Renewable and Sustainable Energy Reviews 82 (2018) 2039–2052

Contents lists available at ScienceDirect

Renewable and Sustainable Energy Reviews

journal homepage: www.elsevier.com/locate/rser

Hybrid renewable microgrid optimization techniques: A review MARK

a,b,⁎ a a,b
Samir M. Dawoud , Xiangning Lin , Merfat I. Okba
a
School of Electrical and Electronics Engineering, Huazhong University of Science and Technology, Wuhan 430074, Hubei, China
b
Department of Electrical Power and Machines Engineering, Tanta University, Egypt

A R T I C L E I N F O A BS T RAC T

Keywords: Power resulted form solar photo voltaic and wind turbine generators are reliant on the climate variations. Both
Hybrid Renewable Energy Systems (HRESs) the wind and solar systems are non-reliable if there are insufficient capacity storage units like storage batteries
Artificial Intelligence (AI) or backup units like diesel generators. The microgrids reliability increases when both systems (wind turbine and
Optimization techniques photo voltaic) are combined with the storage devices. The sufficient storage batteries bank capacity are needed
Wind Turbine Generators (WTGs)
to feed the load demands with power in cloudy and non-windy days. So the optimal placing of the components
Solar Photovoltaics (SPVs)
assigns to the required parts of hybrid microgrid. Also, this study reviews new ways of energy practice of hybrid
sources. It presents the physical modelling of the renewable energy resources with numerous methodologies and
principles of the optimization for the hybrid networks. Additionally, the hybrid sources are gaining popular and
fame in the environmental crises and current scenario of energies. Based on this study, it has introduced a
global survey on the present condition of optimization techniques especially that related to the isolated
microgrid in the presented literature. The current trend of optimization for hybrid renewable sources
demonstrations that artificial intelligence provides worthy optimization for the microgrid operations without
an extensive long-term weather data.

1. Introduction level of a customer, a renewable distributed generation is allowed to

help in an effective usage of the consumer energy, and for reaching to
Energy has a creative role for social and economic developments all the intelligent demands response [5,6].
over the world. Nowadays, the worldwide energy demands are supplied Most of RES such as SPV and WTG are clean and environment-
using fossil fuels considering diesel depletion, growing inhabitants and friendly. Many researchers have been performed studies which con-
increasing load requirements. The electrical power generations have cerning with the hybrid SPV–WTG microgrid. From the investigation,
gone into a new evolution phase which is highly characterized by it has been confirmed that the HRES gives a good performance and a
growing what concerned with weather variations, a transition from a lower cost compared to the individual SPV or WTG system [7–10].
hydrocarbon-based economy, and the effective deployment of the However, the hybrid SPV and WTG still has less disadvantages in
energies [1–4]. comparison with the conventional sources if it is not designed in an
Fossil fuels which represented in coal, oils and natural gasses, are accurate way. For example, the irregular nature of wind speeds and
the principal energy sources of the world. The dependence on fossil solar radiation which cause power fluctuations can be treated by using
fuels in the 20th century decreased heavily the natural reserve of them. storage units like storage battery bank. These batteries can store extra
Most three RESs used widely are hydropower, WTG and SPV. RESs are power and supply loads with power in a case of occurring shortage [11].
considered as the important alternative sources, so they are selected in Adding the storage batteries also support to avoid the SPV and WTG
many countries. The hybrid systems can be SPV, WTG and hydropower sources to be oversize. However, when the storage batteries have
or the mix of these units. Such systems have the backup units like a charged to its maximum value, and there is still excess power has been
diesel engine and/or storage batteries to feed a peak hour demand. The lost from the generation units which have to be avoided somewhere.
characteristics of RES have been introduced in Table 1. Recently, to Diminishing this unutilized excess power could lead to decrease a Cost
make an efficient usage of electrical energy, there is a great interest in of Energy (COE) [12]. Hence, the optimal capacity of each RES is
optimizing a design of urban settlements by the exploitation of necessary for confirming the actual load that is accommodated.
ordinary sources of energy like SPV, WTG systems and developments The electrical energy generation by the alternative sources such as
of building management. From the level of generation station to the WTG and SPV, have developed more attractive attention and are widely

⁎
Corresponding author at: School of Electrical and Electronics Engineering, Huazhong University of Science and Technology, Wuhan 430074, Hubei, China.
E-mail address: [email protected] (S.M. Dawoud).

http://dx.doi.org/10.1016/j.rser.2017.08.007
Received 25 April 2016; Received in revised form 26 May 2017; Accepted 2 August 2017
Available online 08 September 2017
1364-0321/ © 2017 Published by Elsevier Ltd.
S.M. Dawoud et al. Renewable and Sustainable Energy Reviews 82 (2018) 2039–2052

Table 1
Characteristics of renewable system.

Advantages Disadvantages

Renewable energy is well renewable resources free like the sun and the wind as a fuel. RES relies heavily upon the climate for sources of supply: rain, the wind, and
the sunshine,…etc.
Economic Benefits: RES is cheaper in fuel consumption, and Operation and Maintenance (O & Large Capital Cost: Initial investments are quite high in case of building RE
M) Requirement is low plants.
Environmental Benefits: Difficult to produce a large amount of energy as those created by coal stations
No pollution or waste natural resources
Reliable Energy Source: solar and wind plants are distributed over all geographical region and To meet up the large amounts of electricity produced by fossil fuels, large
climate disruptions in one region will not cut off power to any area. amount of SPV panels and WTG farms need to be set up

utilized for replacing fossil fuels in the process of electrical energy since ever necessary. This paper will present a new trend of the world energy
the 1970s because of the crisis oil. Nevertheless, such alternative scenario, and develop the restraint on the energy load demand of the
energy units have a slow development and the conversion into a new future with appropriate references. Also, the energy microgrid perfor-
phase of evolution in the power generation sector seems to be complex mance is usually evaluated by long series of time for climate or load
tasks because of different insights of problems. The environmental and demand profile which record the increase in the method calculus
economic issues not only sufficient reasons but there are other reasons complexity. Several mathematical models like probabilistic approaches
such as social and psychological impacts on people's behavior. [19–21], the Artificial Neural Networks (ANNs) [21,22], the Genetic
Moreover, new directions of technology for the energy productions Algorithms (GAs) [22–24]and the Particle Swarm Optimization (PSO)
from alternative sources have tested a considerable development to [25,26] have been suggested to deal with the multi-FCs, non-linear
improve the microgrid efficiency and electric power generation relia- characteristic response of models components and long time series for
bility nets in alternative sources which consist of the practice of climate variables.
information and communications technology. In this way, the electrical The aim of this study is to perform a brief overview of the present
company is developed to be more dynamical and flexible and support state of methodologies which is used for sizing HRES with the energy
distributed storage units [13]. However, as a stochastic nature in some storage units for both standalone and grid connected systems.
of alternative energies such as SPV and WTG, the transition into low This paper is organized into 12 sections. Section 2 delivers the
carbon society will require a non-single solution. hybrid RES architectures. Section 3 contains the world energy state.
Fossil fuels can be replaced by means of alternative energy Section 4 provides the kinds of small power networks that includes two
resources, the specialists forecast that will be integrated and essential types grid connected and the remote operation. Section 5 presents
for multi-HRESs which are working together such as hydro, geother- HRES that contains SPV, WTG, hydropower station and biomass
mal, Biomass (BM), WTG, SPV, hydrogen and nuclear at key part of energy. Section 6 involves in the future of HRES. Section 7 presents
energy generation and customer level in reorganized RES [9,10,14]. the mathematical modelling of HRES. Section 8 contains the HRES
When the energy system contains two energy supplies or more, it is expansions planning methods. Section 9 presents HRES optimization
called HRES. Sometimes these systems present the lowest costs and the techniques. Section 10 provides the best design criteria of HRES.
highest reliability compared to the other systems which use only one Section 11 includes the issues with HRES. Finally, Section 12
renewable source. Therefore, the penetration of HRES into the energy concludes the conclusions of this paper.
market rest on main methodologies which can be utilized for designing
these kinds of HERS in an optimally way. The optimization problems
are discussed to select the “best” set of system components from a 2. Hybrid RES architectures
search space or set of possible solutions. This includes a selection of
one optimization problem or more, an Objective Function (FC), and The general HRES construction is demonstrated in Fig. 1. This type
problem constraints. of energy systems is called “hybrid” because they include two RESs or
The FC and constraints should be functions of single optimization more to feed a required electrical load, and a commonly AC load. It may
variable or more. Additionally, the problem growths its complexity due also supply a DC load or both at the similar time. Energy sources may
to the non-linear characteristic answered in the system components, be renewable, conventional or energy storage units. In this approach,
stochastic obtained ability in some RES (for example SPV and WTG), lacking some energy units are supplemented by strengthening the other
the considered constraints and optimization variables. Electric power sources in a natural or a controlled way. It can be demonstrated that
has the vital role for the economic and the personal HRES. A known- despite being some alternative sources (such as SPV and WTG)
engineered HRES has a good effect in cost, high reliable and improves a unpredictable availability, they present complementary patterns
quality of life [15]. In most cases, SPV and WTG complement each [27,28]. HRES can operate grid connected or standalone microgrid.
other; however, both of systems are unpredictable as instantaneous
fluctuation for wind speeds and solar radiation. The hybrid systems
contain SPV-WTG and hydropower were realized by references [16,17]
in the rural villages of Nepal. Reference [18] found out the character-
istics for an isolated microgrid HRES and their effects which concern-
ing with microgrid reliability. HRES is seriously reliant on its compo-
nents. Therefore, an accurate modelling of every component for HRES
delivers tools to recognize the performance of model in a good way, and
aid to optimize HRES.
This paper reviews comprehensively the optimization criteria and
various optimal sizing for HPS reliant on small SPV, WTG, hydropower
and battery devices. For understanding in a clear way, Tables have
displayed optimal designs and different optimization approaches. The
Mathematical models, flow charts and Figures also have added wher-
Fig. 1. General HRES architecture.

2040
S.M. Dawoud et al. Renewable and Sustainable Energy Reviews 82 (2018) 2039–2052

Fig. 2. World energy consumption, 1990–2035.

