Superior performances strategies of different hybrid renewable energy systems configurations with energy storage units

Abstract

The essential function of a hybrid renewable energy system is to produce an adequate electrical supply to the load demand with low cost. This paper proposes the optimization and operation of the renewable energy power sources for electrification of isolated rural city in Algeria desert. For this purpose, a system composed by PV panel (PV), Wind Turbine (WT), Battery Bank (BB) for storage of the electrical energy and Fuel Cell (FC) with hydrogen tank (H₂ Tank) and electrolyzer (Elect) system for storage of the chemical energy is used to fulfill the need of the load. In the present paper we are interested in evolutionary algorithms for solving optimization problem of hybrid configuration of power system. However, a new heuristic algorithm namely whale optimization algorithm (WOA) is used to obtained the best solution of multi-objective optimization system of cost of energy (COE), total net present cost (T_NPC) and loss power supply probability (LPSP). Two recent algorithms, particle swarm optimization (PSO) and grey wolf optimizer (GWO) are also implemented in this work. Seven cases studies have been tested for examining the efficient of proposed optimization technique. The suggested whale optimization algorithm, as demonstrated by simulations and comparisons with existing methods, solves the problem of multi-objective optimization of hybrid power system configurations with high accuracy and validity.

Keywords

Cost of energy total net present cost optimization storage units Algeria desert

Introduction

The photovoltaic PV panel and the wind generator turbine energy are pollution free alternative to the conventional sources, both in grid connected and off-grid regions and capable of overcoming the worldwide energy crisis. Grid-connected power systems being replaced with autonomous power systems. They include insolation, wind, and hydraulic energy sources, all of which are virtually limitless. Because of the ever-dwindling supplies of gas and oil, as well as international awareness and initiatives to encourage renewable energy generation systems, the percentage of renewable energy generated is should increase (Infield, 2008).

The year 2020 presents the global energy transition due to the great demand for hydrogen technology on the one hand and the tragic COVID-19 crisis on the other hand. For these reasons several countries in the world have proposed strategies to develop hydrogen as an important energy source.

A hybrid power generating system integrates different energy sources that are linked together to produce synchronized output power. Hybrid power system can be augmented with PV, WT, H₂ Tank, diesel generator (DG), FC and biomass with adjustable of the energy stored in BB and H₂ Tank systems in such away which allows the entire system to satisfy power demands (Emanuele et al., 2020; Fathima and Palanisamy, 2015). The battery bank system can be composed with rechargeable battery or ultra-capacitor (Lokeshgupta and Sivasubramani, 2019). The best solution of this sizing is to attempt to satisfy the load, the hybrid power capacity system rating will be used to fulfill the energy demand. In this way, the cost of the hybrid system is must be very efficient and effective than using fuel cells to supply the non-satisfaction of the load (Singh and Fernandez, 2018; Türkay and Telli, 2011).

Optimization is one of the most important branches of applied mathematics, and much research, both practical and theoretical, has been devoted to it. There are two main approaches to optimization. One is said to be deterministic: the search algorithms always use the same path to arrive at the solution, so we can determine the search steps in advance. The other is random: for given initial conditions, the algorithm will not follow the same path to reach the solution, and can even offer different solutions. It is toward this second approach that our work will be oriented, and more particularly toward a very specific type of random search algorithm, evolutionary algorithms which represents an important tool for solving optimization problems. Moreover, they are increasingly used in multiple fields. They are easy to use and provide excellent performance at low cost (Fathima and Palanisamy, 2015).

