Early Fault Diagnosis of Rolling Bearings by Using CEEMDAN and Parameter-optimized Multiscale Permutation Entropy
-
摘要: 针对滚动轴承的早期故障特征微弱的特点,提出了自适应噪声完备集合经验模态分解(CEEMDAN)与多尺度排列熵(MPE)结合提取故障特征,采用支持向量机(SVM)进行故障状态判别的滚动轴承早期故障诊断方法。利用CEEMDAN将信号分解为若干分量,计算各分量与原信号的相关系数,将大于相关系数阈值的分量重构,对MPE的参数运用PSO算法寻优,计算重构后的信号的MPE值并作为故障特征向量,使用SVM对故障状态进行识别。将该方法运用于XJTU-SY滚动轴承加速寿命试验数据集,并与MPE参数未优化以及未CEEMDAN分解且MPE参数未优化得到的MPE值作为特征向量SVM进行识别的结果进行对比,结果表明本文所提方法的故障识别率分别提高了10.71%和14.28%。
-
关键词:
- 完备集合经验模态分解 /
- 多尺度排列熵 /
- 支持向量机 /
- 故障诊断
Abstract: In view of the weak features of early fault of rolling bearings, an early fault diagnosis method of rolling bearings is proposed, which combines adaptive noise complete set empirical mode decomposition (CEEMDAN) with multiscale permutation entropy (MPE) to extract fault features, and uses support vector machine (SVM) to identify the fault state. The originalsignal is decomposed into several components by CEEMDAN, and the correlation coefficient between each component and the original signal is calculated. The components larger than the threshold of correlation coefficient are reconstructed. The MPE parameters are optimized by PSO algorithm. The MPE value of the reconstructed signal is calculated as the fault feature vector, and the fault state is identified by SVM. The method is applied to XJTU-SY rolling bearing accelerated life test data set, and compared with the results of identifying MPE values as feature vector SVM without optimization of MPE parameters and CEEMDAN decomposition and optimization of MPE parameters, the results show that the fault identification rates of the method proposed in this paper are improved by 10.71% and 14.28%, respectively. -
表 1 XJTU-SY轴承数据集信息
Table 1. Overview of XJTU-SY bearing dataset information
数据集 基本额定寿命 实际寿命 失效位置 Bearing2_1 406.08 ~ 703.56 min 491 min 内圈 Bearing2_2 406.08 ~ 703.56 min 161 min 外圈 Bearing2_3 406.08 ~ 703.56 min 533 min 保持架 表 2 各IMF分量与原信号相关系数
Table 2. Correlation coefficients between each IMF component and original signal
IMFi 数值 IMFi 数值 IMF1 0.7184 IMF7 0.1117 IMF2 0.2403 IMF8 0.0433 IMF3 0.3241 IMF9 0.0374 IMF4 0.4930 IMF10 0.0365 IMF5 0.4397 IMF11 0.0043 IMF6 0.2599 IMF12 0.0049 表 3 PSO寻优的MPE参数
Table 3. MPE parameters optimized by PSO
故障类型 m s t N 正常 5 12 4 885 内圈 6 12 4 958 外圈 5 10 4 777 保持架 4 11 3 983 表 4 故障诊断结果
Table 4. Fault diagnosis results
所用方法 轴承状态 测试样本数 识别结果 平均识别率/% 正常 内圈故障 外圈故障 保持架故障 CEEMDAN-PSO-
MPE-SVM正常 70 66 0 4 0 98.21 内圈故障 70 0 70 0 0 外圈故障 70 0 0 70 0 保持架故障 70 1 0 0 69 CEEMDAN-
MPE-SVM正常 70 61 0 0 9 87.5 内圈故障 70 23 47 0 0 外圈故障 70 0 0 70 0 保持架故障 70 3 0 0 67 MPE-SVM 正常 70 55 0 0 17 83.93 内圈故障 70 27 43 0 0 外圈故障 70 0 0 70 0 保持架故障 70 1 0 0 69 -
[1] TORRES M E, COLOMINAS M A, SCHLOTTHAUER G, et al. A complete ensemble empirical mode decomposition with adaptive noise[C]//2011 IEEE International Conference on Acoustics, Speech and Signal Processing. Prague: IEEE, 2011: 4144-4147 [2] 李怡, 李焕锋, 刘自然. 基于CEEMDAN多尺度熵和SSA-SVM的滚动轴承故障诊断研究[J]. 机电工程, 2021, 38(5): 599-604.LI Y, LI H F, LIU Z R. Fault diagnosis of rolling bearing based on CEEMDAN multi-scale entropy and SSA-SVM[J]. Journal of Mechanical & Electrical Engineering, 2021, 38(5): 599-604. (in Chinese) [3] 杨保俊, 洪荣晶, 潘裕斌. 基于CEEMDAN-EFICA去噪的风电齿轮箱故障诊断[J]. 组合机床与自动化加工技术, 2020(2): 115-118.YANG B J, HONG R J, PAN Y B. Fault diagnosis of wind turbine gearbox based on CEEMDAN-EFICA denoising[J]. Modular Machine Tool & Automatic Manufacturing Technique, 2020(2): 115-118. (in Chinese) [4] 蒋玲莉, 谭鸿创, 李学军, 等. 