See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/321216546

Socio-Techno-Economic Design of Hybrid

Renewable Energy System Using Optimization
Techniques

Article in Renewable Energy · November 2017

DOI: 10.1016/j.renene.2017.11.058

CITATIONS READS

0 81

3 authors:

Yashwant Sawle S.C. Gupta

Maulana Azad National Institute of Technolo… Maulana Azad National Institute of Technolo…
15 PUBLICATIONS 56 CITATIONS 17 PUBLICATIONS 94 CITATIONS

SEE PROFILE SEE PROFILE

Aashish Kumar Bohre

Maulana Azad National Institute of Technolo…
21 PUBLICATIONS 57 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Renewable Energy system View project

All content following this page was uploaded by Yashwant Sawle on 23 December 2017.

The user has requested enhancement of the downloaded file.

 Renewable Energy 119 (2018) 459e472

Contents lists available at ScienceDirect

Renewable Energy
journal homepage: www.elsevier.com/locate/renene

Socio-techno-economic design of hybrid renewable energy system

using optimization techniques
Yashwant Sawle*, S.C. Gupta, Aashish Kumar Bohre
Electrical Engineering Department, MANIT, Bhopal, India

a r t i c l e i n f o a b s t r a c t

Article history: This paper presents the optimal hybrid power system design including various configurations of
Received 27 April 2017 renewable energy generation. To decide the optimal configuration of parameters a new multi-objective
Received in revised form function with six separate objectives of hybrid renewable system is presented using GA, PSO, BFPSO and
15 November 2017
TLBO optimization techniques. The different parameters namely technical (LPSP, Renewable factor),
Accepted 20 November 2017
economical (COE, Penalty & Fuel consumption) and social (Job creation, HDI & PM) features are inves-
Available online 21 November 2017
tigated as objectives simultaneously for optimal design of hybrid system. The design consideration of
hybrid system using a novel PM factor, human health impacts are directly shown whereas pollutant
Keywords:
HDI
emission is measured in the hybrid system design. Based on the minimum value of multi-objective
JCI function optimal values are decided for objective indices. For optimal configuration including various
Particular matter combinations of wind, PV, diesel generator, biomass and battery bank, separate cases from I to VI of
Renewable factor hybrid system are tested. Performance of TLBO is found to be better than BFPSO, PSO and GA as per the
Cost of energy analysis of results for individual cases. Also the case I found to be the most efficient solution among all
HRES cases.
© 2017 Elsevier Ltd. All rights reserved.

1. Introduction
Conventional fuel is the main source of electricity generation
but it produces a large amount of carbon and it's by-products in the
environment. Thus harmful gases are the main concern now a day
in the world which damages the greenhouse system of the envi-
ronment. This problem can be reduced through the renewable
energy sources such as solar, wind and biomass etc. Energy and
ecological security issues are being confronted by most nations
nowadays. The explosion of the human populace on the earth has
brought on a consistent growth in the need for energy, especially
electricity. The developing trend in energy application is 2.8%
increment for each annum. But such an expanding demand for
power on a worldwide scale cannot be taken care of by the current
structure of energy sources [1]. The increasing cost of conventional
power and environmental pollution has led to the consideration of
alternative renewable sources for power generation. Thus renew-
able energy sources are designed as off-grid or grid connected
system. Off-grid hybrid renewable energy sources (HRES) with a
* Corresponding author.
E-mail addresses: yashsawle@gmail.com (Y. Sawle), scg.nit.09@gmail.com
(S.C. Gupta), aashu371984@gmail.com (A.K. Bohre).
diesel generator and battery unit are used for electrifying far off
locations. In literature optimal sizing of the off-grid system is
investigated regarding lowest net present cost (NPC), cost of energy
(COE), loss of power supply probability (LPSP) and maximum
renewable factor (RF) [2,3]. Many studies are also reported for
minimization of carbon dioxide, unmet load, penalty cost of
emission, and maximization of job creation factor (JCF), human
development index (HDI) [4e6]. Optimal sizing of hybrid system
has been also evaluated on the basis of social parameters, technical
parameters and economical parameters with multi-objective and
multi-criteria in the previous studies.
In remote areas where human advancement factor is generally
minimum, a key factor in lessening the poverty is access to elec-
trical power [7,8]. According to report 2014, a large portion of the
world population around 17.8% cannot access the electrical power
which indicates a low human development index [9]. HDI factor is
improved if the living standard of the population increases such as
better use of electricity; better education, higher net earnings per
capita and use of electrical appliances, new industry and jobs. The
technical design has been suggested as an important factor by
many authors due to its effect on the HDI [10,11]. The HDI approach
depends on four key columns that must bolster any vital
advancement activity to achieve what we mean by achievement:

