摘要
为对管道大泄露高频信号进行降噪,提取信号的故障频率,实现对管道泄漏的源定位。引入了VMD方法对收集到的信号进行不同频率之间的分离,突出信号的局部细节。为消除VMD分解中k值与惩罚因子这两个人为因素对分解效果的影响,采用模态分量与原信号的相关系数法,得出最优分解层数k值。使用优化的k值来对信号进行VMD分解之后,用最大峭度值来选择最优的惩罚因子。将通过最优参数得到的VMD分解模态分量再次与原信号进行相关系数比较,选出包含故障信息最多的模态分量进行Hilbert包络分析,得到信号的故障信息,对故障进行诊断及定位。研究发现,该方法能有效分离故障信号成分,从而实现对故障频率的提取。
In order to reduce the noise of the high-frequency signal of large pipeline leakage,extract the fault frequency of the signal,and realize the source location of the pipeline leakage.The VMD method is introduced to separate the collected signals between different frequencies,highlighting the local details of the signal.In order to eliminate the influence of the two human factors of k value and penalty factor on the decomposition effect in the VMD decomposition,the correlation coefficient method between the modal component and the original signal is used to obtain the optimal decomposition layer number k value.After using the optimized k value to perform VMD decomposition of the signal,the maximum kurtosis value is used to select the optimal penalty factor.The VMD decomposition modal components obtained through the optimal parameters are again compared with the original signal with the correlation coefficients,and the modal components containing the most fault information are selected for Hilbert envelope analysis to obtain the fault information of the signal and diagnose and locate the fault.The research found that this method can effectively separate the fault signal components,so as to realize the extraction of the fault frequency.
作者
李志星
王春鹏
鲍慧茹
LI Zhi-xing;WANG Chun-peng;BAO Hui-ru(Beijing Key Laboratory of Performance Guarantee on Urban Rail Transit Vehicles,Beijing University of Civil Engineer-ing and Architecture,Beijing 102612,China;Institute of Mechnical Engineering,Inner Mongolia University of Science and Technology,Inner Mongolia Baotou 014010,China;Baotou Vocational and Technical College,Inner Mongolia Baotou 014030,China)
出处
《机械设计与制造》
北大核心
2022年第7期102-107,共6页
Machinery Design & Manufacture
基金
城市轨道交通车辆服役性能保障北京市重点实验室开放课题
国家自然科学基金青年基金项目(51805275)。
关键词
管道泄漏
声发射
变分模态分解
Hilbert包络谱
Pipeline Leakage
Acoustic Emission
Variational Mode Decomposition
Hilbert Envelope Spectrum
