基于ReliefF和因子分析的管道泄漏源特征识别方法  

Feature Recognition Method of Pipeline Leakage Sources Based on Factor Analysis and ReliefF

作　　者：高琳[1,2] 周剑楠 周小杰 王红 GAO Lin;ZHOU Jiannan;ZHOU Xiaojie;WANG Hong(School of Mechanical Engineering,Inner Mongolia University of Science and Technology,Baotou 014000,Inner Mongolia,China;School of Mechanical Engineering,Beijing University of Science and Technology,Beijing 100083,China)

机构地区：[1]内蒙古科技大学机械工程学院,内蒙古包头014000 [2]北京科技大学机械工程学院,北京100083

出　　处：《噪声与振动控制》2024年第4期174-179,241,共7页Noise and Vibration Control

基　　金：内蒙古自然科学基金资助项目(2021LHMS05027)。

摘　　要：针对管道泄漏源识别方法中由于特征冗余性较高从而影响识别精度的问题,提出一种结合ReliefF和因子分析的特征降维方法。首先,采集管道泄漏源的声发射信号,并从声发射信号中提取出27个频域特征和时域特征;其次,利用ReliefF方法对特征进行筛选,筛选出相关性较高的特征作为敏感特征,并通过因子分析提取敏感特征中的公因子,组成特征集;最后,将降维后的特征集输入支持向量机中进行识别,输出对管道泄漏源形状特征与尺寸大小的识别结果。实验结果表明,该方法能准确识别出管道泄漏源的不同形状特征以及尺寸大小,同时能有效降低特征冗余性和运算时长。In order to solve the issue of high feature redundancy affecting identification accuracy in pipeline leakage source identification,a feature dimension reduction method combining ReliefF and factor analysis is proposed.Firstly,acoustic emission signals from pipeline leakage sources are collected,and 27 frequency domain features and time domain features are extracted from the signals.Then,the ReliefF method is used to screen the features,so that the features with higher relevance are obtained and used as sensitive features.And factor analysis is used to extract common factors from the sensitive features to form a feature set.Finally,the feature set after the dimension reduction is input into the support vector machine for identification,and the recognition results of the shape and size of the pipeline leak source are obtained.Experimental results show that this method can accurately identify shape characteristics and size of pipeline leakage sources,and effectively reduce feature redundancy and computation time.

关 键 词：振动与波 管道泄漏 特征降维 RELIEFF 因子分析 

分 类 号：TN911.6[电子电信—通信与信息系统] TP277[电子电信—信息与通信工程]