考虑潜变量影响的风电机组双馈感应发电机故障检测方法

doi:10.19595/j.cnki.1000-6753.tces.242077

Abstract
Figure/Table
References
Related Citation (15)

Download: PDF (2401 KB) HTML (1 KB)
Export: BibTeX | EndNote (RIS)

Abstract Double-fed induction generators (DFIG) in wind turbines face significant challenges in accurately and promptly detecting early anomalies caused by various faults due to latent variables arising from operational conditions and coupling mechanisms. This paper proposes a fault detection method for DFIG based on dynamic kernel latent variable regression (DKLVR). The aim is to achieve unified monitoring of DFIG under dynamic and nonlinear conditions using multiple variables, thereby robustly and sensitively identifying early fault characteristics in their initial stages.
The mutual information principle is employed to analyze the overall temperature characteristics of the DFIG, identifying external factors that affect its stability. The temperature characteristics of coupling components have a more significant influence on the overall temperature characteristics of DFIG, and this influence exists in an unobservable form, i.e., as latent variables. Thus, a latent variable regression (LVR) model is needed to extract and eliminate the latent variables. However, LVR models have rarely been applied in the wind power field. Moreover, due to the influence of wind turbine operating conditions, SCADA data exhibit dynamic and nonlinear characteristics. This study introduces time-delay data into LVR to describe the dynamic characteristics of wind turbine operations and employs kernel functions to capture the nonlinear characteristics of the process. Additionally, feature vector selection methods are used to mitigate the dimensionality issues arising from high-dimensional mappings.
Based on these enhancements, a DKLVR-based online fault detection framework for DFIG is proposed. The specific process includes:(1) Standardizing SCADA historical data from wind turbines. (2) Training the DKLVR model using historical data to determine key parameters. (3) Constructing generator monitoring statistics based on the residual subspace using PCA and determining control limits. (4) Online detection: If new input data exceeds the control limits, a generator fault is identified. Fault variables are determined through residual contribution analysis. Three typical DFIG fault cases (bearing fault, winding fault, and slip ring fault) were analyzed using two performance indicators: false alarm rate and early warning margin. The results show that the proposed method outperforms DLVR, KLVR, and LVR in better adapting to the dynamic and nonlinear characteristics of wind turbine SCADA data. In three different scenarios, the proposed detection method is more reasonable and practical. Additionally, compared with mainstream detection methods, such as Transformer, SAE, and DKPCA, the proposed method exhibits superior detection performance and is capable of locating the positions of all three types of faults. Finally, a comprehensive analysis is conducted using indicators such as the daily false alarm rate and training time for DKLVR, LVR, Transformer, and SAE. The results demonstrate that the proposed method meets the online detection requirements and exhibits superior detection performance.

Key words： Doubly-fed induction generator fault detection multivariate statistical process control dynamic kernel latent variable regression latent variable

Received: 19 November 2024

PACS:

TM307

	Service

	E-mail this article
	Add to my bookshelf
	Add to citation manager
	E-mail Alert
	RSS
	Articles by authors
	Yin Yanbo
	Xu Boqiang
	Wu Jinglong
	Sun LiLing

Cite this article:

Yin Yanbo,Xu Boqiang,Wu Jinglong等. Fault Detection Method for Wind Turbines Considering Latent Variable Influences Variable Regression[J]. Transactions of China Electrotechnical Society, 2026, 41(2): 527-540.

URL:

https://dgjsxb.ces-transaction.com/EN/10.19595/j.cnki.1000-6753.tces.242077 OR https://dgjsxb.ces-transaction.com/EN/Y2026/V41/I2/527

