Energy Conversion and Management 247 (2021) 114709

Contents lists available at ScienceDirect

Energy Conversion and Management

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

Integrated socio-environmental and techno-economic factors for designing

and sizing of a sustainable hybrid renewable energy system
Anil Kumar Singh Maisanam *, Agnimitra Biswas , Kaushal Kumar Sharma
Department of Mechanical Engineering, National Institute of Technology Silchar, Silchar 788010, India

A R T I C L E I N F O A B S T R A C T

Keyword: Hybrid renewable energy systems (HRESs) are penentrating into the distributed energy market, which is being
PV-BG-PHS system proved to be more economical than extending grid-electricity through long distance thereby incurring trans­
Land cost mission losses. But, the major concern is the accountability issues resulted from socio-environmental factors that
Carbon emission penalty
are still the hindrace to the sustainability of the HRES. In this paper, two major socio-environmental factors, viz.
Levelized cost of energy
Total net present cost
land cost and carbon emission penalty are integrated with the important techno-economic factors for designing
GOA and sizing of an HRES that comprises PV, biogas generator, and pumped hydro storage (PHS) system. The
PSO approach is applied for remote area electrification of an Indian town. The component level modeling of the HRES
is performed considering the ambient conditions of the location. In this work, for the first time GOA based meta-
heuristic optimizer is developed for solving the optimization problem. The simulation results obtained using GOA
are compared with the famous meta-heuristic PSO algorithm to understand the robustness of the former opti­
mizer in terms of computational efforts and statistical performance based on skewness and kurtosis of the total
net present cost (TNPC). It is found that the contribution of the socio-environmental factors in the present case is
6% of the TNPC as compared to 94% in regard to techno-economic factors. Comparision of the two optimizers
demonstrate shorter computational time and lower cost in respect of GOA. The higher kurtosis value of the TNPC
obtained through GOA optimizer indicates the capability of this optimizer to escape from local minima with a
greater capability to approximate the global optimum. From GOA, the optimal size of the HRES is obtained as
57.5 kW (341 m2) of PV system, 8 kW of BG plant, 378.4 kWh of energy storage and 33 kW of converter size. On
comparing the optimized cost of the present HRES with a very recent work under similar load condition, the
present design is found to exhibit lower value of cost of energy.

1. Introduction due to which there are economic and reliability concerns associated with
renewable energy systems. To overcome this, two or more sources are
Energy security plays an important role in improving the socio- combined in a single system such that they complement each other. Such
economic condition of people living in periurban and rural or remote systems are known as hybrid renewable energy system (HRES). HRES
areas [1]. Keeping note of this fact, electrification in such areas in India does not only improve the reliability but also improve overall efficiency
is going on at an aggressive pace under different government schemes to and thus makes the system cost effective [8]. Due to these advantages,
eliminate energy poverty [2,3]. Despite continuous efforts made, there there is a tremendous increase of its standalone application, and hence
still exist many areas that are deprived of electricity [4]. Various chal­ design and estimation of optimal component size have become an
lenges in the electrification of remote locations exist, such as large dis­ essential step to further improve reliability and make the system cost
tance from the national grid, inadequate infrastructure for transmission effective [9]. Even a reliable HRES can face problems from the
and distribution, difficult to reach terrains, low or sparsely distributed accountability issues resulted from the socio-environmental hindraces
population, low income level, etc [567]. One of the options to circum­ toward harness of renewable power. Thus these factors have attracted
vent these issues of power transmission would be the reliance on attention as much as techno-economic factors of the HRES. A successful
standalone renewable energy systems. HRES design can be obtained through optimizing its performance after
The major limitation of renewable energy is its intermittent nature integrating the socio-environmental and techno-economic factors.

* Corresponding author.
E-mail address: [email protected] (A.K.S. Maisanam).

https://doi.org/10.1016/j.enconman.2021.114709
Received 1 June 2021; Accepted 28 August 2021
Available online 10 September 2021
0196-8904/© 2021 Elsevier Ltd. All rights reserved.
A.K.S. Maisanam et al. Energy Conversion and Management 247 (2021) 114709

To obtain optimal system configuration various optimization tech­ considered economic (NPC) and environemental factors (CO2 emission)
niques have been utilized. The ability of an optimization method to to optimize HRES comprising of PV/WT/BG using HOMER pro software.
achieve global optimum is the most essential requirement followed by In this work again the financial implication of emission was neglected.
the rate of convergence [10]. The HRES design optimization is generally One more important factor is the social factor that has been generaly
performed through either using a software tool package such as Hybrid ignored in majority of the research works. In recent years, limited works
Optimization Model for Electric Renewables (HOMER), Hybrid2, have considered social and environmental factors for the optimization of
RETScreen, etc. or utilizing metaheuristic techniques such as genetic HRES design. The social factors such as job creration, human develop­
algorithm (GA), particle swarm optimization (PSO), ant bee colony ment index have been considered in the HRES design [30,31]. Although
(ABC), etc. Since, the HRES optimization is a non-convex and non-linear to the best of the authors knowledge, the estimation of vast area of land
optimization problem, metaheuristic techniques are more effective. required for installation of such project has not been not optimized yet.
Metaheuristics are faster than traditional optimization methods and also Land resource is one of the most important social factor in country like
provide better results [11]. Further, most of the software packages do India where land acquisition is difficult due to social pressure, therefore
not allow usage of user defined function that limits their applicability its optimization is essential. The importance of inclusion of social factor
when certain parameter is not in-built in the package. Therefore, due to is also highlighted by Cuesta et al. [32]. From the above literature, the
the complexity of the problem and flexibility of the metaheuristic consideration of social factor such as land utilization is till now ignored
techniques, they are now being used to solve HRES design problems. in most of the reported work. Also there are limited work that consid­
Along with the solution techniques, parameters considered for the ered CO2 emission and its economic implications on the sizing of the
optimization are also very important. Maleki and Askarzadeh [12] HRES.
optimized PV/Wind/Battery and PV/Wind/Fuel cell system considering Therefore, it is observed that numerous studies were conducted on
economic parameters, such as capital cost, maintenance cost and fuel the design and optimal sizing of HRES. Majority of the studies focussed
cost, though the parameters excluded important economic factors such on technical or economic aspect or both. However, the primary objective
as replacement cost. The economic optimization was performed using is to design a sustainable system, therefore it necessitates integration of
four different metaheuristic techniques (PSO, simulated annealing (SA), all the dependent factors that can result in the development of a sys­
tabu search (TS) and harmony search (HS) algorithm) and results were tematic approach towards designing a sustainable HRES. Hence, the
compared, which show that PSO algorithm is better as well as robust in conventional procedure of considering only economic and technical
solving HRES problem. Ant colony optimization algorithm was used to parameters in the design process is incomplete, as socio-environmental
size a solar, wind and battery system considering capital cost and issues become equally important along with techno-economic consid­
maintenance cost of the components [13]. Maleki and Pourfayaz [14] eration in terms of implementation of the HRES design in the practical
utilized HS algorithm to optimize a PV/Wind/Diesel/Battery system field. There will be concerns of acceptance or non-acceptance of a new
considering only the economic factors such as capital cost, maintenance technology by the people and also the environmental concerns that need
cost and the fuel cost. Further, the results also presented the amount of to be cleared so that there is no hindrance for HRES from becoming a
emission caused by the diesel system, but their implications on the reality. This is the research gap and also the motivation of doing the
overall system cost was ignorned in the study. Alshammari and Asu­ present work. Hence, the main objective is to design and optimally size
madu [15] designed HRES using three metaheuristic algorithms, namely an HRES compring of PV, biogas generator, and PHS by using meta­
HS, Jaya, and PSO considering techno-economic factors. The optimized heuristic techniques, considering integrated socio-environmental and
results show that the HS algorithm is faster and more efficient is HRES techno-economic factors. The adopted approach is applied for remote
sizing. Javed et al. [16] compared four metaheuristic techniques, area electrification of an Indian town.
namely ant colony optimization (ACO), firefly algorithm (FA), GA and The remaining paper is organized as follows. In Section 2, mathe­
PSO to explore their performance in solving HRES optimization problem matical modeling of all the system components is presented. The opti­
and found that the mean efficiency of FA and PSO is same, although PSO mization parameters such as the technical, economic, social and
gives better optimized system. Makhdoomi and Askarzadeh [17] environemental factors are introduced and discussed in Section 3. Sec­
investigated the techno-economic performance of two different HRES tion 4 presents the details of the study location, renewable resources
configurations i.e. PV/Diesel/PHS and PV/diesel/Fuel cell considering availability and the electrical load demand of the study location. The
fixed, single and two axis pv tracker. The results indicate that HRES is optimization algorithm is introduced in Section 5. In Section 6 the en­
not economic with tracking systems and it is also observed that pumped ergy management strategy of the present work is presented in the detail.
hydro storage (PHS) is cost effective and reliable. Kaabeche and Bakelli Finally the results obtained and their implications are discussed in
[18] explored the importance of the selection of proper electrochemical Section 7, and finally the concluding remarks with recommendations are
storage system in HRES design and also compared the performance of presented in the Section 8.
four optimization algorithms, i.e. ACO, grey wolf optimization (GWO),
krill herd optimization (KHO) and Jaya algorithm. Jamshidi et al. [19] 2. Hybrid renewable energy system components
performed a feasibility study considering three main criteria i.e. cost of
energy (COE), deficiency of power supply probability (DPSP), and car­ The HRES in the present study consists of 4 major components i.e. PV
bon dioxide emission, to understand the sizing of HRES without having system, Biogas system, converter and pumped hydro storage system.
detail renewable energy resource data. Various authors considered These components are connected together to complement each other
different scenarios, such as forced outage due to hardware failure [20], and meet the required energy demand. To improve the overall efficiency
state of health of storage systems [21], dispatch strategies [22], of the system, optimization of HRES is a necessary step in the designing
demand–supply energy management [23], etc. to design their HRES. process. The technical and economic characteristics of all the compo­
Most published literatures present techno-economic optimization of nents required in the optimization are presented in Table 1. Most of the
HRES [458 914152024–27]. However, one of the main reasons for data have been collected from literature. However, the economic data of
adoption of HRES is the environmental impact of conventional fuel. the PV panel have been collected from a local vendor through market
Investigation of environemental aspect of HRES system is important to survey. Further, the land requirement information of biogas generator is
understand the advantage of using HRES. Mayer et al. [28] in their work collected from Geon engineers Pvt. Ltd. 1250 (Ground Floor), Mukherjee
presented a multi-objective HRES optimization by considering envi­ Nagar,
ronmental as well as economic factors such as carbon footprint and net Delhi 110009, India. To optimize the system, the component level
present cost (NPC) using GA, although the financial implication of the mathematical modelling of the system is presented in the following
emission was ignored in the reported work. Jahangir and Cheraghi [29] section.

