Modelling of 100% Renewable Energy Systems in Integrated Assessment Models by multi-timeframe regression analysis

Abstract

Working on holistic approaches that aim to capture a wide range of knowledge, researchers are usually faced with phenomena characterized by different time and geographical scales. This is the case of energy systems and Integrated Assessment Models (IAMs). More specifically, the nature of the variable renewable energy supply (VRES) has traditionally posed a barrier to accurately capturing the effects inflicted by VRES in the energy system. This research provides a soft link between an energy system model running with an hourly time step, on the one hand, and a yearly-based IAM, on the other hand, by the implementation of an emulator. The proposal here presented is a bridge, based on different types of knowledge, which successfully allows the flow of information between time scales. Results achieve a 100% renewable energy system on a case of Bulgaria. After a brief literature review on the topic, the method is explained in detail, including some results between EnergyPLAN (energy system model) and MEDEAS (Integrated Assessment Model, IAM) for Bulgaria. Results show that the ability of assessment is notably increased from the previous MEDEAS version. Finally, both results and limitations of this method are discussed. The authors hope this article captures interest in the field of IAMs, especially those which address with energy transition studies.