Keywords
non-stationary time series
operation anomaly detection
multi-turbine dependency
SCADA data
residual reconstruction
renewable energy systems
Abstract
Wind farms generate large-scale operational time series from turbine speed, blade pitch angle, generator temperature, gearbox vibration, nacelle direction, wind speed, power output, and converter status. These signals are strongly affected by wind variability, wake interaction, seasonal weather, turbine aging, and control strategy changes, making anomaly detection under non-stationary conditions challenging. This study proposes a multi-turbine dependency reconstruction method for non-stationary wind farm operation anomaly detection. The method constructs a turbine dependency graph using spatial layout, wake influence, wind-direction alignment, and lagged power correlation. A temporal reconstruction network is then used to learn normal multi-turbine operating patterns under changing wind regimes. An adaptive residual envelope is introduced to detect abnormal deviation without using a fixed global threshold. Experiments are conducted on a wind farm dataset containing 132 turbines, 48 SCADA variables, and 5-minute observations collected over 32 months. The dataset includes 54 million operational records and 1,860 maintenance-confirmed abnormal events, including gearbox overheating, pitch-control failure, yaw misalignment, converter instability, and abnormal power curve deviation. The proposed method reduces median detection delay from 9.3 hours to 2.4 hours compared with an isolated-turbine LSTM baseline. False alerts decrease to 1.7 events per turbine per month. The Matthews correlation coefficient reaches 0.902 under seasonal wind-regime changes, while residual-envelope drift remains below 0.27 standard units across winter and summer conditions. Online inference processes 18,600 SCADA points per second, and one farm-level daily assessment completes in 5.9 minutes. The results demonstrate that multi-turbine dependency reconstruction can improve abnormal operation detection in non-stationary wind farm time series.
References
1. Wu, C., Zhang, F., Chen, H., & Zhu, J. (2025). Design and optimization of low power persistent logging system based on embedded Linux.
2. Karimi Dastgerdi, A., Mercorelli, P., & Nemati, H. (2026). Autoencoder-driven fault detection in drilling machines: a study of hybrid deep learning models. International Journal of Computer Integrated Manufacturing, 1-18.
3. Wang, Y., Chen, J., Arias, R., Wang, Y., & Yin, X. (2026). Development and Validation of a Patient-Friendly Digital Assessment Platform for Precision Screening of Oral Anti-Obesity Medications (AOMs).
4. Alagha, N., Khairuddin, A. S. M., Haitaamar, Z. N., Al-Khatib, O., & Kanesan, J. (2025). Artificial intelligence in wind turbine fault detection and diagnosis: Advances and perspectives. Energies, 18(7), 1680.
5. Chen, X., Xiao, H., Zeng, Z., Zhang, S., & Xiao, R. (2025). Fine-Grained Multivariate Time Series Anomaly Detection via Causal Inference. Knowledge-Based Systems, 114765.
6. Yin, J., Huang, Y., & Rao, H. (2026). A Study on the Dynamic Evolution of Learning Behavior and Outcome Prediction in Digital Educational Environments. Available at SSRN 6607139.
7. Haseeb, F., Peng, M., Arshad, F., & Ali, W. (2026). Uncertainty aware unsupervised fault diagnosis of PWR nuclear power plant using KNN and SHAP method. Progress in Nuclear Energy, 193, 106191.
8. Yang, J. (2026). Computational Analysis of How Digital Non-Clinical Communication Tools Influence Social Participation Among Older Adults in Community-Based Elderly Care. Available at SSRN 6682838.
9. Zhang, Z. (2026). A Study on Multimodal Product Understanding and Assembly Guidance Optimization in e-commerce for Complex Consumer Goods. Available at SSRN 6734559.
10. Gerbino, M. (2025). Data-Driven Diagnosis of Wind Turbines via SCADA-Based Anomaly Detection Methods (Doctoral dissertation, Politecnico di Torino).
11. Qi, C., & Qiao, X. (2026, May). Design Patterns for Multi-Agent Systems in Production. In 2026 International Conference on Intelligent Systems, Automation and Control (ISAC) (pp. 198-202). IEEE.
12. Su, D., & Shi, X. Optimizing Family–School Collaboration to Improve Educational and Social Outcomes for Children with Intellectual Disabilities in Inclusive Public Schools.
