针对新装机组数据量不足和单源域信息有限的问题,提出一种基于双源域自适应的风电机组状态监测方法,通过特定域的数据分布与回归预测对齐,实现机组发电功率的准确预测。首先,通过皮尔逊系数选出与机组发电机相关的特征;其次,通过公共特征提取器提取出目标域和源域机组的公共特征;然后,通过域特定特征对齐与域特定回归器对齐,减少源域和目标域机组的特征分布差异并减少目标域机组经过不同源域机组训练回归器的输出差异;接着,融合回归损失、域特有特征差异损失和域回归器差异损失3个损失函数,联合优化模型的总损失;最后,将目标域机组在线数据输入到模型中预测发电功率,结合功率预测残差分析对机组运行状态进行监测与判断。结果表明所提方法比单源域自适应具有更高的预测精度,可有效实现风电机组的运行状态监测。
Abstract
A wind turbine condition monitoring method based on dual-source domain adaptation is proposed to address the challenges of insufficient data for newly installed turbines and limited data from a single source domain. This method enhances the accuracy of wind turbine power prediction by employing a two-stage framework for domain domain-specific alignment and domain-specific regression prediction alignment. First, features correlated with the turbine generator are selected using the Pearson correlation coefficient. Then, features from both the target and source domain turbines are extracted via a common feature extractor. Domain-specific feature alignment and domain-specific regressor alignment are applied to reduce the feature distribution discrepancy between the source and target domain turbines, as well as to minimize the output differences of regressors that are trained on different source domains for the target turbine. Subsequently, the total loss of the model is jointly optimized by integrating three loss functions： regression loss, domain-specific feature difference loss and domain regressor difference loss. Finally, the model is used to predict the power of the target domain turbine using online data, and the turbine's operational conditions are monitored and evaluated through the analysis of prediction residuals. The results show that the proposed method outperforms single-source domain adaptation in prediction accuracy and enables wind turbine condition monitoring effectively.
关键词
风电机组 /
状态监测 /
单源域自适应 /
双源域自适应
Key words
wind turbines /
condition monitoring /
single-source domain adaptation /
dual-source domain adaptation
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 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 X H, SUN Y, SHAN J H, et al.Fault diagnosis and prognosis for wind turbines: an overview[J]. Chinese journal of scientific instrument, 2017, 38(5): 1041-1053.
[2] 陈雪峰, 郭艳婕, 许才彬, 等. 风电装备故障诊断与健康监测研究综述[J]. 中国机械工程, 2020, 31(2): 175-189.
CHEN X F, GUO Y J, XU C B, et al.Review of fault diagnosis and health monitoring for wind power equipment[J]. China mechanical engineering, 2020, 31(2): 175-189.
[3] 龙寰, 杨婷, 徐劭辉, 等. 基于数据驱动的风电机组状态监测与故障诊断技术综述[J]. 电力系统自动化, 2023, 47(23): 55-69.
LONG H, YANG T, XU S H, 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.
[4] TAUTZ-WEINERT J, WATSON S J.Using SCADA data for wind turbine condition monitoring-a review[J]. IET renewable power generation, 2017, 11(4): 382-394.
[5] 金晓航, 王宇, Zhang Bin.工业大数据驱动的故障预测与健康管理[J]. 计算机集成制造系统, 2022, 28(5): 1314-1336.
JIN X H, WANG Y, ZHANG B.Industrial big data-driven fault prognostics and health management[J]. Computer integrated manufacturing systems, 2022, 28(5): 1314-1336.
[6] SUN Y, ZHOU Q B, SUN L, et al.CNN-LSTM-AM: a power prediction model for offshore wind turbines[J]. Ocean engineering, 2024, 301: 117598.
[7] KISVARI A, LIN Z, LIU X L.Wind power forecasting-A data-driven method along with gated recurrent neural network[J]. Renewable energy, 2021, 163: 1895-1909.
[8] KONG Z Q, TANG B P, DENG L, et al.Condition monitoring of wind turbines based on spatio-temporal fusion of SCADA data by convolutional neural networks and gated recurrent units[J]. Renewable energy, 2020, 146: 760-768.
[9] 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.
