This paper analyzes the simulation of a photovoltaic and battery connected system for electrolysis hydrogen production and storage supplied hydrogen fuel cell power. All the systems stated are connected to evaluate the efficiency to power a continuously running fuel cell. The system is designed to be disconnected from the power grid and provide a continual power source for a given load. The three-day simulation evaluates how the variables limit the system performance and efficiency.
Research Article
Simulated Photovoltaic and Battery Connected Microgrid for Electrolysis Hydrogen Production & Fuel Cell Power
https://doi.org/10.21203/rs.3.rs-3767499/v1
This work is licensed under a CC BY 4.0 License
Version 1
posted
You are reading this latest preprint version
Hydrogen is a quickly growing energy source in the world. The International Energy Agency drafted a report from their analysis citing the global demand on hydrogen. They found that Hydrogen demand reached 94 million tons in 2021, which is equal to about 2.5% of global final energy consumption. Most of the demand was met by producing hydrogen through the use of fossil fuels. In 2021, water electrolysis accounted for only around 0.1% of global hydrogen production; however, the electrolysis market is continuing to increase [2]. With the growing hydrogen market, there is also growing hydrogen fuel cell market.
This analysis aims to evaluate the practicality and efficiency of using a microgrid photovoltaics (PV) and battery system to produce hydrogen and power a fuel cell. The benefit to creating a hydrogen fueled system to produce power is that the output of a system is limited by the quantity of hydrogen produced over a period of time, storage size available for the produced hydrogen, and the peak output of the fuel cell. This system provides a way to manage power delivery similar to that of a battery during off peak solar irradiance hours.
There are a few benefits to creating a fuel cell powered microgrid compared to other energy sources. Fuel cells generally have more than twice the energy conversion efficiency of traditional combustion energies. Additionally, fuel cells provide an improved power quality and reliability source. As you’ll see the following sections, solar power can be intermittent and uncertain, but having the storage capacity to generate a hydrogen fuel source can aid in providing a continuous and reliable power supply. Furthermore, fuel cells are generally considered more flexible with regards to scalability [10].
Found in Figure 1 is a diagram of the evaluated model designed for the three-day simulation. The model evaluated incorporated the use of the MathWorks “Green Hydrogen Microgrid ©” [1]. Changes that were made to the MathWorks model include input parameters of the PV system solar irradiance profile, battery system power and power controller, electrolyzer gas flow components, and the addition of the fuel cell, fuel limiter, and system fuel scopes. The following sections will provide an overview of the design of each component may impact the performance of the other connected components in series.
The designed model evaluates the power variability that would exist with varying solar power and a charging and discharging battery-controlled system while supplying power to the electrolyzer. The simulation provides solar and battery power over a 72-hour time period. The solar irradiance of the PV system is similar for the first and second day with a peak power output of 36 kilowatts (kW) and 34 kW respectively. The solar irradiance on the third day is modeled after a cloudy day where the solar power is slightly over half the power received on the previous days. Figure 2 is the layout of the Simulink PV system.
The existing battery modeled in the simulation is a 672 kWh system that starts in the simulation at a State of Charge (SoC) of 100%. The designed battery continually provides power to the electrolyzer until it reaches a SoC of 35%. Once the battery reaches a SoC of 35%, it will utilize some of the power from the PV system to charge the battery until the battery reaches a SoC of 95%. Figure 3 is the layout of the Simulink PV system.
Both the designed battery and PV system provide power the cathode and anode of the electrolyzer to separate the hydrogen from the supplied water source, and provide fuel to the electrolyzer. Shown in Figure 4 is a graph of the power delivered over a 72-hour period from the battery and PV system and power absorbed by the electrolyzer. After hour 42 is when the battery system shuts off due to reaching the 35% SoC, and the power delivered to the electrolyzer is strictly from the PV system.
Figure 5 displays the SoC of the battery system over time. Due to the low irradiance of the PV system at certain times, power is not delivered to the battery system during the flat periods indicated on the graph. The battery begins to charge when the solar irradiance is greater than 50 amps, which is enough to split the power between that delivered to the electrolyzer and battery system.
Electrolysis is the process of electrochemical decomposition of water by using energy with two electrodes immersed in an electrolyte. There are many factors which impact the production rate of hydrogen, which are not limited by, but include, the voltage applied to the electrodes, total resistance of the electrolyzer system (electrodes and cross-sectional transport area), pH of the water, water temperature and water enthalpy [7]. Figure 6 is the layout of the Simulink electrolyzer and gas flow system.
The power absorbed by the electrolyzer previously shown in Figure 4 resulted in the hydrogen output in Figure 7 below. There was 12.1 kilograms (kg) of hydrogen produced at the end of the 72-hour simulation. The storage level shown in the same figure displays the excess mass of hydrogen produced, stored, and later utilized by the fuel cell which is further explained in the following sections.
The factors for calculating the electrolysis efficiency depend on the simulation input parameters. The system modeled in this simulation achieved a constant result of producing 1 kg of hydrogen per 60 kilowatt-hours (kWh) of energy supplied. 1 kg of hydrogen is capable of providing between 33.3 – 39.4 kWh of electrical energy depending mount of energy required to vaporize a liquid fuel into a gaseous fuel, which would make the system 55.5% - 65.7% efficient [5].