The hybrid PV/WT/BB system and the hybrid PV/WT/Elect/H₂ Tank/FC system have been studied extensively, but few researchers in this field have been used hybrid PV/WT with BB and Elect/H₂ Tank/FC simultaneously. El-Shatter et al. (2002), optimize the size of hybrid PV/Elect/FC generation system. Shapiro et al. (2005), used hybrid PV/Elect/FC system considering reliability parameters for rural electrification. Galli and Stefanoni (1997) studied the performance of a plant in Italy with PV/Elect/H₂ Tank/FC. Paul and Andrews (2008) compared the energy produced by PV array and a Electrolyzer for obtained an optimal sizing of series and parallel connection of PV/Electsystem. Li et al. (2009), studied three autonomous PV power systems with different storage systems and modeled PV modules, FC_s, Elect_s, compressors, H₂Tank_s, and BB_s. Uzunoglu et al. (2009) proposed the combination of PV, FC and ultra-capacitor systems for sustained energy production. Deshmukh and Boehm (2008) focused a review of hydrogen systems driven by renewable energy and associated modeling approaches. The authors Kanase-Patil et al. (2010) offered four scenarios for combining renewable energy systems for seven separate communities in Almara, India, using all types of micro hydropower, biomass, biogas, PV panels, and WT. Two decentralized power stations in Sabah, Malaysia, were studied for a system model and performance evaluation, each with a distinct combination of PV panels, diesel engines, battery banks, and converters (Halabi et al., 2017). Using optimization approaches, Das et al. (2019) was able to establish the appropriate sizing of a stand-alone PV/biogas generator/pumped hydro energy storage/battery system. In Ghenai et al. (2020), the approaches for dispatch control and the best hybrid system design were examined. To establish an efficient and optimal energy management operation for H₂/WT/PV/GMT hybrid power systems, artificial intelligent (AI) controllers were applied in Tabanjat et al. (2018).

The main goal of this paper is to optimize the size of the components of various combinations of stand-alone photovoltaic generators, wind turbines, fuel cells, and water electrolyzes to produce hydrogen energy, hydrogen tank with batteries and converters to meet the energy needs of Timimoun, Algeria. The presented systems in this paper are used for optimal configuration, total net present cost (TNPC) minimization, cost of energy (COE), and loss power supply minimizing (LPSP). Comparative study with seven hybrid systems is investigated in this work (PV//BB, PV/FC, WT/BB, WT/FC, PV/WT/BB, PV/WT/FC, PV/WT/BB/FC). This work also uses three optimization techniques: whale optimization algorithm (WOA), particle swarm optimization (PSO), and grey wolf optimizer (GWO). In comparison to PSO and GWO, the obtained results show that the proposed whale optimization algorithm solves multi-objective optimization problems of hybrid renewable energy systems with high accuracy and validity, because it provides the best compromise between total net present cost of installation, cost of energy, system architecture, and low loss power supply probability.

Operating principle of the proposed system

Photovoltaic panels, wind turbines, and storage devices make up the system depicted in Figure 1. The storage devices are the battery, electrolyzer, hydrogen tank, and fuel cell. The excess energy produced by photovoltaic systems and wind turbines is saved in storage systems when the energy produced exceeds the load requirement. A portion of the excess energy is stored in batteries, while the remainder is sent to the electrolyzer and transformed to hydrogen. The electrolyzer produces hydrogen, which is kept in the H2tank. When the energy produced by PV panels and wind turbines is insufficient to fulfill the load requirement, batteries and fuel cells can step in.

Figure 1.

Architecture of standalone hybrid PV/WT/BB/FC system.

Modeling of system components

Photovoltaic panels

The power of a PV system depending on solar radiation can be calculated using the following formula (Maleki et al., 2017):

P_{PV} = I (t) \times η_{PV} \times S

(1)

Where, $I$ stands for solar radiation, $S$ for PV area, and $η_{PV}$ for PV system efficiency.

Wind turbines

The following equation is used to determine the power generated by the wind turbine at each moment t (Maleki et al., 2017):

P_{WT} = {\begin{matrix} 0 v (t) \leq v_{cut - in} o r v (t) \geq v_{cut - out} \\ p_{r} \frac{v (t) - v_{cut - in}}{v_{r} - v_{cut - in}} v_{cut - in} < v (t) < v_{r} \\ p_{r} v_{r} < v (t) < v_{cut - out} \end{matrix}

(2)

Where, $v (t)$ is the wind speed as a function of time t (m/s), $p_{r}$ is the wind turbine’s nominal power in kW, and $v_{cut - out}$ , $v_{cut - in}$ , and $v_{r}$ are the cut-off and rated speeds of the wind turbine, respectively (m/s).

Electrolyzer/Hydrogen tank/Fuel cell

If the power provided by the wind turbine and PV panel is greater than the load side at any time t in the hybrid PV/Wind and Fuel Cell (FC) system, the hydrogen energy produced by the electrolyzer is used to fill the hydrogen tank. The following equation is used to compute the amount of hydrogen in the hydrogen tank (Ahmadi and Abdi, 2016):

E_{stor} (t) = E_{stor} (t - 1) + [(E_{PV} (t) \times n_{inv}) + ((E_{WT} (t) \times n_{inv}) - (E_{Load} (t)) / n_{inv}] \times n_{elect}

(3)

$E_{stor} (t)$ and $E_{stor} (t - 1)$ are the energy stored in the hydrogen tanks in kWh at time (t) and (t − 1), respectively, E_load is the load side in kWh, η_elect is the inverter efficiency, and η_elect is the electrolyzer efficiency.