It can be used as grid-connected to feed the local energy demand and

supply the grid with any surplus energy, on the other hand, it can be Fig. 4. Energy sources in the US in 2015.
used as a standalone system for making energy autonomously for the
grid utility in the rural regions. The HRES contains SPV or WTG,
auxiliary sources of energy such as battery units, fuel cells or backup
diesel engine can be considered. So, it is aforementioned to bypass a
stochastic availability of these energy resources [29]. The energy
sources availability are very important for reaching to the balance, in
the other ways, the control units which select any energy sources will
supply the electric loads. For evaluating the performance of HRESs,
there are several indexes used in the previous years. Some of them are
presented in section ten.

3. World energy state

A consumption of the world energy is rising about 2.3% for every

year from Energy Information Administration [30]. In Denmark,
Fig. 5. Renewable energy sources in the US in 2015.
WTGs supply about 20% of the total electricity, Globally, WTGs supply
less than 1% of the total electricity, but WTs are a fast growing energy
grow gradually to 15% of the whole generation in 2040.
sources. It is predicted that the world's energy consumption is raised by
In Figs. 4 and 5, the energy generations in the US using various
53% from the year 2008–2035 [30]. Fig. 2 shows the strongest growth
sources are shown. Global demands of RE are continuing to growth
of an overall world energy consumption. In Fig. 2, it can be noticed that
during 2015, providing an estimated 13% of whole final energy
although RE share will rise by few percent, the fossil fuels such as
consumption in 2014. In 2015 globally, WTG accounted about 34%
natural gasses and coal are still the dominate. The overall economy
of RE capacity. It can be noticed that the main share of the electricity
growth depends on climate variations whether the growing energy
produced by RESs in 2015 fed hydroelectric stations by (48%), then
demands can be fed or not. In the world, a fossil fuel is not uniformly
WTGs (34%), BMs (12%), geothermal (3%) and SPVs (3%).
distributed, and local or worldwide conflicts may occur energies crisis if
Additionally, it is demonstrated that 2015 was the first year that the
a global economy is seriously dependent on them. Nowadays the
wind, solar, and another hybrid non-hydro RESs produced more
conventional fuels are used, the global environments have been
electricity per year than the amount generated by hydropower stations.
harmfully affected, and an environmental of the approximate areas
have been damaged severely [31]. Fig. 3 displays the growth of WTG
and SPV variations from 2000 to 2040. In 2013, due to the increase in
output of WTGs and SPV sources, the hydropower RE was almost equal
to hydroelectric generation. In 2040, the output of hydropower RESs 4. Kinds of small power system (SPS)
will account more than 2/3 of the total RE. The total RE share of all
electricity generation growths from 13% in 2013 to 18% in 2040 and as SPS is used for supplying rural areas with power. The increase in
much as 22% in 2040 in the highest fuel cost. Owing to the reduction in small grids has grown in parallel by low prices in SPV, WTG and the
natural gas costs and the rising of diesel costs, the RE shares of the power inverter technologies. Depending on the connections of power
whole electricity generation produce more output but it is predicted to networks to the grid, SPS can be classified to the grid-connected
networks and the isolated power networks. Fig. 6 shows every type of
the power networks kinds in details.

Fig. 3. Renewable electricity generation by fuel type, 2000–2040. Fig. 6. Classification of SPS depend on grid connection.

2041
S.M. Dawoud et al. Renewable and Sustainable Energy Reviews 82 (2018) 2039–2052

Fig. 7. Schematic diagram of HRE systems.

4.1. Stand-alone power systems of Load Probability (LOLP). In this reference, the optimum solution
could be gotten if SPV contributed by 75% of required energy. A
Most of all the SPSs which are planned for feeding rural areas by comparison study between SPV and WTG for their reliability effect on
the power demands are isolated grid systems. Standalone systems have microgrid in a rural area in Egypt is conducted by [10]. This work
not connected to the grid connected. Standalone systems strongly vary showed the WTG which assistances the microgrids to improve its
in the sizing and application in rural areas. energy not fed by 2.92% and improves some interruptions by 1.57%
each year, but, the SPV power improves the interruption duration of
4.2. Grid connected power systems the model by 1.46% each year.
The possibility of using RE to decrease the consumption of fossil
The grid-connected systems are linked to bigger independent fuel to feed load demand of GSM base station in Ikwerre using HOMER
network typically a public grid [13] and supply directly the grid with have existed in [36]. Reference [37] has investigated the lessening in
energy. Feeding a grid by energy needs the conversion of DC to AC fuel consumption by using diesel engine and batteries in an SPV hybrid
power by the inverter. microgrid using HOMER software. The economics operation consider-
ing uncertainty for the microgrid with WTG, SPV generation, diesel
engine and storage units are obtained using PSO technique in [38].
5. Hybrid renewable energy system In [39] the authors analyzed the impacts of change SPV array area,
the battery units and WTG capacity of HRES. The author in [30]
A first village HRES containing SPV and a diesel generator was proposed a decision support method for policy makers about an
fixed in 1978 in the USA. A power resulted from this microgrid was influencing factor in a plan for grid connected SPV-WTG system.
utilized to provide electricity for washing machine, community refrig- This reference used analytically approach process to compute different
erator, pumps, and loads light until an electrical grid was protracted to parameters which cause confusion in the planning of HPS. The study
the loads in 1983 [32]. Nowadays, a combination of RESs is utilized in was dependent on the economic, political and social issues.
HRES. SPV, micro hydro and small WTG sources with storage units or A block diagram for HREs exists in Fig. 7. In this figure, the hybrid
without are extensively used to supply electrical loads with power in converter is used for connecting two sources (SPV-WTG). The battery
rural zones. The benefits of different alternative sources have various storage units are utilized to store the extra power and feed the load with
generation characteristics such as water flow in the river changes power in the case of insufficient power from hybrid microgrid. The
depending on the seasons. The solar radiations are greater in the inverter converts DC to AC voltages to feed the load requirement. The
summer than the winter and high in the day and non-radiation at night instantaneous variation in sun radiation and wind speeds character-
and wind speeds are similar where WT speeds are higher in summer in istics affect the energy produced thus an exact design is necessary for
some area. Two RESs or more can be combined into one system, reliable hybrid microgrid to supply loads under changing weather
dependent on a local RE possible. No emission results from many conditions. An accurate design also should be done to make the
HRES (like SPV, WTG, and hydro integrated). Modular (SPV and WTG microgrid cost low.
system) are simple to be fixed and in most cases needs to minor designs
for house load requirement. Small hybrid units are cheap than large
and complex system such as nuclear system. Fuels for HRES are free, 5.1. Solar photovoltaic system
copious and inexhaustible therefore the electrical energy resulted from
small hybrid units is not based on fuel prices in the region. Combining The solar panel converts solar energy coming from the sun directly
SPV and WTG has benefits for reducing a battery capacity (it is a high to a benefit electrical energy. The countries which have the biggest
cost) and the diesel requirements. Additionally, a good performance of individual SPV power stations are Agua Caliente Photovoltaic Project
the SPV-WTG system, possible sun radiation and WTG energy are the (Arizona, USA) and the California Valley Photovoltaic Ranch (USA).
necessity to the place. Environmental factors, SPV capacity, WTG These two power stations produce above 250 MWP [30]. As the prices
capacity, storage devices capacity, location and size of loads and of solar panels still high, using them limited to 1% of the electricity
generation site play a vital role in the operation, and price of hybrid generation worldwide. Owing to the intermittent nature of SPV, it can
SPV/WTG/diesel system. In reference [33] the authors reported that not produce energy all 24 h/day, so storage units such as battery
evaluation of the overall performance of SPV/WTG hybrid microgrid storages are used. Among systems fixed in 2011, the median stated cost
can be performed by the computer-modelling approach in MATLAB/ was approximately 6.13 $ per Watt for small commercial applications
Simulink software program. The technique used to plan the optimal up to 10 kW, and about 4.87$/W for commercial applications higher
placing of multi-SPV based distributed generators in microgrids than 100 kW [40]. SPV system is described as an efficient solution for
showed in [34]. The authors in reference [35] presented the mathe- price which can provide the rural areas with energy supplies. The
matical modelling for sizing the hybrid SPV units considering the Loss Economics viability of SPV system for decentralizing generation have