2
A.K.S. Maisanam et al. Energy Conversion and Management 247 (2021) 114709

Table 1 where, ηbg is the efficiency of the biogas generator, FBG (t) is the amount
Component characteristics of proposed HRES.
of fuel consumption (m3/h) and LHV BG is the lower heating value of the
Equipment Characteristics Value biogas (kWh/m3). The heating value of the biogas varies according to
PV Panel* Power rating 1 kW the type of fuel used in the plant. In this study, cow manure is considered
Derating Factor 88 % to be the fuel fed in the digester. The LHV of the cow manure is
Capital Cost 929 ($/kW) considered to be 21.78 MJ/m3, assuming methane content to be 60%.
Operation and Maintenance cost 4 ($/year)
Life Time 25 years
Land requirement 5.93 m2/kW 2.3. Pumped hydro storage system modeling
Biogas Generator Overall efficiency 21% [37]
Calorific value of Biogas 5.47 kWh/m3 [37] The pumped hydro storage (PHS) system is the storage unit consid­
Cost of fuel 5 ($/ton) [38]
Capital Cost 2300 ($/kW) [37]
ered in the present study. This system consists of 3 major components,
Operation and Maintenance Cost 2 ($/year) [37] pump, turbine, and upper reservoir unit. The modeling of these units has
Replacement Cost 2300 ($/kW) [37] been discussed in ref. [35].
Lifetime 15,000 hrs [37]
Area utilization* 8.33 m2/kW
2.3.1. Turbine modeling
Converter Inverter efficiency 95 % [22]
Lifetime 15 years [22] When the total generated power by the HRES is insufficient, the
Capital Cost 300 ($/kW) [22] potential energy of water, present in the upper reservoir, is utilized to
Operation and Maintenance Cost 3 ($/year) [22] meet the load demand. The power output of the turbine is given by:
Replacement Cost 300 ($/kW) [22]
Area utilization* 0.15 m2/kW Pturb (t) = ηturb ρghqturb (t) (5)
Pumped Hydro Storage Energy Cost 217 ($/kWh) [39]
Pump/Turbine cost 370 ($/kW) [16] where, ηturb is the conversion efficiency of the turbine, ρ is the density of
Operation and maintenance cost 3 ($/year) [39]
water (1000 kg/m3), g is the acceleration due to gravity (9.81 m/s2), h is
Roundtrip efficiency 80% [16]
Lifetime 60 years [16] the upper reservoir height (m) and qturb is the volumetric flow rate of
Area utilization 4.54 m2/kWh [39] water (m3/s).
*
Data collected from local vendors.
2.3.2. Pump modeling
In case excess energy is available from the HRES, it is utilized to run
2.1. Solar PV system modeling
the pump to transfer water from the lower reservoir to the upper
reservoir. The amount of water stored using the excess power at an
Energy conversion is necessary to convert any form of available en­
instant ‘t’ is given by:
ergy into a usable form of energy. Solar PV panels are the energy con­
version device, which are commonly utilized to convert the solar energy ηpump Pexcess (t)
qpump (t) = (6)
into electricity. To determine the power extracted through the PV panels ρgh
in ‘t’ hours, the following mathematical relationship is utilized [33]:
where, ηpump is the efficiency of the pump and Pexcess (t) is the excess
PPV (t) = ηpv (t) × Itilt (t) × Apv (1) power available at instant ‘t’ (W).

where, ηpv is the efficiency of PV panels, Itilt (t) is the incident radiation
2.3.3. Upper reservoir
(W/m2) on the tilted panel, and Apv is the effective surface area of PV The amount of water storage required is directly proportional to the
panel. The PV panels are sensitive to the operating cell temperature, duration for which the energy availability is lower than the load de­
therefore in this study temperature effect is taken into consideration mand. The available head of water is similar to the term state of charge
while estimating the PV power output. The relationship between the (SOC) in case of a battery storage. Therefore, SOC term is used in the
efficiency and the cell temperature is given by[34]: present paper to represent the level of the storage reservoir. The po­
ηpv (t) = ηrefpv (t) × β(Tcell (t) − Tcell STC ) (2) tential energy (in kW) stored in the upper reservoir is given by [36]:
ηturb ρVgh
where, ηref pv rated efficiency of the PV panel, β is the temperature co­ EUR (t) = (7)
3.6 × 106
efficient (%/◦ C), Tcell STC is the cell temperature at standard test condi­
tion and Tcell (t) is the cell temperature during operation and is given by where V is the volume of water stored in the upper reservoir, and ηturb , ρ,
[34]: g, h have the same meaning as explained in earlier section.
( )
NOCT − 20
Tcell (t) = Tamb + × Itilt (t) (3) 2.4. Converter system modeling
0.8 × ISTC
The HRES consists of both AC and DC equipment, which necessitates
where, Tamb is the ambient temperature, NOCT is the nominal operating
the use of a converter. The conversion efficiency of 95% is assumed
cell temperature and ISTC is the radiation at standard test condition
constant for the converter throughout its lifetime of 20 years and is
which is 1000 W/m2. The value of β and NOCT is usually obtained from
modeled as given below:
the PV panel manufacturer catelogue.
Pinv out = Ppv (t) × ηinv (8)
2.2. Biogas generator modeling
where, Pinv out is the output power of the converter, Ppv (t) is the PV power
The Biogas system is one of the best options for organic waste output at instant ‘t’ and ηinv is the efficiency of the inverter.
management. Organic matters undergo anaerobic decomposition in the
digester releasing the biogas, which is then utilized to produce elec­ 3. Techno-Economic-Socio-Environmental factors for HRES
tricity. The power output of the biogas generator is given by: performance

