基于多熵融合和多尺度卷积神经网络的风电机组轴承故障诊断方法

张天瑞; 周连弘

doi:10.19912/j.0254-0096.tynxb.2024-0478

PDF(2593 KB)

太阳能学报 ›› 2025, Vol. 46 ›› Issue (7) : 429-438. DOI: 10.19912/j.0254-0096.tynxb.2024-0478

第二十七届中国科协年会学术论文

基于多熵融合和多尺度卷积神经网络的风电机组轴承故障诊断方法

张天瑞, 周连弘

作者信息 +

FAULT DIAGNOSIS METHOD FOR WIND TURBINE BEARINGS BASED ON MULTI-ENTROPY FUSION AND MULTI-SCALE CONVOLUTIONAL NEURAL NETWORK

Zhang Tianrui, Zhou Lianhong

Author information +

文章历史 +

摘要

针对风电机组轴承在运行过程中收集到的故障信号较弱、状态特征难以有效表征的难题,提出一种基于多熵融合与多尺度卷积神经网络的风电机组轴承故障诊断新方法。首先对原始信号进行处理,分解出多个模态分量。随后,通过计算这些模态分量的多种熵值,构造出多熵融合矩阵,以充分表征信号的复杂特性。在此基础上,通过在卷积神经网络中集成不同尺寸的并行卷积核,设计一种结合多熵融合与多尺度卷积神经网络的故障诊断模型。结果表明,所提出的模型方法具有较好的诊断与泛化能力。

Abstract

To address the challenge of weak fault signals in wind turbine bearings, which complicates effective state characterization, this paper introduce a fault diagnosis method combining multi-entropy fusion with a multi-scale convolutional neural network. The approach decomposes the original signal into modal components, calculates multiple entropy measures to construct a feature matrix capturing the signal’s complex characteristics, and integrates this matrix into a convolutional neural network featuring parallel convolutional kernels of varying sizes. Experimental results from two datasets demonstrate the method’s enhanced diagnostic accuracy and generalization performance.

导出引用

张天瑞, 周连弘. 基于多熵融合和多尺度卷积神经网络的风电机组轴承故障诊断方法[J]. 太阳能学报. 2025, 46(7): 429-438 https://doi.org/10.19912/j.0254-0096.tynxb.2024-0478

Zhang Tianrui, Zhou Lianhong. FAULT DIAGNOSIS METHOD FOR WIND TURBINE BEARINGS BASED ON MULTI-ENTROPY FUSION AND MULTI-SCALE CONVOLUTIONAL NEURAL NETWORK[J]. Acta Energiae Solaris Sinica. 2025, 46(7): 429-438 https://doi.org/10.19912/j.0254-0096.tynxb.2024-0478