2042
S.M. Dawoud et al. Renewable and Sustainable Energy Reviews 82 (2018) 2039–2052

been achieved and has verified its effectiveness for villages with about then to electrical power by a generator. The first hydroelectric station in
100 families. A size of SPV-hybrid microgrid contains a backup diesel 1882 is placed in Wisconsin, which produces 12.5 kW. The hydro-
generator have been shown in [41]. The starting and the stopping electric power generations over the world have increased gradually
threshold of the backup generator were evaluated with related to a about 3% per year for the last four decades. In 2011, about 16% of a
battery unit capacity. The mathematical model using FORTRAN global electricity had been advanced by the hydropower from more
language to estimate the least number of the SPV arrays and storage than 160 countries. Countries such as Norway, Bhutan, Paraguay,
days considering the preoperating time of diesel units for hybrid SPV Ethiopia, Egypt, Venezuela, and Nepal are getting most of their energy
systems was shown in [42]. A technique to obtain the optimal produced through hydropower [30]. In constant unpredictable and
combinations of SPV and battery unit to feed the load demand has quickly fluctuating of SPV and WTG, hydropower has an extended
been established in [43]. The load and insulation were found out by seasonal cycle. The water passing through the streams and rivers varies
statistical models. The closed procedure solution method to the slowly based on seasons of a year. Therefore, the needs for energy
calculation of loss of power supply probability (LPSP) of the standalone storage units are not required. Hydropower was the greatest utilized
(SPV-battery storage) hybrid microgrid has been considered in [44]. In form of RESs accounting for 16% of the whole electricity production in
the iterative optimization techniques of hybrid SPV system, the optimal 2010.
combinations can be proposed on the basis of the price of electricity
produced that is justified on a root of expansion from nearing power 5.4. Biomass system
line, azimuth and tilt angle. The performance of hybrid SPV structure is
calculated on the base of the sources reliability under extensively Biomass (BM) is the remnants of all organic materials which have
fluctuating load conditions. A technique for calculating the capacity of been recycled to be used for energy generations. There are three main
SPV arrays, batteries, and WTGs with minimum price has been configurations of BM resources which are represented in agricultural
suggested in [7]. In reference [45], the authors evaluated a battery unused from crop remains biodegradable wastes from industrial mud,
capacity in grid-connected SPV system. dung and waste of food, and forest bioenergy resources. Up till date,
BM is still used as a simple source of energy by varieties in the
5.2. Wind turbine system traditional methods of the developing countries. Through making great
efforts towards transferring energy from non-renewable energy to
Nowadays, new techniques have been established to harness the renewable energy, BM is considered one of the popular sources of
required amount of energy by using WTG. The Previous utilization of energy used. Technical and economic feasibility of BM is still under
WTG was to move the boat and pumping water. The primary windmill proving conditions in different countries all over the world. The
utilized for producing the electricity was building in 1887 in Scotland. changeability of power generation schemes is available using BM for
The WTG design is often tailored to specific place characteristics with a generation. BM is exposed to produce heat or steam for generating
low wind speeds having large rotors and high wind speed places having electricity through other operating energy resources such as biodiesel,
small rotors. The limitation of the mechanical loads is continuously bioethanol, BM pellets and bio-methane gas. The most current trend in
developed in high wind conditions at any place so that the maximum some countries contains a combination of heat and power on con-
wind speed will require a small rotor. Many WTGs have been designed structing or community scale for electricity or district heating applica-
in variable pitch or variable speeds to control loads. In a winter of a tions. Using BM for energy generation still depends widely on index
year (1887), Charles was the first one who use the wind-powered factors like an efficiency of the microgrid, availability of the resource,
machines to produce electricity [46]. To use the hybrid WTG economic- social appropriateness, emissions factor, land use strategy, technology
ally and effectively, the selective location should have the worthy and economic costs of the grid. There are some restrictions for example
potential of WTG energy through the year. Nowadays, WTG is coupled economic adversity of land and water practice coupled with soaring
with large and small WTG for different kinds and structures. It has a amount of unsubsidized RE in many developing countries of the world
fast-rising of alternative energy units. Unlike of solar powers, WTGs which have affected the level of diffusion and adoption of new
have a long operating time, and they also can generate power during bioenergy structures practice. However, the diversity of BM resources
the night and cloudy days. Europe only produces approximately about and energy extraction methodologies accounts the vitality in RE
35,000 MW of the electricity using the wind power generation. When application and decarburization of the power sector. Biogas is con-
the wind speeds are low, these WTGs can not generate power. During sidered one of the RES which can be gotten from the group of BMs with
the duration of no wind power generation, other sources are needed to a biodegradable characteristic. For instances, such of these sources are
serve the load. Therefore both the wind and the solar require a storage waste dung which is resulted from local animals, human feces, waste-
device to store excess energy and utilize it when there is insufficient water, a biogenic element of municipal hard waste and food waste from
power generated to feed the load demand. In contrast, the comforting restaurants. Biogas is methane ironic explosive chemicals with typical
probabilities are that consumers can produce their energy to feed the great energy rate. In the past, works have been continued in several
daily load demands by building small SPVs and WTG farms. In developing countries of the world to growth the level of biogas feeding
reference [47] the authors presented effect of windmill constraints on for energy construction. In remote regions of developing countries,
a capacity factor (CF) based on extended period wind turbine speed common of the inhabitants like farmers who could help the community
data. CF of WTG is used for selecting a particular kind of WTG at the participation in the production of bio-feedstock for power generation.
chosen location as a necessary component of the hybrid WT system. A Sympathetically, the fact that the energy density of crop residues is
closed procedure solution method for calculating LPSP of standalone certainly quite small, and most of the power generations using these
WTG with storage units was obtainable by authors in [48]. Authors in residues are typically operated in small-power scale. Therefore, the
[49] suggested a simulation method generating probability index by conclusion explained that the hybrid system involved in BM using
Monte Carlo simulations which helped to determine appropriate WTG agricultural residues could be important for the vicinity of rural areas
penetration in the existing system considering reliability and economic where the resources are available.
aspects.
6. Future of HRE systems
5.3. Hydropower system
In the distributed hybrid networks, power is generated or closed to
Water wheel is the predecessor for the modern turbine in the the stage of the production. Distributed energy sources decrease the
current period utilized to convert a hydraulic power to mechanical and high costs and power loss of a transmission and a distribution network.

2043
S.M. Dawoud et al. Renewable and Sustainable Energy Reviews 82 (2018) 2039–2052

Fig. 8. A diagram showing the steps of this paper.

Therefore, there are requirements to select locations for setting up

SPVs and WTGs, then linking them with a grid to decrease the prices of
electricity without occurring a disturbance of the present network.
RESs play an essential role in energy demands for several countries in
the world.
Nowadays, costs of RES technologies especially WTGs and SPVs
continue to fall, occurring RESs competitive to the other traditional Fig. 9. SPV cell model.
energy technologies gradually. The most main things which concern
with the hybrid units are the system reliability during changing are also common. The model that contains two-diode SPV model was
environmental conditions and the total cost of networks. Many authors shown in [22].
attempted to optimize one of these things or more. Fig. 8 shows the ⎛ ⎛ v + Rs I ⎞ ⎞ v + Rs I
bubble diagram for the field of this article. I = Ip − Io ⎜exp ⎜ ⎟ − 1⎟ −
⎝ ⎝ A ⎠ ⎠ Rsh (1)

7. A mathematical modelling for the HRES ⎛ ⎛ qv ⎞ ⎞

I = Ip − Io ⎜exp ⎜ ⎟ − 1⎟
⎝ ⎝ kT ⎠ ⎠ (2)
The HRES consists of RE conversion systems like WTG, SPV array,
hydro turbine and other traditional generators like the diesel generator, Ns nkT
A=
microturbine and storage systems like batteries and hydrogen storage q (3)
tank. The HRES can use all or parts of these units. For being the sizes
of individual components for the microgrid accurate, it is necessary to Where, I and I0 are an output current in (A) and an inverse saturation
run the simulations of the scheme under the active operating condi- current respectively. Ip is the photocurrent in (A). v is an output voltage
tions such as appropriate weather, insulation, wind speeds, and and Rs is the series resistance. q is an electron charge that equal to
electrical loads. The components of HRES are connected together to 1.6×10–19 C, k is Boltzmann constant, T is SPV temperature (°C), n is a
obtain the optimal system configuration. The plan of hybrid microgrids usual ideality factor, Rsh is the parallel resistance (Ω) and Ns is a
will be dependent on several factors like location, off grid or grid number of SPV linked in series.
connected DC loads or AC loads. Power generated of SPV is resulted from multiplying the current by
Usually, all of the distributed systems are planned to feed power to the voltage. An input power to SPV model is diffuse and normal sun
families or small societies for simple electrical usage as a basic house, irradiation which dependent on a position of the sun and a time of a
street lighting, and small home applications, etc. The main objective of year. The whole irradiation of SPV calculated by:
utilizing the HRESs is to get the maximum proportions of RESs. Other It = Ib Rsh + Id Rd + (Ib + Id ) Rr (4)
factors include the financial investments and power systems reliability.
Where, Ib and Id are a normal radiation and a diffuse solar irradiation
The initial points in the optimization of HRES performance are the
of SPV, Rd and Rr are tilt diffuse and tilt factors of reflected solar
modelling of every component alone. Comprehensive mathematical
irradiations [51]. Hourly SPV output power PSJ in kW that an area ASPV
modelling of HRES includes small hydro, SPV, and WTG systems
in m2 [21]. The whole solar irradiations of ITJ in kW/m2 are incident on
which have been shown in [50]. The models of HRES such as the SPV,
SPV surface.
WTG, diesel units and battery units are explained below:
PSJ = ITJ ASPV μ (5)
7.1. SPV array modelling Where, µ is an efficiency of SPV. The Photovoltaic output power is
mainly related to the light strength and the ambient temperature, while
The most public model utilized for representing the energy con- light intensity is affected by shadows, clouds and others factor which
struction of SPV modelling is one diode circuit as illustrated in Fig. 9 presents a certain level of randomness. This equation considers only an
and a single-diode SPV cell without parallel resistance are defined as effect of light intensity and ambient temperature on solar cell output.
ideal single diode modelling [19–21,50]. Dependent on an availability
PSJ = PSTC GC [1 + K (TC − TSTC )]/ GSTC (6)
of parallel resistance (Rsh), an output current in single diode SPV cells
2
model are showed using Eqs. (1)–(3). However, two-diode SPV cells Where, GC is the light intensity at working point，GSTC=1 kW/m ; PSJ

2044
S.M. Dawoud et al. Renewable and Sustainable Energy Reviews 82 (2018) 2039–2052

is the output power of photovoltaic generation, PSTC is a rated power of

photovoltaic generation, TC is a surface temperature of battery,
TSTC=25 °C; k is a temperature parameter. A simplified modelling of
SPV is proposed by the references [23,24] presented as:
Voc
voc =
NkT / q (7)

⎛ V ⎛ V ⎞⎞
⎜ NkToc/ q − Ln ⎜ NkToc/ q + 0.72⎟ ⎟ ⎛ ⎞ ⎛ ⎞γ
⎜ ⎝ ⎠ ⎟⎛ Rs ⎞ ⎜ Voco ⎟ ⎜ TO ⎟
Fig. 10. Output curve of ideal WTG power.
Pmax =⎜ ⎟⎝⎜ 1 − ⎟ *
1 +
Voc
Voc / ISC ⎠ ⎜⎝ 1 + βLn Go ⎟⎝ T ⎠
⎠
⎜ NkT / q ⎟ G
Each wind turbine model has a specific power curve that can
⎝ ⎠
determine the wind speeds and power that is a function of WTG speeds.
⎛ G ⎞α
ISCO ⎜ ⎟ Usually, they generate power from 4 m/s wind speed to 25 m/s. As
⎝ Go ⎠ (8) wind speed varies in a day, generated power is varying in a day. The
Where, voc is a normalized voltage value in a stage of an open-circuit power becomes higher by rising wind speed from 3.5m/s to 11 m/s;
VOC with related to a thermal voltage which Vt = N k T/q. N is an then it still constant for wind speed which is between 11 m/s and
ideality factor and a value of N between one and two. T is SPV module 25 m/s, after that, it reduces suddenly to zero to protect WTG from the
temperature, α is a factor accountable for nonlinear effects that mechanical damage [8]. The WTG output is related to wind speed, as
photocurrent dependent on it; β is an SPV module technology WTG converts wind mechanical energy into electrical energy. A
specific-related dimensionless coefficients and γ is a factor which mechanical energy obtained by the WTG blades is combined to the
taking all nonlinear temperature voltage impacts. For an actual system, power generator. The power output PWT (kW/m2) from WTG is
a current and voltage are typically scaled up by connecting SPV calculated by:
generating sources in parallel and series to SPV array. A peak output ⎧ 0, V ≤ Vin , V > Vcout
⎪
power of SPV array is expressed by Eq. (9). PWT (t ) = ⎨ A + B + CV 2, Vin < V ≤ Vcout
⎪P ,
PAr = Ns *Np *Pm (9) ⎩ r Vr < V ≤ Vcout (13)