[1] GWEC|Global wind report 2025[EB/OL].GWEC|Global wind report 2025[EB/OL]. https://gwec.net/global-wind-report-2025.
[2] 朱茂林, 刘灏, 毕天姝. 考虑风电场量测相关性的双馈风力发电机鲁棒动态状态估计[J]. 电工技术学报, 2023, 38(3): 726-740.
Zhu Maolin, Liu Hao, Bi Tianshu.Robust dynamic state estimation of doubly-fed induction generator considering measurement correlation in wind farms[J]. Transactions of China Electrotechnical Society, 2023, 38(3): 726-740.
[3] 庞彬, 郑涵升, 周紫烨, 等. 考虑不平衡磁拉力的双馈异步风力发电机轴承外圈故障动力学建模[J].电机与控制应用, 2024, 51(3): 30-37.
Pang Bin, Zheng Hansheng, Zhou Ziye, et al.Dynamic modeling of doubly-fed wind turbine bearing outer ring fault considering unbalanced magnetic pull[J]. Electric Machines & Control Application, 2024, 51(3): 30-37.
[4] 夏志凌, 胡凯波, 刘心悦, 等. 基于变模态分解的异步电机转子断条故障诊断[J]. 电工技术学报, 2023, 38(8): 2048-2059.
Xia Zhiling, Hu Kaibo, Liu Xinyue, et al.Fault diagnosis of rotor broken bar in induction motor based on variable mode decomposition[J]. Transa- ctions of China Electrotechnical Society, 2023, 38(8): 2048-2059.
[5] 张辉, 戈宝军, 韩斌, 等. 基于GAF-CapsNet的电机轴承故障诊断方法[J]. 电工技术学报, 2023, 38(10): 2675-2685.
Zhang Hui, Ge Baojun, Han Bin, et al.Fault diagnosis method of motor bearing based on GAF-CapsNet[J]. Transactions of China Electrotechnical Society, 2023, 38(10): 2675-2685.
[6] 许伯强, 郑泽慧. 基于Duffing系统与APES算法的DFIG定子匝间故障检测新方法[J]. 电力自动化设备, 2019, 39(5): 103-108.
Xu Boqiang, Zheng Zehui.Novel method based on Duffing system and APES algorithm for stator winding inter-turn fault detection of DFIG[J]. Electric Power Automation Equipment, 2019, 39(5): 103-108.
[7] Wang Ziqi, Jin Xiaohang, Xu Zhengguo.An adaptive condition monitoring method of wind turbines based on multivariate state estimation technique and continual learning[J]. IEEE Transactions on Instrumentation and Measurement, 2023, 72: 3513009.
[8] 刘展, 刘健洵, 包琰洋, 等. 基于图正则化堆叠自编码器的风机轴承故障诊断方法[J]. 发电技术, 2024, 45(6): 1146-1152.
Liu Zhan, Liu Jianxun, Bao Yanyang, et al.Bearing faults diagnosis method based on stacked auto- encoder with graph regularization for wind turbines[J]. Power Generation Technology, 2024, 45(6): 1146-1152.
[9] 刘展, 包琰洋, 李大字. 基于重采样降噪与主成分分析的宽卷积深度神经网络风机故障诊断方法[J]. 发电技术, 2023, 44(6): 824-832.
Liu Zhan, Bao Yanyang, Li Dazi.Fault diagnosis method of wind turbines based on wide deep convolutional neural network with resampling and principal component analysis[J]. Power Generation Technology, 2023, 44(6): 824-832.
[10] Xiang Ling, Yang Xin, Hu Aijun, et al.Condition monitoring and anomaly detection of wind turbine based on cascaded and bidirectional deep learning networks[J]. Applied Energy, 2022, 305: 117925.
[11] 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.
[12] 赵洪山, 林诗雨, 孙承妍, 等. 考虑多时间尺度信息的风力发电机滚动轴承故障预测[J]. 中国电机工程学报, 2024, 44(22): 8908-8920.
Zhao Hongshan, Lin Shiyu, Sun Chengyan, et al.Fault prediction of wind turbine rolling bearing con- sidering multi-time scale information[J]. Proceedings of the CSEE, 2024, 44(22): 8908-8920.
[13] 赵洪山, 刘辉海, 刘宏杨, 等. 基于堆叠自编码网络的风电机组发电机状态监测与故障诊断[J]. 电力系统自动化, 2018, 42(11): 102-108.
Zhao Hongshan, Liu Huihai, Liu Hongyang, et al.Condition monitoring and fault diagnosis of wind turbine generator based on stacked autoencoder network[J]. Automation of Electric Power Systems, 2018, 42(11): 102-108.
[14] Jin Xiaohang, Xu Zhuangwei, Qiao Wei.Condition monitoring of wind turbine generators using SCADA data analysis[J]. IEEE Transactions on Sustainable Energy, 2021, 12(1): 202-210.
[15] Wang Yifei, Ma Xiandong, Qian Peng.Wind turbine fault detection and identification through PCA-based optimal variable selection[J]. IEEE Transactions on Sustainable Energy, 2018, 9(4): 1627-1635.
