Graph Neural Networks for Electric and Hydraulic Data Fusion to Enhance Short-term Forecasting of Pumped-storage Hydroelectricity

##plugins.themes.bootstrap3.article.main##

##plugins.themes.bootstrap3.article.sidebar##

Published Jun 27, 2024
Raffael Theiler Olga Fink

Abstract

Pumped-storage hydropower plants (PSH) actively participate in grid power-frequency control and therefore often operate under dynamic conditions, which results in rapidly varying system states. Predicting these dynamically changing states is essential for comprehending the underlying sensor and machine conditions. This understanding aids in detecting anomalies and faults, ensuring the reliable operation of the connected power grid, and in identifying faulty and miscalibrated sensors. PSH are complex, highly interconnected systems encompassing electrical and hydraulic subsystems, each characterized by their respective underlying networks that can individually be represented as graph. To take advantage of this relational inductive bias, graph neural networks (GNNs) have been separately applied to state forecasting tasks in the individual subsystems, but without considering their interdependencies. In PSH, however, these subsystems depend on the same control input, making their operations highly interdependent and interconnected. Consequently, hydraulic and electrical sensor data should be fused across PSH subsystems to improve state forecasting accuracy. This approach has not been explored in GNN literature yet because many available PSH graphs are limited to their respective subsystem boundaries, which makes the method unsuitable to be applied directly. In this work, we introduce the application of spectral-temporal graph neural networks, which leverage self-attention mechanisms to concurrently capture and learn meaningful subsystem interdependencies and the dynamic patterns observed in electric and hydraulic sensors. Our method effectively fuses data from the PSH’s subsystems by operating on a unified, system-wide graph, learned directly from the data, This approach leads to demonstrably improved state forecasting performance and enhanced generalizability.

How to Cite

Theiler, R., & Fink, O. (2024). Graph Neural Networks for Electric and Hydraulic Data Fusion to Enhance Short-term Forecasting of Pumped-storage Hydroelectricity. PHM Society European Conference, 8(1), 11. https://doi.org/10.36001/phme.2024.v8i1.4129
Abstract 143 | PDF Downloads 99

##plugins.themes.bootstrap3.article.details##

Keywords

pumped-storage hydropower plant, hydroelectricity, data fusion, graph neural networks, state forecasting

References
Bronstein, M. M., Bruna, J., LeCun, Y., Szlam, A., & Vandergheynst, P. (2017, July). Geometric Deep Learning: Going beyond Euclidean data. IEEE Signal Processing Magazine, 34(4), 18–42. doi: 10.1109/MSP.2017.2693418

Cao, D., Li, J., Ma, H., & Tomizuka, M. (2021, May). Spectral Temporal Graph Neural Network for Trajectory Prediction. In 2021 IEEE International Conference on Robotics and Automation (ICRA) (pp. 1839–1845). doi: 10.1109/ICRA48506.2021.9561461

Cao, D., Wang, Y., Duan, J., Zhang, C., Zhu, X., Huang, C.,Zhang, Q. (2021, March). Spectral Temporal Graph Neural Network for Multivariate Time-series Forecasting (No. arXiv:2103.07719). arXiv. doi: 10.48550/arXiv.2103.07719

Chen, D., Liu, R., Hu, Q., & Ding, S. X. (2021). Interaction-Aware Graph Neural Networks for Fault Diagnosis of Complex Industrial Processes. IEEE Transactions on Neural Networks and Learning Systems, 1–14. doi: 10.1109/TNNLS.2021.3132376

Fatah, M. G. A., Claessens, B. J., & Schoukens, M. (2021, August). Integrating Power Grid Topology in Graph Neural Networks for Power Flow. Eindhoven University of Technology, 11.

Grigsby, J., Wang, Z., & Qi, Y. (2022, May). Long-Range Transformers for Dynamic Spatiotemporal Forecasting (No.arXiv:2109.12218). arXiv.

Guo, W., Che, L., Shahidehpour, M., & Wan, X. (2021, January). Machine-Learning based methods in short-term load forecasting. The Electricity Journal, 34(1), 106884. doi: 10.1016/j.tej.2020.106884

Halder, M. (2018, September). Power Demand Management – Smart Grid @ SBB. Innotrans.

Jeddi, A. B., & Shafieezadeh, A. (2021, December). A Physics-Informed Graph Attention-based Approach for Power Flow Analysis. In 2021 20th IEEE International Conference on Machine Learning and Applications (ICMLA) (pp. 1634–1640). doi: 10.1109/ICMLA52953.2021.00261

Kim, H., Jeong, J., & Kim, C. (2022, November). Daily Peak-Electricity-Demand Forecasting Based on Residual Long Short-Term Network. Mathematics, 10(23), 4486. doi: 10.3390/math10234486

Kundacina, O., Cosovic, M., & Vukobratovic, D. (2022, April). State Estimation in Electric Power Systems Leveraging Graph Neural Networks (No. arXiv:2201.04056). arXiv.