If the energy provided by the photovoltaic panel and wind turbine is insufficient to fulfill the load demand, the fuel cell is used to meet the need. This is accomplished by estimating the amount of hydrogen energy stored in the H₂ tanks at time t as follows:

E_{stor} (t) = E_{stor} (t - 1) - ([((E_{Load} (t) / n_{inv}) - (E_{PV} (t) \times n_{inv}) + (E_{WT} (t) \times n_{inv})]) / (n_{fc} \times n_{inv})

(4)

Where, $n_{fc}$ is the efficiency of the fuel cell.

For full optimization, fuel cell generation is simulated throughout the year. We can see that the production of fuel cells coincides with a low production of wind turbines and photovoltaic panels at times, but this will adjust the output power. Water electrolysis is the process of splitting water into hydrogen and gaseous oxygen using electrodes to conduct a direct current through it. This process can be explained by the following equation (Mohammadi and Mehrpooya, 2018):

H_{2} O \to 2 H_{2} + O_{2}

(5)

Battery bank

When the total energy produced by the solar panel and the wind turbine is greater than the load side, the battery is charged by the excess energy produced by photovoltaic and wind power. At any moment t, the battery’s state of charge is calculated as follows (Acakpovi et al., 2020; Maleki et al., 2017):

E_{bat} = E_{bat} (t - 1) \times (1 - σ) + [(E_{PV} (t) \times n_{inv}) + (E_{WT} (t) \times n_{inv}) - (\frac{E_{Load} (t)}{n_{inv}})] \times n_{bc} / n_{inv}

(6)

Where $E_{bat}$ and $E_{bat} (t - 1)$ represent the amount of charge from battery storage at time t and (t − 1) in kWh, $n_{bc}$ represents the battery storage system charge efficiency, and $σ$ is the hourly self-discharge When the PV panel and wind turbine combined power is less than the load side, the battery discharges and fills the gap.

Optimization problem formulation

The ideal design of a stand-alone hybrid photovoltaic, wind turbine, electrolyzer, fuel cell power system with battery storage and hydrogen tank to meet the load requirement of the city of Timimoun in south Algeria is the subject of this paper. The proposed optimal design research focuses on cost-effectiveness, with the best scaling of the hybrid renewable energy system based on a simulation model created with Matlab software. With genuine weather data, the load profile of Timimoun city is used.

Objective function

The cost of energy (COE) is used as the objective function in this study to determine the optimal benchmark for system cost analysis. Total net present cost (T_NPC), capital recovery factor (CRF), and total generated power are all included in COE. The objective function, on the other hand, includes the calculation of reliability. In order to determine the probability of a power supply failure, the dependability is assessed using fundamental probabilistic concepts. Only total capital cost (T_CC), total replacement cost (T_RC), and total operation and maintenance cost (T_NPC) are included in T_NPC (T_OMC). PV panels, wind turbines, converters, battery banks, electrolyzers, hydrogen tanks, and fuel cells are all components to consider. Then, COE can be calculated as follow (Hossain et al., 2017; Xu et al., 2013):

COE = \frac{T_{NPC . CRF (r, n)}}{\sum_{t = 1}^{744} E_{L}}

(7)

Where, E_L is the load demand, r is the interest rate, that equal 0.06 in this work and n represent the life span of each component.

