Modelling and Simulation of Hydrogen Electrolyzers for Power System Applications

Sammanfattning: The rapid increase in the share of intermittent renewables in the energy mix is exacerbating the complexity of the grid stability. To facilitate the inclusion of these green technologies and achieve the climate targets set by the EU, the generation and storage of hydrogen through electrolysis may be crucial. Moreover, the hydrogen can later be used as a replacement for polluting components in several industries, helping to reduce the CO2 emissions. In order to be able to couple the hydrogen electrolyzers with the grid, accurate models need to be developed, where all the physical characteristics and the empirical behavior of the cells are encapsuled. Hence, this thesis will explore the different models available in the literature for the three main technologies of electrolyzers: proton exchange membrane (PEMEC), alkaline (AEC) and solid oxide cells (SOEC). Even though the alkaline is the most mature and established technology, the aforementioned exhibit important features, such as faster response times or lower electricity requirements, that may enable them to overtake alkaline electrolyzers in terms of presence in the grid support services. The models are reproduced and compared using MATLAB/Simulink so that a clear overview of the current state of the electrolyzers and the ground for their expansion is prepared. Three core modalities are examined: electrochemical equations, electrical analogues and thermal submodels. The equations are found to reliably replicate the behavior obtained in the experiments. Meanwhile, the thermal influence is usually often disregarded due to the significantly slower rate at which temperature changes occur compared to the response times of the electrical signals. So far, SOEC are still in development so a clear electrical modelling is yet not available, however, for the already commercial PEMEC and AEC, several different equivalent circuits can be found. The models show good agreement with the experimental data, being the PEMEC faster in response and having a higher degree of degradation rate.