Congestion management in power grids using multi-agent systems and particle swarm optimization

Congestion management is a critical challenge in modern power grids, especially as grids become more complex with the increasing integration of renewable energy sources. Efficiently alleviating power grid congestion requires dynamic, flexible, and decentralized solutions that can adapt to rapidly changing grid conditions without incurring excessive computational costs. In this context, a novel approach is introduced using a multi-agent system (MAS) to address power grid congestion. The proposed system consists of generator agents, coordinated by a manager agent, that optimize their outputs through Particle Swarm Optimization (PSO). The IEEE 30-bus system is employed for simulation, demonstrating significant improvements in Available Transfer Capability (ATC), with over 70% enhancement observed after re-dispatch, and a reduction in generation cost by more than 25%. By utilizing the decentralized nature of MAS, the approach enables scalable and adaptable congestion management, where each generator agent operates independently and contributes to a global optimization objective. The PSO technique facilitates rapid convergence (in 25 iterations, 15.4s) and effective navigation of complex solution spaces, making the method computationally efficient and suitable for real-time applications. The study highlights the effectiveness and adaptability of combining MAS with PSO for managing grid congestion in diverse operational scenarios.