Usage of GAMS-Based Digital Twins and Clustering to Improve Energetic Systems Control

With the increasing constraints on energy and resource markets non-decreasing trend in demand, need for relevant clean generation storage solutions is growing gradually reaching individual home. However, small-scale still an expensive investment 2022 risk/reward ratio not yet attractive enough homeowners. One solution homeowners to store excess individually but produce hydrogen mutual use. In this paper, a collective production of daily filling bus considered. Following our previous work subject, investigation consists finding optimal buy/sell rule grid, use with additional objective: mobility. The dominant technique community reinforcement learning, which however difficult when learning data limited, as study. We chose less data-intensive technically well-documented approach. Our results show that rulebooks, different more interesting than usual robust rule, exist can be cost-effective. some cases, they even it worth punctually missing H2 requirement exchange higher economic performance. require fine-tuning deteriorate system