Master's thesis and internship : Development and Improvement of the power system module within the European Integrated Assessment Model MEDEAS

Abstract

This master's thesis is situated within the broader framework of global climate change and the European Green Deal, which sets the ambitious goal of achieving net-zero greenhouse gas (GHG) emissions in Europe by 2050. In particular, this research focuses on the Integrated Assessment Model (IAM) MEDEAS. This model aims to address the challenges of the energy transition within the European Union (EU) by providing comprehensive assessments of the potential impacts and mitigation strategies associated with various policy measures.

The master's thesis aims to propose a new version of MEDEAS which incorporates a machine learning-based surrogate model (SM) to improve the predictive potential of the IAM, particularly in simulating the European electrical power grid's curtailment and load shedding dynamics. This surrogate model was developed in previous works and is an efficient and flexible tool mirroring Dispa-SET unit commitment and economic dispatch model.

The other key advancements include the integration of additional data from PyPSA-EUR, enabling both the integration of the SM and new investment assessments of renewable energy sources (RES), grid reinforcement, and storage installations. Additionally, new feedback mechanisms inspired by PID control theory simulate instantaneous societal responses aimed at reducing energy curtailment and load shedding.

A comparative analysis against the previous MEDEAS version and a practical case study demonstrate the enhanced model's utility in exploring new energy scenarios and providing meaningful insights for policymakers.