Anomaly detection of wind turbines is a key technology for achieving intelligent operation and maintenance management. Existing approaches mainly focus on monitoring individual components, requiring separate detection models for each part. This not only reduces overall system efficiency but also neglects interactions among components. Moreover, in the early stage of a fault, it is often difficult to accurately determine the fault type. To address these challenges, an anomaly detection approach based on an autoencoder was proposed. First, a system-wide parameter reconstruction model was built using autoencoders, with the learning ability enhanced by using turbine data under similar operating conditions. Next, dynamic detection threshold curves were designed for each parameter to improve anomaly identification. Furthermore, a correlation-based identification strategy was introduced to locate key parameters most likely to cause anomalies by analyzing out-of-tolerance errors, thereby supporting subsequent operation and maintenance decisions. Finally, real operation data from a Huanghua wind farm in Hebei, China were used for validation. The results demonstrate that the proposed approach achieves reliable performance in anomaly detection and early fault warning of wind turbines.
状态监测通过实时采集风电机组的运行数据,对各个部件的工况进行监控。根据数据来源的不同,相关研究主要可分为基于数据采集与监控系统(supervisory control and data acquisition, SCADA)和基于状态监测系统(condition monitoring system, CMS)两种技术路径[11-12]。CMS通过在齿轮箱、发电机、叶片等关键部件上加装传感器获取监测信息并进行诊断分析[13]。尽管该方法在理论创新、实验验证与小规模应用中取得了一定成效,但其部署成本较高,且难以实现对风机整体运行状态的全面覆盖。相比之下,SCADA系统作为风电机组的标准配置,能够对核心运行参数、环境信息、故障停机及维修记录等进行远程实时监测与存储,无需增设额外硬件,有效控制了运维投入[14]。因此,基于SCADA数据的风电机组状态监测方法近年来逐渐受到行业与研究者的关注[15]。SCADA系统所积累的海量运行数据,也为深度学习技术在异常检测中的应用提供了基础[16-17]。深度学习凭借其多层非线性映射的网络结构,能够从输入数据中自动提取有效特征,尤其适用于风电机组这类具有强非线性和参数耦合特性的复杂系统。
BashirM B A. Principle parameters and environmental impacts that affect the performance of wind turbine: an overview[J]. Arabian Journal for Science and Engineering, 2022, 47(7): 7891-7909.
[2]
International Renewable Energy Agency. Renewable energy cost analysis: wind power[R]. Abu Dhabi: International Renewable Energy Agency, 2012.
[3]
YanJ, LiuY, LiL, et al. Wind turbine condition monitoring using the SSA-optimized self-attention BiLSTM network and change point detection algorithm[J]. Sensors, 2023, 23(13): 5873.
ZhaoJinsong, WangZiqi, LiuChangliang. Wind turbine gearbox condition monitoring based on Bagging ensemble strategy and multivariate state estimate technique[J]. Science Technology and Engineering, 2020, 20(20): 8180-8186.
SunJian, YangYubing. Icing fault detection of wind turbine blades based on feature extraction and an optimal weighted ensemble strategy[J]. Science Technology and Engineering, 2024, 24(11): 4501-4509.
ShenYanjie, LiPengfei, WuYingjie, et al. A phenomenological model-based gearbox fault monitoring method using stator current analysis for doubly-fed induction wind turbines[J]. Chinese Journal of Scientific Instrument, 2025, 46(9): 348-359.
XiaBo, LiChunyang, WanLulu, et al. Review of wind turbine fault warning methods based on deep learning[J]. Science Technology and Engineering, 2023, 23(9): 3577-3587.
CuiKai, XuYifei, LiXuesong, et al. Wind turbine performance prediction model and early warning of abnormal condition based on GRNN[J]. Science Technology and Engineering, 2020, 20(32): 13220-13228.
LongHuan, YangTing, XuShaohui, et al. Review of data-driven condition monitoring and fault diagnosis technologies for wind turbines[J]. Automation of Electric Power Systems, 2023, 47(23): 55-69.
YuHaoshuai, TangBaoping, ZhangKai, et al. Fault diagnosis method of wind turbine gearboxes mixed with attention prototype networks under small samples[J]. China Mechanical Engineering, 2021, 32(20): 2475-2481.