https://doi.org/10.1016/j.renene.2017.11.058
0960-1481/© 2017 Elsevier Ltd. All rights reserved.
460 Y. Sawle et al. / Renewable Energy 119 (2018) 459e472

Nomenclature T
PWIND;each Power produced by turbine
vR Nominal speed of the wind turbine
HRES Hybrid renewable energy sources V Wind speed
GA Genetic algorithm
vo Cut out speed
PSO Particle swarm optimization
vJ Cut in speed
TLBO Teaching-Learning-Based Optimization
PR Wind Generator rated power
COE Cost of energy
PTPV:each Power produced by PV system
LPSP Loss of power supply probability
PRS PV panel rated power
Pc Emission cost
R Radiation factor
HDI Human development index
Rcr certain radiation at 150 W/m2
JCF Job creation factor
Rsrs Standard solar radiation at 1000 W/m2
Cc Carbon content engaged
hBIO Efficiency of the biomass
RF Renewable factor
CBBIO Calorific value of the rise husk
PM Particular matter
ETRC is Tradable renewable energy certificate cost in $/kWh
NPC Net present cost
r1 & r2 Arbitrary figures value
PWIND Power generated through the wind plant in MW
VSDG Variable speed diesel generator
hbf Discharging efficiency of battery bank
CRF Capital recovery factor k Iteration number
hinverter Efficiency of the inverter W Inertia weight
N No. Of swarm fmaxSE Element of maximum surplus energy which may be
fmaxELOAD Element to increase yearly AC load in order that the utilized for AC additional load
maximum surplus energy eLY=C Energy yearly electricity Consumption per capita in
eSE Yearly surplus energy of the system in kWh/yr kWh/yr/person
nHUMAN A number of humans which consume the generated JCBIOMASS Number of job/MWP through the biomass plant
power through the hybrid system PBIOMASS Power generated through the biomass plant in MW
JCPV Number of job/MWP through the PV plant JCD Number of created through diesel generator in job/
PPV Peak power generation of PV plant in MWP GWh/yr
JCWIND Number of job/MWP through the wind plant ED Yearly energy generated by diesel generator GWh/yr
DG Diesel generator JCBATTERY Job created by battery unit in job/MWh

opening the doors of growth for individuals and changing them

into masters of their own advancement. These four standards are as
following [12]:

➢ Equity and diversity

➢ Sustainability
➢ Empowerment
➢ Productivity

For the design of HRES, many authors have considered the Job
Creation (JC) factor [28e32]. Clean energy can lead to many do-
mestic job employment and a great deal of these jobs would remain
local as they include infrastructure development. The economy can
be stimulated by diverting the investment from energy expenses to
adopting energy productivity measures which would increase job
creation. A number of studies have found that better utilization of
sustainable power sources and methods focused on energy effi-
ciency gives financial advantages through employment creation,
while also protecting the economy from political and monetary
risks related with over-dependence on a limited set of energy ad-
vancements and fuels. We concentrate on the power region in this
review as it is the biggest essential energy segment and further-
more the rapid developing division and most employment/job
creation studies have been done around here [30].
The working model of off-grid PV/Biomass/wind with DG and
battery unit which is connected to AC load, which is shown in Fig. 1.
Generated power of hybrid system is excessive as compared to load
demand; this excessive power is sold and utilized to fulfill load
demand of another consumer. So that it creates new jobs because of
extra works needed and also increases the human development
index because of earning money. In hybrids system dump load also
Fig. 1. General block diagram of HRES.
connected with AC side which consumes surplus energy produced
from renewable sources after cover the AC load of design location,
storage battery unit and additional AC load.
For the economic power supply to the consumer, dispatch
strategy plays an important role. Through a proper dispatch strat-
egy of DG, a reliable and cost-efficient operation can be done for
any HRES [13,27]. A variable speed diesel generator (VSDG) has a
great impact on HRES for obtaining a techno-economic system
which is first introduced by Manwell [14]. Compared to other
existing control dispatch strategies an optimal strategy with VSDG
has many advantages as is discussed in this paper.
In the hybrid system proposed in this paper particulate matter
(PM) is also considered which has not received much attention in
 Y. Sawle et al. / Renewable Energy 119 (2018) 459e472 461