PBG = ηbg × FBG (t) × LHV BG (4)

The four important parameters identified in the design of HRES are

3
A.K.S. Maisanam et al. Energy Conversion and Management 247 (2021) 114709

the technical, economic, social and environemental factors. In the pre­ 3.2.3. Annualized replacement cost
sent problem, technical factor is represented by the reliability factor- The replacement cost is incurred when the life of an equipment is less
loss of power supply probability (LPSP); the factors like annualized than the project life. This investment must be included during the
capital cost, operation and maintenance cost and replacement cost are calculation of system cost. The total number of times an equipment
considered under the economic factors. The social factors include the needs to be replaced during the project life (N) is determined by Rno. =
( )
land cost required for the installation of the system, and the emission N
from the system is considered under the environmental factor, which is Lcomp − 1, where Lcomp is the component lifetime. The evaluation of

believed to responsible for the global warming and pollution. Since there replacement cost is performed using the Eq. (14):
are more than one parameter, the problem is a multi-objective pro­ ∑
gramming problem. Therefore, to simplify the problem all the parame­ Carep = Crep comp × EF (14)
ters are converted into a cost criteria. The Total net present cost (TNPC)
comp

of the system is evaluated using the capital recovery factor (CRF). Thus,
where EF is the Escalation factor and is estimated by:
by employing the optimization algorithms, the HRES design is optimized
( )x
to obtain the minimum TNPC subject to various constraints. ∑ 1 + IR
EF = (15)
( ) 1+i
Lproj
3.1. Loss of power supply probability (LPSP) x= j
Rno.
j = 1, 2, ..Rno.
LPSP is the reliability factor, it can take any value within the range of
0 to 1. A LPSP value of 0 implies that the system can meet demand all the
3.2.4. Annualized fuel cost
time and the LPSP value of 1 impiles that the system is unable to meet
The HRES consists of biogas generator, which requires raw material
any load. Hence it is the factor which store the instances when the
to generate biogas. To obtain the raw material, investment must be
system was unable to meet the load and also records the total amount of
made for continuous generation of electricity. To evaluate the total cost
deficit energy and it is mathematically expressed as:
incurred for procuring raw materials, Eq. (16) is utilized, which is given
∑8760 as:
DP(t)
LPSP = ∑8760 t=1 (9) ( )
t=1 Eactualload (t) × Δt ( ) ∑8760 Ebg (t)
Cafc = xbg × FCbg . t=1 (16)
ηbg × LHV bg
3.2. Economic factors
3.2.5. Land resource utilization
The economic factors are discussed below- The land is a scarce resource and its utilization for any type of
commercial activity requires the prior consent of the people living there.
3.2.1. Annualized capital cost It is one of the important factors, which is generally ignored in the
The annual overall system capital cost is estimated using Eq. (10): optimization problem but impacts the lives of the farmers earning their
[ ] livelihood through farming. The land area is locally measured in a unit
Cacap = Apv × Cpv + Pbg × Cbg + Ephs × Cphs + Pinv × Cinv × CRF (10) called Katha which translate into an area equal to 720 ft2. The price of 1
Katha is found to be $ 357, when converted into per m2 cost of land it
where Apv is the area of the PV in m2, Cpv is the cost of PV per m2, Pbg is
translates to $ 5.33/m2. Therefore, in this study Total land cost (TLC) is
the power of biogas generator (kW), Cbg is the cost of biogas generator included and is defined as the cost of land resource utilized for the
per unit kW, Ephs is the total energy capacity of the PHS (kWh), Cphs is the installation of all the components of the HRES.
cost of energy storage per unit kWh, Pinv is the rating of the converter
and Cinv is the cost of inverter per unit kW. CRF is the capital recovery Calc = (PVarea cost + BGarea cost + PHSarea cost ) × CRF (17)
factor which converts the initial cost of system into annual cost. It is
given by: 3.2.6. Carbon emission tax (CET)
n
CET can be defined as the amount that is penalized for producing
CRF =
ir (1 + ir )
(11) CO2 from the system. In the present system, the biogas system produces
(1 + ir )n − 1 CO2 during its operation. Hence, it is penalized for producing CO2,
which is given by:
i − IR
ir = (12)
1 + IR (Ebg × CEF × costpertonofCO2 emission × n)
CET = × CRF (18)
1000
IR is the inflation rate ir is the real discount rate, i is the nominal
discount rate, n is the lifetime of the project in years. where CEF is the Carbon emission factor (in kg of CO2 ), which is given
by Eq. (19):
3.2.2. Operation and maintenance cost
The cost incurred during the operation of the HRES and the regular CEF = (%CH4 × ρCH4 × 2.75 + ρCO2 (1 − %CH4 ) ) (19)
cost of maintenance is considered under operation and maintenance
cost. Complexity of each equipment is different, hence the operation and where %CH4 is the percentage of methane content by volume in the
maintenance cost also differs. It is thus necessary to minimize this cost biogas (60%), ρCH4 is the density of methane (0.65 kg/m3), ρCO2 is the
which will result in minimization of overall cost of the HRES. It is density of carbon dioxide (1.80 kg/m3), 2.75 represents the mass of CO2
evaluated using Eq. (13) : produced after complete combustion of 1 kg of methane [40]. The car­
( ) bon tax is assumed to be $ 0.015/kg of CO2 [19].
∑ ∑N (1 + IR)j
Caomc = OM Comp (13)
j=1 1+i 3.3. Total net present cost

where “Comp” represents all the components of the HRES, i = interest It is the present worth of the lifetime cost of the project, which in­
rate, IR = Inflation rate cludes all the costs that occur within the lifetime of the project for every
N= = project life system configuration. It is calculated mathematically by the following

4
A.K.S. Maisanam et al. Energy Conversion and Management 247 (2021) 114709

formula: access to grid-electricity, for their socio-economic development and the