The following equation is used to compute the total net current cost:

T_{NPC} = T_{CC} + T_{RC} + T_{OMC}

(8)

CRF stands for capital recovery factor, which is a ratio used to evaluate the present value of a sequence of equal yearly cash flows (Notton et al., 2011; Olatomiwa et al., 2018; Yahiaoui et al., 2016):

CRF = \frac{d {(1 + d)}^{T}}{{(1 + d)}^{T} - 1}

(9)

Loss power supply probability

The loss power supply probability gives precise results when evaluating the level of reliability in power systems. To predict component outages, this method use probabilistic analysis. It can be characterized in terms of LPSP, which is calculated using statistical data and allows determining if generated power is unable to meet load demand owing to technical reasons or the failure of renewable energy sources to produce the required power. It expresses the rate of non-satisfaction of the load by presenting the fraction of unmet demand on that demanded by the load. As a result, the ratio between the sum of the energy not given defines this probability (ENS) and the total energy required by the load throughout the course of a period of operation (Hadidian-Moghaddam et al., 2016; Kharrich et al., 2019):

LPSP = \frac{ENS}{\sum_{t = 0}^{T} E_{L}}

(10)

The energy not supplied $(ENS$ ) can be obtained by following equation:

ENS = \sum_{t = 0}^{T} (E_{L} - (P_{PV} + P_{WT}))

(11)

Table 1 summarizes the economic and technical parameters of all components considered in the system optimization.

Table 1.

Specification of all components considered in the system optimization.

Components	PV panel	WT	Battery bank	Electrolyzer	H₂-tank	FC	Converter
Capital cost ($/unit)	275.072	33,985.02	179.83	1134.72	1000	27,062	1491.3057
Replacement cost ($/unit)	0	0	10	0	15	0	0
Operation and maintenance cost ($/year)	35.6	20	1.5	25	30	1	8
Life time (year)	25	20	10	15	15	15	20
Efficiency (%)	90	90	85	75	95	50	90
Power	0.33 kW	1 kW	6.7 kWh	0.2 kW	1 kW	6 kW	2.5 kW

Results

On the Matlab software we have defined each element of our installation based on all the characteristics and data provided mentioned previously. In this work, we perform the optimization of PV/Wind/Battery bank/Electrolyzer/H₂ tank/Fuel cell hybrid system in order to choose the best system architecture, in terms of cost to supply the isolated site of Timimoun.

In order to build this electrical system, we must have all the information necessary from the chosen site. Typical information is: the load profile, solar radiation and temperature for photovoltaic energy production, wind speed for production wind energy, the initial cost of each component, cost and life of the project. Then with this data we can run simulations to obtain the best configuration of hybrid renewable energy system for the site studied. Analysis of the variation of temperature presented in Figure 2 shows that the hottest month is July and August with a temperature average of around 34°C, while the coldest month is January with an average temperature of around −2°C, the average monthly temperature is around 22°C. The solar radiation values every hour for the whole year as shown in Figure 3.

Figure 2.

The temperature profile.

Figure 3.

The solar radiation profile.

Figure 4 depicts the wind speed profile over the four seasons of the year. The average monthly wind speed on the Timimoun site is always greater than 5 m/s at a height of 10 m from the ground, and since most wind turbines start at a wind speed greater than 3 m/s, wind energy extraction is advantageous for this location. The Timimoun site has a lot of potential for wind energy and can potentially be used for high-power installations. Figure 5 depicts energy use for a single day in the year 2020.

Figure 4.

The wind speed profile.

Figure 5.

The load profile.

We note that the energy consumed is then approximately 2.5527 MWh/day in the year 2020 with a peak power of 298.1 KW. According to Figure 4 the electricity consumption of the Timimoun site differs hour to hour. This change in consumption caused by the difference in operation of the household appliances, between night and day. Figure 6 shows a presentation of energy production by the PV/WT/FC isolated hybrid system.

Figure 6.

Yearly operation of PV/WT/FC hybrid system: (a) PV panel output power, (b) wind turbine output power, (c) electrolyzer output power, (d) mass of hydrogen stored, and (e) fuel cell output power.

Figure 7 depicts the convergence curves of the WOA, GWO, and PSO algorithms for the seven potential configurations. The number of iterations in WOA is 100, and the population size is 20. For GWO, the number of iterations is 100, the number of search agents is 20, and the number of iterations is reduced linearly from 2 to 0. The number of iterations in PSO is 100, the population size is 20, and C1 and C2 are 2.2 and 2.2, respectively. The number of PV panels (N_PV), the number of wind turbines (N_WT), and the number of battery banks are the best sizing parameters (N_BB), the rated power of electrolyzer (P_Elect), the mass of the hydrogen gas tank (M_{H2 Tank}), the rated power of the fuel cell (P_FC), the rated power of the converter (P_Conv) for the seven suggested configurations with different optimization techniques are presented in Tables 2 to 4. We can see from these tables that WOA generates the lowest COE and TNPC of all the configurations of the systems presented in this research work with optimal energy loss probability compared to GWO and PSO algorithms, this demonstrates the superiority of the WOA method for the optimization of hybrid renewable energy systems. Moreover, we notice that for all the systems the optimal values of Tnpc is obtained before 40 iterations, the 100 iteration used as a criterion of termination of our algorithms in order to show the effectiveness of our results.

