An Open Source Energy Modelling Framework Comparison of OSeMOSYS and oemof

Sammanfattning: Well-founded research is needed to gain sound insights into the relationship between energy and the environment in order to ensure a basis for responsible policy-making in the face of a climate crisis. Transparency in code and data is a prerequisite for reproducible research in energy modelling and its quality assurance and is embedded in the philosophy of open source energy modelling frameworks (OSEMFs). While the focus was on improving the quality of input data and comparing models to compare results, there has been a lack of research on the effects of the structural assumptions across different frameworks. This thesis analyses the structural differences and structural assumptions between the OSEMFs OSeMOSYS and oemof. The methodology applies a wholly harmonised experimental setup to allocate output differences to diverging contextual assumptions in the underlying different mathematical equations among the OSEMFs. Diverging results between OSEMFs based on different structural assumptions occurred in this thesis and traced back to their origin. OSeMOSYS's and oemof's underlying energy balance equations lead to the same results though, when diverging structural assumptions are removed. Already a single undocumented opaque assumption in the underlying mathematical equations among frameworks can have a noticeable influence on the optimisation results. The documentation of OSEMFs is still not comprehensive enough to facilitate investigations of this kind efficiently, and the documentation of assumptions embedded in OSEMFs’ mathematical equations needs to be more detailed. The thesis could identify opportunities to enhance the performance of OSeMOSYS and highlights the importance of correct framework documentations to ensure their usability in practice.