风力机齿轮箱因长期处于多噪声、高转速工况下运行,振动信号呈现非线性特性,致故障信息难以准确有效提取。基于此,提出自适应白噪声平均集成经验模态分解（CEEMDAN）与卷积神经网络（CNN）联合故障辨别与诊断方法。利用CEEMDAN较强的非线性特征分解能力将振动信号分解,多重相关系数筛选有效故障特征分量组并剔除冗余分量,再将最佳分量组输入CNN实现故障诊断。结果表明：不同故障状态和信噪比下,较EMD-CNN与EEMD-CNN方法,均突显了所提方法良好的鲁棒性与泛化性。
Abstract
The gearbox of wind turbine runs in multi-noise and high speed for a long time, the vibration signal presents nonlinear characteristics, which makes it difficult to extract the fault information accurately and effectively. Based on this, a fault diagnosis method based on the fusion of complete ensemble empirical mode decomposition with adaptive noise（CEEMDAN）, and convolutional neural network is proposed. CEEMDAN’s strong nonlinear feature decomposition ability is used to decompose vibration signals, and multiple correlation coefficients are used to screen effective fault feature components and eliminate redundant components, and then the optimal component group is input into CNN to realize fault diagnosis. The results show that comparing with EMD-CNN and EEMD-CNN, the proposed method has better robustness and generalization under different fault states and SNR.
关键词
风电机组 /
故障诊断 /
信噪比 /
卷积神经网络 /
相关性分析法
Key words
wind turbines /
fault diagnosis /
signal to noise ration /
CNN /
correlation analysis methods
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] MARUGÁN A P, MARQUEZ F P G. Advanced analytics for detection and diagnosis of false alarms and faults: a real case study[J]. Wind energy, 2019, 22(11): 1622-1635.
[2] HU Z X, WANG Y, GE M F, et al.Data-driven fault diagnosis method based on compressed sensing and improved multiscale network[J]. IEEE transactions on industrial electronics, 2020, 67(4): 3216-3225.
[3] WANG L, LIU Z W, MIAO Q, et al.Time-frequency analysis based on ensemble local mean decomposition and fast kurtogram for rotating machinery fault diagnosis[J]. Mechanical systems and signal processing, 2018, 103: 60-75.
[4] WANG D, ZHAO Y, YI C, et al.Sparsity guided empirical wavelet transform for fault diagnosis of rolling element bearings[J]. Mechanical systems and signal processing, 2018, 101: 292-308.
[5] LEI Y G, HE Z J, ZI Y Y.Application of the EEMD method to rotor fault diagnosis of rotating machinery[J]. Mechanical systems and signal processing, 2009, 23(4): 1327-1338.
[6] LI S P, QIN N, HUANG D R, et al.Damage localization of stacker’s track based on EEMD-EMD and DBSCAN cluster algorithms[J]. IEEE transactions on instrumentation and measurement, 2020, 69(5): 1981-1992.
[7] DING F, LI X, QU J X.Fault diagnosis of rolling bearing based on improved CEEMDAN and distance evaluation technique[J]. Journal of vibroengineering, 2017, 19(1): 260-275.
[8] WANG F A, JIANG H K, SHAO H D, et al.An adaptive deep convolutional neural network for rolling bearing fault diagnosis[J]. Measurement science and technology, 2017, 28(9): 095005.
[9] LIU X L, TIAN Y, LEI X H, et al.Deep forest based intelligent fault diagnosis of hydraulic turbine[J]. Journal of mechanical science and technology, 2019, 33(5): 2049-2058.
[10] 李东东, 王浩, 杨帆, 等. 基于一维卷积神经网络和Soft-Max分类器的风电机组行星齿轮箱故障检测[J]. 电机与控制应用, 2018, 45(6): 80-87, 108.
LI D D, WANG H, YANG F, et al.Fault detection of wind turbine planetary gear box using 1D convolution neural networks and Soft-Max classifier[J]. Electric machines & control application, 2018, 45(6): 80-87, 108.
[11] XIA M, LI T, XU L, et al.Fault diagnosis for rotating machinery using multiple sensors and convolutional neural networks[J]. IEEE/ASME transactions on mechatronics, 2018, 23(1): 101-110.
[12] TORRES M E, COLOMINAS M A, SCHLOTTHAUER G, et al.A complete ensemble empirical mode decomposition with adaptive noise[C]//Proceedings of 2011 IEEE International Conference on Acoustics, Speech and Signal Processing(ICASSP), Prague, Czech Republic, 2011.
[13] 李彦冬, 郝宗波, 雷航. 卷积神经网络研究综述[J]. 计算机应用, 2016, 36(9): 2508-2515, 2565.
LI Y D, HAO Z B, LEI H.Survey of convolutional neural network[J]. Jouraal of computer applications, 2016, 36(9): 2508-2515, 2565.
[14] ZHAO B, ZHANG X M, ZHAN Z H, et al.A robust construction of normalized CNN for online intelligent condition monitoring of rolling bearings considering variable working conditions and sources[J]. Measurement, 2021, 174: 108973.
[15] KUO C C J. Understanding convolutional neural networks with a mathematical model[J]. Journal of visual communication and image representation, 2016, 41: 406-413.
[16] KRIZHEVSKY A, SUTSKEVE I, HINTON G E.ImagNet classification with deep convolutional neural networks[J].Communications of the ACM, 2017, 60(6): 84-90.
[17] TANG S, YE M, ZHU C, et al.Adaptive pedestrian detection using convolutional neural network with dynamically adjusted classifier[J]. Journal of electronic imaging, 2017, 26(1): 013012.
[18] JIN K H, MCCANN M T, FROUSTEY E, et al.Deep convolutional neural network for inverse problems in imaging[J]. IEEE transactions on image processing, 2017, 26(9): 4509-4522.
[19] SHENG S W.Wind turbine gearbox vibration condition monitoring benchmarking datasets[R]. National Renewable Energy Laboratory, Colorado, USA, 2012.
基金
国家自然科学基金（51976131; 52006148）; 上海市“科技创新行动计划”地方院校能力建设项目（19060502200）
PDF(2885 KB)
