Modelling and Control of Hydrogen Electrolyzer Systems for Frequency Regulation of Low-Inertia Grids

QC 20260508

Abstract

Hydrogen has the potential to become a key technology for the green energy transition. Hydrogen electrolyzers are employed to produce hydrogen from the decomposition of water using electric power provided by Power Electronics Interfaces (PEIs) connected to the grid. Hydrogen storage between the electrolyzer and the downstream application decouples electric consumption from hydrogen demand. Therefore, large-scale electrolyzers have the flexibility to provide balancing services such as frequency regulation to low-inertia grids with a high share of Variable Renewable Energy Sources (VRES).

The provision of balancing services implies that electrolyzers need to operate in a wide operational range. At the electrolyzer side of the PEI, the current contains a main Direct Current (DC) component and a ripple. At partial load, the current ripple fed to the electrolyzer can significantly increase depending on the PEI topology. Therefore, an evaluation approach is proposed to assess the impact of power converter topology on the electrolyzer performance. The input to the evaluation approach are realistic large-scale current waveforms for industrial applications with a broad range of thyristor and transistor-based power converter topologies. These waveforms are scaled down to perform the evaluation on a small-scale setup. Simulations are performed with a novel electrolyzer stack model, and simulation results are validated with experimental results.

At the grid side of the PEI, the electrolyzer systems need to deliver a fast response for the provision of frequency regulation. The speed of response of electrolyzer systems depends on the electrolyzer technology employed. Alkaline Electrolyzers (AELs) are the most mature and cost-effective technology, but their response is slower compared to Proton Exchange Membrane Electrolyzers (PEMELs). Therefore, both technologies can be combined in AEL/PEMEL Hybrid Hydrogen Electrolyzer Systems (HHESs) to make the most of the speed of the PEMELs, and reduce costs thanks to AELs. To achieve this, a decentralized dynamic power sharing control strategies is proposed, where the slow dynamics of the AEL are respected while the PEMEL compensates with a fast response. Compared to a centralized control approach, the decentralized approach enhances the reliability and scalability of the system. This way there is no risk of single point of failure due to the only use of local communications, and no modification of existing control architectures is required when the system is expanded. The control strategies are simulated in large-scale HHESs, and validated in a small-scale experimental setup.

The response speed of electrolyzer systems can be further improved by adding other fast-acting devices such as Energy Storage Systems (ESSs). A decentralized control strategy is proposed for electrolyzer/ESS Hybrid Hydrogen Systems (HHSs), making the most of each technology. The AEL dynamics are respected thanks to the fast response of ESSs, and PEMELs can be added to provide an intermediate response and reduce the size of the ESS. Furthermore, apart from the increased reliability and scalability compared to a centralized approach, the proposed decentralized approach includes DC bus voltage and ESS State of Charge (SoC) recovery to minimize the use of the ESS. Large-scale simulations are performed, as well as validation with experimental results.

Due to MW or even GW level of upcoming large-scale electrolyzer systems, Transmission System Operators (TSOs) are implementing Fault-Ride Through (FRT) requirements for the grid integration of these systems. To evaluate the FRT performance of electrolyzer systems, a novel control strategy is proposed considering operational limitations of electrolyzers such partial load and ramp rate limits. Simulations are performed for a large-scale system under different faulty conditions, and the use of a supercapacitor to enhance FRT performance is explored.

Overall, the work conducted in this thesis provides evaluation methods and models for the minimization of electrolyzer performance losses due to current ripple of power converter topologies, as well as for grid integration with FRT requirements. Furthermore, it provides control strategies for enhancement of response speed for frequency regulation using AEL/PEMEL HHESs, or electrolyzer/ESS HHSs. Therefore, the conducted work provides a comprehensive overview for the provision of frequency regulation with hydrogen electrolyzer systems in low-inertia grids.

Link to DiVA