Feasible Capacity Ratio of H2 storage Using Electrolyzers and Batteries in Distributed Energy Systems Under Large-Scale Implementation of Solar Cells

Abstract

Introduction The power grid controls their electric power supply by maintaining voltage, frequency, and transient stability, by predicting the electricity demand at any given time. However, recent large-scale implementation of photovoltaics (PV) makes it difficult to meet them because the gap between power supply and demand will be widening. The suppression of any power sources including solar power generation with sacrificing some electricity is sometimes adopted to avoid large-scale blackouts, and the situation is expected to be a common event at the 2030s. An effective solution to such situations is called as “leveling power”, which is the ability to maintain a balance between power supply and demand. Leveling power needs to be introduced not only into the centralized power grid but also into distributed energy systems. The major candidate of leveling power at distributed energy systems is electrochemical energy storage technology such as secondary batteries or hydrogen Power-to-Gas-to-Power (P2G2P). A smart energy system called Ene-swallow has been developed by our research group, and applied to Ookayama campus at the Tokyo Institute of Technology. This system manages 1400 kW-PV, 100 kW-FC, 70 kW gas engine and 100 kWh storage battery with a 10,000 kW-class electricity demand. Ene-Swallow also accumulates the power demand data in 1-sec intervals and PV power generation data in 1-min intervals. We previously carried out a techno-economic analysis of a hydrogen energy storage system to level the annual gap of power supply and demand [1]. Here we evaluated the potential of electrochemical energy storage as a leveling power with using the annual data from those data. Method We analyzed the current situation requirements for leveling power (analysis-1) and the applicability of a distributed energy storage system (analysis-2). For analysis-1, we analyzed 1-hour data of power generation and consumption published from power companies across Japan (Fig. 1a), and 1-min data of PV power generation (Fig. 1b) and 1-sec data of power demand (Fig. 1c) at Ookayama campus. For analysis-2, we set a tentative distributed grid with a 10,000 kW-class electricity demand, a given amount of primary power generation by PV and the power supply from the grid. In this work, we set no leveling power by the grid, which supplied constant power. Then, we determined the entire gap between power supply and demand during a single year based on the actual 1-minute data of PV power generation and power demand from April, 2017 to March, 2018. Then we estimated the required amount of energy storage, and thus determined the required number of P2G2P devices with alkaline electrolysis cell (AEC), solid oxide fuel cells (SOFC), H2 tanks and compressors, and lithium-ion batteries (LIB). In the system, the balance of power generation and consumption in a day is controlled as shown in Fig. 2. During the daytime, the surplus power from PV is leveled by an AEC. If AEC capacity is not sufficient, it is leveled by LIB. At night, the required power for demand is generated by an SOFC and/or an LIB. By optimizing the capacity of each component device in the distributed system, we evaluated the total system cost. We specified the expected technical parameters in the future (for the year 2030), such as unit price, efficiency and lifetime as shown in Table 1. Results From analysis-1, the 1-hour data showed that the supply fluctuation of PV power generation during the day strongly impacts the power grid when the energy mix ratio of PV to total annual demand reached about 10% when keeping nuclear power generation constant. The 1-minute data showed large fluctuations in PV generation within in a short time period (< 1 hr) during weather such as a cloudy day (Fig. 1b). Power fluctuations with different time scales have to be leveled by using electrochemical energy storage technologies. Analysis-2 showed that the total system cost could be reduced by using a combined system of P2G2P and LIB by utilizing their different characteristics. The combined system of P2G2P can reduce their total energy cost with dividing the role by power generation and the storage when the cost targets shown in Table 1 can be achieved. [Acknowledgements] A part of this study is supported by the New Energy and Industrial Technology Development Organization (NEDO) and MIRAI program of Japan Science and Technology Agency (JST). Reference [1] T.Okubo et al., 234th ECS Meeting AiMES 2018, (September 30-October 4, 2018) Figure 1