基于GAN-FCNN数据增强的特种车辆声品质预测研究

doi:10.19562/j.chinasae.qcgc.2025.08.015

Abstract

Abstract:

For the special vehicle sound quality prediction， the cost of collecting noise samples is high， and only a small number of sample sets can be obtained after sample processing， lacking sufficient noise samples， which affects the model accuracy during the training of various prediction models. In this paper， a GAN-FCNN network is established， and a four-layer fully connected layer is used to construct a generator and discriminator for adversarial training， and a pseudo-sample set is generated. The enhanced sample set is introduced into the LASSO linear regression model and the RF， BP and PSO optimization models respectively for regression prediction. Through verification， the prediction accuracy and performance of the models are improved. Compared with the traditional oversampling algorithm， the GAN-FCNN network has higher accuracy， which is more suitable for sample expansion in the establishment of special vehicle sound quality prediction model.

Key words: data enhancement, special vehicle, sound quality, GAN, oversampling

Kun Qian,Xikang Du,Yanfu Wang,Jiying Duan,Ke Liu,Jing Tan,Zhenghua Shen,Jian Zhao. Research on Sound Quality Prediction of Special Vehicles Enhanced with GAN-FCNN Data[J].Automotive Engineering, 2025, 47(8): 1596-1606.

Figures/Tables 22

References 26

[1]	陈剑，蒋丰鑫，肖悦. 挖掘机驾驶室低频噪声分析与控制［J］. 中国机械工程， 2014， 25（15）： 2124-2130.
	CHEN J， JIANG F X， XIAO Y. Analysis and control of low frequency noise in excavator cab ［J］. China Mechanical Engineering， 2014， 25（15）： 2124-2130.
[2]	柳亚，张键，杨鄂川，等. 某特种车辆车内噪声声品质分析［J］. 重庆理工大学学报（自然科学）， 2017， 31（3）： 34-42.
	LIU Y，ZHANG J，YANG E C，et al. Analysis of sound quality of interior noise of a special vehicle ［J］. Journal of Chongqing University of Technology （Natural Sciences）， 2017， 31（3）： 34-42.
[3]	张勇，王坤祥，欧健，等. 基于RBF神经网络的车内声品质预测及分析［J］. 控制工程， 2019， 26（11）： 2025-2030.
	ZHANG Y， WANG K X， OU J， et al. Interior sound quality prediction and analysis based on RBF neural network ［J］. Control Engineering， 2019， 26（11）： 2025-2030.
[4]	欧健，彭沸潭，张庆庭，等. 极限梯度提升声品质预测模型在车内噪声主动控制中的运用［J］. 振动工程学报， 2023， 36（5）： 1349-1355.
	OU J， PENG F T， ZHANG Q T，et al. Application of limit gradient enhanced sound quality prediction model in active vehicle noise control ［J］. Journal of Vibration Engineering， 2023， 36（5）： 1349-1355.
[5]	张勇，孟天，王坤祥，等. 车内声品质主观评价模型及中频噪声优化［J］. 汽车工程， 2020， 42（5）： 651-657，664.
	ZHANG Y， MENG T， WANG K X， et al. Subjective evaluation model of interior sound quality and intermediate frequency noise optimization ［J］.Automotive Engineering，2020，42（5）：651-657，664.
[6]	MüNDER M， CARBON C C. Howl， whirr， and whistle： the perception of electric powertrain noise and its importance for perceived quality in electrified vehicles ［J］. Applied Acoustics， 2022， 185： 108412.
[7]	LIU Z， XIE L， HUANG T， et al. The objective quantification of door closing sound quality based on multidimensional subjective perception attributes ［J］. Applied Acoustics， 2022， 192： 108748.
[8]	XIE L， LU C， LIU Z， et al. The evaluation of automobile interior acceleration sound fused with physiological signal using a hybrid deep neural network ［J］. Mechanical Systems and Signal Processing， 2023， 184： 109675.
[9]	HAN H， WANG W Y， MAO B H. Borderline-SMOTE： a new over-sampling method in imbalanced data sets learning ［C］. Proceedings of the International Conference on Intelligent Computing， F， 2005. Springer.
