Computing energy system transformation pathways with high spatial and temporal resolution

Abstract

Decarbonisation of the energy system requires substantial efforts along all sectors. This necessitates the allocation of carbon budgets among both energy sectors as well as the years along the transformation pathway. Additionally state-of-the-art energy system models need to represent the increasing complexity of the energy system from decentralized generation, increasing electrification in the heating and transport sector and demand for flexibility options. The downside of this additional complexity shows itself in the numerical difficulties making these models computational intractable for most solvers. The BEAM-ME project addresses this need for efficient solution strategies for complex energy system models. This goal is achieved by bringing energy system model to high performance computers, a process that requires interdisciplinary expertise from the fields of energy systems analysis, mathematics, operations research, and informatics. By using a specialized solver adapted to the specific problem structure of energy system models the parallel computation power can significantly speed up the optimization process. This talk provides an overview in preparing the energy system model REMix to utilize the parallel interior-point solver PIPS-IPM. The adjusted model is parametrized considering both transition pathways and high spatial as well as temporal resolution. First insights in the resulting problem structure are discussed and performance improvements are presented.