[22]
Tautz-WeinertJ, WatsonS J. Using SCADA data for wind turbine condition monitoring-a review[J]. IET Renewable Power Generation, 2017, 11(4): 382-394.
[23]
FengY, QiuY, CrabtreeC J, et al. Use of SCADA and CMS signals for failure detection and diagnosis of a wind turbine gearbox[C]// European Wind Energy Conference and Exhibition 2011. Brussels: European Wind Energy Association, 2011: 17-19.
WangZhongxing, ZhouYuangui, ZhangXueguang. Review of artificial intelligence algorithms-based wind turbine condition monitoring and fault diagnosis techniques[J]. Journal of Northeast Electric Power University, 2024, 44(1): 42-51.
[26]
WenY, RahmanM F, XuH, et al. Recent advances and trends of predictive maintenance from data-driven machine prognostics perspective[J]. Measurement, 2022, 187: 110276.
[27]
ChenW, QiuY, FengY, et al. Diagnosis of wind turbine faults with transfer learning algorithms[J]. Renewable Energy, 2021, 163: 2053-2067.
[28]
LiH, LiuC, DuQ, et al. Optimizing resource allocation for wind turbine maintenance through process and fault data fusion[J]. Information Fusion, 2025: 103678.
[29]
ChatterjeeJ, DethlefsN. A dual transformer model for intelligent decision support for maintenance of wind turbines[C]// 2020 International Joint Conference on Neural Networks. Glasgow: IEEE, 2020: 1-10.
[30]
ZhangG, LiY, JiangW, et al. A fault diagnosis method for wind turbines with limited labeled data based on balanced joint adaptive network[J]. Neurocomputing, 2022, 481: 133-153.
[31]
ChenH, LiuH, ChuX, et al. Anomaly detection and critical SCADA parameters identification for wind turbines based on LSTM-AE neural network[J]. Renewable Energy, 2021, 172: 829-840.
[32]
FuJ, ChuJ, GuoP, et al. Condition monitoring of wind turbine gearbox bearing based on deep learning model[J]. IEEE Access, 2019, 7: 57078.
[33]
XiaoC, LiuZ, ZhangT, et al. Deep learning method for fault detection of wind turbine converter[J]. Applied Sciences, 2021, 11(3): 1280.
[34]
ZhengM, ManJ, WangD, et al. Semi-supervised multivariate time series anomaly detection for wind turbines using generator SCADA data[J]. Reliability Engineering & System Safety, 2023, 235: 109235.
ZhuYanmin, LiZhonghu, YangLiqing, et al. Fault prediction of wind turbine based on convolutional bidirectional long short term memory model[J]. Science Technology and Engineering, 2024, 24(25): 10790-10797.
[37]
WangH, WangH, JiangG, et al. Early fault detection of wind turbines based on operational condition clustering and optimized deep belief network modeling[J]. Energies, 2019, 12(6): 984.
[38]
RenströmN, BangaloreP, HighcockE. System-wide anomaly detection in wind turbines using deep autoencoders[J]. Renewable Energy, 2020, 157: 647-659.
[39]
BaldiP. Autoencoders, unsupervised learning, and deep architectures[C]// Proceedings of ICML Workshop on Unsupervised and Transfer Learning. Bellevue: JMLR Workshop and Conference Proceedings, 2012: 37-49.
[40]
SreenathaM, MallikarjunaP B. A fault diagnosis technique for wind turbine gearbox: an approach using optimized blstm neural network with undercomplete autoencoder[J]. Engineering, Technology & Applied Science Research, 2023, 13(1): 10170-10174.
NiWeijian, SunYujian, LiuTong, et al. Business process remaining time prediction using bidirectional recurrent neural networks with attention[J]. Computer Integrated Manufacturing System, 2020, 26(6): 1564-1572.
[43]
WęglarczykS. Kernel density estimation and its application[C]// ITM Web of Conferences. Kielce: EDP Sciences, 2018, 23: 00037.
[44]
GuoN, LiuC, LiC, et al. Explainable and effective process remaining time prediction using feature-informed cascade prediction model[J]. IEEE Transactions on Services Computing, 2024, 17(3): 949-962.