artificial intelligence; machine learning; deep neural networks; random forests; security constrained optimal power flow; reproducibiliy; proxies; robustness
Abstract :
[en] In this paper, we focus on the robustness of machine learning based proxies used to speed up, alone or jointly with state-of-the-art mathematical optimization methods, optimal power flow and security-constrained optimal power flow calculations. On data sets for the Nordic32 alternative current security-constrained optimal power flow benchmark, we evaluate the robustness of proxies with respect to load distribution, power factors, on-line generators and network topology, and generator costs. We show that simplified random load sampling procedures that are used in most published academic studies, are insufficient to yield robust machine learnt proxies, and consequently limit their usefulness in the real world. Based on these results, we formulate recommendations for future research.
Disciplines :
Electrical & electronics engineering
Author, co-author :
Popli, Nipun ; Université de Liège - ULiège > Montefiore Institute of Electrical Engineering and Computer Science
Davoodi, Elnaz; LIST - Luxembourg Institute of Science and Technology [LU] > ERIN
Capitanescu, Florin; LIST - Luxembourg Institute of Science and Technology [LU] > ERIN
Wehenkel, Louis ; Université de Liège - ULiège > Département d'électricité, électronique et informatique (Institut Montefiore) > Méthodes stochastiques
Language :
English
Title :
On the robustness of machine-learnt proxies for security constrained optimal power flow solvers
Publication date :
28 December 2024
Journal title :
Sustainable Energy, Grids and Networks
ISSN :
2352-4677
Publisher :
Elsevier, United Kingdom
Volume :
37
Pages :
101265
Peer reviewed :
Peer Reviewed verified by ORBi
Tags :
CÉCI : Consortium des Équipements de Calcul Intensif
Capitanescu, F., Martinez Ramos, J., Panciatici, P., Kirschen, D., Marano Marcolini, A., Platbrood, L., Wehenkel, L., State-of-the-art, challenges, and future trends in security constrained optimal power flow. Electr. Power Syst. Res. 81:8 (2011), 1731–1741.
Velloso, A., Van Hentenryck, P., Combining deep learning and optimization for preventive security-constrained DC optimal power flow. IEEE Trans. Power Syst. 36:4 (2021), 3618–3628.
Chen, W., Park, S., Tanneau, M., Van Hentenryck, P., Learning optimization proxies for large-scale security-constrained economic dispatch. Electr. Power Syst. Res., 213, 2022, 108566.
Liu, S., Guo, Y., Tang, W., Sun, H., Huang, W., Hou, J., Varying condition SCOPF optimization based on deep learning and knowledge graph. IEEE Trans. Power Syst., 2022, 1–12.
X. Pan, T. Zhao, M. Chen, DeepOPF: Deep Neural Network for DC Optimal Power Flow, in: 2019 IEEE International Conference on Communications, Control, and Computing Technologies for Smart Grids, SmartGridComm, 2019, pp. 1–6.
K. Baker, Learning Warm-Start Points For Ac Optimal Power Flow, in: 2019 IEEE 29th International Workshop on Machine Learning for Signal Processing, MLSP, 2019, pp. 1–6.
Misra, S., Roald, L., Ng, Y., Learning for constrained optimization: Identifying optimal active constraint sets. INFORMS J. Comput. 34 (2022), 463–480.
Zhou, M., Chen, M., Low, S.H., DeepOPF-FT: One deep neural network for multiple AC-OPF problems with flexible topology. IEEE Trans. Power Syst., 2022, 1–4.
Chen, L., Tate, J.E., Hot-starting the AC power flow with convolutional neural networks. 2020 arXiv.
W. Dong, Z. Xie, G. Kestor, D. Li, Smart-PGSim: Using Neural Network to Accelerate AC-OPF Power Grid Simulation, in: SC20: International Conference for High Performance Computing, Networking, Storage and Analysis, 2020, pp. 1–15.
