Stochastic and risk averse optimization for renewable hydrogen production

Details

Supervisors

Glineur, François ; de Maere d'Aertrycke, Gauthier ; Binter, Florian

Faculty

Ecole polytechnique de Louvain

Degree label

Master [120] : ingénieur civil en mathématiques appliquées, à finalité spécialisée

Abstract

Hydrogen is increasingly being put forward as a key solution for decarbonizing sectors such as aviation or maritime transport, where electrification would be impossible. Hydrogen would be used to produce synthetic fuels directly used in airplane and boat engines. In a recent delegated act, the European Commission defined rules for certification of Renewable Fuels of Non Biological Origin (RFNBOs). In this master thesis, we investigate hydrogen production in the framework of RFNBO certification. Specifically, hydrogen must be produced through electrolysis with renewable electricity and electrolysers must produce at the same time as renewables. The major challenge for modelling renewable hydrogen production comes from stochasticity of renewable electricity production. In this master thesis, we model the renewable hydrogen production problem as a two-stage stochastic program and analyze the impact of variability on profit and investment decisions. Additionally, we test risk averse approaches to the problem and compare results with optimizing profit expectation.