Where, Ns is SPV arrays linked in series, Np is SPV arrays connected in Where, A, B, C are the power characteristic curve parameters of WTG,
parallel and Pm is the SPV module power. slightly different for different wind turbines, Pr is WTG rated power.
Vin, Vr, Vcout are the cut-in, rated, and cut-out WTG speed and the
7.2. Wind turbine modelling values of these speeds are 4 m/s, 14 m/s, and 24 m/s respectively. The
curve between the wind speeds and the output power production of
The output of WTG at a certain region dependent on WTG speed WTG exists in Fig. 10.
design and the values of wind speeds at a hub height. Eq. (10) estimates P = PWin AWin η (14)
the value of wind speed vs at a hub height.
Where, Awin is a swept area, η is an efficiency of WTG and PWin is the
h x
vs = vi [ ] rating of WT.
hi (10)
Where, vi and vs are wind speeds with heights hi and h respectively. x is 7.3. Engine generator modelling
power laws exponent that is a function of atmospheric and stability in a
layer. x is estimated to be suitable and underlying surface character- Diesel generator contains three main components like diesel
istics. For instance, the value of x in open land equals 1/7 [52]. The engines, synchronous AC generator, and control system. It works like
main equation for calculating a mechanical output power of WTG is a voltage stability maintainer during the isolated micro-grid operation.
written as: When the wind or solar resources are lacking, diesel generators
ρA increase its output to compensate the power shortage. During a time
Pm = Cp (λ , β ) VW 3 t, the relationship between the output power of a single diesel engine
2 (11)
and its fuel consumptions can be written by the next formula below.
Where, Pm is a mechanical output power of WTG in Kw, ρ and A are an
air density in kg/m3, and the WTG swept the area in m2 respectively, Vfuel (t ) = a + bPengine (t ) (15)
Vw and ß are wind speed with a unit m/s and pitch blade angle in deg. λ
Where, the unit of Vfuel(t) in (L/h), a is a diesel generator at no-load
is tip speed ratio of rotor blade which equals a tip speed to a wind
fuel consumption in (L/h), and b is the slope of the relationship curve
speed. Cp is a performance coefficient of WTG. The value of λ is written
between the diesel fuel and output power (L/kW h).
as:
ωr
λ= 7.4. Modelling of Battery Banks
V (12)
Where, ω and r are the angular velocity and the radius of WTG (m) The battery storage units are backup energy storage units which are
respectively. Fig. 10 displays characteristics of WTG output power at designed accurately to feed the electric loads in times of insufficient
rated wind speed. Cut in speed is a small wind turbine speed of WTG energy from HRES. There are some factors which affect the battery
first rotate and start to produce power. Cut-out speed is the highest sizing, for example, the capacity of battery storage, battery lifetime and
wind turbine speed at the forces on WTG configuration which is very temperature correction. When the distributed power generation in
high so there are dangers of damaging the wind rotor when wind micro-grid is sufficient, the batteries charge; when power generation is
turbine speed increases more than the cutout speed. To prevent lacking, the batteries discharge. Thus they represent a fundamental
damage, the braking systems are engaged to reach the wind rotor to role in a standalone microgrid operation. According to the energy
a standstill. Rated output speed is a speed between cut-in and cut-out conservation theorem, a charging and a discharging of the energy
speed where output power approaches the maximum limits in which storage units at any known time are calculated with the equations
the electric generators are sufficient to feed the loads. below.

2045
S.M. Dawoud et al. Renewable and Sustainable Energy Reviews 82 (2018) 2039–2052

Table 2
Kinds of the optimization techniques.

Study Techniques Energy Sources Ref. Comments

Graphical Construction Battery storage and SPV [52,58,60–62] This method used for two parameters.
Deterministic Approach Stand-alone SPV with [63,64] Based on statistical approach of data collection
battery storage
Alternative Approach Hybrid SPV with WTG Based on LOLP to get the mixture of Hybrid SPV with WT system and depend on the FC and the
1. Hill Climbing [27,61,65–67] constraint of the system.
2. DP [68]
3. LP [69,70]
4. Multi-FC [69,71]
Artificial Intelligent Hybrid SPV-WTG [72–76] Real-coded GA depends on the mechanism of natural selection and natural GA Based on proposed
1. GA technique and FC of the system.
with battery storage [77,78] PSO can effort the problems of scattering and optimization. Considered probabilistic parameters
with generating training
4. PSO [79] Input-output data for an uncertain function. In hybrid method, optimization ability can be simply
5. FL [80,81] done.
6. ANN [82,83]
7. Hybrid of AI
Homer Software All RE element [7,53,84–86] HOMER model micro power systems with single or multiple sources of energy: SPV, WTG, BM
Diesel and other Cogeneration Microturbines Batteries.

Eb, t = Eb, t −1 (1 − σ ) + [Eg, t − EL, t / μinv ] μbat (16) equations that utilize sets of algorithmic steps, for example, SPV/WTG
microgrid for getting a better optimal size and operation. There are
Where, Eb,t and Eb,t-1 are the energies stored in batteries in an hour (t)
numerous parameters which are considered such as WT type, WT
and in a previous hour respectively. EL,t is a load demand in an hour (t).
capacity, SPV modules, the best tilt angle, minimum system capital
µbat and µinv are efficiencies of battery and an inverter respectively. σ is
cost, replacement cost, O & M cost, the reduction of the fuel consump-
the self-discharge rate for each an hour. When the electric loads are
tion considering the reliability requirement and CO2 emission limit, …,
greater than the existing energy generated, the battery storages will be
etc. as presented in Section 9.
discharged. The discharging of an energy storage battery can be
estimated as [53]:
8.3. Enumeration
Eb, t = Eb, t (1 − σ ) − [EL, t / μinv − Eg, t ] (17)
When the calculated energies generated and stored in batteries are Programming algorithm investigates several kinds of techniques to
not enough to feed the electric load demand for an hour t, this deficit get a solution for FC and an expansion policy for the given time. There
called the loss of power source (LPSs) for an hour t calculated as [53]: are many computational methods and processes for reaching to the
optimized solution as stated in Section 9.
LPSt = EL, t − [Eg, t + Eb, t −1 − Eb min ] (18)
LPSP for a period T is a ratio of all LPS values for this period to the 9. HRES optimization techniques
sum of a load in the similar period and it is defined by [53]:
The good-designed simulation programs permit to obtain the
LPSP = pr {Eb, t ≤ Eb min, for t ≤ T (19)
optimal sizing for the battery banks, SPV array, WTG, Hydropower
Where, Eb,t is energy stored in batteries at an hour t, Ebmin is a generations and other systems for an independent or grid connected
minimum storage battery permissible energy level. Let Eb Max is a HRES to the required load and the desired LPSP which dependent on
nominal battery capacity with the rate of Battery State of Charge (SOC) numerous criteria. The LPSP is a measurable probability of electrical
as: loads which have been fed. The loads will be supplied when the energy
produced from RE and the energy supplied from a battery are greater
Eb min = (1 − DOD ) Eb max (20)
than a required energy to feed the loads. In an ideal system where the
Where, DOD is the depth of a discharge equal 30–50% loads are fed by the supply LPSP equal zero. On the other hand, the
system where loads are never fed, LPSP equal one. Some of this criteria
8. HRES expansion planning methods are classified as the minimum system costs, system capacities, number
of battery units, LPSPs, LOLPs and maximum power generations. The
This paper also reviews the hybrid generation expansion planning different optimization techniques like the graphical construction,
for the isolated and grid-connected microgrids which classified as: [38,53–57] probabilistic approach, iterative technique, AI, dynamic
programming (DP), linear programming (LP) and multiple-objective
8.1. Reliability analysis were studied by the researchers to improve the hybrid SPV/WTG
model. Table 2 displays details of the optimization methods used by
Reliability analysis of the microgrid gathers all the constraints and different techniques.
limitations which are applied in the microgrids. This design to satisfy
the balance between electric sources and electric loads. There are 9.1. Graphical construction method
several factors for the reliability evaluation which are related to the
probability of the imbalance between supply and electric load. The In the Graphical Construction, a problem of the two plan variables
main factors are LPSP, Expected Energy not Supplied (ENNS), LOLP, is solved graphically by observing how it varies from one to other. The
COE and Life Cycle Costs (LCCs) as presented in Section 10. constraint functions have been plotted on the same chart. Through
visual inspections of a feasible area, the optimum point on a graph is
8.2. Optimization known after drawing the FC contours. In [58] the authors used the long
series time of solar radiation, so the optimal sitting was obtained by the
It contains methods of FC representation in the optimization of superposition of the contributions from climate cycles of a low solar

2046
S.M. Dawoud et al. Renewable and Sustainable Energy Reviews 82 (2018) 2039–2052

irradiation per day. In [59] the authors showed a method of optimal solution for an optimization problem till termination criterions are
size for hybrid SPV/WTG system. The performances of hybrid SPV/ achieved. If a number of the optimization variables increases, the
WTG microgrid was determined dependent on an hour by making the estimated time exponentially increases by using this method. The
WTG capacity constant. A yearly LOLP with various capacity of SPV authors in reference [67] has used this method to optimize SPV-WTG-
array and a battery storage was intended, and optimum configurations battery HRES dependent on a minimization of LCC.
(costs and LPSPs) were obtained by drawing the tangent to a trade-off
curve. 9.5. AI techniques models

