基于声发射技术的液压滑阀内漏识别方法研究  

Research on identification method of hydraulic spool valve internal leakage based on acoustic emission technology

作　　者：彭利坤[1] 何毓明 宋飞[1] PENG Li-kun;HE Yu-ming;SONG Fei(College of Power Engineering,Naval Univ.of Engineering,Wuhan 430033,China)

机构地区：[1]海军工程大学动力工程学院,武汉430033

出　　处：《海军工程大学学报》2020年第4期80-85,共6页Journal of Naval University of Engineering

基　　金：湖北省自然科学基金资助项目(2016CFB614)。

摘　　要：为找到液压滑阀内漏信号特征,寻求最优的声发射检测算法,以5组滑阀为研究对象,搭建滑阀内泄漏声发射检测实验台。采集5组滑阀1~12 MPa压力下的声发射信号,对信号进行EMD分解并提取的能量特征,初步找到内漏滑阀和正常滑阀信号的区别。结合SVM和LSSVM对信号做出模式识别,融入群智能算法对原算法进行优化,结果表明:基于LSSVM的优化算法识别时间最短,基于SVM的优化算法准确率较高,其中EMD-ABC-SVM准确率达到100%,用时2.0 s,最适用于作为滑阀内漏的声发射检测算法。In order to find the signal characteristics of hydraulic slide valve internal leakage and establish the optimal acoustic emission detection algorithm,five groups of slide valves were taken as the research objects,and an acoustic emission testing platform for internal leakage of slide valves was built.Acoustic emission(AE)signals of 5 groups of sliding valves under 1~12 MPa pressure were collec-ted.The signals were decomposed by EMD and the energy characteristics were extracted.The diffe-rence between the signals of internal leakage sliding valves and normal sliding valves was preliminarily found.Combining SVM and LSSVM,the original algorithm is optimized with swarm intelligence algorithm.The results show that the optimization algorithm based on LSSVM has the shortest recognition time,and the optimization algorithm based on SVM has the higher accuracy.The accuracy of EMD-ABC-SVM reaches 100%and the recognition takes only 2.0 s,which is most suitable for acoustic emission detection of sliding valve leakage.The analysis mode and optimization algorithm adopted in this paper are useful attempts for acoustic emission testing of spool valve leakage.

关 键 词：液压滑阀 内泄漏 声发射检测 EMD 优化算法 

分 类 号：TH137[机械工程—机械制造及自动化]