overall development of the country, it would be very significant to map
Annualizedcostofthesystem(ACS)
TNPC = (20) the load demand of villages having houses, schools, and similar entities
CRF
with electricity generated from a sustainable HRES design. Such typical
where ACS comprises of the economic factors of system, i.e. the annu­ load demand can be comparable to similar load demands as will be
alized capital cost, system maintenance cost, component replacement viable in any rural and peri-urban locations in the world.
cost alongwith the biogas plant raw material cost (fuel cost), further the
socio-environmental factors such as the land utilization cost and the 4.3. Load profile
cabon emission tax is also included to it. It can be expressed as:
Two types of loads have been considered in this study i.e. residential
ACS = Cacap + Caomc + Carep + Cafc + Calc + CET (21) load and community load. Under the community load, load of a primary
school is considered. The monthly load demand of both the loads is
3.4. Levelized cost of energy presented in Fig. 1 and Fig. 2. From the observation of residential load
demand, four different types load patterns have been obtained. The
Levelized cost of energy is the cost per unit of energy. The equation of residential load demand is lowest during the winter months, i.e.
LCOE is as follows [41]: December to February. The demand increases during the months of April
to September. Fig. 1 presents the monthly load demand of a typical
(TNPC × CRF)
LCOE = (22) house in the studied location. Load demand of a school is considered
Totalenergygenerated
under the community load type. The variation of load due to summer
and winter vacation is taken into consideration and monthly load de­
4. Description of the study region mand curve is generated and presented in Fig. 2. It is found that for a
typical house in the studied village, the monthly average daily
The study region is described as below- maximum load demand for the months of January, February, March,
April, May, June, July, August, September, October, November, and
4.1. Location December is 8.54 kW, 8.04 kW, 14.06 kW, 19.68 kW, 18.64 kW, 19.04
kW, 19.46 kW, 20.21 kW, 19.16 kW, 14.41 kW, 13.81 kW, and 8.4 kW,
This study has been conducted in Borakhai tea estate at Silchar, a respectively. From the load diagram, it can be observed that the load for
northeastern town of India. It is located at a Northern Latitude of 24◦ 50′ the house peaks during the evening hours (mainly from 1800 hrs – 2200
2′ ’, Eastern Longitude of 92◦ 46′ 45′ ’, and elevation of 25 m above sea hrs). In case of school, the monthly average daily maximum load de­
level. As the region is nearly 155 km away from the Tropic of Cancer, it mand for the months of January, February, March, April, May June,
enjoys a significant amount of solar radiation round the year and the July, August, September, October, November, and December is 4.28 kW,
climatic condition is almost tropical by nature. As a result, it has a very 4.42 kW, 5.51 kW, 5.11 kW, 5.22 kW, 5.61 kW, 3.97 kW, 5.59 kW, 5.28
hot and humid summer and cold but dry winter. Monsoon generally kW, 4.51 kW, 4.46 kW and 3.99 kW, respectively. The load diagram of
starts from June and lasts until the end of September. the school reveals that the load is almost even throughout the year. The
highest load occurs during the day time (mainly between 0600 hrs –
4.2. Selection criteria of the location 1000 hrs). Hence, it is observed that the peak hour for a typical house
load and a typical school load occurs at different times. Also another
The criteria to select this study area fall within the scope of the characteristic to be observed is that the monthly loads as well as the
factors considered, i.e. technical, economic, social and environmental yearly average load for a typical house in the village are higher than the
factors. As in this study, an integrated approach of socio-environmental same for a typical school in the same village. These two different load
and techno-economic factors is used to design a sustainable hybrid demands have been combined together and a single hourly load demand
renewable energy system; it is needed to find a location where such a curve is synthesized with the help of HOMER pro software as performed
system can be studied with respect to the considered factors. In this in the ref [42]. The peak demand is found to be 31.86 kW and the
respect, a typical Indian village in rural location is selected that has average load demand per day is found to be 191.93 kWh/day. The day to
water resources, land resources and biomass resources in addition to day variability and the time step variability is taken as 10 % and 20 %,
solar resources. The studied village has all such resources in abundance. respectively, which is the default setting of the software. The hourly load
The villagers of the study area are mainly small farmers, tea estate demand curve is presented in Fig. 5.
workers, daily-wage based workers, etc., which is the characteristic of
any typical Indian village. In order to assess the socio-environment 4.4. Resource assessement
factors, there is nothing better than such a typical Indian village,
where there is a case of social acceptance of any design that is to be The HRES consists of solar and biomass as the primary source of
established on their farming land masses and that which will use the energy generation. Therefore, in this section the assessment of the
free-flowing water stream in the river to run the pumped hydro storage available energy resource is discussed.
system. Along with that the environmental pollution from burning of
biomass is another factor under environmental factor, which is to be also 4.4.1. Solar resource
considered. So to summarise, mainly these socio-environmental factors The solar resource analysis is essential before implementing a proj­
of a typical Indian village is the main criteria to select this study area. ect. The solar radiation data is collected from the solar radiation
However, techno-economic factors are to be always integrated to the resource assessment (SRRA) station located in the NIT Silchar, Assam,
socio-environmental factors as no design can sustain unless the same is India. The data station records the data for every minute throughout the
techno-economically designed. Irrespective of location or place, any year. Using the recorded data, further analysis is performed to extract
design should be techno-economically feasible. the solar radiation information. Fig. 3 shows the distribution of monthly
The above socio-environmental issues are of paramount importance average hourly solar radiation data. It is observed that the most suitable
in similar cases globally. Thus the methodology developed in this work period during a day is between 1000 hrs to 1300 hrs, during which
would be a new paradigm shift in the studies on HRES with most sig­ maximum average solar radiation is available. The overall monthly
nificant application being the other similar villages of the third world average solar energy is found to be 4.25 kWh/m2/day.
countries. As people living in rural places and villages have less or no The ambient temperature of the site is also recorded as the efficiency

5
A.K.S. Maisanam et al. Energy Conversion and Management 247 (2021) 114709

Fig. 1. Monthly Load demand of a typical house.

Fig. 2. Monthly load demand of school.

of the PV panels also depends on the operating temperature. Hence, essential. In the present study, cow dung is considered as the raw ma­
Fig. 4 depicts daily average ambient temperature for a year. It can be terial for the plant. The raw material is sourced from the location itself to
observed that the temperature ranges from a minimum of 17.2 ◦ C to a reduce the fuel cost. It is found that nearly 1000 kg of cow manure is
maximum of 32.2 ◦ C. The average temperature of the studied locaton is available daily. It is found that 15–25 kg of manure is required to pro­
found to be 25.9 ◦ C. Using the ambient temperature information, the PV duce 1 m3 of biogas. Therefore, total daily biogas availability is
panel cell temperature has been calculated using Eq. (3). Avalability of approximately 97.5 m3/day. The calorific value of biogas with 55%
actual site data is very important as it increases the accuracy, reduces methane content is taken as 20 MJ/m3 [40]. The conversion efficiency of
the risk involved in the viability of the project before its implementation the biogas generator is assumed to be 21% [43].
and also boost the investor’s confidence in such projects.
5. Description of the optimization algorithm
4.4.2. Biomass resource
The second primary resource used in the HRES is the biogas plant. Grasshopper optimization algorithm (GOA) is a nature-inspired
For proper functioning of the plant, availability of raw material is optimization algorithm developed by Saremi et al. [44]. Grasshoppers

6
A.K.S. Maisanam et al. Energy Conversion and Management 247 (2021) 114709

Fig. 3. Monthly average hourly Solar radiation.

Fig. 4. Daily average ambient temperature.

Fig. 5. Hourly load demand curve.

7
A.K.S. Maisanam et al. Energy Conversion and Management 247 (2021) 114709

are herbivorous insects and can form large swarms. The recent locust ⎛ ⎞
attack in Pakistan and India showed that the swarm size can be very ⎜ ∑ ⎟
ubd − lbd (⃒⃒ d ⃒)
large and is a serious threat to agricultural produce. This swarming ⎜ N ⎟
d ⃒ xj − xi ⎟
Xid = C⎜ C S ⃒x − x ⃒ ̂
⎟ + Td (25)
behavior is the motivation behind the development of this optimization ⎜
⎝j=1 2 j i
dij ⎠
algorithm. Swarming behavior is observed in both the nymph and
j∕
=i
adulthood. The two essential functions of the nature-inspired search
process are exploitation and exploration, both the characteristics are
where, Xid is the position of ith grasshopper with dth dimension, ubd and
naturally present in the grasshopper, which help in locating the global
lbd is the upper and lower boundary in the dth dimension, respectively.
optima. There is continuous communication between the grasshoppers
xdj and xdi are the positions of the ith and jth grasshoppers and dij is the
to find the food. Due to this particular network of interaction, their
interaction is separated into three zones i.e. attraction zone, repulsion normalized distance between them. C is the coefficient that helps in
zone, and comfort zone, which is defined by a social force function (S) as attaining the global optima by balancing the exploration and exploita­
given by [44]: tion. This coefficient reduces the comfort zone, which depends on the
current iteration and maximum iteration. The mathematical represen­
(23) tation of the parameter is given by Eq. (26) [44].
− r r
S(r) = fe l − e−
where f indicates the intensity of attraction and l indicates the Cmax − Cmin
C(t) = Cmax − t (26)
attraction length scale. In the present study, the value of f and l is tmax
assumed to be 0.5 and 1.5, respectively. To generate the ‘N’ search
In Eq. (25), there are two terms, the first term simulates the inter­
agents to initialize the optimization problem, the following equation can
be used [44]: action between the other grasshoppers and the second term, T ̂ d moves
th
the swarm towards the target i.e. food in the d dimension. The flow­
Gi = (ubi − lbi )rand(1, N) + lbi (24) chart of the GOA is presented in Fig. 6.
There are mainly two unique features of GOA. First, each search
where, Gi is the artificial grsasshopper at ith position and i = 1, 2, 3⋯ agent is defined completely by only one position vector. Second, the
dimension. updation of the position of a search agent (Eq. (25)) is performed by
To simulate the swarming behavior of the grasshoppers, which act as considering its current position, global best and position of all other
search agent, it is mathematically modeled as given below [44]: search agents [45]. However, in case of PSO, which is considered to be
one of the best swarm based algorithm, the search agent requires two

Fig. 6. Grasshopper optimization algorithm flowchart.

