Run-time reduction for linear optimizing energy system models

Abstract

The large numbers of variables and constraints and interdependencies in energy system models makes these large and complex. In order to keep model running times in feasible limits, the resolution in the dimensions time, space and technologies frequently has to be adapted to the computing infrastructure capacity. The error that the applied simplifications bring about is mostly unknown. In the BEAM-ME project (funded by the German Federal Ministry for Economic Affairs and Energy) a consortium of energy system modellers, mathematicians, computer scientists and software developers has implemented and assessed methods for run time reduction. In this talk the methods and assessments are presented, a short introduction on running energy system models on high performance computers is given and the open access best practice guide that is a result of the BEAM-ME project is introduced.