针对风电机组齿轮箱故障振动信号的非线性、特征信息混叠和诊断精度低等问题,结合时域特征分析与多传感器信息融合技术,提出一种无阈值递归图（NRP）与深度学习相结合的早期故障识别方法。首先,将时域指标作为特征参数并采用特征级与数据级融合技术构建数据信息;其次,采用NRP将一维数据信息转换为二维彩色可视特征图;然后,构建一种AlexNet-SENet网络结构,利用嵌入SENet注意力机制的改进AlexNet使其自适应选择并重点搜索结构与节点的关键特征信息。最后,以华中科技大学行星齿轮箱动力学试验平台采集的振动数据集为实例,结果表明该方法的诊断准确率为99%,能更有效提取故障特征信息,具有更高的分类诊断精度。
Abstract
Aiming at the issues of nonlinearity, aliasing of characteristic information, and low diagnostic accuracy of vibration signals in the wind turbine gearbox, by integrating time-domain characteristic analysis and multi-sensor information fusion technology, an early fault identification approach based on non-threshold recurrence plot （NRP） and deep learning was proposed. First, the time-domain indices is used as the feature parameter, and the feature-level and data-level fusion technology is used to construct the dataset. Secondly, NRP converts one-dimensional data into two-dimensional colored visual feature maps. Then, an AlexNet-SENet network structure is built, and the improved AlexNet embedded with SENet's attention mechanism makes it adaptive to focus on the key structural and nodal feature. Finally, taking the vibration data set collected by the planetary gearbox dynamics test platform of Huazhong University of Science and Technology as an example, the results show that the diagnosis accuracy of this method is 99%, which can extract fault characteristics more effectively and has higher classification and diagnosis accuracy.
关键词
风电机组齿轮箱 /
故障诊断 /
深度学习 /
信息融合 /
无阈值递归图 /
注意力机制
Key words
wind turbine gearbox /
fault diagnosis /
deep learning /
information fusion /
non-threshold recurrence plot /
attention mechanism
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 张振海, 王维庆, 王海云, 等. 基于HCS-GWO-MSVM的风电机组齿轮箱复合故障诊断研究[J]. 太阳能学报, 2021, 42(10): 176-182.
ZHANG Z H, WANG W Q, WANG H Y, et al.Research on compound fault diagnosis of wind turbine gearbox based on HCS-GWO-MSVM[J]. Acta energiae solaris sinica, 2021, 42(10): 176-182.
[2] 吴岚, 柳亦兵, 吴仕明, 等. 基于VMD倒频变换的风电机组齿轮箱复合故障诊断[J]. 振动与冲击, 2023, 42(24): 221-227, 256.
WU L, LIU Y B, WU S M, et al.Composite fault diagnosis of wind turbine gearboxes based on VMD cepstral transform[J]. Journal of vibration and shock, 2023, 42(24): 221-227, 256.
[3] 谭启瑜, 马萍, 张宏立, 等. 基于图注意力网络的风力发电机齿轮箱故障诊断[J]. 太阳能学报, 2024, 45(1): 265-274.
TAN Q Y, MA P, ZHANG H L, et al.Fault diagnosis for wind turbine gearbox based on graph attention networks[J]. Acta energiae solaris sinica, 2024, 45(1): 265-274.
[4] TANG M Z, ZHAO Q, WU H W, et al.Review and perspectives of machine learning methods for wind turbine fault diagnosis[J]. Frontiers in energy research, 2021, 9: 751066.
[5] 张品杨, 陈长征. 基于U-net和随机森林的齿轮箱振动时频分析和故障诊断[J]. 机电工程, 2023, 40(8): 1210-1217.
ZHANG P Y, CHEN C Z.Time-frequency analysis and fault identification of wind turbine gearbox based on U-net and random forest[J]. Journal of mechanical & electrical engineering, 2023, 40(8): 1210-1217.
[6] 陈萱, 杨永超, 袁博洋, 等. NGO-VMD和SSNGO-RF算法在风机齿轮箱故障诊断中的应用[J]. 湖北民族大学学报(自然科学版), 2023, 41(4): 520-529.
