Operation planning of an independent microgrid for cold regions by the distribution of fuel cells and water electrolyzers using a genetic algorithm

https://doi.org/10.1016/j.ijhydene.2011.08.004Get rights and content

Abstract

An energy system using a microgrid was examined in this work. The motivations for this study are to promote green energy usage, discuss concerns regarding energy supply during disasters, and improve the efficacy of waste heat usage. To create a society based on clean hydrogen energy, this paper studied the use of a microgrid to supply energy to six houses in a cold region. The proposed microgrid consisted of photovoltaics, a water electrolyzer, a fuel cell, and a heat pump; furthermore, this microgrid was not accompanied by any external energy supply. In this paper, the optimized calculation results obtained from the genetic algorithm (GA) were compared between a system operated using one set of large capacity equipment (a concentrated system) and a system operated using two or more pieces of distributed small-capacity equipment (a distributed system). From this comparison, the operation efficiency of each set of equipment was characterized using the difference in the load factor of the fuel cell and that of the water electrolyzer of each system. Moreover, the optimal capacities of the solar cell, fuel cells, water electrolyzers, and heat pumps while operating an energy-independent microgrid with the concentrated system and the distributed system were presented.

Highlights

► Optimizing calculation obtained from a genetic algorithm is compared for system operation. ► Introduced into a proposed microgrid consists of photovoltaics, a water electrolyzer, a fuel cell. ► Operation of a concentrated system and a distributed system. ► Optimal capacity of the solar cell area, fuel cells, water electrolyzers, and heat pumps.

Introduction

Integrating distributed energy resources can reduce power transmission losses, improve the efficacy of waste heat usage, and utilize green energy. To promote the spread of green energy, improve the method of supplying energy in times of disaster, and utilize waste heat more efficiently, a distributed energy system using a microgrid was examined in this study [1], [2], [3]. The energy supply system in the microgrid does not require equipment to be installed in all of the houses in the microgrid. Therefore, a distributed energy system has lower equipment cost than a standalone system due to the centralization of equipment. The goal of this paper was to model a society based on clean hydrogen energy by considering the operation of an independent microgrid in cold regions. The microgrid consisted of fuel cells, heat pumps, and water electrolyzers that used photovoltaics. Until now, only compound systems consisting of photovoltaics and water electrolyzers have been investigated [4], [5], [6]. The dynamic characteristics of a system with photovoltaics and a compound water electrolyzer were investigated [7], [8], and an analysis and physical testing of those optimizations [9] were reported. Distributing the energy equipment within a microgrid is simple, and the selection of the type and capacity of the pieces of equipment is flexible. In this paper, a microgrid was operated by one set of large-scale fuel cells as a concentrated system. In contrast, a distributed system consists of two or more small-capacity fuel cells. Because the load factors of the fuel cell for each system differ when they are introduced into a microgrid with large power load changes, the power generation efficiency is also expected to differ. Similarly, the installation capacity for the water electrolysis equipment is expected to affect the production efficiency. The production efficiency of hydrogen and oxygen differs between a concentrated system and a distributed system.

This paper discusses the construction of an independent microgrid that uses only the electric power of photovoltaics. In addition, the relationship between the load factor and efficiency of each piece of equipment placed in the microgrid is investigated using a numerical analysis that utilizes a genetic algorithm (GA). In the proposed system, the electricity demand and thermal demand of an entire microgrid were satisfied using the electric power obtained from two or more arrays of photovoltaics. Hydrogen and oxygen were generated by distributed water electrolyzers using the surplus power from the photovoltaics. These storage gases were supplied to a proton-exchange membrane fuel cell (PEFC) at an arbitrary time. Two or more electric heat pump systems were introduced into a proposed microgrid assuming that the proposed system would be installed in cold regions with a high thermal energy demand. Moreover, because the electric power of the photovoltaics could be stored by a water electrolyzer, an expensive battery was not installed in the proposed system. In an energy-independent microgrid using photovoltaics, optimizing the area of the solar cell is important. Furthermore, it is necessary to clarify the differences in the operation method between a concentrated system and a distributed system for a fuel cell and a water electrolyzer. Therefore, the construction of an independent microgrid for cold regions supplied only with the electric power of photovoltaics is dependent upon two factors: the equipment capacities and the operation plan for both the concentrated system and the distributed system. Furthermore, the partial load performances of a fuel cell and a water electrolyzer were considered for both the concentrated and distributed systems.

