Next generation energy modelling - benefits of applying parallel optimization and high performance computing

Abstract

Quantitative energy system modelling based on primarily linear mathematical optimization provide a basis to answer political, technical, and economic questions regarding the future development of electricity systems. In order to capture the complexity of future decentralized electricity system, models are required to become increasingly complex. This presentation outlines how high performance computing (HPC) can be applied to energy system modelling and describes challenges and benefit that arise when complex energy models based on linear and non-linear algorithms are applied to parallel computing. The presentation draws on results from the project BEAM-ME funded by the German Federal Ministry for Economic Affairs and Energy. The research project aims at improving computational performance of energy system models. Within this project the consortium of researchers from different research fields (system analysis, mathematics, operations research and informatics) develop new strategies to increase computational performance of energy system models and to apply energy system models to high performance computing. This presentation presents intermediate results from a benchmark analysis when applying the energy system model REMix to HPC. The presentation concludes with a discussion on future challenges for energy system modelling and how HPC can enable energy modelers to improve scenarios, market design and policy advice.