基于VMD-多尺度排列熵和SVM的船用空压机故障诊断方法  

Fault diagnosis of marine air compressor based on VMD multi-scale permutation entropy and SVM

作　　者：胡以怀[1] 李从跃 沈威 崔德馨 张成 芮晓松 HU Yihuai;LI Congyue;SHEN Wei;CUI Dexin;ZHANG Cheng;RUI Xiaosong(Shanghai Maritime University,Shanghai 201306,China;China Merchants Dingheng Shipbuilding Co.,Ltd.,Yangzhou 225217,China)

机构地区：[1]上海海事大学,上海201306 [2]招商局鼎衡造船有限公司,江苏扬州225217

出　　处：《中国测试》2024年第6期20-27,共8页China Measurement & Test

基　　金：上海市科技计划(20DZ2252300)。

摘　　要：船用机械振动信号存在非线性、非平稳性问题,故障特征难提取,通过变分模态分解(variational mode decomposition,VMD)多尺度排列熵(multiscale permutation entropy,MPE)与支持向量机(support vector machine,SVM)融合的故障诊断方法,对振动信号进行研究。以空压机为例,首先,模拟6种空压机工况,对各工况的热工参数进行测试,分析各工况热工参数的变化程度,并对采集的振动信号进行频域分析。然后通过VMD对振动信号进行分解,得到一系列固有模态分量,计算与原始信号的互相关系数筛选敏感固有模态分量。最后计算出敏感固有模态分量的多尺度排列熵,将其作为特征向量,输入到SVM中,进行故障辨识。实验结果表明:VMD多尺度排列熵与SVM融合的空压机故障辨识方法,能有效地识别故障类型,整体准确率可保持在98.6667%,将此方法与其他方法进行对比,证明此方法有效。Marine machinery vibration signals have nonlinear and non-stationary problems,and fault features are difficult to extract.Through variational mode decomposition multiscale permutation entropy and support vector machine fusion fault diagnosis method to study the vibration signal.Taking the air compressor as an example,first,6 kinds of air compressor working conditions are simulated,the thermal parameters of each working condition are tested,the degree of change of the thermal parameters of each working condition is analyzed,and the collected vibration signals are analyzed in the frequency domain.Then the vibration signal is decomposed by VMD to obtain a series of natural modal components,and the cross-correlation coefficient with the original signal is calculated to screen out the sensitive natural modal components.Finally,the multi-scale permutation entropy of the sensitive natural mode components is calculated,which is used as a feature vector and input into the SVM for fault identification.The experimental results show that the air compressor fault identification method based on the fusion of VMD multi-scale arrangement entropy and SVM can effectively identify the fault type,and the overall accuracy rate can be maintained at 98.6667%.The comparison of this method with other methods proves the effectiveness of this method.

关 键 词：船用往复式空压机 变分模态分解 多尺度排列熵 故障诊断 

分 类 号：TB9[一般工业技术—计量学] U664.51[机械工程—测试计量技术及仪器]