8
A.K.S. Maisanam et al. Energy Conversion and Management 247 (2021) 114709

vectors i.e. position and velocity vector, to define its location. Further, the deficit energy is calculated, the biogas generator (BG) power output
the position of the search agent is updated by considering the current is estimated at an instant ‘t’. In this scenario, two different cases can be
position, personal best, and global best. Hence the updation procedure observed:
do not have any contribution of remaining search agents, whereas in In first case, As soon as the the system detects as PV generation lower
case of GOA the next position is determined using the information of all than the demand, the biogas generator is turned on. If Ppv (t) +
the search agents. In the present study performance of GOA is compared Pbg (t) ≤ Pload (t), then deficit power (Pdef ) is calculated. Further, the
with the PSO algorithm. The details of PSO algorithm can be found in the energy available in the UR is calculated by:
ref [12].
Pdef
URavail = (30)
ηturb × ρwater × g × h
6. Energy management strategy
If the available water stored in the UR is larger than the required
Optimization of HRES requires proper energy management to amount, then the PHS starts discharging. Further, LPSP and TNPC are
improve the performance of the system. The present system consists of evaluated for this condition. The UR volume at instant ‘(t + 1)’ is
two generating units i.e. PV and biogas generator, and one storage updated using Eq. (31) as given below:
system in the form of PHS and the energy flow takes place as shown in
URt+1 = URt − qturb (t) (31)
Fig. 7. The energy management strategy can be understood by consid­
ering two scenarios, which are discussed below: where qturb (t) is the amount of discharge at an instant ‘t’.
Scenario A- Ppv (t) ≥ Pload (t) On the other hand, if the available volume of water is less than
When the power output of PV panels (PPV(t)) is greater than the load required but above the minimum UR volume, then the PHS discharge
demand (PLoad, i.e. Eload (t)/ɳinv), there is a surplus of energy, so the takes place till the UR volume reaches the minimum value. It is math­
excess energy (Pex) is calculated. The state of charge (SOC) of the upper ematically represented as given below:
reservoir (UR) is then checked. If UR at an intant ‘t’ is less than the
maximum UR limit (URvol(t) < URmax), than the pump is turned on to Pphs = (URt − URmin ) × ρwater × g × h × ηturb (32)
charge the UR till it reaches the maximum SOC. The charging of PHS at t
In this case, Pdef and the loss of power suppy (LPS) are calculated.
time is modeled as given below-
One more possibility is if the available UR volume is equal to the min­
Pex (t) × Δt × ηpump imum reservoir volume, then the deficit power cannot be fulfilled by
URvol (t) = URvol (t − 1) × (1 − βt ) + (27) PHS and hence, the LPS is estimated.
Qpump (t) × 3600
In the second case, if the total energy generation is equal to the load
Where demand then the LPSP and TNPC are calculated for the system. The
simplified flow chart of the energy management strategy is presented in
Eload (t)
Pload (t) = (28) Fig. 8.
ηinv
and URvol (t) is the UR volume of PHS at time t, URvol (t − 1) is the UR 7. Results and discussion
volume at time t-1, ηpump is the efficiency of the pump, and βt is the self-
discharge rate. The UR volume at time t is given as- The energy management strategy developed in this study has been
implemented to design and optimally size the HRES for standalone
URvol (t) = URvol (t − 1) + Qpump (t) (29) micro-load electrication of a remote area of the studied location. Opti­
mization of the PV-BG-PHS system has been carried out using the GOA
where Qpump (t) is the pump flow rate in m3/sec at time t. The LPSP in this and compared with PSO. The optimized configuration thus obtained
case is zero. using the GOA comprises of 57.5 kW PV system, 8 kW BG system, PHS
Scenario B- Ppv (t) ≤ Pload (t)When the PV panel power output is not upper reservoir requirement is found to be 1913 m3 and the converter
sufficient to meet the demand, then the deficit energy is calculated. Once

Fig. 7. HRES configuration.

9
A.K.S. Maisanam et al. Energy Conversion and Management 247 (2021) 114709

Fig. 8. Energy management Strategy.

size is 33 kW. The detail discussion on the obtained results are presented 7.1. Energy management curve
in the subsequent sections. Now, to simplify the problem certain as­
sumptions were considered, which are as follows: The available energy is not always sufficient to meet the demand.
Therefore, the management of generated and stored energy is important
i) The simulation time step is considered as 1 h. to meet the demanded efficiently. This energy management is performed
ii) The project life is taken 25 years which is the lifetime of the Solar using the strategy discussed in the previous section. Fig. 10 presents the
PV system. management of available energy for a period of 72 h. It can be observed
iii) All the components of HRES complete their full operational life. that during the day time PV panel is most of the time sufficient in
iv) The discount rate and the inflation rate is considered 10% and meeting the load even excess energy is generated as the demand is lower
4%, respectively. and solar radiation is higher during that time period. However, during
v) The capacity shortage of the system is considered 2%. the absence of sunlight, demand is met by using the BG plant and the
vi) Carbon emission during the production of components is energy storage. Whenever there is deficit energy, BG plant starts to mend
neglected. the demand–supply gap. Although BG plant alone cannot meet all the
deficit energy, PHS system is utilized to meet the remaining amount of
energy.

Fig. 9. Convergence curve.

10
A.K.S. Maisanam et al. Energy Conversion and Management 247 (2021) 114709

Fig. 10. Energy management curve.

7.2. Energy analysis The detailed energy balance is given in Table 2.

The PV-BG-PHS system is found to meet the load demand of the 7.3. State of charge for pumped hydro storage system
residential and the community load. The detail of the energy in and out
of the hybrid system is represented through a bar graph in Fig. 11. It PHS is the most mature storage system. Its integration with a hybrid
represents the energy balance of the system considering one year time system makes the HRES highly reliable. Therefore, it becomes essential
period. The PV output is consistent with the monthly average solar en­ to focus on the operating condition of the storage unit. The hourly
ergy available at the location and it is observed that generation is highest variation of the PHS charge is shown in Fig. 12. The SOC of PHS varies
during March and April. As the monsoon season arrives, the PV output between 6% and 100% during the whole year. The variation in the
decreases and the BG output increases. The highest utilization of the hourly SOC during the winter months (Nov-Feb) is less as the energy
storage unit is observed during the summer and monsoon season as the demand decreases during these months. As the summer months arrive
power demand increases. However, with the arrival of the winter season the load demand increases tremendously due to which the utilization of
(during Nov-Feb), the load demand decreases due to which utilization of storage unit increases, and hence high fluctuation is observed during
storage unit decreases to some extent. The total energy output of the PV
and BG system is found to be 80,279 kWh/yr and 26,574 kWh/yr, Table 2
respectively. Out of the total generated energy, 57,880 kWh/yr is used Annual energy balance of the optimal system.
for charging the PHS system to meet the demand whenever energy Parameter Energy data (kWh/yr)
generation is low. The total discharge from the storage unit is found to
Total load Demand 73,057
be 33,167 kWh/yr. The energy balance reveals that 9083 kWh/yr of
PHS In 57,880
surplus energy is available, which can be utilized for the deferrable load. PV output 80,279
The availability of surplus energy creates an opportunity for small BG output 26,574
businesses to grow and also to establish new businesses, which can PHS Output 33,167
create jobs and financial security to the local population. Excess 9083

Fig. 11. Monthly averaged energy balance.