[10] ZHUANG F Z, QI Z Y, DUAN K Y, et al.A comprehensive survey on transfer learning[J]. Proceedings of the IEEE, 2021, 109(1): 43-76.
[11] JIN X H, PAN H T, YING C Z, et al.Condition monitoring of wind turbine generator based on transfer learning and one-class classifier[J]. IEEE sensors journal, 2022, 22(24): 24130-24139.
[12] ZHU Y C, ZHU C C, TAN J J, et al.Anomaly detection and condition monitoring of wind turbine gearbox based on LSTM-FS and transfer learning[J]. Renewable energy, 2022, 189: 90-103.
[13] WANG A Q, QIAN Z, PEI Y, et al.A de-ambiguous condition monitoring scheme for wind turbines using least squares generative adversarial networks[J]. Renewable energy, 2022, 185: 267-279.
[14] CHEN P, LI Y, WANG K S, et al.A threshold self-setting condition monitoring scheme for wind turbine generator bearings based on deep convolutional generative adversarial networks[J]. Measurement, 2021, 167: 108234.
[15] 吴宣勇, 黄忠全, 李琪康, 等. 无源数据约束下多源域自适应的风电齿轮箱故障诊断方法[J]. 太阳能学报, 2024, 45(4): 238-246.
WU X Y, HUANG Z Q, LI Q K, et al.Multi-source domain adaptive fault diagnosis method of wind turbine gearbox under no-accessing source data constraints[J]. Acta energiae solaris sinica, 2024, 45(4): 238-246.
[16] LI Q K, TANG B P, DENG L, et al.Cross-attribute adaptation networks: distilling transferable features from multiple sampling-frequency source domains for fault diagnosis of wind turbine gearboxes[J]. Measurement, 2022, 200: 111570.
[17] SUN S L, SHI H L, WU Y B.A survey of multi-source domain adaptation[J]. Information fusion, 2015, 24: 84-92.
[18] JIN X H, LV S Y, KONG Z Q, et al.Graph spatio-temporal networks for condition monitoring of wind turbine[J]. IEEE transactions on sustainable energy, 2024, 15(4): 2276-2286.
[19] JIN X H, XU Z W, QIAO W.Condition monitoring of wind turbine generators using SCADA data analysis[J]. IEEE transactions on sustainable energy, 2021, 12(1): 202-210.
[20] KIRANYAZ S, INCE T, ABDELJABER O, et al.1-D convolutional neural networks for signal processing applications[C]//ICASSP 2019-2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). Brighton, UK, 2019: 8360-8364.
[21] 田淼, 苏晓明, 陈长征, 等. 基于领域自适应的风力机发电机轴承故障诊断方法研究[J]. 太阳能学报, 2023, 44(11): 310-317.
TIAN M, SU X M, CHEN C Z, et al.Research on fault diagnosis method of wind turbine generator bearings based on domain adaptation[J]. Acta energiae solaris sinica, 2023, 44(11): 310-317.
[22] PAN S J, YANG Q.A survey on transfer learning[J]. IEEE transactions on knowledge and data engineering, 2010, 22(10): 1345-1359.
[23] 王晋东, 陈益强. 迁移学习导论[M]. 2版. 北京: 电子工业出版社, 2022.
WANG J D, CHEN Y Q.Introduction to transfer learning[M]. 2nd ed. Beijing: Publishing House of Electronics Industry, 2022.
[24] SAARI J, STRÖMBERGSSON D, LUNDBERG J, et al. Detection and identification of windmill bearing faults using a one-class support vector machine (SVM)[J]. Measurement, 2019, 137: 287-301.
[25] 王梓齐. 基于正常行为建模的风电机组状态监测方法研究[D]. 北京: 华北电力大学, 2022.
WANG Z Q.Research on condition monitoring method of wind turbine based on normal behavior modeling[D]. Beijing: North China Electric Power University, 2022.
[26] ZHU Y C, ZHUANG F Z, WANG D Q.Aligning domain-specific distribution and classifier for cross-domain classification from multiple sources[C]//Proceedings of the AAAI Conference on Artificial Intelligence Honolulu, USA, 2019: 5989-5996.
基金
国家重点研发计划（2022YFE0198900）; 国家自然科学基金（62473336）; 浙江省自然科学基金（LZ25F030004）
PDF(5937 KB)