9.2. Probabilistic approach technique AI is a branch of the computer sciences that improves the AI
software and machines. AI includes many branches like ANN, Fuzzy
In this approach, randomness is a current method which is Logic (FL), GA and hybrid techniques that contain two of this branches
dependent on the collected data. Therefore, the variable state is not or more. The suitable use of AI technologies primes to appropriate
known for the individual values, but rather it is preferred to use one of systems with a good performance of AI or other features that may not
the statistical tools. The optimal size of SPV/WTG hybrid system is be suitable for traditional techniques.
computed per an hour. Hybrid SPV/WTG can also be evaluated per
month during the days which SPV has minimum solar radiation and 9.5.1. Optimization of GA
low wind power. Two benefits of this technique are that costs of a GA is classified as the global search heuristics. It is a particular class
system and the load's data collections are minimal. of the evolutionary algorithms which employs techniques stimulated by
the evolutionary biology for example mutation, selection, and combi-
9.2.1. Hourly average generated capacity technique nation. A typical GA needs both variables to be known:
An hourly average data of wind, insulation, and power demands are
used to obtain the optimal sizes of a system. These calculations are 1. A genetic demonstrate for the solution domains.
reliant to an average yearly, a month data of the winds and sun. The 2. Fitness Function (FF) to estimate the solution domains.
following equation gives the FC:
GA may be useful for problem domains which have the complex FF
FC = CC + CM (21)
landscape with the traditional hill climbing method which may fail.
Where, CM and Cc are an annual maintenance cost and capital cost Fig. 11 shows a flow chart for genetic algorithm optimization process of
respectively. The difference power (ΔP) between (PGen) power genera- HRES. In [76] the authors used GA and the elitist strategies for
tion and power consumed by the load (PDem) reduced to have a balance optimum sizing of stand-alone hybrid SPV/WTG system per a year
over the time. (8760 h). Their major objectives were decreased the total initial capital
ΔP = PGen − PDem (22) system costs with limited LPSP.

9.5.2. Optimization of fuzzy logic

9.2.2. Most unfavorable monthly method The FL established models are highly used because it maps the
The unfavorable wind speeds and sun irradiation month are exact situation to the best level. This method is used when the experts'
evaluated dependent on the existing data in reference [87]. Based on reaction is a fuzzy nature. Many rounds are showed among profes-
this approach, the sizes of SPV and WTG is evaluated in most of an sionals to reach a consensus. In fuzzy regression, the data for
unfavorable month. The size A1 in (m2) of SPV or WTG component is dependent and non-dependent variables are captured in a fuzzy
estimated by: manner, and the resultant regression equation estimates the effects
⎛ ELoad , M ⎞ of the independent variables on the dependent variable. Fuzzy gray
A1 = max ⎜ ⎟ prediction is similar to regression approach where the fuzziness are
⎝ EI , M ⎠ (23) used to present the old area in the variables considered for the
Where, M is a month of a year and the value of M between 1 to 12, dependency prediction. Fuzzy AHP and ANP are utilized for finding a
ELoad is a load of a particular month. EI,M is a monthly energy resulted relative status of the variables. Fuzzy approach aids in correctly
from (SPV or WTG) system per unit area. Eqs. (24) and (25) express capturing the fuzziness in the attentions of people while ranking the
the whole energy gotten from the generators (SPV, WTG) and fed the variables, the range is given to the variables for helping clearly in
load: defining the clusters and draw boundaries. These approaches are used
with related to a problem domain. For the forecast purposes, fuzzy
∑ E1 A1 = ELoad (24) regression and fuzzy gray prediction are utilized. Fuzzy AHP and ANP
are utilized for obtaining a relative importance of energy resources.
E1 *A1 = F1 *ELoad (25)
Fuzzy clustering is used for collecting resources which are dependent
Where, F1 is a fraction of load fed by a generator 1. on selected criteria such as price, obtainability, pollution,…, etc. These
techniques categorized as ‘simple’ considering its complexity which
9.3. Deterministic approach method involved in the technique, and they are utilized for prediction or grade
the significance of electrical utility.
Based on this technique, each set of the variable state is individually
evaluated by a parameter in a model and puts the value of these 9.5.3. Optimization of ANN
variables with a value of earlier states. So, there is continuously a ANN is a structured group of AI neurons that practices the
unique optimal solution for the known parameters unlike of the mathematical models for processing information dependent on the
probabilistic approach. The authors [87] have evaluated the system connectionist method for making the computation. In reference [88],
sizes and prices of SPV installation system in Nepal. the authors applied the ANN-based method for using preventive
control approaches for big HRES. ANNs are an important part which
9.4. Iterative approach method are better than common statistical methods in the dynamic safety
pattern class and also calculate the level of safety. The reference [89]
An iterative method is a mathematical technique performed using proposed ANN control strategy for multi-energy common DC bus
the computer that computed the sequence of approximate refining hybrid power supply by examining the distinctiveness of SPV, WTG. In

2047
S.M. Dawoud et al. Renewable and Sustainable Energy Reviews 82 (2018) 2039–2052

Fig. 11. Flow chart of typical GA containing of SPV-WTG - hydro models.

Table 3 scheme by accounting the initial, replacement, operation & mainte-

Methods of the optimal sizing of HRES. nance, fuel, and interest costs. The hourly energy flows through each
component, and annual cost can be obtained. In [91], Hrayshat makes
Optimization Based on FC Ref.
the detailed techno-economical analysis by HOMER to plan an
Economy Capital Cost [92,93] optimum hybrid SPV-diesel-battery system model in Jordan for the
Annualized Cost [66] rural houses. Further information for optimizing methods employed in
Reliability LPSP [84,85] HRES. Table 2 presents the detailed references of every type of the
EENS [94]
LOLP [95]
optimization methods.
Techno-Economic Analysis System Cost [36,96]
Yearly-Monthly-Hourly [61] 10. The design criteria of HRES
Average Method
Most Unfavorable Month Method [87,97]
Electrical Load Utilizing of HRES sizing is obtaining the cheapest combinations of
1. Constant Load [97] all sources size (renewables and conventional generators) and a storage
2. Variable Load [98] battery that will encounter the anticipated demands load with a less
satisfactory level of safety. For optimal HRES designing, researchers
have adopted some design criteria dependent on the economic,
reference [90], the author improved a hybrid model for an hourly
reliability and electrical loads. Table 3 shows the details of optimum
forecast of the SPV-WTG RES and used the computational intelligence
HRES plan criteria. In this section, normally plans criteria used which
of PSO for calculating different definitions of the forecast error.
are briefly shown.

9.6. HOMER software 10.1. Cost of energy

HOMER aids us to plan off-grid and grid-connected systems. In the hybrid microgrid, the lifetime of a plant is supposed to be
HOMER can do analyses to solve a range of the project questions: 25years, and a storage battery is normally changed every five years. The
COE depends on depreciation periods, initial, operation & maintenance
• What are the sizes of components should be? costs and energy generated in a year. Eq. (26) evaluates the energy cost
• Which are systems cost most operative? of any system.
• What will occur to the scheme's economics if the prices or the loads
R
vary? CE = CcA * *COM
• Is the RES adequate?
ETOT (26)
Where, CE and CCA are an energy cost and the initial cost for the HRES
HOMER software of the National RE Laboratory (NREL)/USA is respectively, R is a yearly discount rate of initial costs, ETOT and COM
the general commercial software used for planning and examining are the whole energy produced and annual operation & maintenance
HPS. Solar insulation, electrical load, hybrid microgrid technical costs respectively. Many researchers have widely used the price of
details, prices, constraints, controls, and kind of the strategy dispatch electricity generated by a hybrid to estimate HRES configuration at a
have been used as an input to HOMER software. It does energy balance predefined of LOLP. COE has been computed which is important for
computations for every one hour per year. HOMER software compares the calculation of HRES. COE in $/kWh is written as:
the electrical and the thermal loads in every hour to energy that a
CTOT
system can feed in this hour. For systems that contain storage batteries, COST / kwh ($) =
ETOT (kwh ) (27)
HOMER selects for every an hour how to control the units and whether
the storage battery can be charged or discharged. If the system feeds Where, CTOT is the total yearly costs incurred to generate the total
the loads for the entirely year, HOMER estimates the total cost of a yearly electricity generation ETOT in kWh. HOMER estimated COE as

2048
S.M. Dawoud et al. Renewable and Sustainable Energy Reviews 82 (2018) 2039–2052

in reference [84]. To calculate COE, HOMER software divides annual- charge of a battery, the energy is not utilized for the productive
ized costs of generating electricity by a whole useful energy generation microgrid.
as known in the next equation: 2. During deficiency of power generations by HRESs, the stored energy
COE = Can, tot /(Eprim + Edef ) in storage units is used for supplying the electric load demand and
(28)
the new SOC is estimated till the battery units capacity descents to a
Where, Can,tot is the whole annualized cost, Eprim and Edef are the whole minimum level till hardware disconnects battery units from the
primaries and deferrable load respectively that the system can feed electric load. Dependent on SOC of battery storages, LPSP evaluated
electrical loads per a year. The denominator of Eq. (28) is an expression by:
of the whole amount of suitable energy which the system can generate
LPSP = pr {Eb, t ≤ Eb min, for t ≤ T (34)
yearly.
Where, Eb(t) is a battery energy stored for any time (t). A probability of
10.2. Life cycle cost SOC at any time t in the period T is fewer or equivalent to the low level
of supplied energy in a storage system Ebmin.
The whole life cost of HRES signs to the whole costs incurred
through the prolific lifetime of a system. LCC has been utilized to select 10.4. Loss of load probability
the system economic feasibility. The system which has the minimum
LCC is chosen. LCC is evaluated without considering system deprecia- The LOLP is an ability of HRES to feed a maximum load per day.
tion [99]. LOLP happens when the electric loads exceed the existing generation
T capacities. The total probability that will cause a loss of power is named
Ck LOLP considering the days per a year, hours in a day or the percentage
PV = ∑
K =1
(1 + I )k (29) of the time. Eqs. (35), (36) and (37) calculate LOLP [100].
Where, PV is a current value of system without deprecation, T N
∑T =1 HOUR (ISUP < INED (T ))
represents the analysis time, and I is an interest rate of one year. LOLP =
N (35)
LCC is a total cost of installing and operating a component or a system
over a definite time span, and it is usually several years. For the L (T ) − PW (T ) − PPV (T )
INED (T ) = μ (IBAT (T ))
economic estimation of the project, LCC is calculated as [84]: VL (36)

