Application of Multi-parameter and Gaussian Process Classification in Gearbox Fault Diagnosis
-
摘要: 由于齿轮箱振动信号的非平稳非线性等问题加大了故障诊断的难度,本文提出了一种基于互补集合经验模态分解(CEEMD)和多尺度排列熵(MPE)、样本熵(SE)相结合的故障特征提取方法。首先对齿轮箱振动信号进行互补集合经验模态分解,并根据相关系数原则对各模态分量进行筛选和重构,再利用多尺度排列熵对筛选出的模态分量进行特征提取,同时对重构后的信号提取其样本熵作为特征值;最后将提取出的多种故障特征融合输入到高斯过程分类器中进行实验验证,实验结果表明该方法提取齿轮箱振动信号的故障特征是有效的,高斯过程分类能快速准确地分辨出故障结果。
-
关键词:
- 齿轮箱 /
- 互补集合经验模态分解 /
- 多尺度排列熵 /
- 高斯过程分类 /
- 故障诊断 /
Abstract: Considering the non-stationary and nonlinear characteristics of the gearbox vibration signal, it is difficult in signal processing and fault detection. A fault feature extraction method based on complementary ensemble empirical mode decomposition (CEEMD), multi-scale permutation entropy (MPE) and sample entropy (SE) is proposed in this paper. Firstly, the gearbox vibration signal is subjected to complementary ensemble empirical mode decomposition, and the modal components are filtered and reconstructed according to the correlation coefficient principle. Then, the multi-scale permutation entropy is used to extract the modal components and extract the features. Simultaneously the sample entropy of reconstructed signals is extracted as the features. Finally, the extracted fault features are merged into the Gaussian process classifier for experimental verification. The experimental results show that the method can effectively extract the fault characteristics of the gearbox vibration signal, and Gaussian classifiers can perform fault detection accurately and quickly.-
Key words:
- gearbox /
- fault detection /
- vibrations /
- signal processing /
- feature extraction /
- classifiers /
- MPE /
- SE /
- CEEMD /
- multi-scale permutation entropy
-
表 1 各模态分量与原始信号相关系数
模态分量 IMF1 IMF2 IMF3 IMF4 IMF5 相关系数 0.763 8 0.574 5 0.258 5 0.056 1 0.026 6 -
[1] 李耀华, 王星州.直升机齿轮箱故障诊断方法[J].机械科学与技术, 2018, 37(10):1628-1633 doi: 10.13433/j.cnki.1003-8728.20180059Li Y H, Wang X Z. A new method for fault diagnosis of helicopter gearbox[J]. Mechanical Science and Technology for Aerospace Engineering, 2018, 37(10):1628-1633(in Chinese) doi: 10.13433/j.cnki.1003-8728.20180059 [2] 陈法法, 汤宝平, 苏祖强.基于局部切空间排列与MSVM的齿轮箱故障诊断[J].振动与冲击, 2013, 32(5):38-42, 47 doi: 10.3969/j.issn.1000-3835.2013.05.008Chen F F, Tang B P, Su Z Q. Gearbox fault diagnosis based on local target space alignment and Multi-kernel support vector machine[J]. Journal of Vibration and Shock, 2013, 32(5):38-42, 47(in Chinese) doi: 10.3969/j.issn.1000-3835.2013.05.008 [3] 李莎, 潘宏侠, 张君东, 等.基于多重分形和PSO-SVM的齿轮箱故障诊断[J].机械传动, 2015, 39(2):132-136 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jxcd201502034Li S, Pan H X, Zhang J D, et al. Fault diagnosis of gearbox based on Multi-fractal and PSO-SVM[J]. Journal of Mechanical Transmission, 2015, 39(2):132-136(in Chinese) http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jxcd201502034 [4] 李建毅, 石林锁, 滕明春, 等.基于EMD细化包络谱分析的齿轮箱故障诊断[J].机械传动, 2013, 37(3):15-18, 44 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jxcd201303005Li J Y, Shi L S, Teng M C, et al. Zooming envelope spectrum analysis of gearbox fault diagnosis based on EMD[J]. Journal of Mechanical Transmission, 2013, 37(3):15-18, 44(in Chinese) http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jxcd201303005 [5] 林旭泽, 蔡艳平, 王新军.自适应的EEMD及其在滚动轴承故障诊断中的应用[J].机械科学与技术, 2016, 35(11):1727-1732 doi: 10.13433/j.cnki.1003-8728.2016.1115Lin X Z, Cai Y P, Wang X J. Adaptive EEMD and Its applications to rolling bearing fault diagnosis[J]. Mechanical Science and Technology for Aerospace Engineering, 2016, 35(11):1727-1732(in Chinese) doi: 10.13433/j.cnki.1003-8728.2016.1115 [6] 李嘉康, 赵颖, 廖洪林, 等.