Figure 7.

Different optimization strategies converge: (a) PV/Battery system, (b) PV/FC system, (c) Wind/Battery system, (d) Wind/FC system, (e) PV/Wind/FC system, (f) PV/Wind/Battery system, and (g) PV/Wind/Battery/FC system.

Table 2.

System component optimization results based on WOA for different configuration of hybrid system.

	PV/Battery	PV/FC	Wind/Battery	Wind/FC	PV/Wind/Battery	PV/Wind/FC	PV/Wind/Battery/FC
Number of PV panels	91.0544	60.0051	-	-	69.2689	53.9499	148.3481
Number of wind turbines	-	-	92.6142	49.0744	61.2377	83.9221	81.1905
Number of battery banks	103.1919	-	60.3948	-	96.3740	-	97.2281
Electrolyzer (kW)	-	48.004	-	48.2854	-	42.8317	54.798
Mass of the hydrogen tank (kg)	-	45.6038	-	41.0426	-	40,6902	52.0581
Fuel cell (kW)	-	22.8019	-	20.5213	-	21.9795	13,0305
DC/AC converter (kW)	497.9239	36.5610	384.5709	58.2193	562.0829	102.1791	683.8752
COE ($/kWh)	0.1342	0.1521	0.4981	0.4982	0.1632	0.1635	0.1639
T_NPC ($)	1,615,917.8015	1,832,218.5928	5,997,311.3742	5,998,184.0549	1,965,568.8293	1,968,895.3756	1,973,494.8510

Table 3.

System component optimization results based on GWO for different configuration of hybrid system.

	PV/Battery	PV/FC	Wind/Battery	Wind/FC	PV/Wind/Battery	PV/Wind/FC	PV/Wind/Battery/FC
Number of PV panels	73.9680	76.7788	-	-	29.7875	71.9230	51.7891
Number of wind turbines	-	-	56.5893	59.9827	50.8039	55.7015	50.0506
Number of battery banks	74.2710	-	85.8226	-	49.0029	-	54.9603
Electrolyzer (kW)	-	77.9013	-	55.3313	-	61.2276	42.4020
Mass of the hydrogen tank (kg)	-	58,426	-	41.4985	-	45.9207	31.8015
Fuel cell (kW)	-	55.5047	-	39.4236	-	43.6246	30.2114
DC/AC converter (kW)	469.8226	70.4772	505.5667	172.25523	333.641913	343.961531	388.5315
COE ($/kWh)	0.1521	0.1522	0.4982	0.49826	0.16331	0.16380	0,49817
T_NPC ($)	1,831,371.4012	1,832,384.7029	5,997,750.3426	5,998,350.1650	1,966,027.6326	1,971,950.1636	1,997,311.3742

Table 4.

System component optimization results based on PSO for different configuration of hybrid system.

	PV/Battery	PV/FC	Wind/Battery	Wind/FC	PV/Wind/Battery	PV/Wind/FC	PV/Wind/Battery/FC
Number of PV panels	40	71	-	-	31	77	117
Number of wind turbines	-	-	69	58	53	54	84
Number of battery banks	50	-	78	-	57		95
Electrolyzer (kW)	-	120	-	80	-	61.0526	69.4736
Mass of the hydrogen tank (kg)	-	114	-	76	-	58	66
Fuel cell (kW)	-	57	-	38	-	29	33
DC/AC converter (kW)	266.967	70.2783	400.599	81.18	343.3698	90.4221	615.9366
COE ($/kWh)	0.1615	0.1533	0.4983	0.5043	0.1634	0.1638	0.1660
T_NPC ($)	1,943,757.6609	1,845,733.6901	5,998,569.9704	6,071,567.9991	1,967,294.3754	1,971,981.9979	1,998,066.2760

The costs of the various components for each hybrid system by WOA algorithm is presented in Table 5, the total net present cost over the year of our project costs (capital cost, replacement cost, Operation and Maintenance cost) is 1,615,917.8015$ for PV/BB system, 1,832,218.5928$ for PV/FC system, 5,997,311.3742$ for WT/BB system, 5,998,184.0549$ for WT/FC system, 1,965,568.8293$ for PV/WT/BB system, 1,968,895.3756$ for PV/WT/FC system and 1,973,494.8510$ for PV/WT/BB/FC system. We find that the highest cost in each hybrid system is the total capital cost (T_CC), after the total operation and maintenance cost (T_OMC), after the total replacement cost (T_RC).

