基于形态差值算子和特征能量比的液压泵故障分离方法  

Method Based on Morphological Difference Operator and Feature Energy Ratio for Fault Signal Separation of Hydraulic Pump

作　　者：郑直 姜万录[2,3] 王宝中 郭洋 ZHENG Zhi;JIANG Wan-lu;WANG Bao-zhong;GUO Yang(College of Mechanical Engineering,North China University of Science and Technology,Tangshan,Hebei 063210;Hebei Provincial Key Laboratory of Heavy Machinery Fluid Power Transmission and Control,Yanshan University, Qinhuangdao,Hebei 066004;Key Laboratory of Advanced Forging&Stamping Technology and Science(Yanshan University), Yanshan University,Qinhuangdao,Hebei 066004;College of Qinggong,North China University of Science and Technology,Tangshan,Hebei 064000)

机构地区：[1]华北理工大学机械工程学院,河北唐山063210 [2]燕山大学河北省重型机械流体动力传输与控制重点实验室,河北秦皇岛066004 [3]先进锻压成形技术与科学教育部重点实验室(燕山大学),河北秦皇岛066004 [4]华北理工大学轻工学院,河北唐山064000

出　　处：《液压与气动》2018年第9期7-14,共8页Chinese Hydraulics & Pneumatics

基　　金：国家自然科学基金(51475405);华北理工大学博士科研启动基金(0088/28412499)

摘　　要：针对液压泵滑靴和斜盘磨损复合故障信号的分离问题,提出了一种基于形态差值算子与特征能量比相结合的方法。首先,将若干种不同长度的结构元素和复合故障信号的形态特征进行匹配,利用形态差值算子提取出若干个信号;其次,分别对上述信号计算两种故障的特征能量比;最后,找出两种故障的最大特征能量比,他们所对应的即为最优匹配结构元素长度,且基于该两种长度的形态差值算子所提取出的两个信号分别为最优分离出的滑靴和斜盘磨损故障信号。通过对实测液压泵复合故障信号的实验验证,表明所提方法能够根据信号形态特征的多样性有效地实现对复合故障信号的最优分离,且比Robust ICA方法有效和优越。Aimed at compound fault signal separation of slipper wear and swash plate wear of hydraulic pump,a method based on morphological difference operator and feature energy ratio is proposed.Firstly,some structure elements with different lengths are matched with morphological features of compound fault signal respectively,and then some signals can be extracted by morphological difference operator.Secondly,feature energy ratios of slipper wear and those of swash plate wear are computed based on the extracted signals;Finally,two maximum values of feature energy ratios can be collected,and they correspond to the best matching structure element lengths,then morphological difference operators with structure elements of the two lengths can optimally separate slipper wear fault signal and swash plate wear fault signal from the compound fault signal.The fault signals of hydraulic pump are tested and verified,and the result shows that the proposed method can optimally separate compound fault signals based on morphological feature diversity of signal.Furthermore it is more effective and superior than RobustICA.

关 键 词：复合故障 形态差值算子 特征能量比 最优分离 液压泵 

分 类 号：TH137[机械工程—机械制造及自动化] TP277[自动化与计算机技术—检测技术与自动化装置]