LCC = Initial Cost + O&M Cost + Fuel Cost + replacement Cost 0.2SOC SOC (T ) σ − SOCMIN
ISUP (T ) = MIN (IMAX = , )
ΔT ΔT (37)
− Salvage value (30)
Where, INED(t) is a current required to a load at an hour t, ISUP(t) is a
HOMER computes the capital initial, replacement, O & M and fuel
supply current of the HRES at time t. VL is nominal voltage necessary
costs, with the salvage values and any other costs for every component
for a system. N is some samples. L(T) is electric loads demand at time
to obtain the component's annualized cost. The operating cost is an
T. PW(T) is output power of WTG at time T. PPV(T) is power generation
annualized value of all costs such as replacement, fuel, O & M and
of SPV modules at time T. When the LOLPs are low, the costs of model
revenues other than initial cost. Eq. (31) calculates the operating cost.
will be high.
Cop = Can, tot − Can, cap (31)
10.5. Expected energy not supplied
Where, Cop is the operation cost, Can,tot is the whole annualized cost
and Can,cap is the overall annualized initial capital cost. HOMER
EENS is a probabilistic reliability index which is discussed in
comprises the overall costs and the revenues which occurred through
reference [94] due to the increase in the available capacity of load.
the project period. Additionally, all costs and revenues include the
According to load demand (L), and power generation from hybrid
initial costs of system components, the costs of replacing every
energy system (PH), EENS is written as:
component that happened within the project period, fuel cost, main-
tenance cost, and costs of buying power from the utility. The following ⎧ PHM
⎪ L > PHMAX L− ∫ PH *FPH (PH ) dPH
equation gives the NPC: ⎪ PHMI
EENS (L , PH ) = ⎨ P PHM
NPC = Can, tot / CRF(i, Rpro) (32) ⎪ HMI < L < PHM ∫PHMI (1 − PH )*FPH (PH ) dPH
⎪
⎩ L < PHMI 0 (38)
Where, Can,tot is the overall annualized cost, Rpro signifies to a project
lifetime, and CRF is the initial capital recovery factor that expressed by Where, PHM is a maximum power generation from hybrid energy
the next equation: system, PHMI is the minimum energy generation from HES and the
CRF = i (1 + i ) N /[i (1 + i ) − 1] (33) value of it is zero, and FPH(PH) is the probability density function for
the output power of HES.
Where, i and N are the annual real interest rate and the number of
years respectively. 11. Issues with HRE microgrid

10.3. Loss of power supply probability However the HRE microgrid has bundle of advantages, there are
some issues and problems related to HRE microgrid should be
The power source reliability is dependent on a concept of prob- mentioned:
ability of LPS from HRES which is not able to feed load demands. LPSP
equal one refers to that an electric load is never fed while LPSP equal 1. A majority of HRE microgrid needs storage units which the batteries
zero refers to that the electric load is continuously fed [84]. The are heavily utilized. These storage batteries need monitoring and
methodology of LPSP is discussed in the following two steps: replacing, so its cost will raise because the lifetime of the storage
batteries is limited to few years. It is informed that the battery
1. Through the extra power generation from HRES, surplus powers are storage lifetime would increase to around years for the economic use
used to charge the storage battery, and the new SOC is estimated in HRE microgrid.
until the storage capacity unit is full. Additionally, after a maximum 2. Owing to the dependence of RESs included the HRE microgrid on

2049
S.M. Dawoud et al. Renewable and Sustainable Energy Reviews 82 (2018) 2039–2052

climate results in the load demands sharing between the various International Conference on: IEEE; 2011. p. 703-8. 〈http://dx.doi.org/10.1109/
ICCEP.2011.6036357〉.
sources employed for power generation, the optimum power dis- [10] Beshr E. Comparative study of adding PV/wind energy systems to autonomus
patch, and the calculations of cost per unit generation are not easy. micro grid. Electric Power and Energy Conversion Systems (EPECS), 2013 3rd
3. The power generation reliability of HRE microgrid can be guaran- International Conference on: IEEE; 2013. p. 1-6. 〈http://dx.doi.org/10.1109/
EPECS.2013.6713072〉.
teed by incorporating independent weather sources of energy such [11] Barton JP, Infield DG. Energy storage and its use with intermittent renewable
as diesel unit or fuel cell. energy. IEEE Trans Energy Convers 2004;19:441–8. http://dx.doi.org/10.1109/
4. For the stability issue, a sudden variation in the output power from TEC.2003.822305.
[12] Dufo-Lopez R, Bernal-Agustín JL, Contreras J. Optimization of control strategies
some sources or a sudden variation in the load demand can clearly for standalone renewable energy system with hydrogen storage. Renew Energy
affect the microgrid stability because a power generation of numer- 2007;32:1102–26. http://dx.doi.org/10.1016/j.renene.2006.04.013.
ous power sources of HRE microgrid is comparable. [13] Kaundinya DP, Balachandra P, Ravindranath N. Grid-connected versus stand-alone
energy systems for decentralized power-A review of literature. Renew Sustain
5. Individual sources of the HRE microgrid should be worked at a point
Energy Rev 2009;13:2041–50. http://dx.doi.org/10.1016/j.rser.2009.02.002.
that gives the greatest sufficient generation. In fact, this will not be [14] Liu G, Rasul M, Amanullah M, Khan MMK. Feasibility study of standalone PV-
happened because that load demand sharing is often not related to wind-biomass hybrid energy systems in Australia. Power and Energy Engineering
the ratings of the power sources. Many factors elect load demand Conference (APPEEC), 2011 Asia-Pacific: IEEE. p. 1-6; 2011. 〈http://dx.doi.org/
10.1109/APPEEC.2011.5749125〉.
sharing such as the power source reliability, economy of use, the [15] Enslin JH, Wolf MS, Snyman DB, Swiegers W. Integrated photovoltaic maximum
required switching between the power sources, availability of fuel, power point tracking converter. IEEE Trans Ind Electron 1997;44:769–73. http://
etc. So, it is selected to estimate the schemes for raising the efficiency dx.doi.org/10.1109/41.649937.
[16] Bhandari B, Lee K-T, Lee G-Y, Cho Y-M, Ahn S-H. Optimization of a hybrid
to a high probable level. renewable energy power systems: a review. Int J Precis Engine Manuf-Green
Technol 2015;2:99–112. http://dx.doi.org/10.1007/s40684-015-0013-z.
12. Conclusions [17] Bhandari B, Lee K-T, Lee CS, Song C-K, Maskey RK, Ahn S-H. A novel off-grid
hybrid power systems comprised of solar photovoltaic, wind, and hydro energy
sources. Appl Energy 2014;133:236–42. http://dx.doi.org/10.1016/j.ape-
Nowadays, HRES is gradually common for rural areas electrifica- nergy.2014.07.033.
tion due to lower cost of SPV and WTG. It is supposed that it will be [18] Ahn S-H, Lee K-T, Bhandari B, Lee G-Y, Lee CS-Y, Song C-K. Formation strategy of
renewable energy sources for high mountain Off-grid system considering sustain-
helpful in supplying electricity for more of one billion populations who
ability. J Korean Soc Precis Eng 2012;29:958–63. http://dx.doi.org/10.7736/
are depressed of electricity in developing countries. Instead of the KSPE.2012.29.9.958.
single generation systems such as SPV or WTG, HRES can produce [19] Chouder A, Silvestre S, Sadaoui N, Rahmani L. Modeling and simulation of a grid
connected PV system based on the evaluation of main PV module parameters.
energy even the potential of one unit is minimal which increases the
Simul Model Pract Theory 2012;20:46–58. http://dx.doi.org/10.1016/j.sim-
power reliability. The present overall energies scenario and main pat.2011.08.011.
contributions of HRE microgrids have been systematically involved [20] Das D, Esmaili R, Xu L, Nichols D. An optimal design of grid connected hybrid
in this article. A compact mathematical modelling of several RES and wind/photovoltaic/fuel cell system for distributed energy production. Industrial
Electronics Society, 2005 IECON 2005. In: Proceedings of the 31st Annual
battery unit has been presented. The comprehensive review of optimi- Conference of IEEE: Ieee. p. 6; 2005. 〈http://dx.doi.org/10.1109/IECON.2005.
zation methods for HRE microgrids containing RES is described in 1569298〉.
details. Numerous optimization techniques and designs criteria are [21] De Soto W, Klein S, Beckman W. Improvement and validation of a model for
photovoltaic array performance. Sol Energy 2006;80:78–88. http://dx.doi.org/
discussed highly in details. The review involved in this study provides a 10.1016/j.solener.2005.06.010.
short yet sensibly which is a very worthy idea for tools that researchers [22] Salam Z, Ishaque K, Taheri H. An improved two-diode photovoltaic (PV) model for
have used within several decades for improving the HRE microgrid PV system. Power Electronics, Drives and Energy Systems (PEDES) & 2010 Power
India, 2010 Joint International Conference on: IEEE; 2010. p. 1-5. 〈http://dx.doi.
whatever the grid connected or not. Additionally, the presence of the org/10.1109/PEDES.2010.5712374〉.
scope diagram for the simplified visualization of this article contents [23] Scheffler M, Colombo P. Cellular ceramics: structure, manufacturing, properties
presents an obvious visualization. Both two tables contain the details of and applications. John Wiley & Sons; 2006.
[24] Markvart T. Solar electricity. John Wiley & Sons; 2000.
the optimization approaches and HRE microgrids dependent on [25] Habib M, Said S, El-Hadidy M, Al-Zaharna I. Optimization procedure of a hybrid
several significant criteria which have been done seriously by all photovoltaic wind energy system. Energy 1999;24:919–29. http://dx.doi.org/
researchers. 10.1016/S0360-5442(99)00042-0.
[26] Evans D. Simplified method for predicting photovoltaic array output. Sol Energy
1981;27:555–60. http://dx.doi.org/10.1016/0038-092X(81)90051-7.
References [27] Yang H, Lu L, Burnett J. Weather data and probability analysis of hybrid
photovoltaic–wind power generation system in Hong Kong. Renew Energy
[1] Eastin J, Grundmann R, Prakash A. The two limits debates: “limits to growth” and 2003;28:1813–24. http://dx.doi.org/10.1016/S0960-1481(03)00015-6.
climate change. Futures 2011;43:16–26. http://dx.doi.org/10.1016/j.fu- [28] Tina G, Gagliano S. Probabilistic analysis of weather data for a hybrid solar/wind
tures.2010.03.001. energy system. Int J Energy Res 2011;35:221–32. http://dx.doi.org/10.1002/
[2] Al-Saleh Y. Renewable energy scenarios for major oil-producing nations: the case of er.1686.
Saudi Arabia. Futures 2009;41:650–62. http://dx.doi.org/10.1016/j.fu- [29] Celik AN. Techno-economic analysiss of autonomous PV-wind hybrid energy
tures.2009.04.005. systems using different sizing methods. Energy Convers Manag 2003;44:1951–68.
[3] Roth S, Hirschberg S, Bauer C, Burgherr P, Dones R, Heck T, et al. Sustainability of http://dx.doi.org/10.1016/S0196-8904(02)00223-6.
electricity supply technology portfolio. Ann Nucl Energy 2009;36:409–16. http:// [30] Rahman S. Economic impact of integrating photovoltaics with conventional electric
dx.doi.org/10.1016/j.anucene. 2008.11.029. utility operation. IEEE Trans Energy Conv 1990;5:422–8. http://dx.doi.org/
[4] Bouffard F, Kirschen DS. Centralised and distributed electricity systems. Energy 10.1109/60.105264.
Policy 2008;36:4504–8. http://dx.doi.org/10.1016/j.enpol.2008.09.060. [31] Wang C. Modeling and control of a hybrid wind/photovoltaic/fuel cell distributed
[5] Castillo-Cagigal M, Gutiérrez A, Monasterio-Huelin F, Caamaño-Martín E, Masa D, generation systems. Montana State University-Bozeman, College of Engineering;
Jiménez-Leube J. A semi-distributed electric demand-side management system 2006.
with PV generation for self-consumption enhancement. Energy Convers Manag [32] Bifano W, Ratajczak A, Bahr D, Garrett B. Social and economic impact of solar
2011;52:2659–66. http://dx.doi.org/10.1016/j.enconman.2011.01.017. electricity at Schuchuli Village; 1979.
[6] Järventausta P, Repo S, Rautiainen A, Partanen J. Smart grid power system control [33] Nehrir MH, LaMeres BJ, Venkataramanan G, Gerez V, Alvarado L. An approach to
in distributed generation environment. Annu Rev Control 2010;34:277–86. http:// evaluate the general performance of standalone wind/photovoltaic generating
dx.doi.org/10.1016/j.arcontrol.2010.08.005. systems. IEEE Trans Energy Conv 2000;15:433–9. http://dx.doi.org/10.1109/
[7] Dawoud SM, Lin XN, Sun JW, Okba MI, Khalid MS, Waqar A. Feasibility study of 60.900505.
isolated PV-wind hybrid system in Egypt. Adv Mater Res: Trans Tech Publ [34] Dawoud SM, Xiangning L, Flaih FM, Okba MI. PSO algorithm for optimal
2015:145–51. http://dx.doi.org/10.4028/www.scientific.net/AMR.1092-1093.145. placement of multiple SPV based distributed generators in Microgrids. Power and
[8] Celik A. Optimisation and techno-economic analysises of autonomous photovol- Energy Engineering Conference (APPEEC), 2016 IEEE PES Asia-Pacific: IEEE. p.
taic–wind hybrid energy systems in comparison to single photovoltaic and wind 125-9; 2016. p. 125–9. 〈http://dx.doi.org/10.1109/APPEEC.2016.7779483〉.
systems. Energy Convers Manag 2002;43:2453–68. http://dx.doi.org/10.1016/ [35] Notton G, Muselli M, Louche A. Autonomous hybrid photo voltaic power plant
S0196-8904(01)00198-4. using a back-up generator: a case study in a Mediterranean Island. Renew Energy
[9] Kusakana K, Vermaak H. Hybrid Photovoltaic-Wind system as power solution for 1996;7:371–91. http://dx.doi.org/10.1016/0960-1481(96)00016-X.
network operators in the DR Congo. Clean Electrical Power (ICCEP), 2011 [36] Aris AM, Shabani B. Sustainable power supply solutions for off-grid base stations.

