Lehna, MalteMalteLehnaHolzhüter, Clara JulianeClara JulianeHolzhüterTomforde, SvenSvenTomfordeScholz, ChristophChristophScholz2024-09-232024-09-232024https://publica.fraunhofer.de/handle/publica/47548210.1016/j.segan.2024.101510With the growth of Renewable Energy (RE) generation, the operation of power grids has become increasingly complex. One solution could be automated grid operation, where Deep Reinforcement Learning (DRL) has repeatedly shown significant potential in Learning to Run a Power Network (L2RPN) challenges. However, most existing DRL algorithms have only considered individual actions at the substation level. In contrast, we propose a more holistic approach by proposing specific Target Topologies (TTs) as actions. These topologies are selected based on their robustness. In this paper, we present a search algorithm to find the TTs and upgrade our previously developed DRL agent CurriculumAgent (CAgent) to a novel topology agent. We compare our upgrade with the CAgent and significantly increase its L2RPN score by 10%. Further, we achieve a 25% better median survival time with our TTs included. Later analysis shows that almost all TTs are close to the base topology, explaining their robustness.enTopology optimizationElectricity gridsDeep reinforcement learningLearning to run a power networkProximal policy optimizationjournal article