Using automated algorithm configuration to improve the optimization of decentralized energy systems modeled as large-scale, two-stage stochastic programs

The optimization of decentralized energy systems is an important practical problem that can be modeled using stochastic programs and solved via their large-scale, deterministic equivalent formulations. Unfortunately, using this approach, even when leveraging a high degree of parallelism on large high-performance computing (HPC) systems, finding close-to-optimal solutions still requires long computation. In this work, we present a procedure to reduce this computational effort substantially, using a stateof-the-art automated algorithm configuration method. We apply this procedure to a well-known example of a residential quarter with photovoltaic systems and storages, modeled as a two-stage stochastic mixed-integer linear program (MILP). We demonstrate substantially reduced computing time and costs of up to 50% achieved by our procedure. Our methodology can be applied to other, similarly-modeled energy
systems.