2050
S.M. Dawoud et al. Renewable and Sustainable Energy Reviews 82 (2018) 2039–2052

Energies 2015;8:10904–41. http://dx.doi.org/10.3390/ en81010904. Trans Energy Conv 1998;13:70–5. http://dx.doi.org/10.1109/60.658206.

[37] Bhandari Y, Chalise S, Sternhagen J, Tonkoski R. Reducing fuel consumption in [67] Li J, Wei W, Xiang J. A simple sizing algorithm for standalone PV/wind/battery
microgrids using PV, batteries, and generator cycling. Electro/Information hybrid microgrids. Energies 2012;5:5307–23. http://dx.doi.org/10.3390/
Technology (EIT), IEEE International Conference on: IEEE; 2013. 〈http://dx.doi. en5125307.
org/10.1109/EIT.2013.6632692〉. [68] De A, Musgrove L. The optimization of hybrid energy conversion systems using the
[38] Sobu A, Wu G. Optimal operation planning method for isolated micro grid dynamic programming model—Rapsody. Int J Energy Res 1988;12:447–57.
considering uncertainties of renewable power generations and load demand. http://dx.doi.org/10.1002/er.4440120309.
Innovative Smart Grid Technologies-Asia (ISGT Asia), 2012 IEEE; 2012. p. 1-6. [69] Chedid R, Rahman S. Unit sizing and control of hybrid wind-solar power systems.
〈http://dx.doi.org/10.1109/ISGT-Asia.2012.6303271〉. IEEE Trans Energy Conv 1997;12:79–85. http://dx.doi.org/10.1109/60.577284.
[39] Chedid R, Akiki H, Rahman S. A decision support technique for the design of hybrid [70] Morais H, Kadar P, Faria P, Vale ZA, Khodr H. Optimal scheduling of a renewable
solar and wind power systems. IEEE Trans Energy Convers. 1998;13:76–83. micro-grid in an isolated load area using mixed-integer linear programming. Renew
http://dx.doi.org/10.1109/60.658207. Energy 2010;35:151–6. http://dx.doi.org/10.1016/j.renene.2009.02.031.
[40] Feldman D. Photovoltaic (PV) pricing trends: historical, recent, and near-term [71] Chedid R, Karaki S, Rifai A. A multi-objective design methodology for hybrid
projections; 2014. renewable energy systems. Power Tech, 2005 IEEE Russia: IEEE; 2005. p. 1-6.
[41] Muselli M, Notton G, Poggi P, Louche A. PV-hybrid power systems sizing 〈http://dx.doi.org/10.1109/PTC.2005.4524339〉.
incorporating battery storage: an analysis via simulation calculations. Renew [72] Yang H, Wei Z, Chengzhi L. Optimal design and techno-economic analysis of a
Energy 2000;20:1–7. http://dx.doi.org/10.1016/S0960-1481(99)00094-4. hybrid solar–wind power generation system. Appl Energy 2009;86:163–9. http://
[42] El-Hefnawi SH. Photovoltaic diesel-generator hybrid power system sizing. Renew dx.doi.org/10.1016/j.apenergy.2008.03.008.
Energy 1998;13:33–40. http://dx.doi.org/10.1016/ S0960-1481(97)00074-8. [73] Yang H, Zhou W, Lu L, Fang Z. Optimal sizing method for stand-alone hybrid
[43] Shrestha G, Goel L. A study on optimal sizing of standalone photovoltaic stations. solar–wind system with LPSP technology by using genetic algorithm. Sol Energy
IEEE Trans Energy Conv 1998;13:373–8. http://dx.doi.org/10.1109/60.736323. 2008;82:354–67. http://dx.doi.org/10.1016/j.solener.2007.08.005.
[44] Abouzahr I, Ramakumar R. Loss of power supply probability of standalone [74] Koutroulis E, Kolokotsa D, Potirakis A, Kalaitzakis K. Methodology for optimal
photovoltaic systems: a closed form solution approach. IEEE Trans Energy Convers sizing of stand-alone photovoltaic/wind-generator systems using genetic algo-
1991;6:1–11. http://dx.doi.org/10.1109/60.73783. rithms. Sol Energy 2006;80:1072–88. http://dx.doi.org/10.1016/j.sol-
[45] Ru Y, Kleissl J, Martinez S. Storage size determination for grid-connected ener.2005.11.002.
photovoltaic systems. IEEE Trans Sustain Energy 2013;4:68–81. http:// [75] Mohammadi M, Hosseinian S, Gharehpetian G. GA-based optimal sizing of
dx.doi.org/10.1109/TSTE.2012.2199339. microgrid and DG units under pool and hybrid electricity markets. Int J Electr
[46] Price TJ. James Blyth—Britain's first modern wind power pioneer. Wind Eng Power Energy Syst 2012;35:83–92. http://dx.doi.org/10.1016/
2005;29:191–200. http://dx.doi.org/10.1260/030952405774354921. j.ijepes.2011.09.015.
[47] Salameh ZM, Safari I. The effect of the windmill's parameters on the capacity factor. [76] Xu D, Kang L, Chang L, Cao B. Optimal sizing of stand alone hybrid wind/PV power
IEEE Trans Energy Conv 1995;10:747–51. http://dx.doi.org/10.1109/60.475848. systems using genetic algorithms. Electrical and computer engineering, Canadian
[48] Abouzahr I, Ramakumar R. Loss of power supply probability of standalone wind conference on: IEEE; 2005. p. 1722-5. 〈http://dx.doi.org/10.1109/CCECE.2005.
electric conversion systems: a closed form solution approach. IEEE Trans Energy 1557315〉.
Conv 1990;5:445–52. http://dx.doi.org/10.1109/60.105267. [77] MoghaddasTafreshi S, Hakimi S. Optimal sizing of a standalone hybrid power
[49] Karki R, Billinton R. Cost-effective wind energy utilization for reliable power system via particle swarm optimization (PSO). Power Engineering Conference,
supply. IEEE Trans Energy Convers 2004;19:435–40. http://dx.doi.org/10.1109/ 2007 IPEC 2007 International: IEEE. p. 960-965; 2007. 〈http://dx.doi.org/
TEC.2003.822293. 10028197〉.
[50] Dalton G, Lockington D, Baldock T. Case study feasibility analysis of renewable [78] Boonbumroong U, Pratinthong N, Thepa S, Jivacate C, Pridasawas W. Particle
energy supply options for small to medium-sized tourist accommodations. Renew swarm optimization for AC-coupling stand alone hybrid power systems. Sol Energy
Energy 2009;34:1134–44. http://dx.doi.org/10.1016/j.renene.2008.06.018. 2011;85:560–9. http://dx.doi.org/10.1016/ j.solener.2010.12.027.
[51] Duffie JA, Beckman WA. Solar engineering of thermal processes; 1980. [79] La Terra G, Salvina G, Tina G. Optimal sizing procedure for hybrid solar wind
[52] Borowy BS, Salameh ZM. Methodology for optimally sizing the combination of a power systems by fuzzy logic. Electrotechnical conference, MELECON 2006 IEEE
battery bank and PV array in a wind/PV hybrid system. IEEE Trans Energy Convers Mediterranean: IEEE. 2006. p. 865–8. 〈http://dx.doi.org/10.1109/MELCON.
1996;11:367–75. http://dx.doi.org/10.1109/60.507648. 2006.1653235〉.
[53] Soni VK, Khare R. Optimal sizing of HRES for small sized institute using HOMER. [80] Bilgili M, Sahin B. Comparative analysis of regression and artificial neural network
Electrical Energy Systems (ICEES), 2014 IEEE 2nd International Conference on: models for wind speed prediction. Meteorol AtmosvPhys 2010;109:61–72. http://
IEEE; 2014. p. 77-81. 〈http://dx.doi.org/10.1109/ICEES.2014.6924145〉. dx.doi.org/10.1007/s00703-010-0093-9.
[54] Johnston L, Díaz-González F, Gomis-Bellmunt O, Corchero-García C, Cruz- [81] O’Hair E, Giesselmann MG. Comparative analysis of regression and artificial neural
Zambrano M. Methodology for the economic optimisation of energy storage network models for wind turbine power curve estimation; 2001. http://dx.doi.org/
