148610 20260217205523.0 doi 10.1109/TPWRS.2024.3394371 sideral 141954 ART-2024-141954 eng Lopez-Garcia, Tania B. Optimal Power Flow With Physics-Informed Typed Graph Neural Networks 2024 Access copy available to the general public Unrestricted This work describes a new way to solve the optimal power flow problem applying typed graph neural networks. Typed graph neural networks allow the representation of different elements of transmission systems with different types of nodes, thus adding accuracy and interpretability to the solutions obtained, in comparison to results obtained with conventional feed-forward neural network models. The proposed graph neural network architecture is trained without the need of training data, through a physics informed loss function which incorporates not only the optimization objective, but also operational constraints of the physical system. Results are comparable with those obtained with the interior point method, and it is shown that the calculation time is greatly reduced. info:eu-repo/grantAgreement/ES/MICINN/PID2019-104711RB-100 info:eu-repo/semantics/openAccess by https://creativecommons.org/licenses/by/4.0/deed.es 7.2 2024 ENGINEERING, ELECTRICAL & ELECTRONIC 39 / 368 = 0.106 2024 Q1 T1 3.629 2024 Energy Engineering and Power Technology 2024 Q1 Electrical and Electronic Engineering 2024 Q1 15.9 2024 info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion Domínguez-Navarro, José Antonio Universidad de Zaragoza (orcid)0000-0002-4770-0069 5009 535 Universidad de Zaragoza Dpto. Ingeniería Eléctrica Área Ingeniería Eléctrica 40, 1 (2024), 381-393 IEEE trans. power syst. IEEE TRANSACTIONS ON POWER SYSTEMS 0885-8950 3075196 http://zaguan.unizar.es/record/148610/files/texto_completo.pdf Versión publicada 3509408 http://zaguan.unizar.es/record/148610/files/texto_completo.jpg?subformat=icon icon Versión publicada oai:zaguan.unizar.es:148610 articulos driver 2026-02-17-20:29:19 ARTICLE