基于改进EMD算法和BP神经网络的SST预测研究[J].气候与环境研究, 2017, 22(5):587-600 http://d.old.wanfangdata.com.cn/Periodical/qhyhjyj201705006Li J K, Zhao Y, Liao H L, et al. SST Forecast based on BP neural network and improved EMD algorithm[J]. Climatic and Environmental Research, 2017, 22(5):587-600(in Chinese) http://d.old.wanfangdata.com.cn/Periodical/qhyhjyj201705006 [7] 李晓莉, 李成伟.改进的自适应噪声总体集合经验模态分解在光谱信号去噪中的应用[J].光学精密工程, 2016, 24(7):1754-1762 http://d.old.wanfangdata.com.cn/Periodical/gxjmgc201607026Li X L, Li C W. Application of improved complete ensemble empirical mode decomposition with adaptive noise in spectral signal denoising[J]. Optics and Precision Engineering, 2016, 24(7):1754-1762(in Chinese) http://d.old.wanfangdata.com.cn/Periodical/gxjmgc201607026 [8] 贺彬, 刘泉.基于EMD-MPE与HMM的滚动轴承故障诊断[J].组合机床与自动化加工技术, 2016(12):76-79 http://d.old.wanfangdata.com.cn/Periodical/zhjc201612021He B, Liu Q. Fault diagnosis of rolling bearing based on EMD-MPE and HMM[J]. Modular Machine Tool & Automatic Manufacturing Technique, 2016(12):76-79(in Chinese) http://d.old.wanfangdata.com.cn/Periodical/zhjc201612021 [9] 王余奎, 李洪儒, 叶鹏.基于多尺度排列熵的液压泵故障识别[J].中国机械工程, 2015, 26(4):518-522, 523 doi: 10.3969/j.issn.1004-132X.2015.04.017Wang Y K, Li H R, Ye P. Fault identification of hydraulic pump based on Multi-scale Permutation Entropy[J]. China Mechanical Engineering, 2015, 26(4):518-522, 523(in Chinese) doi: 10.3969/j.issn.1004-132X.2015.04.017 [10] 张蒙, 朱永利, 张宁, 等.基于变分模态分解和多尺度排列熵的变压器局部放电信号特征提取[J].华北电力大学学报, 2016, 43(6):31-37 doi: 10.3969/j.ISSN.1007-2691.2016.06.06Zhang M, Zhu Y L, Zhang N, et al. Feature extraction of transformer partial discharge signals based on varitional mode decomposition and Multi-scale Permutation Entropy[J]. Journal of North China Electric Power University, 2016, 43(6):31-37(in Chinese) doi: 10.3969/j.ISSN.1007-2691.2016.06.06 [11] 郑近德, 程军圣, 杨宇.多尺度排列熵及其在滚动轴承故障诊断中的应用[J].中国机械工程, 2013, 24(19):2641-2646 doi: 10.3969/j.issn.1004-132X.2013.19.017Zheng J D, Cheng J S, Yang Y. Multi-scale Permutation Entropy and Its applications to rolling bearing fault diagnosis[J]. China Mechanical Engineering, 2013, 24(19):2641-2646(in Chinese) doi: 10.3969/j.issn.1004-132X.2013.19.017 [12] 郭学卫, 申永军, 杨绍普.基于样本熵和分数阶傅里叶变换的滚动轴承故障特征提取[J].振动与冲击, 2017, 36(18):65-69 http://d.old.wanfangdata.com.cn/Periodical/zdycj201718010Guo X W, Shen Y J, Yang S P. Application of sample entropy and Fractional Fourier transform in the fault diagnosis of rolling bearings[J]. Journal of Vibration and Shock, 2017, 36(18):65-69(in Chinese) http://d.old.wanfangdata.com.cn/Periodical/zdycj201718010 [13] 肖红军, 刘乙奇, 黄道平.高斯过程建模方法在工业过程中的应用[J].华南理工大学学报, 2016, 44(12):36-43, 52 doi: 10.3969/j.issn.1000-565X.2016.12.006Xiao H J, Liu Y Q, Huang D P. Application of Gaussian Process modeling method in industrial processes[J]. Journal of South China University of Technology, 2016, 44(12):36-43, 52(in Chinese) doi: 10.3969/j.issn.1000-565X.2016.12.006 [14] Nickisch H, Rasmussen C E. Approximations for binary gaussian process classification[J]. Journal of Machine Learning Research, 2008, 9:2035-2078 [15] 尹金良, 朱永利, 俞国勤, 等.基于高斯过程分类器的变压器故障诊断[J].电工技术学报, 2013, 28(1):158-164 doi: 10.3969/j.issn.1000-6753.2013.01.023Yin J L, Zhu Y L, Yu G Q, et al. Fault diagnosis of transformers based on gaussian process classifier[J]. Transactions of China Electrotechnical Society, 2013, 28(1):158-164(in Chinese) doi: 10.3969/j.issn.1000-6753.2013.01.023