基于CEEMDAN排列熵与SVM的螺旋锥齿轮故障识别[J]. 振动、测试与诊断, 2021, 41(1): 33-40.JIANG L L, TAN H C, LI X J, et al. Fault diagnosis of spiral bevel gear based on CEEMDAN permutation entropy and SVM[J]. Journal of Vibration, Measurement & Diagnosis, 2021, 41(1): 33-40. (in Chinese) [5] 徐乐, 邢邦圣, 郎超男, 等. LMD能量熵和SVM相结合的滚动轴承故障诊断[J]. 机械科学与技术, 2017, 36(6): 915-918.XU L, XING B S, LANG C N, et al. Fault diagnosis of rolling bearing combined LMD energy entropy and SVM[J]. Mechanical Science and Technology for Aerospace Engineering, 2017, 36(6): 915-918. (in Chinese) [6] 郑近德, 程军圣, 杨宇. 基于LCD和排列熵的滚动轴承故障诊断[J]. 振动、测试与诊断, 2014, 34(5): 802-806. doi: 10.3969/j.issn.1004-6801.2014.05.003ZHENG J D, CHENG J S, YANG Y. A rolling bearing fault diagnosis method based on LCD and permutation entropy[J]. Journal of Vibration, Measurement & Diagnosis, 2014, 34(5): 802-806. (in Chinese) doi: 10.3969/j.issn.1004-6801.2014.05.003 [7] AZIZ W, ARIF M. Multiscale permutation entropy of physiological time series[C]//Proceeding of 2005 Pakistan Section Multitopic Conference. Karachi: IEEE, 2005 [8] 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 [9] 陈东宁, 张运东, 姚成玉, 等. 基于FVMD多尺度排列熵和GK模糊聚类的故障诊断[J]. 机械工程学报, 2018, 54(14): 16-27.CHEN D N, ZHANG Y D, YAO C Y, et al. Fault diagnosis based on FVMD multi-scale permutation entropy and GK fuzzy clustering[J]. Journal of Mechanical Engineering, 2018, 54(14): 16-27. (in Chinese) [10] 何园园, 张超, 朱腾飞. ELMD熵特征融合与PSO-SVM在齿轮故障诊断中的应用[J]. 机械科学与技术, 2019, 38(2): 271-276.HE Y Y, ZHANG C, ZHU T F. Application of ELMD entropy feature fusion and PSO-SVM in gear fault diagnosis[J]. Mechanical Science and Technology for Aerospace Engineering, 2019, 38(2): 271-276. (in Chinese) [11] COLOMINAS M A, SCHLOTTHAUER G, TRRES M E, et al. Noise-assisted EMD methods in action[J]. Advances in Adaptive Data Analysis, 2012, 4(4): 1250025. doi: 10.1142/S1793536912500252 [12] 程铁栋, 易其文, 吴义文, 等. 改进EWT_MPE模型在矿山微震信号特征提取中的应用[J]. 振动与冲击, 2021, 40(9): 92-101.CHEN T D, YI Q W, WU Y W, et al. Application of improved EWT_MPE model in feature extraction of mine micro-seismic signals[J]. Journal of Vibration and Shock, 2021, 40(9): 92-101. (in Chinese) [13] 王望望, 邓林峰, 赵荣珍, 等. 基于QPSO-MPE的滚动轴承故障识别方法[J]. 振动、测试与诊断, 2021, 41(1): 62-68. doi: 10.16450/j.cnki.issn.1004-6801.2021.01.009WANG W W, DENG L F, ZHAO R Z, et al. Rolling bearing fault identification based on quantum-behaved particle swarm optimization and multi-scale permutation entropy[J]. Journal of Vibration, Measurement & Diagnosis, 2021, 41(1): 62-68. (in Chinese) doi: 10.16450/j.cnki.issn.1004-6801.2021.01.009 [14] 张志霞. 基于RS-SVM的无线传感器网络入侵检测模型研究[J]. 智能计算机与应用, 2019, 9(3): 319-320.ZHANG Z X. Research on intrusion detection model of wireless sensor networks based on RS-SVM[J]. Intelligent Computer and Applications, 2019, 9(3): 319-320. (in Chinese) [15] WANG B, LEI Y G, LI N P, et al. A hybrid prognostics approach for estimating remaining useful life of rolling element bearings[J]. IEEE Transactions on Reliability, 2020, 69(1): 401-412. doi: 10.1109/TR.2018.2882682 [16] 雷亚国, 韩天宇, 王彪, 等. XJTU-SY滚动轴承加速寿命试验数据集解读[J]. 机械工程学报, 2019, 55(16): 1-6. doi: 10.3901/JME.2019.16.001LEI Y G, HAN T Y, WANG B, et al. XJTU-SY rolling element bearing accelerated life test datasets: a tutorial[J]. Journal of Mechanical Engineering, 2019, 55(16): 1-6. (in Chinese) doi: 10.3901/JME.2019.16.001