作者：王余奎, 朱倩, 张磊, 朱臣, 滕伟

作者单位：空军勤务学院，江苏 徐州 221000

关键词：滚动轴承;局部特征尺度分解;谱熵;退化状态识别

摘要：

针对滚动轴承在出现故障时其振动信号呈现出非线性、非平稳特性，以及退化特征难以提取等问题，将局部特征尺度分解法应用到轴承振动信号分析中，并与信息熵理论融合提出局部特征尺度分解谱熵的滚动轴承退化特征指标。该方法首先对不同故障程度的轴承振动信号做局部特征尺度分解，基于得到的内禀尺度分量计算振动信号得能谱熵、奇异谱熵和包络谱熵用于表征轴承故障程度，仿真信号分析结果表明以上特征指标能够较好地反映滚动轴承的退化状态。对内圈故障和外圈故障模式下不同程度故障的轴承振动信号进行分析，结果表明该文提出的退化特征能够有效表征轴承的退化状态，并采用灰关联分析法构建轴承退化状态识别模型，可有效实现轴承退化状态识别。

LCD spectrum entropy and its application on the degradation state identification of rolling bearing
WANG Yukui, ZHU Qian, ZHANG Lei, ZHU Chen, TENG Wei
Air Force Logistics College, Xuzhou 221000, China
Abstract: Aiming at the nonlinear and non-stationary characteristic of bearing vibration signal, the local characteristic scale decomposition was introduced and used in the analysis of its vibration signal. The local characteristic scale decomposition spectrum entropy as a novel degradation feature extraction method was proposed based on the combination of local characteristic scale decomposition arithmetic and information entropy theory. The local characteristic scale decomposition was performed to the rolling bearing vibration signal in different fault levels. The power spectrum entropy, singular spectrum entropy and the envelope spectrum entropy were extracted from the obtained intrinsic scale components. The analysis results of simulation signal demonstrated the good performance of the proposed method. The degradation feature vector was composed with the three features and the performance of it was tested by perform LCDSE to the practical vibration signal of bearing with inner circle point eclipse and inner circle point eclipse in different fault levels. The standard degradation mode matrix was built, and the degradation state testing samples of the two fault pattern were used to perform grey incidence analysis with the standard degradation mode matrix, and the incidence degree was used to judge the degradation state of bearing.
Keywords: rolling bearing;LCD;spectrum entropy;degradation state identification
2020, 46(3):135-142  收稿日期: 2018-11-21;收到修改稿日期: 2019-01-15
基金项目: 空军装备科研重点项目（KJ20172A05171, KJ2016A2162）； 国家自然科学基金青年科研基金项目（51705530）
作者简介: 王余奎(1987-),男,河南太康县人,讲师,博士,研究方向为装备状态监测与故障预测
参考文献
[1] PATILl M S, MATHEW J, RAJENDRA P K. Bearing signature analysis as a medium for fault detection: a review[J]. Journal of Tribology, 2008, 130(1): 014001-1
[2] 张龙, 黄文艺, 熊国良. 基于多尺度熵的滚动轴承故障程度评估[J]. 振动与冲击, 2014, 33(9): 185-189
[3] LI H, WANG Y, WANG B, et al. The application of a general mathematical morphological particle as a novel indicator for the performance degradation assessment of a bearing[J]. Mechanical Systems & Signal Processing, 2016(82): 490-502
[4] GLOWACZ A, GLOWAC W, GLOWAC Z, et al. Early fault diagnosis of bearing and stator faults of the single-phase induction motor using acoustic signals[J]. Measurement, 2018(113): 1-9
[5] 王冰, 李洪儒, 许葆华. 基于多尺度形态分解谱熵的电机轴承预测特征提取及退化状态评估[J]. 振动与冲击, 2013, 32(22): 124-128
[6] 向丹, 葛爽. 基于EMD样本熵-LLTSA的故障特征提取方法[J]. 航空动力学报, 2014, 29(7): 1535-1542
[7] 孙洁娣, 肖启阳, 温江涛, 等. 基于LMD包络谱熵及SVM的天然气管道微小泄漏孔径识别[J]. 机械工程学报, 2014, 50(20): 18-25
[8] COSTA M, GOLDBERGER A L, PENG C K. Multi-scale entropy analysis of biological signals[J]. Physical Review E, 2005, 71: 1-18
[9] SHUEN D W, CHIU W W, KUNG Y L, et al. Modified multi-scale entropy for short-term time series analysis[J]. Physical A, 2013, 392: 5865-5873
[10] 程军圣, 郑近德, 杨宇. 一种新的非平稳信号分析方法-局部特征尺度分解法[J]. 振动工程学报, 2012, 25(2): 215-220
[11] 程军圣, 杨怡, 杨宇. 局部特征尺度分解方法及其在齿轮故障诊断中的应用[J]. 机械工程学报, 2012, 48(9): 64-71
[12] 苏文胜, 王奉涛, 张志新等. EMD降噪和谱峭度法在滚动轴承早期故障诊断中的应用[J]. 振动与冲击, 2010, 29(3): 18-21
[13] 张志刚, 石晓辉, 施全, 等. 基于改进EMD和谱峭度法滚动轴承故障特征提取[J]. 振动、测试与诊断, 2013, 33(3): 478-482
[14] WANG W K, LI H R, PENG Y. Fault feature extraction of hydraulic pump based on CNC De-noising and HHT[J]. Journal of Failure Analysis and Prevention, 2015, 15(1): 139-151
[15] 沈路, 周晓军. 基于形态滤波与灰色关联度的滚动轴承故障诊断[J]. 振动与冲击, 2009, 28(11): 17-20
[16] 黎奇志, 胡国平, 赵红言. 加权灰关联分析在故障诊断中的应用研究[J]. 微计算机信息, 2012, 28(7): 28-30