An optimal solution for reactive power re-dispatch agency using open-source restructuring learning semantic networks

This paper presents a novel model for optimal reactive power re-dispatch in interconnected power systems using open-source restructuring learning semantic networks. The proposed framework addresses key operational challenges, such as voltage regulation and transmission losses, by enabling dynamic, knowledge-driven coordination among network agents. Designed with applicability in developing regions, the model is evaluated in the Ecuadorian transmission system, where long east-to-west corridors require resilient and efficient dispatch solutions. An Optimal Power Flow (OPF) formulation is used to minimize losses while complying with national voltage standards. Unlike conventional methods, our approach incorporates adaptive semantic structures that restructure agent interactions based on system conditions, supporting real-time reasoning and interoperability. The results of two case studies (Milagro-Molino and Santa Rosa-Totoras) confirm improved voltage profiles, significant loss reductions, and computational efficiency. By aligning with emerging themes in intelligent power systems, such as semantic modelling, decentralized decision-making, and open replicable tools, this work contributes to IJEPES’s ongoing discourse on resilient and scalable energy solutions. • Novel reactive power re-dispatch model using semantic learning networks. • Model applied to Ecuador’s grid, addressing regulatory and geographical challenges. • OPF and learning networks cut losses and improve voltage profiles. • Open-source tools ensure replicability, cost-efficiency, and accessibility. • Scalable solution for resource-constrained power systems in developing nations.