Journal of Jilin University (Engineering and Technology Edition), 2006,S2:41-45.
[3] 申秀敏，左曙光，何容，等. 燃料电池轿车声品质客观评价参量的权重[J]. 振动与冲,2011，30(1):91-96.
SHEN Xiu-min, ZUO Shu-guang, He Rong, et al. Weight of objective evaluation parameters for fuel cell vehicle sound quality [J]. Journal of Vibration and Shock, 2011, 30(1):91-96.
[4] 苏丽俐,王登峰,王倩. 基于GRNN的车内噪声品质预测[J]. 吉林大学学报(工学版)，2011,S2 :82-86.
    SU Li-li, WANG Deng-feng, WANG Qian. Sound quality pre-diction of vehicle interior noise based on GRNN[J], Journal of Jilin University(Engineering and Technology Edition), 2011,S
2:82:86.
[5] 高印寒,唐荣江，梁杰，等. 基于径向基函数神经网络的车内噪声品质评价系统[J].吉林大学学报（工学版），2012,4   2(6):1378-1383.
GAO Yin-han, TANG Rong-jiang, LIANG Jie,et al. Vehicle i-nterior noise sound quality System based on RBF neural net-work [J]. Journal of Jilin University(Engineering and Technol-
ogy Edition), 2012,42(6):1378-1383.
[6] 高印寒,唐荣江，梁杰,等.汽车声品质的GA-BP神经网络预测与权重分析[J].光学精密工程,2013,21(2):462-468.
GAO Yin-han, TANG Rong-jiang, LIANG Jie,et al. Sound qua-lity prediction and weight analysis of vehicles based on GA-BP neural network[J].Optics and Precision Engineering,2013,21(2):462-468.
[7] 吴小珊.变速器声品质室内评价方法研究[D].重庆：重庆理工大学，2015.
WU Xiao-shan. Laboratory sound quality evaluation Of trans-mission [D]. Chong Qing: Chong Qing University of Technolo-gy, 2015.
[8] 梁杰. 基于双耳听觉模型的车内声品质分析与评价方法研究[D].长春:吉林大学汽车工程学院,2007.
    LIANG Jie. Research on analysis and evaluation Method of vehicle interior sound quality based on binaural auditory mod-el [D]. College of Automotive Engineering, Jilin University,
2007.
[9] Noumura K, Yoshida J. Perception Modeling and Quantification of Sound Quality in Cabin[J]. SAE international, 2003-01-1514
[10] ALEBOYEH A, KASIRI M B. Prediction of azo dye decolori-zation by UV/H2O2 using artificial neural networks[J]. Dyesa-nd Pigments, 2008(77): 288-294.