中图分类号： TH212 TH213.3

参考文献

[1] 王琦, 邓林峰, 赵荣珍. 基于一维CNN迁移学习的滚动轴承故障诊断[J]. 振动·测试与诊断, 2023, 43(1): 24-30, 195.
WANG Q, DENG L F, ZHAO R Z.Fault diagnosis of rolling bearing based on one-dimensional convolutional neural network transfer learning[J]. Journal of vibration, measurement & diagnosis, 2023, 43(1): 24-30, 195.
[2] LIU Y S, KANG J S, BAI Y J, et al.Research on the health status evaluation method of rolling bearing based on EMD-GA-BP[J]. Quality and reliability engineering international, 2023, 39(5): 2069-2080.
[3] 于德介, 程军圣. EMD方法在齿轮故障诊断中的应用[J]. 湖南大学学报(自然科学版), 2002, 29(6): 48-51.
YU D J, CHENG J S.Application of empirical mode decomposition method to gear fault diagnosis[J]. Journal of Hunan University(natural science), 2002, 29(6): 48-51.
[4] 张雨萍, 张志刚, 周庆华, 等. EMD端点效应问题解决方法的研究[J]. 信息技术, 2024, 48(1): 52-58, 64.
ZHANG Y P, ZHANG Z G, ZHOU Q H, et al.Research on processing method of end effect of empirical mode decomposition[J]. Information technology, 2024, 48(1): 52-58, 64.
[5] LEI Y G, ZUO M J.Fault diagnosis of rotating machinery using an improved HHT based on EEMD and sensitive IMFs[J]. Measurement science and technology, 2009, 20(12): 125701.
[6] 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 (ICASSP), Prague, Czech Republic, 2011: 4144-4147.
[7] 张燕飞, 李赟豪, 王东峰, 等. 基于多源信息融合的滚动轴承故障监测方法[J]. 轴承, 2022(12): 59-65.
ZHANG Y F, LI Y H, WANG D F, et al.Fault monitoring method for rolling bearings based on multi-source information fusion[J]. Bearing, 2022(12): 59-65.
[8] 王浩天, 段修生, 单甘霖, 等. 一种基于关联熵融合与改进回声状态网络的故障预测方法[J]. 振动与冲击, 2019, 38(2): 226-233.
WANG H T, DUAN X S, SHAN G L, et al.Prognostic algorithm based on the relative entropy fusion and modified ESN[J]. Journal of vibration and shock, 2019, 38(2): 226-233.
[9] 肖俊青, 金江涛, 李春, 等. 基于CEEMDAN模糊熵CNN轴承故障诊断研究[J]. 机械强度, 2023, 45(1): 26-33.
XIAO J Q, JIN J T, LI C, et al.Research on bearing fault diagnosis based on CEEMDAN fuzzy entropy amd convolutional neural network[J]. Journal of mechanical strength, 2023, 45(1): 26-33.
[10] 齐晓轩, 都丽, 张国山. 小波包近似熵特征的机动车声识别方法[J]. 南京理工大学学报, 2020, 44(1): 67-73, 79.
QI X X, DU L, ZHANG G S.Vehicle type recognition by acoustic signal based on wavelet packet decomposition and approximate entropy[J]. Journal of Nanjing University of Science and Technology, 2020, 44(1): 67-73, 79.
[11] 李俊卿, 胡晓东, 王罗, 等. 考虑数据不足和基于合作博弈模型融合的风电机组轴承故障诊断方法[J]. 太阳能学报, 2024, 45(1): 234-241.
LI J Q, HU X D, WANG L, et al.Bearing fault diagnosis method for wind turbine considering insufficient data and based on cooperative game model fusion[J]. Acta energiae solaris sinica, 2024, 45(1): 234-241.
[12] 王天泽, 徐永海, 于荣跃, 等. 基于熵特征融合的光伏组串直流电弧故障辨识方法[J/OL]. 华北电力大学学报(自然科学版), 2023: 1-11. (2023-11-27). https://kns.cnki.net/kcms/detail/13.1212.TM.20231124.1259.002.html.
WANG T Z, XU Y H, YU R Y, et al. Fault identification method of photovoltaic series DC arc based on Entropy Feature Fusion[J/OL]. Journal of North China Electric Power University (natural science edition), 2023: 1-11. (2023-11-27). https://kns.cnki.net/kcms/detail/13.1212.TM.20231124.1259.002.html.
[13] PENG X Y, WEI L P, GAO W.Application of CNN-based machine learning in the study of motor fault diagnosis[J]. Computational intelligence and neuroscience, 2022, 2022: 9635251.
[14] DU C Y, ZHONG R, ZHUO Y S, et al.Research on fault diagnosis of automobile engines based on the deep learning 1D-CNN method[J]. Engineering research express, 2022, 4(1): 015003.
[15] 雷春丽, 夏奔锋, 薛林林, 等. 基于MTF-CNN的滚动轴承故障诊断方法[J]. 振动与冲击, 2022, 41(9): 151-158.
LEI C L, XIA B F, XUE L L, et al.Rolling bearing fault diagnosis method based on MTF-CNN[J]. Journal of vibration and shock, 2022, 41(9): 151-158.
[16] ACIKGOZ H, KORKMAZ D.MSRConvNet: Classification of railway track defects using multi-scale residual convolutional neural network[J]. Engineering applications of artificial intelligence, 2023, 121: 105965.
[17] 安文杰, 陈长征, 田淼, 等. 基于迁移学习的风电机组轴承故障诊断研究[J]. 太阳能学报, 2023, 44(6): 367-373.
AN W J, LU C Z, TIAN M, et al.Research on bearing fault diagnosis of wind turbines based on transfer learning[J]. Acta ener giae solaris sinica, 2023, 44(6): 367-373.
[18] 周连弘. 基于多信息融合和深度学习的滚动轴承健康管理研究[D]. 沈阳: 沈阳大学, 2024.
ZHOU LH.Research on Rolling Bearing Health Management Based on Multi-information Fusion and Deep Learning[D]. Shenyang: Shenyang University, 2024.
[19] GAO Z H, LIU Y, WANG Q J, et al.Ensemble empirical mode decomposition energy moment entropy and enhanced long short-term memory for early fault prediction of bearing[J]. Measurement, 2022, 188: 110417.
[20] 李卫民, 马继召, 雷晓柱. 基于近似熵与支持向量机的异步电机故障诊断研究[J]. 机床与液压, 2021, 49(5): 173-176, 155.
LI W M, MA J Z, LEI X Z.Fault diagnosis of asynchronous motor based on approximate entropy and support vector machine[J]. Machine tool & hydraulics, 2021, 49(5): 173-176, 155.
[21] CUESTA-FRAU D.Permutation entropy: influence of amplitude information on time series classification performance[J]. Mathematical biosciences and engineering, 2019, 16(6): 6842-6857.
[22] 宋春雷, 路晓亚, 何笑笑. 基于多尺度模糊熵的时间序列特征提取算法[J]. 无线互联科技, 2023, 20(23): 111-114.
SONG C L, LU X Y, HE X X.Feature extraction algorithm of time series based on multi-scale fuzzy entropy[J]. Wireless internet science and technology, 2023, 20(23): 111-114.
[23] 许子非, 金江涛, 李春. 基于多尺度卷积神经网络的滚动轴承故障诊断方法[J]. 振动与冲击, 2021, 40(18): 212-220.
XU Z F, JIN J T, LI C.Fault diagnosis method of rolling bearing based on multi-scale convolutional neural network[J]. Journal of vibration and shock, 2021, 40(18): 212-220.
[24] 张健, 张子阳, 起雪梅, 等. 基于EMD和VPMCD的滚动轴承故障诊断[J]. 机械设计, 2019, 36(S2): 91-94.
ZHANG J, ZHANG Z Y, QI X M, et al.Fault diagnosis of rolling bearing based on EMD and VPMCD[J]. Journal of machine design, 2019, 36(S2): 91-94.
[25] Case Western Reserve University. Bearing data center [EB/OL].https://csegroups.case.edu/bearingdatacenter.
[26] 吴静然, 刘建华, 崔冉. 子域适应无监督轴承故障诊断[J]. 振动与冲击, 2021, 40(15): 34-40.
WU J R, LIU J H, CUI R.Sub-domain adaptive unsupervised bearing fault diagnosis[J]. Journal of vibration and shock, 2021, 40(15): 34-40.