基于经验小波变换的振动信号特征量提取  

Feature extraction of vibration signals based on empirical wavelet transform

作　　者：王茜[1,2] 田慕琴[1,2] 宋建成[1,2] 贺颖[1,2] 冯君玲[1,2] 吝伶艳[1,2] WANG Xi;TIAN Muqin;SONG Jiancheng;HE Ying;FENG Junling;LIN Lingyan(Nationaland Local Joint Engineering Laboratory of Mining Intelligent Electrical Apparatus Technology,Taiyuan University of Technology,Taiyuan 030024,China;Shanxi Key Laboratory of Mining Electrical Equipmentand Intelligent Control,Taiyuan University of Technology,Taiyuan 030024,China)

机构地区：[1]太原理工大学,矿用智能电器技术国家地方联合工程实验室,太原030024 [2]太原理工大学,煤矿电气设备与智能控制山西省重点实验室,太原030024

出　　处：《振动与冲击》2021年第16期261-266,共6页Journal of Vibration and Shock

基　　金：国家自然科学基金(U1510112);山西省重点研发项目(201803D121008)。

摘　　要：为解决岩巷掘进机动载荷识别困难问题,提出了一种基于经验小波变换(EWT)和相关性阈值去噪相结合的掘进机截割头振动信号特征量提取方法。不同岩壁硬度下的掘进机截割头振动信号经过EWT处理变为若干个分量信号;利用相关性阈值去噪对振动信号各个分量进行去噪处理;计算不同岩壁硬度下各分量与原始信号的相关性,根据选定阈值提取包含振动信息较多的分量,构建振动信号特征矢量,从而实现掘进机截割头振动信号的特征量提取。通过仿真试验表明,EWT能够有效提取不同岩壁硬度下掘进机截割头振动信号的特征量,且其性能优于奇异值分解特征量提取方法。In order to solve the problem of difficulty in identifying the maneuvering load of rock tunnel driving,a method for extracting the feature vector of the vibration signal of the head of the roadheader based on the combination of empirical wavelet transform(EWT)and correlation threshold denoising was proposed.First,the vibration signals of the heading machine under different rock wall hardness are converted into several component signals after EWT processing;thereafter,the correlation threshold denoising method is used to denoise each component of the vibration signal;finally,the correlation coefficient of each component and the original signal under different rock wall hardness is calculated.According to the selected threshold,the components containing more vibration information are extracted,and the vibration signal feature vector is constructed,so as to realize the feature vector extraction of the vibration signal of the cutting head of the roadheader.Simulation experiments show that EWT can effectively extract the feature vector of the vibration signal of the cutting head of the roadheader under different rock wall hardness,and its performance is better than the singular value decomposition feature quantity extraction method.

关 键 词：掘进机 振动信号 经验小波变换(EWT) 去噪 特征量提取 

分 类 号：TD421.5[矿业工程—矿山机电]