previous works for HRES design. Particulate matter (A combination data of the day and season of the year. The value of friction coef-
of smoke, residue particles, and fluid beads that can bring about ficient usually assumes around 1/7. The output power of turbine is
respiratory issues) emitted in the process of carbon combustion evaluated through the given [25,52].
which is injurious to human health and promotes the environ- 8
mental degradation [15,16]. It is characterized in two forms as PM2.5 >
> 0 V  vJ or V  vo
>
>  
and PM10. The PM production in HRES through the DG and Biomass < V  vJ
T
generator unit which are utilized to design a system. The accom- PWIND;each ¼ PRw vJ < V < v (1)
>
> vR  vJ
panying diagram demonstrates the estimation of the PM emission >
>
:
factor for DG the design value 50 kW - 450 kW. The calculation of PRw vR  V < vo
PM is shown in Fig. 2. The wellspring of this information is an in-
ward National Renewable Energy Laboratory description through where
Erin Kassoy characterized “Modeling diesel exhaust emissions in
T
diesel retrofits ". The default an incentive for the generator's PM, PWIND;each¼ power produced by turbine
emissions variable is proportionate to the ordinary incentive be- vR ¼ Nominal speed of the wind turbine
tween 50% and 100% load [39]. V ¼ wind speed
vo ¼ Cut out speed
2. Renewable resources modelling vJ ¼ Cut in speed
PR ¼ Wind generator rated powwer
2.1. Load profile

If the considered study area is small for distribution generation

system design then electrical load demand is also less. The daily 2.3. PV energy
energy consumption is 110.6 kWh/day as per the assessment. The
load profile for distribution generation system in this proposed The solar radiation is available abundantly in the environment.
paper is on an hourly basis. As shown in Fig. 3 the load profile The potential of solar radiation is dependent on location or region
covers 24 h of a day. The calculation of average load or load factor of [49,50]. The distribution generation system is proposed for the
distributed generation system can be characterized as the propor- location of 22.71 N latitude and 75.85E longitude India [20]. Solar
tion of the region under load curve in p.u. to aggregate span [17,20]. radiation average yearly is 5.531 kW/m2 for design location.
So that average load of the system is 4.61 kW and peak load of the Compared to winter season solar radiation is higher in summer
day 7.8 kW. season as seen in Fig. 4. The information related to the cost and
specification of PV panel is shown in Table 1. The output power of
2.2. Wind energy PV panel is calculated through the given equations [21,24].
8  
The design location of distributed generation system has a high >
> R2
>
> Prs 0  R < Rcr
average yearly resource for wind energy (wind speed is 4.5 m/s). >
> Rsrs Rcr
<
The nature of wind speed is variable with time and seasons. Fig. 4  
PPV; each ¼ R (2)
shows the wind speed data with respective hours [18,19]. The >
> Prs Rcr < Rsrs
>
> Rsrs R
friction coefficient has also different names such as Hellmann >
>
:
exponent, power-law exponent or wind gradient. Friction coeffi- Prs Rsrs  R
cient depends on various factors such as wind speed, irregularity of
terrain, altitude above ground, the temperature of the area, hourly where

0.8
PM Emission (g / L)

0.6

0.4

0.2

0
0 20 40 60 80 100
Power Output (% of Rated capacity)

Fig. 2. Particular matter calculation.

