基于零样本学习的风力机故障诊断方法

doi:10.16183/j.cnki.jsjtu.2023.375

摘要/Abstract

摘要：

在工程实践中,风力机故障诊断面临训练故障与实际故障类别不同的情况,为实现对风力机未知故障的诊断,需要将训练过程中习得的故障特征信息迁移至未知故障中.不同于直接建立故障样本与故障类别间映射关系的传统方法,提出一种基于零样本学习的风力机故障诊断方法来完成故障特征迁移.通过描述每种故障的属性建立故障属性矩阵,将其嵌入故障样本空间与故障类别空间之中;并基于卷积神经网络建立故障属性学习器,基于欧氏距离建立故障分类器,形成从故障样本预测故障属性进而分类故障的诊断流程.最后通过与其他零样本学习方法的对比验证了所提故障诊断方法的有效性和优越性.

关键词: 风力机故障诊断, 零样本学习, 卷积神经网络, 知识-数据混合驱动

Abstract:

In engineering practice, wind turbine fault diagnosis encounters situations where the fault category in the training data is different from the actual one. To diagnose unknown wind turbine faults, it is necessary to transfer the fault feature information learned during training to the unknown fault category. Unlike traditional methods that directly establish mapping between fault samples and fault categories, a zero-shot learning (ZSL) method for wind turbine fault diagnosis based on fault attributes is proposed to enable fault feature migration. A fault attribute matrix is established by describing the attributes of each fault, which is embedded into the fault sample space and fault category space. Then, a fault attribute learner is developed based on convolutional neural network (CNN), and a fault classifier is established based on Euclidean distance, forming the diagnosis process where fault attributes are predicted from fault samples and then classified. Finally, the effectiveness and superiority of the proposed fault diagnosis method are validated by comparing it with other zero-shot learning methods.

Key words: wind turbine fault diagnosis, zero-shot learning (ZSL), convolutional neural network (CNN), knowledge-data hybrid driven

中图分类号:

TM315

潘美琪, 贺兴. 基于零样本学习的风力机故障诊断方法[J]. 上海交通大学学报, 2025, 59(5): 561-568.

PAN Meiqi, HE Xing. A Fault Diagnosis Method for Wind Turbines Based on Zero-Shot Learning[J]. Journal of Shanghai Jiao Tong University, 2025, 59(5): 561-568.