11
A.K.S. Maisanam et al. Energy Conversion and Management 247 (2021) 114709

Fig. 12. Hourly variation of the state of charge of PHS system.

these months. Variation of SOC throughout the whole year indicates that
Table 3
the component is being used to the fullest throughout its operational life.
Cost summary (in $) of the proposed system.
Most of the time, the SOC is above 50%, although during few months
SOC drops below 10%. According to the developed methodology for the PV BG PHS Converter Total
present study, the charging takes place whenever there is a production of Capital cost ($) 52,929 18,400 93,581 10,044 174,954
excess amount of energy from the renewable resources. Once the storage Replacement cost ($) 0 11,280 0 3271 14,551
Operation and 3116 218 822 286 4442
unit is fully charged and surplus energy exists, then the remaining en­
maintenance cost ($)
ergy can be utilized in meeting a deferable load or must be dumped. Fuel cost ($) 0 2238 0 0 2238
Land cost ($) 1821 533 10,433 27 12,814
7.4. PV output Carbon emission 0 17,838 0 0 17,838
penalty ($)
TNPC ($) 57,866 50,507 104,836 13,628 226,837
The PV system generates energy throughout the year, although it has
the limitation as the system can work only in the presence of sunlight.
The PV output trend represented in Fig. 13 shows that majority of energy not required throughout the project life. However, the biogas system and
is generated before and after the arrival of monsoon. The reason for that converter require replacement as the life span is less than the project
is the cloud cover due to which the average solar radiation availability is lifetime. Life of the biogas system is 15 years while that of the converter
less during the monsoon. It is observed that the highest power generated is 20 years. Hence, both the systems are replaced once during the entire
is 53 kW. Most of the generated energy is stored in PHS system as the project life. The results show that the most cost intensive component of
load demand is low during the day and highest during the night. the system is the PHS system, which requires 46% of the entire invest­
ment. The remaining components such as PV system, BG system and the
7.5. Cost summary converter contributes 26 %, 22% and 6 %, respectively of the total net
present cost of the system as represented in Fig. 14. The carbon emission
The cost of the system is divided into capital cost, replacement cost, tax (environmental factor) incurred due to operation of biogas plant is
operation and maintenance cost, fuel cost, land cost, and carbon emis­ found to be $ 17838. The land cost is the parameter considered under
sion penalty. The complete information related to the financial detail of the social factor and it contributes $ 12,814 of the TNPC of the system.
the system is presented in Table 3. The HRES is designed for a lifetime of The total contribution of the socio-environmental factor in the TNPC is
25 years, which is same as the lifetime of PV system and less than the found to be 6%. The remaining 94 % of the cost is contributed by the
lifetime of PHS system. Therefore replacement of these components are techno-economic factors. If only techno-economic factor is considered

Fig. 13. Hourly PV output curve.

12
A.K.S. Maisanam et al. Energy Conversion and Management 247 (2021) 114709

Fig. 14. Contribution of HRES components in the system cost.

then the TNPC thus obtained is $ 196185. The LCOE of the HRES be­ synonymous with the inertial weight in PSO and the maximum and
comes 0.207 $/kWh, which indicates a reduction of 15.9 % in LCOE. minimum value of the coefficient ranges from 1 and 0.00001, respec­
This change in the LCOE is an indicator of the significance of considering tively. All the controlling parameters and their corresponding values
the socio-environmental factor alongwith the conventional techno- used in these algorithms are given in Table 4. The variation in the value
economic factors. of the objective function in each iteration is depicted through the
convergence curve presented in Fig. 9. The number of search agents
7.6. Land area requirement of the system considered in the problem is equal to 50. The vertical axis represents the
value of the objective function and the horizontal axis represents the
Minimum utilization of Land area is an major step towards making iteration number. The convergence curve represents that the objective
these systems more attractive and socially acceptable in the rural or function converges to the optimum value after approximately 250 iter­
remote areas. Hence the optimization of the land utilization is very ations. Therefore the maximum number of iteration is fixed at 300. From
important for successful implementation of such projects. The results Fig. 9, it is observed that convergence is achieved faster in case of GOA.
obtained in the present study show that the installation of PV requires The detail results of the optimization are presented in Table 5.
341 m2, BG is supposed to utilize 100 m2 area, converter area require­
ment is 5 m2 and setting up PHS will require the highest area of 1716 m2.
7.9. Statistical analysis of TNPC results
The installation of the whole system will require a total area of 2162 m2.
The area required is converted into financial terms according to the land
The statistical analysis is performed to determine the effectiveness of
value of the location. Its impact on the TNPC is shown in Table 3, which
the nature inspired algorithm GOA in comparision to PSO. For the
highlights all the cost components of the system.
analysis, 30 test runs have been carried out for both GOA and PSO. As
mentioned earlier 300 iterations have been performed to determine the
7.7. Carbon foot print of the system optimum TNPC of the system. The results of the statistical analysis is
presented in Table 5. The value of skewness and the kurtosis gives the
Generally renewable energy systems are considered to be clean information regarding the distribution pattern of the obtained results.
source of energy. However, the proposed system utilizes biogas setup The positive value of the skewness in both the algorithms implies that
that generates energy through the process of combustion of the biogas. the right-hand tail of the distribution is longer or flatter. It is observed
Hence, their exists some form of carbon emission during the operation of that skewness value in both the cases is more than 1, which represents
biogas plant. In the present system the results show that total energy that the distribution is highly skewed. From the value of Kurtosis, it can
output from the BG plant is 26,574 kWh/yr, which produces 1189 ton of be observed that the simulation results represent a leptokurtic curve.
CO2 during the project life time. The financial implication on the TNPC The kurtosis value is higher for the GOA based approach, which in­
due to the emission is found to be $ 17838. dicates that the algorithm has the ability to escape local minima and find
the global minima more effectively. The standard deviation is higher in
7.8. Comparison of GOA with PSO simulation results
Table 4
The optimal sizing of HRES for electrification is done in the Matlab Control parameters of optimization.
environment using GOA based approach. The performance of GOA is Sl. Algorithm Parameter Value
analyzed to determine the effectiveness of the algorithm in solving an NO.
HRES design problem. Further, performance comparison of GOA with
1 GOA Swarm balancing coefficient (C) Cmin = 0.00001 Cmax =
the Particle swarm optimization (PSO) algorithm is carried out. The 1
parameters such as the initial population size and the stopping criteria Attractive length (l) 1.5
are kept the same for both the algorithms. The mutation rate in the PSO Attractive intensity between 0.5
algorithm is controlled by the acceleration constants and their values are swarms (f)
considered to be 1.5 and 2. The inertial weight factor retains the pre­ 2 PSO Inertial weight 1
vious flight direction information so that sudden changes in the flight Inertial weight damping ratio 0.99
direction is avoided. Its value is assumed to be 1, which decreases as the Personal learning coefficient 1.5
Global learning coefficient 2
iteration progresses. In the case of GOA, the decreasing coefficient C is