CHEN X, YANG Y C, YUAN B Y, et al.Application of NGO-VMD and SSNGO-RF algorithms in fault diagnosis of wind turbine gearboxes[J]. Journal of Hubei Minzu University (natural science edition), 2023, 41(4): 520-529.
[7] 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, 42(1): 24-30.
WANG H N, CUI B Z, PENG Z H, et al.Fault diagnosis of planetary gearbox using ICEEMDAN and SVM[J]. Mechanical science and technology for aerospace engineering, 2023, 42(1): 24-30.
[8] 刘秀丽, 王鸽, 吴国新, 等. VMD及PSO优化SVM的行星齿轮箱故障诊断[J]. 电子测量与仪器学报, 2022, 36(1): 54-61.
LIU X L, WANG G, WU G X, et al.Fault diagnosis method of planetary gear box based on variational modal decomposition and particle swarm optimization support vector machine[J]. Journal of electronic measurement and instrumentation, 2022, 36(1): 54-61.
[9] 张龙, 胡燕青, 赵丽娟, 等. 采用递归图编码技术与残差网络的滚动轴承故障诊断[J]. 西安交通大学学报, 2023, 57(2): 110-120.
ZHANG L, HU Y Q, ZHAO L J, et al.Fault diagnosis of rolling bearings using recurrence plot coding technique and residual network[J]. Journal of Xi'an Jiaotong University, 2023, 57(2): 110-120.
[10] 常淼, 沈艳霞. 基于贝叶斯优化CNN的风电轴承故障诊断策略[J]. 噪声与振动控制, 2021, 41(6): 77-83.
CHANG M, SHEN Y X.Fault diagnosis strategy of wind turbine bearings based on Bayesian optimized CNN[J]. Noise and vibration control, 2021, 41(6): 77-83.
[11] 赵志宏, 李春秀, 窦广鉴, 等. 基于MTF-CNN的轴承故障诊断研究[J]. 振动与冲击, 2023, 42(2): 126-131.
ZHAO Z H, LI C X, DOU G J, et al.Bearing fault diagnosis method based on MTF-CNN[J]. Journal of vibration and shock, 2023, 42(2): 126-131.
[12] WANG D F, GUO Y, WU X, et al.Planetary-gearbox fault classification by convolutional neural network and recurrence plot[J]. Applied sciences, 2020, 10(3): 932.
[13] 韩亮, 罗统军, 蒲秀娟, 等. 使用多尺度递归图和视觉转换器的心律失常分类[J]. 仪器仪表学报, 2022, 43(12): 149-157.
HAN L, LUO T J, PU X J, et al.Arrhythmia classification using multi-scale recurrence plot and vision transformer[J]. Chinese journal of scientific instrument, 2022, 43(12): 149-157.
[14] WANG M H, HUNG C C, LU S D, et al.Fault diagnosis for PV modules based on AlexNet and symmetrized dot pattern[J]. Energies, 2023, 16(22): 7563.
[15] 吴春志, 江鹏程, 冯辅周, 等. 基于一维卷积神经网络的齿轮箱故障诊断[J]. 振动与冲击, 2018, 37(22): 51-56.
WU C Z, JIANG P C, FENG F Z, et al.Faults diagnosis method for gearboxes based on a 1-D convolutional neural network[J]. Journal of vibration and shock, 2018, 37(22): 51-56.
[16] 雷春丽, 薛林林, 焦孟萱, 等. 结合改进ResNet与迁移学习的风力机滚动轴承故障诊断方法[J]. 太阳能学报, 2023, 44(6): 436-444.
LEI C L, XUE L L, JIAO M X, et al.Fault diagnosis method of wind turbines rolling bearing based on improved resnet and transfer learning[J]. Acta energiae solaris sinica, 2023, 44(6): 436-444.
基金
国家自然科学基金（52377185）; 湖南省教育厅科学研究项目（22B0590; 23B0537）
PDF(1609 KB)