Table 5.

Economic results of different hybrid system components for different configuration of hybrid system using WOA.

Components	PV/Battery	PV/FC	Wind/Battery	Wind/FC	PV/Wind/Battery	PV/Wind/FC	PV/Wind/Battery/FC
T_CC ($)	1,615,188.1105	1,831,154.5370	5,996,259.8240	5,996,916.7052	1,964,429.3779	1,967,328.0499	1,971,396.8715
T_RC ($)	78.4944	111.7899	499.0547	374.5047	408.2492	558.8292	645.0819
T_OMC ($)	651.1966	952.2658	552.4955	892.8449	731.2021	1008.4965	1452.8976

The contribution of all costs of each proposed hybrid system to the T_CC T_RC and T_OMC in the optimal design using WOA is provided in Figures 8 to 10 respectively. In Figure 8 the total capital costs are broken regarding each component in all proposed hybrid systems in this paper. It can be observed that the WT/BB system has the highest value of T_CC of wind turbines (5,411,886.252$) within all configurations systems. From Figure 9 we can noticed that in the hybrid PV/WT/BB/FC system the total replacement cost of battery banks has the highest value (449.4299$). From Figure 10 we can observed that the hybrid PV/WT/BB/FC system has the highest value of the total operation and maintenance cost of PV panels (528.11923$).

Figure 8.

The capital costs of the various components of each hybrid system.

Figure 9.

The replacement costs of the various components of each hybrid system.

Figure 10.

The operation and maintenance costs of the various components of each hybrid system.

Figure 11 represents the results obtained within 200 hours of the amount of energy generated by each component of all proposed hybrid systems compared to the amount of electrical energy consumed using the optimized parameters by WOA algorithm, where we note that storage batteries and hydrogen tanks are important pieces of equipment in this electric power system and they are used continuously throughout the day, during the days of the year, especially at night and this because of the lack of renewable energy (weather conditions) on the other hand for various reasons, among these reasons: the difficulty delivery and liaison with this remote region. And the charge level varies between 180 and 200 kg during the days for hydrogen tanks. It is used as the essential and fundamental element for the generation of energy.

Figure 11.

Operation of the hybrid systems on a certain day (200 hours) compared to the load using the results of WOA: (a) PV/Battery system, (b) PV/FC system, (c) Wind/FC system, (d) Wind/Battery system, (e) PV/Wind/ FC system, (f) PV/Wind/Battery system, and (g) PV/Wind/Battery/FC system.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Adel Yahiaoui

Abdelhalim Tlemçani

References

Acakpovi

Adjei

Nwulu

, et al. (2020) Optimal hybrid renewable energy system: A comparative study of wind/hydrogen/fuel-cell and wind/battery storage. Journal of Electrical and Computer Engineering 2020: 1–15.

Ahmadi

Abdi

(2016) Application of the Hybrid Big Bang–Big Crunch algorithm for optimal sizing of a stand-alone hybrid PV/wind/battery system. Solar Energy 134: 366–374.

Das

Singh

MAK

Biswas

(2019) Techno-economic optimization of an off-grid hybrid renewable energy system using metaheuristic optimization approaches – Case of a radio transmitter station in India. Energy Conversion and Management 185: 339–352.

Deshmukh

Boehm

(2008) Review of modeling details related to renewably powered hydrogen systems. Renewable and Sustainable Energy Reviews 12(9): 2301–2330.

El-Shatter

Eskandar

El-Hagry

(2002) Hybrid PV/fuel cell system design and simulation. Renewable Energy 27(3): 479–485.

Emanuele

Herib

Raul

, et al. (2020) Green hydrogen cost reduction. Abu Dhabi: International Renewable Energy Agency.

Fathima

Palanisamy

(2015) Optimization in microgrids with hybrid energy systems – A review. Renewable and Sustainable Energy Reviews 45: 431–446.

