文章摘要
杨东坡,王孝兰,王岩松,郭辉,刘宁宁.高速工况下乘员耳侧噪声信号重构方法[J].声学技术,2019,38(3):340~347
高速工况下乘员耳侧噪声信号重构方法
The reconstruction method of occupant's ear side noise under high speed condition
投稿时间:2018-01-20  修订日期:2018-03-04
DOI:10.16300/j.cnki.1000-3630.2019.03.018
中文关键词: 高速工况|车内噪声|经验模态分解|BP神经网络|信号重构
英文关键词: high speed condition|vehicle interior noise|empirical mode decomposition(EMD)|BP neural network|signal reconstruction
基金项目:国家自然科学基金(51675324、51175320)
作者单位E-mail
杨东坡 上海工程技术大学汽车工程学院, 上海 201620  
王孝兰 上海工程技术大学汽车工程学院, 上海 201620 jlu_wangxiaolan@aliyun.com 
王岩松 上海工程技术大学汽车工程学院, 上海 201620  
郭辉 上海工程技术大学汽车工程学院, 上海 201620  
刘宁宁 上海工程技术大学汽车工程学院, 上海 201620  
摘要点击次数: 1001
全文下载次数: 438
中文摘要:
      高速工况下,车内噪声信号具有随机性和波动性的特征。将一种基于经验模态分解(Empirical Mode Decomposition,EMD)和反向传输(Back Propagation,BP)神经网络的算法,用于重构车内乘员耳侧噪声信号。首先通过对车内乘员耳侧噪声贡献量分析,确定关键噪声源信号;其次对选择的噪声源信号进行EMD分解,得到有限个相对平稳的固有模态函数(Intrinsic Mode Function,IMF)分量;然后采用极值点划分法,按各个分量的波动情况进行重新划分,将信号分量重构为高频、中频和低频3个分量;最后对不同频段的部分建立相应BP神经网络模型,并将不同频段分量的重构结果叠加作为原信号的重构结果。以在某轿车采集到的5个噪声信号源为基础,利用该方法进行乘员耳侧噪声信号重构,并对其进行分析。结果表明:提出的噪声重构方法可以实现高速工况乘员耳侧噪声信号的重构,并具有良好的性能。
英文摘要:
      Vehicle interior noise signals have the characteristics of randomness and volatility under high speed condition. An algorithm based on Empirical Mode Decomposition (EMD) and BP neural network is applied to reconstructing the occupants' ear side noise in this paper. Firstly, the critical noise source signal is determined by the analysis of the contributing to vehicle interior occupants' ear side noise. Secondly, the selected noise source signal is decomposed into finite relatively stable IMF components by EMD decomposition. Then, the extreme point division method is adopted to divide the signals according to the fluctuation of each component, and the signal components are reconstructed into high frequency components, middle frequency components and lower frequency components. Finally, the corresponding BP neural network models are established for the signals in different frequency bands, and the reconstruction results in different frequency bands are superposed as the reconstruction result of the original signal. Based on five noise signal sources collected in a car, the noise signals in occupants' ears side are reconstructed and the reconstruction results are analyzed. The results show that the noise reconstruction method proposed in this paper can realize the reconstruction of occupants' ear side noise under high speed condition and has good performance.
查看全文   查看/发表评论  下载PDF阅读器
关闭