基于信号共振稀疏分解与HFE的往复压缩机故障诊断  

Fault Diagnosis of Reciprocating Compressor on the Resonance-based Sparse Signal Decomposition and Hierarchical Fuzzy Entropy(HFE)

作　　者：王金东[1] 卜庆超 杨松山[1] 欧凌非 

机构地区：[1]东北石油大学机械科学与工程学院,黑龙江大庆163318

出　　处：《压缩机技术》2017年第5期19-25,共7页Compressor Technology

基　　金：黑龙江省自然科学基金资助项目(E2015037;E2016009)

摘　　要：往复压缩机结构复杂,内部激励源众多,振动信号呈现强烈的非平稳、非线性的特性。针对这一问题,提出了信号共振稀疏分解与层次模糊熵(HFE)结合的往复压缩机故障诊断方法。与传统的频带划分信号分解方法不同,信号共振稀疏分解方法根据品质因子的不同,可将中心频率相近且频带相互重叠的信号有效分离,形成高、低共振分量;层次模糊熵既分析信号的低频成分,又分析信号的高频成分,可准确、全面的描述信号特征。利用信号共振稀疏分解方法提取故障信号,然后使用层次模糊熵构建特征向量;以SVM作为模式分类器,诊断故障类型。实验及结果表明,该方法可有效诊断往复压缩机轴承故障。The structure of reciprocating compressor is complex, which occur many internal excitation sources, and the vibration signal presents a strong non-stationary, non-linear characteristics. Aiming at this problem, a fault diagnosis method of reciprocating com-pressor combined with signal resonance sparse decomposition and hierarchical fuzzy entropy （HFE） is proposed. Different from the tra- ditional frequency division signal decomposition method, the signal resonance sparse decomposition method can effectively separate the signals with the center frequency close to each other and form the high and low resonance components according to the different quality factors. The hierarchical fuzzy entropy analyzes the low frequency of the signal component analysis of the high -frequency components of the signal, can accurately and comprehensively describe the signal characteristics. In this paper, the signal resonance sparse decomposition method is used to extract the fault signal, and then the feature vector is constructed by using the hierarchical fuzzy entropy; SVM is used as the model classifier to diagnose the fault type. Experiments and results show that this method can ef-fectively diagnose the reciprocating compressor bearing failure.

关 键 词：共振稀疏分解 层次模糊熵(HFE) 往复压缩机 轴承 

分 类 号：TH457[机械工程—机械制造及自动化]