Galli

Stefanoni

(1997) Development of a solar-hydrogen cycle in Italy. International Journal of Hydrogen Energy 22(5): 453–458.

Ghenai

Salameh

Merabet

(2020) Technico-economic analysis of off grid solar PV/Fuel cell energy system for residential community in desert region. International Journal of Hydrogen Energy 45(20): 11460–11470.

10.

Hadidian-Moghaddam

Arabi-Nowdeh

Bigdeli

(2016) Optimal sizing of a stand-alone hybrid photovoltaic/wind system using new grey wolf optimizer considering reliability. Journal of Renewable and Sustainable Energy 8: 035903.

11.

Halabi

Mekhilef

Olatomiwa

, et al. (2017) Performance analysis of hybrid PV/diesel/battery system using HOMER: A case study Sabah, Malaysia. Energy Conversion and Management 144: 322–339.

12.

Hossain

Mekhilef

Olatomiwa

(2017) Performance evaluation of a stand-alone PV-wind-diesel-battery hybrid system feasible for a large resort center in South China Sea, Malaysia. Sustainable Cities and Society 28: 358–366.

13.

Infield

(2008) An overview of renewable energy technologies with a view to stand alone power generation and water provision. In: Presented at the water and sanitation in international development and disaster relief (WSIDDR) International Workshop, Edinburgh, Scotland, 28–30 May.

14.

Kanase-Patil

Saini

Sharma

(2010) Integrated renewable energy systems for off grid rural electrification of remote area. Renewable Energy 35: 1342–1349.

15.

Kharrich

Mohammed

OHM

Akherraz

(2019) Assessment of renewable energy sources in morocco using economical feasibility technique. International Journal of Renewable Energy Research 9(4): 1856–1864.

16.

Zhu

Cao

, et al. (2009) Dynamic modeling and sizing optimization of stand-alone photovoltaic power systems using hybrid energy storage technology. Renewable Energy 34(3): 815–826.

17.

Lokeshgupta

Sivasubramani

(2019) Multi-objective home energy management with battery energy storage systems. Sustainable Cities and Society 47: 101458.

18.

Maleki

Pourfayaz

Hafeznia

, et al. (2017) A novel framework for optimal photovoltaic size and location in remote areas using a hybrid method: A case study of eastern Iran. Energy Conversion and Management 153: 129–143.

19.

Mohammadi

Mehrpooya

(2018) A comprehensive review on coupling different types of electrolyzer to renewable energy sources. Energy 158: 632–655.

20.

Notton

Diaf

Stoyanov

(2011) Hybrid photovoltaic/wind energy systems for remote locations. Energy Procedia 6: 666–677.

21.

Olatomiwa

Blanchard

Mekhilef

, et al. (2018) Hybrid renewable energy supply for rural healthcare facilities: An approach to quality healthcare delivery. Sustainable Energy Technologies and Assessments 30: 121–138.

22.

Paul

Andrews

(2008) Optimal coupling of PV arrays to PEM electrolysers in solar–hydrogen systems for remote area power supply. International Journal of Hydrogen Energy 33(2): 490–498.

23.

Shapiro

Duffy

Kimble

, et al. (2005) Solar-powered regenerative PEM electrolyzer/fuel cell system. Solar Energy 79(5): 544–550.

24.

Singh

Fernandez

(2018) Modeling, size optimization and sensitivity analysis of a remote hybrid renewable energy system. Energy 143: 719–731.

25.

Tabanjat

Becherif

Hissel

, et al. (2018) Energy management hypothesis for hybrid power system of H2/WT/PV/GMT via AI techniques. International Journal of Hydrogen Energy 43(6): 3527–3541.

26.

Türkay

Telli

(2011) Economic analysis of standalone and grid connected hybrid energy systems. Renewable Energy 36: 1931–1943.

27.

Uzunoglu

Onar

Alam

(2009) Modeling, control and simulation of a PV/FC/UC based hybrid power generation system for stand-alone applications. Renewable Energy 34(3): 509–520.

28.

Ruan

Mao

, et al. (2013) An improved optimal sizing method for wind-solar-battery hybrid power system. IEEE Transactions on Sustainable Energy 4(3): 774–785.

29.

Yahiaoui

Benmansour

Tadjine

(2016) Control, analysis and optimization of hybrid PV-diesel-battery systems for isolated rural city in Algeria. Solar Energy 137: 1–10.