Page 155 - 《应用声学》2022年第4期

第 41 卷 第 4 期 贾尚帅等： 利用卷积网络的高速列车主观声品质预测 653

数与车内噪声的响度负相关系数最大，而与抖动度 Shen Xiumin, Zuo Shuguang, Han Le, et al. Interior ve-
的负相关系数最小。 hicle noise quality prediction using support vector ma-
chines[J]. Journal of Vibration, Measurement & Diagno-
(4) 建立了基于 CNN 的声品质预测模型，将同
sis, 2011, 31(1): 55–58, 128.
时包含车内噪声时域和频域信息的时 -频分布图作 [7] Fang Y, Zhang T. Sound quality investigation and im-
为模型输入，模型更具有真实物理意义，预测精度 provement of an electric powertrain for electric vehicles[J].
IEEE Transactions on Industrial Electronics, 2018, 65(2):
比 BP 神经网络模型更高，更适宜用于指导高速列
1149–1157.
车车内声品质的优化设计。 [8] Liu H, Zhang J H, Guo P, et al. Sound quality predic-
tion for engine-radiated noise[J]. Mechanical Systems and
Signal Processing, 2015, 56–57: 277–287.
参 考 文 献 [9] Xing Y F, Wang Y S, Shi L, et al. Sound quality recogni-
tion using optimal wavelet-packet transform and artificial
[1] 代文强, 郑旭, 郝志勇, 等. 采用能量有限元分析的高速列 neural network methods[J]. Mechanical Systems and Sig-
车车内噪声预测 [J]. 浙江大学学报 (工学版), 2019, 53(12): nal Processing, 2016, 66–67: 875–892.
2396–2403. [10] Pietila G, Lim T C. Intelligent systems approaches to
Dai Wenqiang, Zheng Xu, Hao Zhiyong, et al. Predic- product sound quality evaluations-a review[J]. Applied
tion of high-speed train interior noise using energy finite Acoustics, 2012, 73(10): 987–1002.
element analysis[J]. Journal of Zhejiang University (Engi- [11] Zhang E L, Hou L, Shen C, et al. Sound quality prediction
neering Science), 2019, 53(12): 2396–2403. of vehicle interior noise and mathematical modeling using
[2] Luo L, Zheng X, Hao Z Y, et al. Sound quality evaluation a back propagation neural network (BPNN) based on par-
of high-speed train interior noise by adaptive Moore loud- ticles warm optimization (PSO)[J]. Measurement Science
ness algorithm[J]. Journal of Zhejiang University-Science and Technology, 2016, 27(1): 015801.
A: Applied Physics & Engineering, 2017, 18(9): 690–703. [12] Cao J W, Cao M, Wang J Z, et al. Urban noise recog-
[3] 鞠龙华, 葛剑敏. 强噪声环境下高速列车内语言清晰度评价与 nition with convolutional neural network[J]. Multimedia
分析 [J]. 同济大学学报 (自然科学版), 2017, 45(7): 994–999. Tools and Applications, 2019, 78(20): 29021–29041.
Ju Longhua, Ge Jianmin. Evaluation and analysis of [13] Huang H B, Wu J H, Lim T C, et al. Pure electric vehi-
speech intelligibility in high-speed train compartments un- cle nonstationary interior sound quality prediction based
der strong noise environment[J]. Journal of Tongji Univer- on deep CNNs with an adaptable learning rate tree[J].
sity (Natural Science), 2017, 45(7): 994–999. Mechanical Systems and Signal Processing, 2021, 148:
[4] Park B, Jeon J, Choi S, et al. Short-term noise annoy- 107170.
ance assessment in passenger compartments of high-speed [14] Otto N, Amman S, Eaton C, et al. Guidelines for jury
trains under sudden variation[J]. Applied Acoustics, 2015, evaluations of automotive sounds[J]. SAE Technical Pa-
97: 46–53. per 1999-01-1822, 1999.
[5] Li J, Li L, Zhang Y X, et al. Annoyance evaluation of [15] 孟凡雨. 高速列车车内声品质评价研究 [D] . 哈尔滨: 哈尔滨
noise emitted by urban substation[J]. Journal of Low Fre- 工业大学, 2013.
quency Noise, Vibration and Active Control, 2021, 40(4): [16] Ferreira M D, Corrêa D C, Nonato L G, et al. Design-
2106–2114. ing architectures of convolutional neural networks to solve
[6] 申秀敏, 左曙光, 韩乐, 等. 基于支持向量机的车内噪声声品 practical problems[J]. Expert Systems with Applications,
质预测 [J]. 振动、测试与诊断, 2011, 31(1): 55–58, 128. 2018, 94: 205–217.