[16] Zhang Kai, Tang Baoping, Deng Lei, et al.Fault detection of wind turbines by subspace reconstruction- based robust kernel principal component analysis[J]. IEEE Transactions on Instrumentation and Measure- ment, 2021, 70: 3515711.
[17] Liu Xingchen, Du Juan, Ye Zhisheng.A condition monitoring and fault isolation system for wind turbine based on SCADA data[J]. IEEE Transactions on Industrial Informatics, 2022, 18(2): 986-995.
[18] 陈鹏, 谢颖, 李道璐. 感应电机定子匝间短路故障温升特性研究[J]. 电工技术学报, 2023, 38(18): 4875-4888.
Chen Peng, Xie Ying, Li Daolu.Research on the temperature rise characteristics of induction motors with stator inter-turn fault[J]. Transactions of China Electrotechnical Society, 2023, 38(18): 4875-4888.
[19] Zhu Qinqin.Latent variable regression for supervised modeling and monitoring[J]. IEEE/CAA Journal of Automatica Sinica, 2020, 7(3): 800-811.
[20] Zhu Qinqin, Joe Qin S, Dong Yining.Dynamic latent variable regression for inferential sensor modeling and monitoring[J]. Computers & Chemical Engin- eering, 2020, 137: 106809.
[21] Zhang Haitian, Zhu Qinqin.Concurrent multilayer fault monitoring with nonlinear latent variable regression[J]. Industrial & Engineering Chemistry Research, 2022, 61(3): 1423-1442.
[22] Bazan G H, Scalassara P R, Endo W, et al.Stator short-circuit diagnosis in induction motors using mutual information and intelligent systems[J]. IEEE Transactions on Industrial Electronics, 2019, 66(4): 3237-3246.
[23] 孔祥玉, 曹泽豪, 安秋生, 等. 偏最小二乘线性模型及其非线性动态扩展模型综述[J]. 控制与决策, 2018, 33(9): 1537-1548.
Kong Xiangyu, Cao Zehao, An Qiusheng, et al.Review of partial least squares linear models and their nonlinear dynamic expansion models[J]. Control and Decision, 2018, 33(9): 1537-1548.
[24] 董峰, 李昭, 李凌涵, 等. 多模态动态核主成分分析的气液两相流状态监测[J]. 自动化学报, 2022, 48(3): 762-773.
Dong Feng, Li Zhao, Li Linghan, et al.Flow state monitoring of gas-liquid two-phase flow using multiple dynamic kernel principle component analysis[J]. Acta Automatica Sinica, 2022, 48(3): 762-773.
[25] 周平, 刘记平, 梁梦圆, 等. 基于KPLS鲁棒重构误差的高炉燃料比监测与异常识别[J]. 自动化学报, 2021, 47(7): 1661-1671.
Zhou Ping, Liu Jiping, Liang Mengyuan, et al.KPLS robust reconstruction error based monitoring and anomaly identification of fuel ratio in blast furnace ironmaking[J]. Acta Automatica Sinica, 2021, 47(7): 1661-1671.
[26] Zhang Qi, Li Peng, Lang Xun, et al.Improved dynamic kernel principal component analysis for fault detection[J]. Measurement, 2020, 158: 107738.
[27] 李军, 张观东. FVS-MSVM方法在机器人建模与辨识中的应用[J]. 振动与冲击, 2018, 37(20): 67-74.
Li Jun, Zhang Guandong.Robot modeling and identi- fication based on the FVS-MSVM methodmethod[J]. Journal of Vibration and Shock, 2018, 37(20): 67-74.
[28] Wei Lu, Qian Zheng, Zareipour H.Wind turbine pitch system condition monitoring and fault detection based on optimized relevance vector machine regression[J]. IEEE Transactions on Sustainable Energy, 2020, 11(4): 2326-2336.
[29] Yang Luoxiao, Zhang Zijun.A conditional con- volutional autoencoder-based method for monitoring wind turbine blade breakages[J]. IEEE Transactions on Industrial Informatics, 2021, 17(9): 6390-6398.
[30] 许伯强, 王彪, 孙丽玲, 等. 基于改进EEMD和强化视觉Transformer模型的风电机组故障预警[J]. 电工技术学报, 2025, 40(20): 6537-6551.
Xu Boqiang, Wang Biao, Sun Liling, et al.Fault warning of wind turbine based on improved EEMD and enhanced visual transformer model[J]. Transa- ctions of China Electrotechnical Society, 2025, 40(20): 6537-6551.
[31] 马然, 栗文义, 齐咏生. 风电机组健康状态预测中异常数据在线清洗[J]. 电工技术学报, 2021, 36(10): 2127-2139.
Ma Ran, Li Wenyi, Qi Yongsheng.Online cleaning of abnormal data for the prediction of wind turbine health condition[J]. Transactions of China Electro- technical Society, 2021, 36(10): 2127-2139.