Multi-Time-Step Voltage Control in Transmission Grids Using Genetic Algorithms

Traditional approaches for transmission system voltage control (TSVC) rely heavily on manual action or mathematical optimization methods, which face challenges when addressing the combinatorial nature of TSVC at large scales. In this paper, we explore the potential of genetic algorithms (GAs) as a meta-heuristic approach for optimizing voltage control over multiple time steps. Our study evaluates GAs on three different grid models, including a real transmission grid, and benchmarks their performance against brute-force and greedy algorithms where feasible. Furthermore, we incorporate topological control actions into the optimization process, filling a gap in existing research. The results indicate that GAs offer a flexible approach to TSVC, effectively handling discrete control decisions over multiple time steps. Their scalability and adaptability make them a promising alternative to classical methods. This study highlights the potential of GAs for automated voltage management in increasingly complex power systems.