13
A.K.S. Maisanam et al. Energy Conversion and Management 247 (2021) 114709

Table 5 HRES. It is concluded that the skewness is lower in case of PSO, which
Statistical analysis of TNPC. indicates that the distribution is closer to the normal distribution curve.
PSO GOA The kurtosis value is higher in case of GOA, which shows that the al­
gorithm escapes from local minima and has the ability to better
Mean 30.79 28.74
Standard deviation 36.66 33.17 approximate the global optimum.
Minimum 16.52 15.89 The simulation results indicate that the energy management of the
Maximum 240.34 255.61 HRES is efficient and the developed strategy is capable of meeting the
Skewness 3.91 4.11 load demand. The majority of the energy is produced by PV system,
Kurtosis 15.97 18.73
which produces 80,279 kWh/year, although most of energy is stored, as
the demand is lower during the day time. BG plant and the PHS unit
case of the PSO algorithm, which indicates that drastic changes in serve as the components for meeting the peak load demand. The opti­
objective function value occur during the iteration process. mized system comprises of 57.5 kW (341 m2) PV system, 8 kW BG plant,
378.4 kWh storage capacity and 33 kW of converter size. The majority of
7.10. Comparison of results with literature investment is required for the purchase and replacement of the com­
ponents during the HRES life time. A total 46 % of TNPC is incurred in
The comparison of the obtained results is performed with the pub­ installing the energy storage system. The socio-environmental factor
lished work. The comparison of HRES system is performed on the basis contribution is approximately 6% of the TNPC and 94% of TNPC is
of the nature of load demand, i.e. for comparison, only micro-load de­ contributed from the techno-economic factors. It is also found that
mand was considered and the peak load in the present work is found to contribution of socio-environmental factor in the LCOE is approximately
be 32 kW and the work reported by Kahwash et al. [46] was 16.53 kW. 16 % which cannot be ignored. The optimized results are also compared
Comparison on the basis of exact configuration of the HRES is not always with other HRES design under similar micro-load condition, and it is
possible due to difference in the components; however, it is thought that found that the proposed system is having lower LCOE than the available
comparison on the besis of similar load condition is quite possible. The GA assisted micro-load HRES design.
optimized system results using the GOA is compared with PSO, and GA, Economic analysis suggest that the investment required for procur­
which is depicted in Table 6. It is observed that better minimization of ing component is the primary contributor in the value of LCOE. The
the objective function takes place in case of GOA based approach. The policy makers have a critical role to play in cost reduction of various
optimum design of the system obtained through GOA simulation con­ renewable energy related components through various subsidies,
sists of 57.5 kW PV system, 8 kW BG system, 21.6 kW converter with the incentivizing the manufacturers of the components, etc. Government
PHS system consisting of 31.6 kW pump/turbine along with a storage interference in creating awareness regarding the benefits of renewable
capacity of 378.4 kWh. The levelized cost of the energy of the designed energy utilization is required for improving acceptability of such sys­
system is found to be $ 0.240/kWh. The results of the simulation using tems. Such steps can substantially increase the penetration of renewable
the PSO based approach indicate that a 62.2 kW PV panel along with a energy systems in periurban and rural/remote area. It will not only
8.6 kW BG system makes the main generating unit. The optimum provide sufficient and reliable electricity but can also create many op­
amount of storage required to meet the load is found to be 385.6 kWh, portunities for the local population, which can improve the overall
which is required to meet the load demand, and DC output of the solar socio-economic conditon of the area.
PV is converted using a 33 kW converter. The designed system shows Finally, the authors believe that this work may create interest in the
that the LCOE of the system is $0.0.261/kWh. The LCOE of other micro- S&T community who are engaged in similar research works to think of
load configuration reported by Kahwash et al. [46] is found to be 0.287 integrating the socio-environmental factors with the mandatory techno-
$/kWh. economic factors in the HRES designing process. Further, in this work,
only the land mass resource under social factor and carbon emission
under environmental factor. There are other social and environmental
8. Conclusions and recommendations
factor, e.g. job creation, human development index, health hazard, etc.,
which are also there to be studied for further investigations in this area
The energy generated from the conventional power plant lacks sus­
for the concerned S&T community. These are the recommendations for
tainability. In such a scenario, HRES will play essential role in increasing
future work, which have been reported at the end of the conclusion
the penetration of renewable energy based systems in the near future.
section.
Hence, for the first time in the present work, GOA based methodology is
developed considering four important factors namely, technical, eco­
CRediT authorship contribution statement
nomic, social and environmental factors to optimally size an HRES
system comprising of PV, BG and PHS unit. The developed strategy is
Maisanam Anil Kumar Singh: Conceptualization, Methodology,
applied for remote area electrification of an Indian town.
Software, Validation, Visualization, Investigation, Writing - original
The simulation results obtained using GOA is compared with those
draft, Writing - review & editing. Agnimitra Biswas: Conceptualization,
obtained from PSO algorithm. It is found that the convergence rate of
Methodology, Validation, Writing - original draft, Writing - review &
GOA is faster than the PSO algorithm in solving optimal sizing of HRES.
editing. Kaushal Kumar Sharma: Methodology, Validation, Writing -
The convergence to final optimum solution occurs after 100 iterations in
original draft, Writing - review & editing, Supervision.
case of GOA and after 150 iterations in case of PSO. The convergence
characteristics reveal that GOA method reduces the computational time
as well as the associated computational resources. The robustness of the
algorithms is tested using the statistical analysis of the TNPC of the

Table 6
Comparison of simulation results.
Method Configuration WT (kW) PV (kW) BG (kW) DG (kW) PHS (kWh) Bat (kWh) Converter (kW) LCOE ($/kWh)

GOA PV/BG/PHS – 57.5 8 – 378.4 – 33 0.240

PSO PV/BG/PHS – 62.2 8.6 – 385.6 – 33 0.261
GA [46] WT/PV/Battery/DG 15 32.8 – 11 – 548 – 0.287

14
A.K.S. Maisanam et al. Energy Conversion and Management 247 (2021) 114709

Declaration of Competing Interest Convers Manag 2020;221:113192. https://doi.org/10.1016/j.

