基于经验模态分解的RV减速器运动参数降噪研究  

Denoising of the kinetic parameters of a RV reducer based on an EMD algorithm

作　　者：宋雷 游东东[2] 郑振兴 周玉山 陈龙崇 SONG Lei;YOU Dongdong;ZHENG Zhenxing;ZHOU Yushan;CHEN Longchong(School of Mechanical and Electrical Engineering,Guangdong Polytechnic Normal University,Guangzhou 510635,China;School of Mechanical&Automotive Engineering,South China University of Technology,Guangzhou 510640,China;Institute of Automotive Engineering Research,Guangzhou Automobile Group Co.,Ltd.,Guangzhou 510640,China)

机构地区：[1]广东技术师范大学机电学院,广州510635 [2]华南理工大学机械与汽车工程学院,广州510640 [3]广州汽车集团股份有限公司汽车工程研究院,广州510640

出　　处：《振动与冲击》2022年第18期266-272,290,共8页Journal of Vibration and Shock

基　　金：国家自然科学基金(51875209);广东省自然科学基金(2018A030313509)。

摘　　要：RV(rotate vector)减速器是现代智能装备中关键传动机构之一,其运动参数的检测信号噪声成分较高,影响了装备的运行精度和稳定性,提出了一种以经验模态分解(empirical mode decomposition,EMD)为基础的信号降噪方法,有效提取了RV减速器运动参数的准确信号。该方法采用连续均方误差(consecutive mean square error,CMSE)和l 2范数两个指标将EMD分解得到的固有模态函数(intrinsic mode function,IMF)区分为噪声IMF、噪声与信息相混合IMF、信息IMF 3个部分,采用不同策略对这三部分IMF进行处理并结合部分重构(part reconstruction,PR)策略实现了信号的降噪处理。采用该文章提出的降噪方法对RV40E减速器的检测信号进行处理,降噪后的信号信噪比得到明显改善,验证了该降噪方法的有效性。The rotate vector(RV)reducer is one of key transmission mechanisms in modern intelligent equipment.There is usually a great proportion of noise in the measurement signals of its kinetic parameters,which affects the operation accuracy and stability of equipments.A signal-denoising approach based on EMD(empirical mode decomposition) was proposed,which can extract accuratly kinetic parameter signals of the RV reducer effectively.In the approach,the IMFs(intrinsic mode functions) derived from EMD were divided into 3 parts,namely,noise IMFs,noise and information,mixed IMFs and information IMFs,with 2 indexes,CMSE(consecutive mean square error) and l 2-norm.Different processing strategies were applied in the 3 parts of IMFs and combining with the PR(part reconstruction) to fulfill the process of denoising.The denoising approach presented has been used in the denoising process of the torque signals of a RV40 E reducer.The results show that the SNRs of the denoised signals are improved observably,and the effectiveness of the approach is validated.

关 键 词：经验模态分解(EMD) RV减速器 连续均方误差(CMSE) l 2范数 信号降噪 

分 类 号：TH1323[机械工程—机械制造及自动化]