Yan, Z., Xu, Y., A hybrid data-driven method for fast solution of security-constrained optimal power flow. IEEE Trans. Power Syst. 37:6 (2022), 4365–4374.
Robson, A., Jamei, M., Ududec, C., Mones, L., Learning an optimally reduced formulation of OPF through meta-optimization. 2019 arXiv.
Chatzos, M., Mak, T.W.K., Hentenryck, P.V., Spatial network decomposition for fast and scalable AC-OPF learning. IEEE Trans. Power Syst. 37:4 (2022), 2601–2612.
Guha, N., Wang, Z., Wytock, M., Majumdar, A., Machine learning for AC optimal power flow. 2019.
Nellikkath, R., Chatzivasileiadis, S., Physics-informed neural networks for AC optimal power flow. Electr. Power Syst. Res., 212, 2022, 108412.
Yan, Z., Xu, Y., Real-time optimal power flow: A Lagrangian based deep reinforcement learning approach. IEEE Trans. Power Syst. 35:4 (2020), 3270–3273.
Baker, K., Emulating AC OPF solvers with neural networks. IEEE Trans. Power Syst. 37:6 (2022), 4950–4953.
Capitanescu, F., Suppressing ineffective control actions in optimal power flow problems. IET Gener. Transm. Distrib. 14:13 (2020), 2520–2527.
Van Cutsem, T., et al. Test systems for voltage stability studies. IEEE Trans. Power Syst. 35:5 (2020), 4078–4087.
Davoodi, E., Capitanescu, F., Wehenkel, L., A methodology to evaluate reactive power reserves scarcity during the energy transition. IEEE Trans. Power Syst. 38:5 (2023), 4355–4368.
Louppe, G., Lectures for INFO8010 - Deep Learning. 2022, Université de Liège accessed: Spring 2022.
I.J. Goodfellow, Y. Bengio, A. Courville, Deep Learning, MIT Press.
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., et al. Scikit-learn: Machine learning in python. J. Mach. Learn. Res. 12 (2011), 2825–2830.
Cayton, L., Algorithms for Manifold Learning. 2005, University of California, San Diego accessed: December 2022.
Fefferman, C., Mitter, S., Narayanan, H., Testing the manifold hypothesis. J. Amer. Math. Soc. 29:4 (2016), 983–1049.
Allen, M., Poggiali, D., Whitaker, K., Marshall, T., van Langen, J., Kievit, R., Raincloud plots: a multi-platform tool for robust data visualization. Wellcome Open Res., 4(63), 2021.
RainCloud plot script. 2020.
Capitanescu, F., Evaluating reactive power reserves scarcity during the energy transition toward 100% renewable supply. Electr. Power Syst. Res., 190, 2021, 106672.
Popli, N., Davoodi, E., Capitanescu, F., Wehenkel, L., Machine learning based binding contingency pre-selection for AC-PSCOPF calculations. IEEE Trans. Power Syst., 2023, 1–4, 10.1109/TPWRS.2023.3338971.
B. Donon, B. Donnot, I. Guyon, A. Marot, Graph Neural Solver for Power Systems, in: International Joint Conference on Neural Networks, IJCNN, 2019, pp. 1–8.
Donon, B., Deep Statistical Solvers & Power Systems Applications. (Ph.D. dissertation), 2022, Université Paris-Saclay.
Müller, S., Development of Nordic 32 System Model and Performance Analysis Based on Real Operation Statistics. 2019, KTH, School of Electrical Engineering and Computer Science (EECS).
Alsac, O., Stott, B., Optimal load flow with steady-state security. IEEE Trans. Power Appar. Syst. PAS-93:3 (1974), 745–751.
Monticelli, A., Pereira, M.V.F., Granville, S., Security-constrained optimal power flow with post-contingency corrective rescheduling. IEEE Trans. Power Syst. 2:1 (1987), 175–180.
