设计研究 Design&Research
基于LMD与多尺度排列熵的往复压缩机故障特征提取方法
娄燕敏1，胡进旭”，岳磊”，陈桂娟
(1.东北石油大学机械科学与工程学院，黑龙江大庆163318；2.中国石油庆阳石化公司，甘肃庆阳745002)
[摘要]：往复压缩机是用于压缩和输送气体的机械设备，针对其振动信号特征，提出基于LMD与多尺度排列端的往复压缩机轴承间隙故障特征提取方法。利用具有保形特性的Hermite插值法替代传统LMD中滑动平均法构造均值与包络函数，提高LMD对非平稳信号的分解精度；以改进的LMD方法分解各状态下的振动信号，依据相关性系数筛选包含故障状态主要信息的PF分量；利用多尺度排列满对各PF 分量进行定量描述，并以平均类间样本距离对尺度因子进行优选，得出可分性良好的特征向量；利用SVM识别轴承间隙款障的类型，以识别准确率为依据，通过
与不同方法所提取的特征向量进行对比，验证了方法的有效性。[关键词]：往复压缩机；LMD；多尺度排列炳：轴承；故障诊断
中图分类号：TH457
文献标志码：A
文章编号：1006-2971(2015)04-0006-05
AFaultFeatureExtractionMethodBasedonLMDMultiscaleEntropy and SVMforBear-ingofReciprocatingCompressor
LOU Yanmin', HU Jinxur, YUE Lei, CHEN Guijuan
(1. Mechanical Seience and Engineering Institute, Northeast Petroleum University, Daqing 163318, China; 2. China Petroleum Qingyang Petrochemi-cal Company, Qingyang 745002, China)
Abstract:Reciprocating compressor is the machinery equipment for compressing and delivering gas, based on the nonstationary and nonlinearity characteristics of reciprocating compressor vibration signal, a feature extraction method for bearing fault of reciprocating compressor based on LMD and multiscale permutation entropy is proposed. To improve the decomposition accuracy of LMD for non-stationary signals, the cubic Hermite interpolation method is used to construct the means and envelope function instead of traditional LMD sliding average method. Vibration signals in each state are decomposed into a series of PF components with the improved LMD method, and the PF components which contain the main information of fault state were chosen according to the relative coefficient. Multiscale permutation entropy of the selected PF components was calculated, and the optimized scale factor was selected based on the maximum of average distances between different states, so the eigenvectors which have the best divisibility were extracted. Taken SVM as patterm classifier, the type of bearing clearance fault was diagnosed, and superiority of this method is verified by comparing
the eigenvectors extracted by LMD and permutation entropy and LMD and sample entropy method. Key words: reciprocating compressor; LMD; multiscale permutation entropy; bearing; fault diagnosis
收稿日期：2015-0205
压缩机技术
06
Compressors
2015年04期（总第252期）翼