[10]	CHAWLA N V. Data mining for imbalanced datasets： an overview ［M］. Data Mining and Knowledge Discovery Handbook， 2010： 875-886.
[11]	DABLAIN D， CHAWLA N. Towards understanding how data augmentation works with imbalanced data ［J］. ArXiv， 2023， abs/2304.05895.
[12]	YAQIN A， RAHARDI M， ABDULLOH F F. Accuracy enhancement of prediction method using SMOTE for early prediction student's graduation in XYZ university ［J］. International Journal of Advanced Computer Science and Applications， 2022， 13（6）.
[13]	CHAWLA N V， BOWYER K W， HALL L O， et al. SMOTE： synthetic minority over-sampling technique ［J］. Journal of Artificial Intelligence Research， 2002， 16： 321-357.
[14]	GOODFELLOW I， POUGET-ABADIE J， MIRZA M， et al. Generative adversarial nets ［J］. Advances in Neural Information Processing Systems， 2014， 27.
[15]	中华人民共和国国家质量监督检验检疫总局. 声学汽车车内噪声测量方法： GB/T 18697—2002［S］. 北京：中国标准出版社， 2002.
	General Administration of Quality Supervision， Inspection and Quarantine of the People's Republic of China. Acoustics-Measurement of noise inside motor vehicles： GB/T 18697—2002［S］. Beijing： China Standard Press， 2002.
[16]	GUASTADISEGNI G， DE PINTO S， CANCELLI D， et al. Ride analysis tools for passenger cars： objective and subjective evaluation techniques and correlation processes-a review ［J］. Vehicle System Dynamics， 2024， 62（7）： 1876-1902.
[17]	钱堃，刘珂，王言夫，等. 电动汽车车内声品质评价研究进展［J］. 汽车工程， 2024， 46（8）： 1431-1446.
	QIAN K， LIU K， WANG Y F， et al. Research progress of in-car sound quality evaluation for electric vehicles ［J］. Automotive Engineering， 2024， 46（8）： 1431-1446.
[18]	SINGH RAWAT S， MISHRA A K. Review of methods for handling class-imbalanced in classification problems ［J］. arXiv e-prints， 2022： arXiv： .
[19]	杨易，高骏，谷正气，等. 基于GA-BP的汽车风振噪声声品质预测模型［J］. 机械工程学报， 2021， 57（24）： 241-249.
	YANG Y， GAO J， GU Z Q， et al. Prediction model of automobile wind noise quality based on GA-BP ［J］. Journal of Mechanical Engineering， 2021， 57（24）： 241-249.
[20]	刘哲，高云凯，解馥荣. 电动汽车关门声品质预测模型研究［J］. 汽车工程， 2021， 43（12）： 1858-1864.
	LIU Z， GAO Y K， XIE F R. Research on prediction model of electric vehicle door closing sound quality ［J］. Automotive Engineering， 2021， 43（12）： 1858-1864.
[21]	黄海波，黄晓蓉，苏瑞强，等. 基于EEMD与GA-小波神经网络的传动系声品质预测［J］. 振动与冲击， 2017， 36（9）： 130-137.
	HUANG H B， HUANG X R， SU R Q， et al. Transmission line sound quality prediction based on EEMD and GA-wavelet neural network ［J］. Vibration and Shock， 2017， 36（9）： 130-137.
[22]	ZHAO J， LING H， LIU J， et al. Machine learning for predicting battery capacity for electric vehicles ［J］. eTransportation， 2023， 15： 100214.
[23]	DAS K， KUMAR R， KRISHNA A. Analyzing electric vehicle battery health performance using supervised machine learning ［J］. Renewable and Sustainable Energy Reviews， 2024， 189： 113967.
[24]	GAUTHIER P A， SCULLION W， BERRY A. Sound quality prediction based on systematic metric selection and shrinkage： comparison of stepwise， lasso， and elastic-net algorithms and clustering preprocessing ［J］. Journal of Sound and Vibration， 2017， 400： 134-153.
[25]	彭沸潭. 某特种车车内声品质预测与主动控制研究［D］. 重庆：重庆理工大学，2023.
	PENG F T. Research on interior sound quality prediction and active control of a special vehicle ［D］. Chongqing： Chongqing University of Technology，2023.
[26]	JAIN M， SAIHJPAL V， SINGH N， et al. An overview of variants and advancements of PSO algorithm ［J］. Applied Sciences， 2022， 12（17）： 8392.