Li, S., Pandey, A., Hooi, B., Faloutsos, C., & Pileggi, L. (2022, September). Dynamic Graph-Based Anomaly Detection in the Electrical Grid. IEEE Transactions on Power Systems, 37(5), 3408–3422. doi: 10.1109/TPWRS.2021.3132852

Liao, W., Bak-Jensen, B., Radhakrishna Pillai, J., Wang, Y., & Wang, Y. (2022). A Review of Graph Neural Networks and Their Applications in Power Systems. Journal of Modern Power Systems and Clean Energy, 10(2), 345–360. doi: 10.35833/MPCE.2021.000058

Liao, W., Yang, D., Wang, Y., & Ren, X. (2021, March). Fault diagnosis of power transformers using graph convolutional network. CSEE Journal of Power and Energy Systems, 7(2), 241–249. doi: 10.17775/CSEEJPES.2020.04120

Lin, W., Wu, D., & Boulet, B. (2021, November). Spatial-Temporal Residential Short-Term Load Forecasting via Graph Neural Networks. IEEE Transactions on Smart Grid, 12(6), 5373–5384. doi: 10.1109/TSG.2021.3093515

Ma, Y., Liu, X., Shah, N., & Tang, J. (2023, July). Is Homophily a Necessity for Graph Neural Networks? (No.arXiv:2106.06134). arXiv.

Nauck, C., Lindner, M., Sch ̈urholt, K., Zhang, H., Schultz, P., Kurths, J., Hellmann, F. (2022, April). Predicting Basin Stability of Power Grids using Graph Neural Networks. New Journal of Physics, 24(4), 043041. doi: 10.1088/1367-2630/ac54c9

Nguyen, B., Vu, T., Nguyen, T.-T., Panwar, M., & Hovsapian, R. (2022, October). Spatial-Temporal Recurrent Graph Neural Networks for Fault Diagnostics in Power Distribution Systems (No. arXiv:2210.15177). arXiv. doi: 10.48550/arXiv.2210.15177

Nicolet, C., Greiveldinger, B., Herou, J. J., Kawkabani, B., Allenbach, P., Simond, J.-J., & Avellan, F. (2007, November). High-Order Modeling of Hydraulic Power Plant in Islanded Power Network. IEEE Transactions on Power Systems, 22(4), 1870–1880. doi: 10.1109/TPWRS.2007.907348

Pagnier, L., & Chertkov, M. (2021a, March). Embedding Power Flow into Machine Learning for Parameter and State Estimation (No. arXiv:2103.14251). arXiv. doi: 10.48550/arXiv.2103.14251

Pagnier, L., & Chertkov, M. (2021b, February). Physics-Informed Graphical Neural Network for Parameter & State Estimations in Power Systems (No. arXiv:2102.06349). arXiv. doi: 10.48550/arXiv.2102.06349

Ringsquandl, M., Sellami, H., Hildebrandt, M., Beyer, D., Henselmeyer, S., Weber, S., & Joblin, M. (2021, October). Power to the Relational Inductive Bias: Graph Neural Networks in Electrical Power Grids. In Proceedings of the 30th ACM International Conference on Information & Knowledge Management (pp. 1538–1547). doi: 10.1145/3459637.3482464

Simeunovi ́c, J., Schubnel, B., Alet, P.-J., & Carrillo, R. E. (2022, April). Spatio-Temporal Graph Neural Net-
works for Multi-Site PV Power Forecasting. IEEE Transactions on Sustainable Energy, 13(2), 1210–1220. doi: 10.1109/TSTE.2021.3125200

Simond, J.-J., Allenbach, P., Nicolet, C., & Avellan, F. (Eds.). (2006). Simulation tool linking hydroelectric production sites and electrical networks. Proceedings of 27th Int. Conf. on Electrical Machines, ICEM.

Wang, F., Chen, P., Zhen, Z., Yin, R., Cao, C., Zhang, Y., & Dui ́c, N. (2022, October). Dynamic spatio-temporal correlation and hierarchical directed graph structure based ultra-short-term wind farm cluster power forecasting method. Applied Energy, 323, 119579. doi: 10.1016/j.apenergy.2022.119579

Wu, H., Wang, M., Xu, Z., & Jia, Y. (2022, November). Graph Attention Enabled Convolutional Network for Distribution System Probabilistic Power Flow. IEEE Transactions on Industry Applications, 58(6), 7068–7078. doi: 10.1109/TIA.2022.3202159

Wu, T., Scaglione, A., & Arnold, D. (2022, September). Complex-Value Spatio-temporal Graph Convolutional Neural Networks and its Applications to Electric Power Systems AI (No. arXiv:2208.08485). arXiv. doi: 10.48550/arXiv.2208.08485

Wu, Y., Dai, H.-N., & Tang, H. (2022, June). Graph Neural Networks for Anomaly Detection in Industrial Internet of Things.IEEE Internet of Things Journal, 9(12), 9214–9231. doi: 10.1109/JIOT.2021.3094295

Zhao, M., & Fink, O. (2024, January). DyEdgeGAT: Dynamic Edge via Graph Attention for Early Fault Detection in IioT Systems (No. arXiv:2307.03761). arXiv. doi: 10.48550/arXiv.2307.03761

Zheng, J., Xu, C., Zhang, Z., & Li, X. (2017, March). Electric load forecasting in smart grids using Long-Short-Term-Memory based Recurrent Neural Network. In 2017 51st Annual Conference on Information Sciences and Systems (CISS) (pp.1–6). doi: 10.1109/CISS.2017.7926112

Zhu, J., Song, Y., Zhao, L., & Li, H. (2020, June). A3T-GCN: Attention Temporal Graph Convolutional Network for Traffic Forecasting (No. arXiv:2006.11583). arXiv. doi: 10.48550/arXiv.2006.11583

Zhu, Y., Gu, C., & Li, F. (2020, July). Cross-Domain Data Fusion On Distribution Network Voltage Estimation with D-S Evidence Theory. In 2020 International Joint Conference on Neural Networks (IJCNN) (pp. 1–6). doi: 10.1109/IJCNN48605.2020.9207414
Section
Technical Papers