图/表 8

图1

图2

图3

图4

表1

表2

表3

表4

参考文献 25

[1]	王彩霞, 时智勇, 梁志峰, 等. 新能源为主体电力系统的需求侧资源利用关键技术及展望[J]. 电力系统自动化, 2021, 45(16): 37-48.
	WANG Caixia, SHI Zhiyong, LIANG Zhifeng, et al. Key technologies and prospects of demand-side resource utilization for power systems dominated by renewable energy[J]. Automation of Electric Power Systems, 2021, 45(16): 37-48.
[2]	龙霞飞, 杨苹, 郭红霞, 等. 大型风力发电机组故障诊断方法综述[J]. 电网技术, 2017, 41(11): 3480-3491.
	LONG Xiafei, YANG Ping, GUO Hongxia, et al. Review of fault diagnosis methods for large wind turbines[J]. Power System Technology, 2017, 41(11): 3480-3491.
[3]	曾军, 陈艳峰, 杨苹, 等. 大型风力发电机组故障诊断综述[J]. 电网技术, 2018, 42(3): 849-860.
	ZENG Jun, CHEN Yanfeng, YANG Ping, et al. Review of fault diagnosis methods of large-scale wind turbines[J]. Power System Technology, 2018, 42(3): 849-860.
[4]	杨晓峰, 方逸航, 赵鹏臻, 等. 基于K-means和BPNN的风机状态识别[J]. 中国电力, 2023, 56(6): 158-166.
	YANG Xiaofeng, FANG Yihang, ZHAO Pengzhen, et al. State recognition of wind turbines based on K-means and BPNN[J]. Electric Power, 2023, 56(6): 158-166.
[5]	王宏伟, 孙文磊, 张小栋, 等. 基于优化VMD复合多尺度散布熵及LSTM的风力发电机齿轮箱故障诊断方法研究[J]. 太阳能学报, 2022, 43(4): 288-295. doi: 10.19912/j.0254-0096.tynxb.2020-0457
	WANG Hongwei, SUN Wenlei, ZHANG Xiaodong, et al. Fault diagnosis method of wind turbine’s gearbox based on composite multiscale dispersion entropy of optimised VMD and LSTM[J]. Acta Energiae Solaris Sinica, 2022, 43(4): 288-295. doi: 10.19912/j.0254-0096.tynxb.2020-0457
[6]	魏书荣, 张鑫, 符杨, 等. 基于GRA-LSTM-Stacking模型的海上双馈风力发电机早期故障预警与诊断[J]. 中国电机工程学报, 2021, 41(7): 2373-2383.
	WEI Shurong, ZHANG Xin, FU Yang, et al. Early fault warning and diagnosis of offshore wind DFIG based on GRA-LSTM-Stacking model[J]. Proceedings of the CSEE, 2021, 41(7): 2373-2383.
[7]	王爽心, 郭婷婷, 李蒙. 风电机组变工况变桨系统异常状态在线识别[J]. 中国电机工程学报, 2019, 39(17): 5144-5152.
	WANG Shuangxin, GUO Tingting, LI Meng. On-line abnormal state identification of pitch system based on transitional mode for wind turbine[J]. Proceedings of the CSEE, 2019, 39(17): 5144-5152.
[8]	黄南天, 杨学航, 蔡国伟, 等. 采用非平衡小样本数据的风机主轴承故障深度对抗诊断[J]. 中国电机工程学报, 2020, 40(2): 563-574.
	HUANG Nantian, YANG Xuehang, CAI Guowei, et al. A deep adversarial diagnosis method for wind turbine main bearing fault with imbalanced small sample scenarios[J]. Proceedings of the CSEE, 2020, 40(2): 563-574.
[9]	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. doi: 10.19912/j.0254-0096.tynxb.2022-0085
	AN Wenjie, CHEN Changzheng, TIAN Miao, et al. Research on bearing fault diagnosis of wind turbines based on transfer learning[J]. Acta Energiae Solaris Sinica, 2023, 44(6): 367-373. doi: 10.19912/j.0254-0096.tynxb.2022-0085
[10]	潘崇煜, 黄健, 郝建国, 等. 融合零样本学习和小样本学习的弱监督学习方法综述[J]. 系统工程与电子技术, 2020, 42(10): 2246-2256. doi: 10.3969/j.issn.1001-506X.2020.10.13
	PAN Chongyu, HUANG Jian, HAO Jianguo, et al. Survey of weakly supervised learning integrating zero-shot and few shot learning[J]. Systems Engineering and Electronics, 2020, 42(10): 2246-2256. doi: 10.3969/j.issn.1001-506X.2020.10.13
[11]	冯良骏. 面向零/少样本场景的弱监督学习方法、应用与实现[D]. 杭州: 浙江大学, 2023.
	FENG Liangjun. Methods, applications, and implementation of weakly supervised learning on zero/few-shot scenarios[D]. Hangzhou: Zhejiang University, 2023.
[12]	胡姚刚, 刘怀盛, 时萍萍, 等. 风电机组偏航系统故障诊断与寿命预测综述[J]. 中国电机工程学报, 2022, 42(13): 4871-4884.
	HU Yaogang, LIU Huaisheng, SHI Pingping, et al. Overview of fault diagnosis and life prediction for wind turbine yaw system[J]. Proceedings of the CSEE, 2022, 42(13): 4871-4884.
[13]	金晓航, 孙毅, 单继宏, 等. 风力发电机组故障诊断与预测技术研究综述[J]. 仪器仪表学报, 2017, 38(5): 1041-1053.
	JIN Xiaohang, SUN Yi, SHAN Jihong, et al. Fault diagnosis and prognosis for wind turbines: An overview[J]. Chinese Journal of Scientific Instrument, 2017, 38(5): 1041-1053.
[14]	潘振东, 刘城, 张勤刚, 等. 海上风力发电机组变桨系统故障分析[J]. 船舶工程, 2022, 44(Sup.2): 107-111.
	PAN Zhendong, LIU Cheng, ZHANG Qingang, et al. Failure analysis of pitch system of offshore wind turbine[J]. Ship Engineering, 2022, 44(Sup.2): 107-111.
[15]	RITCHIE A, BALZANO L, KESSLER D, et al. Supervised PCA: A multiobjective approach[DB/OL].(2020-11-10) [2023-08-07]. https://arxiv.org/pdf/2011.05309 .
[16]	徐舒玮, 邱才明, 张东霞, 等. 基于深度学习的输电线路故障类型辨识[J]. 中国电机工程学报, 2019, 39(1): 65-74.
	XU Shuwei, QIU Caiming, ZHANG Dongxia, et al. A deep learning approach for fault type identification of transmission line[J]. Proceedings of the CSEE, 2019, 39(1): 65-74.
[17]	成珂, 孙琦琦, 马晓瑶. 基于主成分回归分析的气象因子对光伏发电量的影响[J]. 太阳能学报, 2021, 42(2): 403-409.
	CHENG Ke, SUN Qiqi, MA Xiaoyao. Influence of meteorological factors on photovoltaic power generation based on principal component regression analysis[J]. Acta Energiae Solaris Sinica, 2021, 42(2): 403-409.
[18]	张大海, 张晓炜, 孙浩, 等. 基于卷积神经网络的交直流输电系统故障诊断[J]. 电力系统自动化, 2022, 46(5): 132-145.
	ZHANG Dahai, ZHANG Xiaowei, SUN Hao, et al. Fault diagnosis for AC/DC transmission system based on convolutional neural network[J]. Automation of Electric Power Systems, 2022, 46(5): 132-145.
[19]	梁舒曼, 谷艳玲, 罗园庆, 等. 基于增强型卷积神经网络的风力发电机行星齿轮箱故障诊断方法[J]. 太阳能学报, 2023, 44(2): 146-152. doi: 10.19912/j.0254-0096.tynxb.2021-1109
	LIANG Shuman, GU Yanling, LUO Yuanqing, et al. Fault diagnosis method of wind turbine planetary gearbox based on enhanced convolutional neural network[J]. Acta Energiae Solaris Sinica, 2023, 44(2): 146-152. doi: 10.19912/j.0254-0096.tynxb.2021-1109
[20]	陈厚合, 张赫, 王长江, 等. 基于卷积神经网络的直流送端系统暂态过电压估算方法[J]. 电网技术, 2020, 44(8): 2987-2999.
	CHEN Houhe, ZHANG He, WANG Changjiang, et al. A method estimating transient overvoltage of HVDC sending-end system based on convolutional neural network[J]. Power System Technology, 2020, 44(8): 2987-2999.
[21]	郑炜, 林瑞全, 王俊, 等. 基于GAF与卷积神经网络的电能质量扰动分类[J]. 电力系统保护与控制, 2021, 49(11): 97-104.
	ZHENG Wei, LIN Ruiquan, WANG Jun, et al. Power quality disturbance classification based on GAF and a convolutional neural network[J]. Power System Protection and Control, 2021, 49(11): 97-104.
[22]	王守相, 郭陆阳, 赵倩宇, 等. 基于参考序列与全卷积网络的风速数据缺失与异常修复方法[J]. 电力系统自动化, 2023, 47(9): 129-136.
	WANG Shouxiang, GUO Luyang, ZHAO Qianyu, et al. Repair method for missed and abnormal wind speed data based on reference sequence and full convolution network[J]. Automation of Electric Power Systems, 2023, 47(9): 129-136.
[23]	陆文安, 朱清晓, 李兆伟, 等. 基于卷积神经网络的新型电力系统频率特性预测方法[J]. 上海交通大学学报, 2024, 58(10): 1500-1512. doi: 10.16183/j.cnki.jsjtu.2023.071
	LU Wen’an, ZHU Qingxiao, LI Zhaowei, et al. A prediction method of new power system frequency characteristics based on convolutional neural network[J]. Journal of Shanghai Jiao Tong University, 2024, 58(10): 1500-1512.
[24]	ROMERA-PAREDES B, TORR P H S. An embarrassingly simple approach to zero-shot learning[C]// Proceedings of the 32nd International Conference on Machine Learning. Lille, France: ACM, 2015: 2152-2161.
[25]	KODIROV E, XIANG T, GONG S G. Semantic autoencoder for zero-shot learning[C]// IEEE Conference on Computer Vision and Pattern Recognition. Honolulu, USA: IEEE, 2017: 4447-4456.

