Fault diagnosis research of gearbox based on ICEEMDAN and PFA-ELM
-
摘要: 齿轮箱是工业设备中常用的传动部件。针对齿轮箱故障特征提取及诊断精度不足的问题,提出一种基于改进的自适应噪声完备集合经验模态分解(ICEEMDAN)及探路者算法(PFA)优化极限学习机(ELM)的故障诊断方法。首先,利用ICEEMDAN对信号进行分解,得到多个本征模态函数(IMF)。其次,基于斯皮尔曼相关系数,筛选出有效的IMF,并计算出每个有效IMF的模糊熵和排列熵作为故障特征向量。最后,利用PFA算法优化ELM中的权值和阈值,构建基于PFA-ELM的故障诊断模型。实验表明,PFA-ELM的故障诊断精度高达98.67%。该方法能够准确描述齿轮箱的工作状态,具有较高的实用价值。
-
关键词:
- 故障诊断 /
- 齿轮箱 /
- 改进的自适应噪声完备集合经验模态分解 /
- 探路者算法 /
- 极限学习机
Abstract: Gearboxes are commonly used transmission components in industrial equipment. To address the problem of insufficient gearbox fault feature extraction and diagnosis accuracy, a fault diagnosis method based on improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN) and pathfinder algorithm (PFA) optimized extreme learning machine (ELM) is proposed. First, the signal is decomposed using ICEEMDAN to obtain intrinsic mode functions (IMFs). Secondly, based on the Spearman correlation coefficient, the valid IMFs are screened out and the fuzzy entropy and permutation entropy of each valid IMF are calculated as the fault feature vector. Finally, the PFA algorithm is used to optimize the weights and biases in ELM and construct a fault diagnosis model based on PFA-ELM. Experiments show that the fault diagnosis accuracy of PFA-ELM is as high as 98.67%. The method can accurately describe the working condition of gearboxes and has high practical value. -
表 1 IMF分量与正常信号斯皮尔曼相关系数
IMF分量名 斯皮尔曼相关系数 IMF1 0.699 6 IMF2 0.401 9 IMF3 0.256 3 IMF4 0.195 6 IMF5 0.139 3 IMF6 0.114 9 IMF7 0.053 4 IMF8 0.046 5 IMF9 0.051 8 IMF10 0.030 0 IMF11 0.005 7 表 2 样本编码及分布
故障编码 故障类型 训练样本数 测试样本数 1 正常状态 35 15 2 小齿轮断齿 35 15 3 小齿轮磨损 35 15 4 大齿轮断齿 35 15 5 大小齿轮均断齿 35 15 表 3 不同算法诊断结果对比
齿轮箱
状态故障
标签标签识别数量 PFA-ELM WOA-ELM GWO-ELM PSO-ELM ELM 正常状态 1 14 14 14 14 14 小齿轮断齿 2 15 13 11 11 11 小齿轮磨损 3 15 15 15 15 12 大齿轮断齿 4 15 13 13 12 11 大小齿轮均
断齿5 15 15 15 15 12 诊断正确率/(%) 98.67 93.33 90.67 89.33 80 -
[1] Colominas M A, Schlotthauer G, Torres M E. Improved complete ensemble EMD: A suitable tool for biomedical signal processing[J]. Biomedical Signal Processing and Control, 2014, 14(11): 19-29. [2] Kou Z M, Yang F, Wu J, et al. Application of ICEEMDAN energy entropy and AFSA-SVM for fault diagnosis of hoist sheave bearing[J]. Entropy, 2020, 22(12): 1347. doi: 10.3390/e22121347 [3] Chaabi L, Lemzadmi A, Djebala A, et al. Fault diagnosis of rolling bearings in non-stationary running conditions using improved CEEMDAN and multivariate denoising based on wavelet and principal component analyses[J]. International Journal of Advanced Manufacturing Technology, 2020, 107(9-10): 3859-3873. doi: 10.1007/s00170-020-05311-z [4] 王浩楠, 崔宝珍, 彭智慧, 等. 应用ICEEMDAN和SVM的行星齿轮箱故障诊断[J]. 