Once the electrolyzer produces hydrogen, it is sent to the hydrogen storage tank and subsequent flow limiter that supplies the fuel source to the fuel cell. Each hydrogen fuel cell has a peak power limit, which restricts the supplied hydrogen fuel to the fuel cell. Any hydrogen produced in excess of the fuel cell limitations remain in the storage container for later use. Storing excess hydrogen does come at a cost. Because the standard condition of hydrogen is a gas state, the gas is compressed to a smaller area, but compressing gas comes at a high investment cost. As the storage pressures are limited, so are the achievable hydrogen storage densities: at 100 bar and 20 °C, the density of hydrogen gas is approximately 7.8 kg per cubic meter [8]. Therefore, costs may not only be associated with the other power controlled and producing components, but also storing and compressing any excess hydrogen produced.
The hydrogen fuel cell parameters utilized in the simulation are modeled from a 6 kW proton exchange membrane fuel cell datasheet [6]. The datasheet outlines the fuel cell stack efficiency can achieve 55% under nominal conditions. The designed fuel cell replicated the nominal conditions outlined on the datasheet aside from the air flow rate. The flow rate provided to the fuel cell is dependent upon the output of the electrolyzer. One the key parameters of the fuel cell output is the V-I and I-P polarization curve (shown in Figure 8), which identifies the nominal voltage, current, and power of the fuel cell. The nominal rates also aid in identifying other nominal operating parameters such as the supplied flow rate and number of cells.
Identifying and maintaining the nominal conditions of a fuel is critical in optimizing fuel cell efficiency. When optimal conditions are not met then fuel cell efficiency can drastically impact expected performance [6]. Equation 1 and Equation 2 below provide examples of how nominal conditions impact cell performance.
Figure 9 is the layout of the fuel cell system utilized in the model. When the electrolyzer produces hydrogen over a specified limit, then the excess rate remains in storage for later use when the battery is charging. Additionally, the cell efficiency is calculated when using the lower heating value of 33.3 kWh of electrical energy per kg. Shown in Figure 10 is the instantaneous fuel cell power delivered over time and Figure 10 displays the instantaneous fuel cell efficiency over time. The varying flow rates show that when an instantaneous and variable flow rate occurs from the rapidly changing power supplied to the electrolyzer, then fuel cell performance is impacted. The lower rates show the fuel cell operating with the least efficiency, but rates near the maximum operating point have the highest efficiency.
The designed system sizing should revolve around the load demand. With the current design, the fuel cell is capable of achieving over 7 kW of power but that power fluctuates over time. Fuel cells can provide energy dependent upon the input fuel flow rate. Equation 3 provides the ideal capacity required to store the hydrogen [9].
Shown in the simulation is the power variability that can exist with a PV system. Therefore, the power rating of the electrolyzer must be dependent upon when power supplied to the electrolyzer will be at the lowest daily average point. The expected lowest daily average occurs when power is supplied to the battery and electrolyzer over the winter months. 
is the PV generation required for producing hydrogen. 
is the power required to charge the battery.
- The overall system design worked to power the fuel cell for the three-day period, but the current sizing of the PV and battery are not large enough to provide a continuously running hydrogen supply much beyond the three-day simulation.
- The instantaneous fuel cell power from the PV system is the most efficient when peak demand is during peak solar irradiance hours.
- Optimizing the hydrogen delivery to match the balancing hydrogen storage capacity and fuel cell hydrogen supply rate is needed for a more efficient system.
- Due to the variability in peak solar power throughout the year, there needs to be an optimized sizing guide for the PV and battery systems to provide a continuous hydrogen fuel supply to the fuel cell.
Ethical Approval: Not applicable.
Funding: Not applicable.
Availability of data and materials: MathWorks model and datasets available upon request.
- MathWorks Inc., "Green Hydrogen Microgrid ©," MathWorks Inc., 2021 - 2023.
- U.S. Energy Information Administration, "Use of Hydrogen: Basics," U.S. Energy Information Administration, Washington, 2023.
- MathWorks, "Getting Started with Simulink for Signal Processing," MathWorks, 2023.
- International Energy Agency, "Global Hydrogen Review 2022," International Energy Agency, 2022.
- College of the Desert, "Hydrogen Properties," Hydrogen Fuel Cell Engines and Related Technologies, 2001.
- MathWorks, Fuel Cell Stack ©, MathWorks, 2023.
- MathWorks, Electrolyzer, MathWorks, 2022.
- J. Andersson and S. Gronkvist, Large-scale storage of hydrogen, Stockholm: Division of Energy Processes, KTH Royal Institute of Technology, 2019.
- N. Li, Z. Lukszo and J. Schmitz, "An approach for sizing a PV–battery–electrolyzer–fuel cell energy system," Renewable and Sustainable Energy Reviews, 2023.
- Z. Li, Z. Zheng, L. Xu and X. Lu, "A review of the applications of fuel cells in microgrids: opportunities and challenges," BMC Energy, 2019.
Table 1 is available in the Supplementary Files section.
No competing interests reported.