systems for frequency support in wind power plants. Appl Energy 2015;137:660–9. 10.1115/1.1413216.
http://dx.doi.org/10.1016/j.apenergy.2014.09.031. [82] Khatib T, Mohamed A, Sopian K. Optimization of a PV/wind microgrid for a rural
[55] Roy B, Basu AK, Paul S. Techno-economic feasibility analysis of a grid connected housing electrification using a hybrid iterative/genetic algorithm: case study of
solar photovoltaic power system for a residential load. Automation, Control, Energy Kuala Terengganu, Malaysia. Energy Build 2012;47:321–31. http://dx.doi.org/
and Systems (ACES), 2014 First International Conference on: IEEE; 2014. p. 1-5. 10.1016/j.enbuild.2011.12.006.
〈http://dx.doi.org/10.1109/ACES.2014.6808005〉. [83] Bajpai P, Dash V. Hybrid renewable energy system for power generation in stand-
[56] Gu H, Hu J, Pei X, Zhang Y. A probabilistic approach for traffic state estimation. alone applications: a review. Renew Sustain Energy Rev 2012;16:2926–39. http://
Intelligent Transportation Systems (ITSC). IEEE Proceedings of the 19th dx.doi.org/10.1016/j.rser.2012.02.009.
International Conference on: IEEE. 2016. ; 2016. p. 2595–600. 〈http://dx.doi.org/ [84] Dawoud SM, Lin X, Sun J, Mohsin QK, Flaih FM, Long P. Reliability study of
10.1109/ITSC.2016.7795973〉. hybrid PV-wind power system to isolated micro-grid. Intelligent Control and
[57] Guasch D, Silvestre S. Dynamic battery model for photovoltaic applications. Information Processing (ICICIP), 2015 Sixth International Conference on: IEEE;
Progress Photovolt: Res Appl 2003;11:193–206. http://dx.doi.org/10.1002/ 2015. p. 431-7. 〈http://dx.doi.org/10.1109/ICICIP.2015.7388210〉.
pip.480. [85] Dawoud SM, Lin X. Study of hybrid PV-wind energy system to isolated micro-grid.
[58] Markvart T, Fragaki A, Ross J. PV system sizing using observed time series of solar Indones J Electr Eng Comput Sci 2015;16:222–31. http://dx.doi.org/10.11591/
radiation. Sol Energy 2006;80:46–50. http://dx.doi.org/10.1016/j.sol- ijeecs.v16.i2.pp221-231.
ener.2005.08.011. [86] Kumar YP, Bhimasingu R. Optimal size of microgrid for an urban community
[59] Bin A, Hongxing Y, Hui S, Xianbo L. Computer aided design for PV/wind hybrid building in south India using HOMER. Power Electronics, Drives and Energy
system. Photovoltaic Energy Conversion. In: Proceedings of 3rd World Conference Systems (PEDES), 2014 IEEE International Conference on: IEEE; 2014. p. 1-6.
on: IEEE; 2003. p. 2411-4. 〈http://dx.doi.org/10.1016/8047650〉. 〈http://dx.doi.org/10.1109/PEDES.2014.7042059〉.
[60] Markvart T. Sizing of hybrid photovoltaic-wind energy systems. Sol Energy [87] Kusakana K, Vermaak HJ. Hybrid renewable power system for mobile telephon
1996;57:277–81. http://dx.doi.org/10.1016/S0038-092X(96)00106-5. base station in developing countries. Renew Energy 2013;51:419–25. http://
[61] Kelogg W, Nehrir M, Venkatarmanan G, Gerez V. Optimal unit sizing for a hybrid dx.doi.org/10.1016/j.renene.2012.09.045.
wind/photovoltaic generating system. Electr Power Syst Res 1996;39:35–8. http:// [88] Fidalgo J, Lopes JP, Miranda V. Neural networks applied to preventive control
dx.doi.org/10.1016/S0378-7796(96)01096-6. measures for the dynamic security of isolated power systems with renewables. IEEE
[62] Chedid R, Saliba Y. Optimization and control of autonomous renewable energy Trans Power Syst 1996;11:1811–6. http://dx.doi.org/10.1109/59.544647.
systems. Int J Energy Res 1996;20:609–24. http://dx.doi.org/10.1002/(SICI) [89] Jifang AL, Tianhao BT, Jingang CH. A neural network control strategy for Multi-
1099. energy common dc bus hybrid power supply. Power Electronics Electrical Drives
[63] Bhuiyan M, Asgar MA. Sizing of a standalone photovoltaic power system at Dhaka. Automation and Motion (SPEEDAM), 2010 International Symposium on: IEEE;
Renew Energy 2003;28:929–38. http://dx.doi.org/10.1016/S0960-1481(02) 2010. p. 1827-31. 〈http://dx.doi.org/10.1109/SPEEDAM.2010.5542240〉.
00154-4. [90] Quan DM, Ogliari E, Grimaccia F, Leva S, Mussetta M. Hybrid model for hourly
[64] Bhandari R, Stadler I. Electrification using solar photovoltaic systems in Nepal. forecast of photovoltaic and wind power. Fuzzy Systems (FUZZ), 2013 IEEE
Appl Energy 2011;88:458–65. http://dx.doi.org/10.1016/j.apenergy.2009.11.029. International Conference on: IEEE; 2013. p. 1-6. 〈http://dx.doi.org/10.1109/
[65] Yang H, Lu L, Zhou W. A novel optimization sizing model for hybrid solar-wind FUZZ-IEEE.2013.6622453〉.
power generation system. Sol Energy 2007;81:76–84. http://dx.doi.org/10.1016/ [91] Hrayshat ES. Techno-economic analysis of autonomous hybrid photo voltaic-
j.solener.2006.06.010. diesel-battery systems. Energy Sust Dev 2009;13:143–50. http://dx.doi.org/
[66] Kelogg W, Nehrir M, Venkatarmanan G, Gerez V. Generation unit sizing and cost 10.1016/j.esd.2009.07.003.
analysis for stand-alone wind, photovoltaic, and hybrid wind/PV systems. IEEE [92] Testa A, De Caro S, La Torre R, Scimone T. Optimal design of energy storage system

2051
S.M. Dawoud et al. Renewable and Sustainable Energy Reviews 82 (2018) 2039–2052

for standalone hybrid wind/PV generators. Power Electronics Electrical Drives 2008;33:366–82. http://dx.doi.org/10.1016/ j.renene.2007.03.013.
Automation and Motion (SPEEDAM), 2010 International Symposium on: IEEE; [97] Kulkarni CB. Design, Evaluation and Techno-Economic Analysis of a Hybrid Solar-
2010. p. 1291-6. 〈http://dx.doi.org/10.1109/SPEEDAM.2010.5542078〉. Wind Power Generations System in UAE: The British University in Dubai (BUiD);
[93] Kaviani AK, Riahy G, Kouhsari SM. Optimal design of a reliable hydrogen-based 2011.
standalone wind/SPV generating systems, considering components outages. Renew [98] Posadillo R, Luque RL. Approaches for developing a sizing method of standalone
Energy 2009;34:2380–90. http://dx.doi.org/10.1016/j.renene.2009.03.020. PV systems with variable demand. Renew Energy 2008;33:1037–48. http://
[94] Tina G, Gagliano S, Raiti S. Hybrid solar/wind power system probabilistic dx.doi.org/10.1016/j.renene.2007.06.004.
modelling for long-term performance assessment. Sol Energy 2006;80:578–88. [99] Ma T, Yang H, Lu L. Study on stand-alone power supply options for an isolated
http://dx.doi.org/10.1016/j.solener.2005.03.013. community. Int J Electr Power Energy Syst 2015;65:1–11. http://dx.doi.org/
[95] Deshmukh M, Deshmukh S. Modeling of hybrid renewable energy system. Renew 10.1016/j.ijepes.2014.09.023.
Sustain Energy Rev 2008;12:235–49. http://dx.doi.org/10.1016/ [100] Calabrese G. Generating reserve capacity determined by a probability method.
j.rser.2006.07.011. Trans Am Inst Electr Eng 1947;66:1439–50. http://dx.doi.org/10.1109/T-
[96] Beccali M, Brunone S, Cellura M, Franzitta V. Energy, economic and environmental AIEE.1947.5059596.
analysis on RET-hydrogen systems in residential buildings. Renew Energy

2052