为实现风电机组的异常状态监测并用于其故障诊断和日常维护,提出一种新的监测方法,该方法基于视觉 Transformer（ViT）模型与长短期记忆（LSTM）网络融合,能有效识别风电机组的运行状态。首先,利用箱线图法和Spearman相关性分析对原始SCADA数据进行预处理,去除无效数据并选择输入参数。然后,构建融合LSTM的ViT预测模型,并引入统计学中KL散度作为检测指标,对目标参数预测值与真实值进行计算分析。最后采用核密度估计确定安全阈值,根据检测指标是否越过安全阈值来识别风电机组异常状态。通过将该模型应用于华北某风场进行实例分析,并与其他深度学习模型对比。结果表明：该方法相较于其他模型能更好识别出风电机组异常状态。
Abstract
In order to realize the abnormal state monitoring of wind turbines and use them for fault diagnosis and routine maintenance, a new monitoring method is proposed in this paper, which is based on the fusion of Vision Transformer model and long short-term memory （LSTM） network, which can effectively identify the operating status of wind turbines. Firstly, the box plot method and Spearman correlation analysis were used to preprocess the original SCADA data, remove the invalid data and select the input parameters. Then, the Vision Transformer prediction model fused with LSTM was constructed, and the KL divergence in statistics was introduced as the detection index to calculate and analyze the predicted value and the real value of the target parameter. Finally, the kernel density estimation is used to determine the safety threshold, and the abnormal state of the wind turbine is identified according to whether the detection index exceeds the safety threshold. The model was applied to a wind farm in North China for case analysis, and compared with other deep Xi models. The results show that the proposed method can better identify the abnormal state of wind turbines than other models.
关键词
风电机组 /
状态监测 /
长短期记忆网络 /
视觉Transformer /
KL散度
Key words
wind turbine /
condition monitoring /
LSTM /
ViT /
KL divergence
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 冉靖, 张智刚, 梁志峰, 等. 风电场风速和发电功率预测方法综述[J]. 数理统计与管理, 2020, 39(6): 1045-1059.
RAN J, ZHANG Z G, LIANG Z F, et al.Review of wind speed and wind power prediction methods[J]. Journal of applied statistics and management, 2020, 39(6): 1045-1059.
[2] ZHANG C, HU D, YANG T.Anomaly detection and diagnosis for wind turbines using long short-term memory-based stacked denoising autoencoders and XGBoost[J]. Reliability engineering & system safety, 2022, 222: 108445.
[3] YAO Q T, LI X Z, XIANG L, et al.A new condition-monitoring method based on multi-variable correlation learning network for wind turbine fault detection[J]. Measurement science and technology, 2023, 34(2): 024009.
[4] 刘家瑞, 杨国田, 杨锡运. 基于深度卷积自编码器的风电机组故障预警方法研究[J]. 太阳能学报, 2022, 43(11): 215-223.
LIU J R, YANG G T, YANG X Y.Research on wind turbine fault warning method based on deep convolution auto-encoder[J]. Acta energiae solaris sinica, 2022, 43(11): 215-223.
[5] 胡爱军, 连俭, 向玲. 基于ACNN和Bi-LSTM的风电机组故障早期识别[J]. 太阳能学报, 2021, 42(12): 143-149.
HU A J, LIAN J, XIANG L.Early fault identification of wind turbine based on ACNN and Bi-LSTM[J]. Acta energiae solaris sinica, 2021, 42(12): 143-149.
[6] ENCALADA-DÁVILA Á, MOYÓN L, TUTIVÉN C, et al. Early fault detection in the main bearing of wind turbines based on gated recurrent unit (GRU) neural networks and SCADA data[J]. IEEE/ASME transactions on mechatronics, 2022, 27(6): 5583-5593.
[7] XIANG L, YANG X, HU A J, et al.Condition monitoring and anomaly detection of wind turbine based on cascaded and bidirectional deep learning networks[J]. Applied energy, 2022, 305: 117925.
[8] ZHANG K, TANG B P, DENG L, et al.A hybrid attention improved ResNet based fault diagnosis method of wind turbines gearbox[J]. Measurement, 2021, 179: 109491.
[9] 黄荣舟, 汤宝平, 杨燕妮, 等. 基于长短时记忆网络融合SCADA数据的风电齿轮箱状态监测[J]. 太阳能学报, 2021, 42(1): 235-239.
HUANG R Z, TANG B P, YANG Y N, et al.Condition monitoring of wind turbine gearbox based on LSTM neural network fusing SCADA data[J]. Acta energiae solaris sinica, 2021, 42(1): 235-239.
[10] QIN S G, TAO J, ZHAO Z L.Fault diagnosis of wind turbine pitch system based on LSTM with multi-channel attention mechanism[J]. Energy reports, 2023, 10: 4087-4096.
[11] LU L, HE Y G, RUAN Y, et al. Wind turbine planetary gearbox condition monitoring method based on wireless sensor and deep learning approach[J]. IEEE transactions on instrumentation and measurement, 1809, 70: 3503016.
[12] LIU J Y, WANG X S, XIE F Q, et al.Condition monitoring of wind turbines with the implementation of spatio-temporal graph neural network[J]. Engineering applications of artificial intelligence, 2023, 121: 106000.
[13] CHEN W H, ZHOU H T, CHENG L S, et al.Condition monitoring of wind turbine using novel deep learning method and dynamic kernel principal components Mahalanobis distance[J]. Engineering applications of artificial intelligence, 2023, 125: 106757.
[14] CHENG X P, LU F, LIU Y H.Lightweight hybrid model based on MobileNet-v2 and Vision Transformer for human-robot interaction[J]. Engineering applications of artificial intelligence, 2024, 127: 107288.
[15] LIAN J, LIU T Y.Lesion identification in fundus images via convolutional neural network-vision transformer[J]. Biomedical signal processing and control, 2024, 88: 105607.
[16] XU S J, ZHANG R Y, MA H, et al.On vision transformer for ultra-short-term forecasting of photovoltaic generation using sky images[J]. Solar energy, 2024, 267: 112203.
[17] 胡正南, 胡立坤. 基于Vision Transformer多模型融合的视觉闭环检测算法[J]. 激光杂志, 2024, 45(6): 75-81.
HU Z N, HU L K.Vision loop closure detection algorithm based on Vision Transformer multi-model fusion[J]. Laser journal, 2024, 45(6): 75-81.
[18] LI R X, MAGBOOL JAN N, HUANG B, et al.Constrained multimodal ensemble Kalman filter based on Kullback-Leibler (KL) divergence[J]. Journal of process control, 2019, 79: 16-28.
[19] 胡阳, 胡耀宗, 程逸, 等. 基于FD-AT-LSTM的大型风电机组变频器温度状态监测[J]. 动力工程学报, 2023, 43(9): 1207-1215.
HU Y, HU Y Z, CHENG Y, et al.Temperature monitoring of heavy wind power unit inverters based on FD-AT-LSTM[J]. Journal of Chinese Society of Power Engineering, 2023, 43(9): 1207-1215.
基金
国家自然科学基金（52075170; 52175092）; 河北省自然科学基金（A2022502005）; 河北省在读研究生创新能力培养资助项目（CXZZBS2024169）
PDF(1492 KB)