评价者	1	2	3	4	5	6	7	8
相关系数	0.707	0.701	0.700	0.723	0.797	0.810	0.854	0.875
评价者	9	10	11	12	13	14	15	16
相关系数	0.877	0.883	0.873	0.877	0.834	0.875	0.877	0.870
评价者	17	18	19	20	21	22	23	24
相关系数	0.865	0.869	0.882	0.875	0.859	0.875	0.877	0.874
评价值	25	26	27	28	29	30	31	32
相关系数	0.871	0.838	0.831	0.881	0.857	0.842	0.877	0.861

模型指标	RF	BP	PSO（RF）	PSO（BP）	LASSO
RMSE	0.298 1	0.333 0	0.147 0	0.238 8	0.935 0
MAE	0.242 3	0.279 9	0.126 2	0.168 8	0.810 5
MSE	0.088 8	0.110 9	0.021 6	0.057 0	0.874 3
相关系数	0.901 0	0.881 0	0.973 6	0.942 4	0.155 9

模型指标	RF	BP	PSO（RF）	PSO（BP）	LASSO
RMSE	0.210 3	0.196 0	0.110 5	0.195 3	0.868 7
MAE	0.148 1	0.165 7	0.084 1	0.167 3	0.776 5
MSE	0.044 2	0.038 4	0.012 2	0.038 1	0.754 7
相关系数	0.959 4	0.963 3	0.988 3	0.964 9	0.270 0

模型指标	RF	BP	PSO（RF）	PSO（BP）	LASSO
RMSE	0.122 6	0.134 6	0.083 0	0.113 9	0.209 9
MAE	0.095 5	0.080 3	0.065 6	0.081 9	0.164 7
MSE	0.015 0	0.018 1	0.006 9	0.012 9	0.044 1
相关系数	0.986 1	0.983 2	0.993 7	0.988 9	0.963 1

Research on Sound Quality Prediction of Special Vehicles Enhanced with GAN-FCNN Data

RichHTML

PDF (PC)

Like

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 22

References 26

Related Articles 12

Metrics

Comments

Recommended 0

[1]	Xiang Wang,Pengbo Liu,Jian Zhao,Kefeng Fan,Linhui Li. Research on Data Enhancement Methods of BEGAN-Based Intrusion Detection in Automotive CAN Networks [J]. Automotive Engineering, 2024, 46(8): 1394-1402.
[2]	Kun Qian,Ke Liu,Yanfu Wang,Haoyang Li,Jing Tan,Zhenghua Shen,Xikang Du,Jiying Duan,Jian Zhao. Progress in Research on Interior Sound Quality Evaluation of Electric Vehicles [J]. Automotive Engineering, 2024, 46(8): 1431-1446.
[3]	Dafang Wang,Jingdong Du,Jiang Cao,Mei Zhang,Gang Zhao. Research on Style Transfer Network for Autonomous Driving Data Generation [J]. Automotive Engineering, 2022, 44(5): 684-690.
[4]	Zhe Liu,Yunkai Gao,Furong Xie. Study on Predictive Models for Door Slamming Sound Quality of an Electric Vehicle [J]. Automotive Engineering, 2021, 43(12): 1858-1864.
[5]	Cao Yuntao, Tang Lechao, Liu Yingjie. Research on Regulation Adaptation for Acoustic Vehicle Alerting System in Electric Vehicles [J]. Automotive Engineering, 2020, 42(8): 1110-1116.
[6]	Bi Fengrong, Huang Yu, Zhang Lipeng, Shen Pengfei, Lü Dali. Research on Subjective Evaluation Method of Vehicle Sound Quality Based on Interval Grey Number Theory [J]. Automotive Engineering, 2020, 42(7): 933-940.
[7]	Zeng Falin, Hu Feng. Optimization of Engine Mount Based on Sound Quality Contribution Factor [J]. Automotive Engineering, 2020, 42(5): 628-635.
[8]	Zhang Yong, Meng Tian, Wang Kunxiang, Han Chenyang, Yang Echuan. Subjective Evaluation Model for Interior Sound Quality and Optimization of Medium-frequency Noise [J]. Automotive Engineering, 2020, 42(5): 651-657.
[9]	Liu Zhien, Huang Tao, Shao Jiongyang, Zheng Qingqing, Du Songze, Li Qiuyue, Qu Shaoju, Liu Hao. Test and Analysis of Rapid Acceleration Intake Noise in Automobiles [J]. Automotive Engineering, 2019, 41(5): 564-570.
[10]	Bai Zhonghao, Li Zhiqiang, Jiang Binhui, Wang Penghui. Real-time Pedestrian Detection in Advanced Driver Assistance Systems Based on Improved YOLOv2 Model [J]. Automotive Engineering, 2019, 41(12): 1416-1423.
[11]	Lu Yuan, Jin Peng, Wang Yulei, Liao Yi. Time-domain Transmission Path Analysis of Vehicle Interior Noise Quality [J]. Automotive Engineering, 2019, 41(12): 1435-1441.
[12]	Zang Libin, Chen Yong, Chen Hua, Liu Hai, Zhou Huidong, Li Kai & Luo Daguo. Simulation and Experimental Study on Temperature Field Characteristics of Coated Gears in Automatic Transmission [J]. , 2018, 40(9): 1054-1061.