13. Sadeghi, R., Parenti, M., Memme, S., Fossa, M., & Morchio, S. (2025). A review and comparative analysis of solar tracking systems. Energies, 18(10), 2553.
14. Shao, W. (2026, March). Interpretable Ensemble Learning for Network Traffic Anomaly Detection: A SHAP-based Explainable AI Framework for Embedded Systems Security. In 2026 14th International Symposium on Digital Forensics and Security (ISDFS) (pp. 1-4). IEEE.
15. Fu, Y., Gui, H., Li, W., & Wang, Z. (2020, August). Virtual Material Modeling and Vibration Reduction Design of Electron Beam Imaging System. In 2020 IEEE International Conference on Advances in Electrical Engineering and Computer Applications (AEECA) (pp. 1063-1070). IEEE.
16. Khan, M. A., Churchfield, M. J., & Watson, S. J. (2026). Dependence of wind-farm-induced gravity waves and wind farm performance on non-dimensional atmospheric parameters and simulation configuration. Wind Energy Science, 11(5), 1631-1652.
17. Chen, H., Ma, X., Mao, Y., & Ning, P. (2025). Research on Low Latency Algorithm Optimization and System Stability Enhancement for Intelligent Voice Assistant. Available at SSRN 5321721.
18. Wasilczuk, F., Kasprzak, P., Rybiński, M., Flaszyński, P., Dzierwa, P., Ziędalska, K., ... & Taler, D. (2025). Positive effect of wind farm control on energy production and revenue in European markets. Energy, 326, 136159.
19. Liang, R., Feifan, F. N. U., Liang, Y., & Ye, Z. (2025). Emotion-Aware Interface Adaptation in Mobile Applications Based on Color Psychology and Multimodal User State Recognition. Frontiers in Artificial Intelligence Research, 2(1), 51-57.
20. Yang, M., Wang, Y., Shi, J., & Tong, L. (2025). Reinforcement Learning Based Multi-Stage Ad Sorting and Personalized Recommendation System Design.
21. Bekbolatova, Z., Grigoryev, D., Minazhova, S., Bekbayev, A., Omarov, Z., & Kozhageldi, B. (2026). Wind Power Forecasting Based on Time Series Autoregressive Moving Average Model Taking into Account the Effect of Wind Speed Dispersion. ES Energy and Environment, 31, 2025.
22. Deng, H. C., Shu, Z. R., Chan, P. W., & He, X. H. (2026). Cross-height causal dependence of horizontal wind components in offshore environment: insights from transfer entropy analysis. Sustainable Energy Technologies and Assessments, 87, 104890.
23. Gao, G., Gao, R., Lu, C., Gao, R., & Kuang, Y. (2026, March). Security Governance Methods and Quantitative Evaluation for Enterprise SMS and Verification Code Systems. In 2026 International Conference on Generative Artificial Intelligence and Information Security (GAIIS) (pp. 455-458). IEEE.
24. Xu, T., Zhang, J., & Zhu, W. (2026). Fairness-Accuracy Trade-offs and Calibration Mechanisms in Machine Learning-Based Subprime Credit Scoring. Available at SSRN 6893759.
25. Gerbino, M. (2025). Data-Driven Diagnosis of Wind Turbines via SCADA-Based Anomaly Detection Methods (Doctoral dissertation, Politecnico di Torino).
26. Xie, Z., Ren, X., Zheng, T., Bai, J., Fan, W., Xu, B., ... & Song, Y. (2026). A Survey on AI Agent Harness. ResearchGate Preprint. DOI, 10.
27. Özkan, R., & Genç, M. S. (2025). Comparison of Numerical Results With Experiment Using Digital Image Correlation System of Multiobjective Structural Optimised Wind Turbine Blade. Wind Energy, 28(9), e70051.
28. Zhang, Z., Tong, Y., & Gao, Y. (2026). Causal Representation Learning for Explainable Early Warning in High-Stakes Decision Systems.
29. Al Hassan, A., & Dao, P. B. (2025). Bridging data and diagnostics: A systematic review and case study on integrating trend monitoring and change point detection for wind turbines. Energies, 18(19), 5166.
30. Qi, C., & Qiao, X. (2026). Building and Operating a Large Scale Multi-Agent System: A Case Study from Industry. Available at SSRN 6795198.