enconman.2020.113192.
[24] Tamoor M, Tahir MS, Sagir M, Bilal M, Iqbal S, Nawaz T. Design of 3 kW integrated
The authors declare that they have no known competing financial power generation system from solar and biogas. Int J Hydrogen Energy 2020;45:
interests or personal relationships that could have appeared to influence 12711–20. https://doi.org/10.1016/j.ijhydene.2020.02.207.
the work reported in this paper. [25] Chauhan A, Saini RP. Discrete harmony search based size optimization of
Integrated Renewable Energy System for remote rural areas of Uttarakhand state in
India. Renew Energy 2016;94:587–604. https://doi.org/10.1016/j.
References renene.2016.03.079.
[26] Chauhan A, Saini RP. Techno-economic optimization based approach for energy
[1] Rajanna S, Saini RP. Development of optimal integrated renewable energy model management of a stand-alone integrated renewable energy system for remote areas
with battery storage for a remote Indian area. Energy 2016;111:803–17. https:// of India. Energy 2016;94:138–56. https://doi.org/10.1016/j.energy.2015.10.136.
doi.org/10.1016/j.energy.2016.06.005. [27] Suresh V, Muralidhar M, Kiranmayi R. Modelling and optimization of an off-grid
[2] CEA. All India Installed Capacity (in MW) of Power Stations. Cent Eectricity hybrid renewable energy system for electrification in a rural areas. Energy Rep
Authority, Minist Power 2020. http://www.cea.nic.in/reports/monthly/ 2020;6:594–604. https://doi.org/10.1016/j.egyr.2020.01.013.
installedcapacity 2016/installed_capacity-03.pdf (accessed May 8, 2020). [28] Mayer MJ, Szilágyi A, Gróf G. Environmental and economic multi-objective
[3] GOI. Power sector at a glance all India. Gov India, Minist Power 2020. https:// optimization of a household level hybrid renewable energy system by genetic
powermin.nic.in/en/content/power-sector-glance-all-india (accessed May 9, algorithm. Appl Energy 2020;269:115058. https://doi.org/10.1016/j.
2020). apenergy.2020.115058.
[4] Ramchandran N, Pai R, Parihar AKS. Feasibility assessment of Anchor-Business- [29] Jahangir MH, Cheraghi R. Economic and environmental assessment of solar-wind-
Community model for off-grid rural electrification in India. Renew Energy 2016; biomass hybrid renewable energy system supplying rural settlement load. Sustain
97:197–209. https://doi.org/10.1016/j.renene.2016.05.036. Energy Technol Assessments 2020;42:100895. https://doi.org/10.1016/j.
[5] Amutha WM, Rajini V. Cost benefit and technical analysis of rural electrification seta.2020.100895.
alternatives in southern India using HOMER. Renew Sustain Energy Rev 2016;62: [30] Sawle Y, Gupta SC, Bohre AK. Socio-techno-economic design of hybrid renewable
236–46. https://doi.org/10.1016/j.rser.2016.04.042. energy system using optimization techniques. Renew Energy 2018;119:459–72.
[6] Bawah U, Addoweesh KE, Eltamaly AM. Comparative study of economic viability https://doi.org/10.1016/j.renene.2017.11.058.
of rural electrification using renewable energy resources versus diesel generator [31] Dufo-López R, Cristóbal-Monreal IR, Yusta JM. Optimisation of PV-wind-diesel-
option in Saudi Arabia. J Renew Sustain Energy 2013;5:042701–42717. https:// battery stand-alone systems to minimise cost and maximise human development
doi.org/10.1063/1.4812646. index and job creation. Renew Energy 2016;94:280–93. https://doi.org/10.1016/j.
[7] Sarkar T, Bhattacharjee A, Samanta H, Bhattacharya K, Saha H. Optimal design and renene.2016.03.065.
implementation of solar PV-wind-biogas-VRFB storage integrated smart hybrid [32] Cuesta MA, Castillo-Calzadilla T, Borges CE. A critical analysis on hybrid
microgrid for ensuring zero loss of power supply probability. Energy Convers renewable energy modeling tools: An emerging opportunity to include social
Manag 2019;191:102–18. https://doi.org/10.1016/j.enconman.2019.04.025. indicators to optimise systems in small communities. Renew Sustain Energy Rev
[8] Petrollese M, Cocco D. Techno-economic assessment of hybrid CSP-biogas power 2020;122:109691. https://doi.org/10.1016/j.rser.2019.109691.
plants. Renew Energy 2020;155:420–31. https://doi.org/10.1016/j. [33] Askarzadeh A, Coelho S. A novel framework for optimization of a grid independent
renene.2020.03.106. hybrid renewable energy system : A case study of Iran. Sol Energy 2015;112:
[9] Sanajaoba S, Fernandez E. Maiden application of Cuckoo Search algorithm for 383–96. https://doi.org/10.1016/j.solener.2014.12.013.
optimal sizing of a remote hybrid renewable energy System. Renew Energy 2016; [34] Forough AB, Roshandel R. Lifetime optimization framework for a hybrid renewable
96:1–10. 10.1016/j.renene.2016.04.069. energy system based on receding horizon optimization. Energy 2018;150:617–30.
[10] Khare R, Kumar Y. A novel hybrid MOL – TLBO optimized techno-economic-socio https://doi.org/10.1016/j.energy.2018.02.158.
analysis of renewable energy mix in island mode. Appl Soft Comput 2016;43: [35] Ma T, Yang H, Lu L, Peng J. Technical feasibility study on a standalone hybrid
187–98. https://doi.org/10.1016/j.asoc.2016.02.044. solar-wind system with pumped hydro storage for a remote island in Hong Kong.
[11] Poonam S, Manjaree P, Laxmi S. Comparison of traditional and swarm intelligence Renew Energy 2014;69:7–15. https://doi.org/10.1016/j.renene.2014.03.028.
based techniques for optimization of hybrid renewable energy system. Renew [36] Ma T, Yang H, Lu L. Feasibility study and economic analysis of pumped hydro
Energy Focus 2020;35:1–9. https://doi.org/10.1016/j.ref.2020.06.010. storage and battery storage for a renewable energy powered island. Energy Convers
[12] Maleki A, Askarzadeh A. Comparative study of artificial intelligence techniques for Manag 2014;79:387–97. https://doi.org/10.1016/j.enconman.2013.12.047.
sizing of a hydrogen-based stand-alone photovoltaic/wind hybrid system. Int J [37] Singh S, Singh M, Kaushik SC. Feasibility study of an islanded microgrid in rural
Hydrog 2014;39(19):9973–84. https://doi.org/10.1016/j.ijhydene.2014.04.147. area consisting of PV, wind, biomass and battery energy storage system. Energy
[13] Fetanat A, Khorasaninejad E. Size optimization for hybrid photovoltaic–wind Convers Manag 2016;128:178–90. https://doi.org/10.1016/j.
energy system using ant colony optimization for continuous domains based integer enconman.2016.09.046.
programming. Appl Soft Comput 2015;31:196–209. https://doi.org/10.1016/j. [38] Das M, Singh MAK, Biswas A. Techno-economic optimization of an off-grid hybrid
asoc.2015.02.047. renewable energy system using metaheuristic optimization approaches – Case of a
[14] Maleki A, Pourfayaz F. Sizing of stand-alone photovoltaic/wind/diesel system with radio transmitter station in India. Energy Convers Manag 2019;185:339–52.
battery and fuel cell storage devices by harmony search algorithm. J Energy https://doi.org/10.1016/j.enconman.2019.01.107.
Storage 2015;2:30–42. https://doi.org/10.1016/j.est.2015.05.006. [39] Maisanam AKS, Biswas A, Sharma KK. An innovative framework for electrical
[15] Alshammari N, Asumadu J. Optimum unit sizing of hybrid renewable energy energy storage system selection for remote area electrification with renewable
system utilizing harmony search, Jaya and particle swarm optimization algorithms. energy system: Case of a remote village in India. J Renew Sustain Energy 2020;12
Sustain Cities Soc 2020;60:102255. https://doi.org/10.1016/j.scs.2020.102255. (2):024101. https://doi.org/10.1063/1.5126690.
[16] Javed MS, Ma T, Jurasz J, Ahmed S, Mikulik J. Performance comparison of [40] Cuéllar AD, Webber ME. Cow power : the energy and emissions benefits of
heuristic algorithms for optimization of hybrid off-grid renewable energy systems. converting manure to biogas. Environ Res Lett 2008;3:034002 (1-8). 10.1088/
Energy 2020;210:118599. https://doi.org/10.1016/j.energy.2020.118599. 1748-9326/3/3/034002.
[17] Makhdoomi S, Askarzadeh A. Impact of solar tracker and energy storage system on [41] Maleki A, Askarzadeh A. Artificial bee swarm optimization for optimum sizing of a
sizing of hybrid energy systems: A comparison between diesel/PV/PHS and diesel/ stand-alone PV/WT/FC hybrid system considering LPSP concept. Sol Energy 2014;
PV/FC. Energy 2021;231:120920. https://doi.org/10.1016/j.energy.2021.120920. 107:227–35. https://doi.org/10.1016/j.solener.2014.05.016.
[18] Kaabeche A, Bakelli Y. Renewable hybrid system size optimization considering [42] Singh SS, Fernandez E. Modeling, size optimization and sensitivity analysis of a
various electrochemical energy storage technologies. Energy Convers Manag 2019; remote hybrid renewable energy system. Energy 2017;143:719–39. https://doi.
193:162–75. https://doi.org/10.1016/j.enconman.2019.04.064. org/10.1016/j.energy.2017.11.053.
[19] Jamshidi S, Pourhossein K, Asadi M. Size estimation of wind/solar hybrid [43] Singh S, Singh M, Chandra S. Feasibility study of an islanded microgrid in rural
renewable energy systems without detailed wind and irradiation data: A feasibility area consisting of PV, wind, biomass and battery energy storage system. Energy
study. Energy Convers Manag 2021;234:113905. https://doi.org/10.1016/j. Convers Manag 2016;128:178–90. https://doi.org/10.1016/j.
enconman.2021.113905. enconman.2016.09.046.
[20] Sanajaoba S, Fernandez E. Maiden application of Cuckoo Search algorithm for [44] Saremi S, Mirjalili S, Lewis A. Grasshopper Optimisation Algorithm: Theory and
optimal sizing of a remote hybrid renewable energy System 2016;96. application. Adv Eng Softw 2017;105:30–47. https://doi.org/10.1016/j.
[21] Merei G, Berger C, Uwe D. Optimization of an off-grid hybrid PV – Wind – Diesel advengsoft.2017.01.004.
system with different battery technologies using genetic algorithm. Sol Energy [45] Barik AK, Das DC. Expeditious frequency control of solar photovoltaic/biogas/
2013;97:460–73. https://doi.org/10.1016/j.solener.2013.08.016. biodiesel generator based isolated renewable microgrid using grasshopper
[22] Ramesh M, Saini RP. Dispatch strategies based performance analysis of a hybrid optimisation algorithm. IET Renew Power Gener 2018;12(14):1659–67. https://
renewable energy system for a remote rural area in India. J Clean Prod 2020;259: doi.org/10.1049/rpg2.v12.1410.1049/iet-rpg.2018.5196.
120697. https://doi.org/10.1016/j.jclepro.2020.120697. [46] Kahwash F, Maheri A, Mahkamov K. Integration and optimisation of high-
[23] Mokhtara C, Negrou B, Bouferrouk A, Yao Y, Settou N, Ramadan M. Integrated penetration Hybrid Renewable Energy Systems for fulfilling electrical and thermal
supply – demand energy management for optimal design of off – grid hybrid demand for off-grid communities. Energy Convers Manag 2021;236:114035.
renewable energy systems for residential electri fication in arid climates. Energy https://doi.org/10.1016/j.enconman.2021.114035.

15