Developing and Controlling a Hybrid Renewable Energy System with Solar PV and Modular Alkaline Electrolysis

Abstract

Renewable energy resources like wind and solar have the potential to revolutionize our energy infrastructure and enable a decarbonized society. The intermittent nature of renewables poses a challenge to ensuring a consistent and reliable electricity supply. Hydrogen technology is emerging as a promising solution for stabilizing renewable energy systems. There is still significant technological development required to cost-effectively integrate hydrogen with renewable energy assets. Conventional PV solar installations require power electronics for their operation, like charge controllers and inverters. Not only are these power electronic components costly and in high demand, but they also degrade faster than PV solar panels. Removing power electronic components could significantly lower the investment costs associated with a PV solar park.

This work focuses on how PV solar panels can be directly coupled to a modular alkaline electrolyzer, without grid-based buffering or the use of an inverter. Literature research revealed that hydrogen technology has seen little exploration in on-grid Hybrid Renewable Energy Systems (HRES) and no exploration in off-grid HRES. To appropriately investigate whether a directly-coupled HRES would be technically feasible, a megawatt-scale system was modelled and simulated. All elements of the HRES were modelled, duly accounting for physical limits and constraints. Components were sized and configured to complement one another, optimizing for maximum hydrogen production. To experimentally verify the validity of the proposed HRES, a 5 kW pilot system was constructed. To control the HRES, a new algorithm was developed using the Incremental Conductance maximum power point tracking algorithm as a basis. Within the new Maximum Hydrogen Production (MHP) algorithm, the step sizing was discretized and a variable step size was implemented which can be applied to any target slope. This allows for the system to target operational points which optimize hydrogen yield instead of electricity yield. Furthermore, the addition of tracking bias helped adjust for the asymmetric nature of the interaction between electrolyzer stacks and the PV solar park.

Simulation results in The Netherlands demonstrated that the feasibility of the HRES is dependent on the configuration of the PV solar park and on the number of electrolyzer stacks in the system. Compared to industrial and research benchmarks, the proposed HRES increased hydrogen production by 14.9% and 4.2%, respectively. Dynamic 'm-tracking' of the MHP algorithm goal increased hydrogen production by 0.8% in months of high irradiance. Months with a lower average irradiance experienced an artefact in the MHP algorithm, resulting in prolonged periods of zero power output. An experimental setup confirmed the simulation results, showing that it is possible to control a system of PV solar panels directly coupled to a modular alkaline electrolyzer. Experimental results revealed the need for moving average filtering to prevent fluctuations due to changing conditions of the electrolyzer and the weather from causing poor algorithm tracking ability. The low performance of the experimental setup can be attributed to a low iteration and measuring frequency, which increase the likelihood of a tracking error due to rapidly changing operating conditions. Economic analysis of the proposed HRES yielded an LCOH of €3.44 per kg, 20% and 13% lower than industrial and research benchmarks, respectively. Therefore, an HRES featuring PV solar and modular alkaline electrolysis is technologically and economically viable without the use of charge controllers and inverters.