编辑推荐 0

Metrics

阅读次数

全文

1151

HTML			PDF

最新录用	在线预览	正式出版	最新录用	在线预览	正式出版
0	0	10	911	0	230

来源	本网站	其他网站

次数	138	1013
比例	12%	88%

摘要

1615

最新录用	在线预览	正式出版

343	0	1272

来源	本网站	其他网站

次数	123	1492
比例	8%	92%

卷积层数	准确率/%
卷积层数	A组	B组	C组	平均值
3	70.63	77.88	74.88	74.46
4	71.13	78.25	75.00	74.79
5	71.38	78.50	75.38	75.09
6	72.13	79.13	75.88	75.71
7	71.75	78.63	75.50	75.29
8	71.50	78.49	75.45	75.15
9	71.09	78.33	75.32	74.91

池化方式	准确率/%
池化方式	A组	B组	C组	平均值
平均池化	63.88	66.00	69.13	66.34
最大池化	66.50	72.63	67.13	68.75
最大池化+平均池化	72.13	79.13	75.88	75.71

属性学习算法	准确率/%
属性学习算法	A组	B组	C组	平均值
KNN	52.38	50.88	61.00	54.75
Bagging	53.75	67.38	60.50	60.54
CNN	72.13	79.13	75.88	75.71

零样本学习方法	准确率/%
零样本学习方法	A组	B组	C组	平均值
ESZSL	54.00	51.38	58.63	54.67
SAE	65.25	71.00	70.50	68.92
本文方法	72.13	79.13	75.88	75.71