Section snippets

System schematic

Fig. 1 depicts the system flow of an independent microgrid with solar water electrolysis. This microgrid supplies electric power and heat to six houses in total. Fig. 2 explains the details of these load characteristics. As shown in Fig. 1, the electric power obtained by the photovoltaics (photovoltaics (1)–(4)) can be supplied to a power grid, the heat pumps (heat pumps (1) and (2)), and the water electrolyzers (water electrolyzers (1)–(3)). In this paper, the operation of one set of

Chromosome model

The power output from a fuel cell and the power input to a water electrolyzer, which were both used in the balanced equation (Eqs. (1-1), (1-2)), are expressed by Eqs. (11), (12), respectively. A fuel cell and a water electrolyzer were introduced into three sets of equipment at the maximum installation numbers. The terms αt, βt and χt in Eq. (11) and ξt, ψt and ζt in Eq. (12), respectively, are the output rates of fuel cells and the power input rates of water electrolyzers, respectively. Here,

Operation efficiencies of the fuel cell and the water electrolyzer

  • (1)

    Operation efficiency of a fuel cell

Fig. 7 shows the analysis results of the energy balance of the system. Below, the operation analysis details of the system are examined. Fig. 8 shows the analysis of output efficiencies for the operation of the fuel cells and the water electrolyzers that were introduced into the concentrated system in January and July. The time interval during which the efficiency was zero for each month gave the electricity demand for the photovoltaics. As shown in Fig. 3(b),

Conclusions

  • (1)

    The capacity of each piece of equipment introduced into the proposed microgrid that supplied energy to six houses was determined. As a result, the total capacity of the fuel cells and water electrolyzers was 30 and 105 kW, respectively. Moreover, the area of the installed solar cells was 346 m2 during winter (January), when there was a high energy demand. Considering the winter conditions, the system required a heat pump system with a total capacity of 100 kW.

  • (2)

    The change in the load factor and

Acknowledgment

This work was partially supported by a Grant-in-Aid for the Fundamental Research Developing Association Shipbuilding Offshore (REDAS), 2010. We appreciate the support to this research by the REDAS.

Cited by (9)

  • Ensemble learning for optimal active power control of distributed energy resources and thermostatically controlled loads in an islanded microgrid

    2018, International Journal of Hydrogen Energy
    Citation Excerpt :

    In order to shorten the computation time, all the sub-optimizers and the learning concentrator are executed in parallel based on multi-core CPU. The performance of EL for OAPC is thoroughly evaluated on an islanded microgrid, which is compared with that of five algorithms, including proportional (PROP) method [38], GA [33], PSO [34], group search optimizer (GSO) [39], and interior point method (IPM), where PROP is an engineering method with fixed participation factors; the population size and the maximal iteration number of three heuristic algorithms are set to be 150 and 150, respectively. Through the trial-and-error, the main parameters of EL are given in Table 2.

  • A hybrid harmony search algorithm with differential evolution for day-ahead scheduling problem of a microgrid with consideration of power flow constraints

    2016, Applied Energy
    Citation Excerpt :

    Hence, the tendency of employing intelligent optimization methods, which can search for the optimum from different initial points, is increasing in recent years. Many population algorithms such as the genetic algorithm (GA) [62–67], particle swarm optimization (PSO) [68–76], gravitational search algorithm (GSA) [14,77–79], differential evolution (DE) [70,80,81] and harmony search algorithm (HSA) [82] have been used to solve the optimal scheduling problems of microgrids. The above intelligent methods are capable of solving nonlinear and non-convex optimization problem.

  • The influence of distributed generation penetration levels on energy markets

    2013, Energy Policy
    Citation Excerpt :

    In the Brazilian energy context, the National Agency for Electric Energy—ANEEL (2010) includes other sources of energy in a DG system even if they are not from renewable sources, defining DG broadly as small-scale generators distributed along the grid. Micro-grid is a small distribution network which is composed of distributed generation (DG), load, distributed storage (DS), power conversion equipment and control system, and can realize flexible energy management (Obara et al., 2011a, 2011b; Houwing et al., 2008). A typical micro-grid model is shown in Fig. 1.

  • Optimal power and heat scheduling of microgrids under renewable generation uncertainties

    2020, 2020 2nd International Conference on Smart Power and Internet Energy Systems, SPIES 2020
View all citing articles on Scopus
View full text