机械科学与技术, 2023(1): 24-30. [5] 别锋锋, 都腾飞, 庞明军, 等. 基于ICEEMDAN-GRNN神经网络的往复泵故障诊断方法研究[J]. 机械设计与制造, 2021(3): 127-131. doi: 10.3969/j.issn.1001-3997.2021.03.028 [6] 管一臣, 童攀, 冯志鹏. 基于ICEEMDAN方法和频率解调的行星齿轮箱故障电流信号特征分析[J]. 振动与冲击, 2019, 38(24): 41-47. doi: 10.13465/j.cnki.jvs.2019.24.006 [7] Huang G B, Zhu Q Y, Siew C K. Extreme learning machine: theory and applications[J]. Neurocomputing, 2006, 70(1-3): 489-501. doi: 10.1016/j.neucom.2005.12.126 [8] Huang G B, Wang D H, Lan Y. Extreme learning machines: a survey[J]. International Journal of Machine Learning and Cybernetics, 2011, 2(2): 107-122. [9] Lim J Y, Ji P S. Development of fault diagnosis algorithm using correlation analysis and ELM[J]. The Transactions of the Korean Institute of Electrical Engineers, 2016, 65(3): 204-209. doi: 10.5370/KIEEP.2016.65.3.204 [10] Liu X Y, Huang H Z, Xiang J W. A personalized diagnosis method to detect faults in gears using numerical simulation and extreme learning machine[J]. Knowledge-Based Syst, 2020,195: 105653. [11] 闫鹏程, 张超银, 孙全胜, 等. LIF技术与ELM算法的电力变压器故障诊断研究[J]. 光谱学与光谱分析, 2022, 42(5): 1459-1464. doi: 10.3964/j.issn.1000-0593(2022)05-1459-06 [12] 姚峰林, 谢长开, 吕世宁, 等. 基于小波包变换和ELM的滚动轴承故障诊断研究[J]. 安全与环境学报, 2021, 21(6): 2466-2472. doi: 10.13637/j.issn.1009-6094.2020.0999 [13] Li S M, Chen H L, Wang M J, et al. Slime mould algorithm: A new method for stochastic optimization[J]. Future Generation Computer Systems-the International Journal of Escience, 2020, 111: 300-323. doi: 10.1016/j.future.2020.03.055 [14] 王莹, 王亚慧, 安允. 基于PSO-KPCA-LVQ的燃气调压器故障诊断[J]. 现代电子技术, 2020, 43(24): 67-71. doi: 10.16652/j.issn.1004-373x.2020.24.018 [15] Chen W, Wang Z, Xie H, et al. Characterization of surface EMG signal based on fuzzy entropy[J]. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 2007, 15(2): 266-272. doi: 10.1109/TNSRE.2007.897025 [16] Bandt C, Pompe B. Permutation entropy: a natural complexity measure for time series[J]. Physical Review Letters, 2002, 88(17): 174102. doi: 10.1103/PhysRevLett.88.174102 [17] Cheng J, Chen J J, Guo Y N, et al. Adaptive CCR-ELM with variable-length brain storm optimization algorithm for class-imbalance learning[J]. Natural Computing, 2021, 20(1): 11-22. doi: 10.1007/s11047-019-09735-9 [18] Xiao C W, Ye J Q, Esteves R M, et al. Using Spearman's correlation coefficients for exploratory data analysis on big dataset[J]. Concurrency and Computation-Practice & Experience, 2016, 28(14): 3866-3878. [19] Mirjalili S, Lewis A. The whale optimization algorithm[J]. Advances in Engineering Software, 2016, 95: 51-67. doi: 10.1016/j.advengsoft.2016.01.008 [20] Mirjalili S, Mirjalili S M, Lewis A. Grey wolf optimizer[J]. Advances in Engineering Software, 2014, 69: 46-61. doi: 10.1016/j.advengsoft.2013.12.007