摘要
以某增程式电驱动总成为研究对象,提出基于联合算法的噪声分离识别模型。首先,采用互补集合经验模态分解(complementary ensemble empirical mode decomposition,CEEMD)联合快速独立分量分析(fast independent component analysis,FastICA)方法提取纯电模式稳态工况下单一通道噪声信号特征,利用复Morlet小波变换及FFT对各分量信号时频特性进行识别。其次,采用阶次分析法和声能叠加法对稳态分量信号对应的各瞬态响应阶次能量进行对比分析,并结合皮尔逊积矩相关系数(Pearson product moment correlation coefficient,PPMCC)相似性识别确定不同噪声激励源贡献度。结果表明:减速齿副啮合噪声对该增程式电驱总成纯电模式运行噪声整体贡献度最大。
Taking an extended-range electric drive assembly as the research object,this paper proposes a noise separation and identification model based on a joint algorithm.Firstly,the characteristics of single channel noise signal in pure electric mode are extracted by using complementary ensemble empirical mode decomposition(CEEMD)combined with fast independent component analysis(FastICA).At the same time,complex Morlet wavelet transform and FFT are used to identify the time-frequency characteristics of each component signal.And then,the order analysis method and the sound energy superposition method are used to compare and analyze the energy of each response order under transient conditions,and combine with the similarity recognition method of the Pearson product moment correlation coefficient(PPMCC)to determine the contribution of different noise excitation sources.The results show that the meshing noise of the gear pair of the reducer has the greatest contribution to the overall operating noise of the extended-range electric drive assembly in the pure electric mode.
作者
张威
景国玺
武一民
杨征睿
高辉
ZHANG Wei;JING Guoxi;WU Yimin;YANG Zhengrui;GAO Hui(Tianjin Key Laboratory of Power Transmission and Safety Technology for New Energy Vehicles,Hebei University of Technology,Tianjin 300400,China;CATARC(Tianjin)Automotive Engineering Research Institute Co.,Ltd.,Tianjin 300300,China;China Automotive Technology&Research Center Co.,Ltd.,Tianjin 300300,China)
出处
《中国测试》
CAS
北大核心
2024年第4期144-152,共9页
China Measurement & Test
基金
国防重点实验室基金资助项目(6142212200309)。
关键词
电驱动总成
噪声源识别
互补集合经验模态分解
快速独立分量分析
连续小波变换
阶次分析
electric drive assembly
noise source identification
complementary ensemble empirical mode decomposition
fast independent component analysis
continuous wavelet transform
order analysis