462 Y. Sawle et al. / Renewable Energy 119 (2018) 459e472

Seasonal Load Profile

10

9 Spring Fall
Summer
Winter
8

7
Load(kW)

2
1 24 48 72 96
Hours

Fig. 3. Hourly Load profile curve.

expected in the various studies that the rice husk production is 25%
of the paddy and immature paddy production is 3% of the paddy. So
that the total biomass production of rice husk is 115 kg/d [19,22,51].
Rice husk has a calorific value of 12.1e15.2 MJ/kg. The output power
through the Biomass generator is estimated by given equation [23].
Fig. 5 shows hourly load profile of the biomass generator curve.
  !
  Total rice available t  1000  CVBIO  hBIO
kWh yr
PBM ¼
yr 876  ðoperating hours=dayÞ
(3)

where.
Fig. 4. Resource data of wind speed and solar radiation.

PTPV:each ¼ Power produced by PV system. hBIO ¼ Efficiency of the biomass

Prs ¼ PV panel rated power. CVBIO ¼ Calorific value of the husk
R ¼ Radiation factor.
Rcr ¼ certain radiation at 150 W/m2
Rsrs ¼ Standard solar radiation at 1000 W/m2
2.5. Battery bank

The battery is used in a hybrid system for storage and maintains

2.4. Biomass energy the equilibrium demand and electricity supply. Evaluation of state
of charge with respective productivity and time consumption as:
The rice husk is considered as a source of biomass energy in this
T þ PT T T
paper. In India rice is one of the main cereals as it accounts for over a) PPV WIND þ PBio ¼ PDEMAND When battery stable and not
30% of its cropped area and 40% of ground grain production. India's changed.
T T T T
share in world rice production is 21%. The distribution generation b) PPV þ PWIND þ PBio > PDEMAND In this situation, total renewable
system location has per day rice paddy produce of 460 Kg. It is hybrid energy of the system is higher than the load demand. In

Table 1
Of Cost of parameters [37,47,48].

S$N Parameters Specification S$N Parameters Specification

1 Wind capacity 1 kW 14. DG cost 550$

2 Wind turbine cost 1800$ 15. DG O&M cost 0.008/kWh
3 Wind O&M cost 0.02/kWh 16. DG life hours 15000hr
4 Wind turbine life years 20yr 17. Battery capacity 2.76 kWh
5 PV capacity 0.260 kW 18. Battery cost 150$
6 PV cost 468$ 19. Battery O&M cost 10$
7 PV O&M cost 0.005/kWh 20. Battery life years 5yr
8 PV life 20yr 21. Inverter capacity 8 kW
9 Biomass capacity 1.3 kW 22. Inverter cost 500$
10 Biomass cost 500$ 23. Inverter O&M cost 0/kWh
11 Biomass O&M cost 0.02/kWh 24. Inverter life years 10yr
12 Biomass life hours 15000hr 25. Real interest 6
13 DG capacity 1.8 kW 26. Project life years 20
 Y. Sawle et al. / Renewable Energy 119 (2018) 459e472 463

1.2

1.1
Summer
1
Fall

0.9
Spring
Winter
0.8
Biomass Power (kW)

0.7

0.6

0.5

0.4

0.3

0.2
1 24 48 72 96
Hours

Fig. 5. Hourly Load profile of the biomass generator curve.

this condition battery is in charging position and charged inside restricted cut-off points at load levels. VSDG units modify the
quantity of the battery at time (T) is given as; speed of DG relying upon the load and utilize electrical power to
" # guarantee that the yield voltage and frequency are kept up steady
  PlT irrespective of the motor speed [26]. The advantage of VSDG as
T
Ebattery T1
Ebattery  tÞ þ T
PWIND þ T
PPV þ T
PBio  hbc
hinverter
$ð1
compared to constant speed DG is that it provides an extensive
working range for different output power at a greatly lower mini-
(4)
where. mum loading. In a VSDG, the machine is operated at comparatively
low speeds for low power requirement and at higher speeds for
T
Ebattery T1
and Ebattery ¼ Charge quantities of the battery higher power requirement. The load demand through the VSDG is
T  1 and T ¼ Battery bank charge times. determined to rely on the operating approach. In a situation when
PlT ¼ Energy demand of the particular hour demand shortage to be met is lesser than the least loading of the
T
PWIND ¼ Power produced by turbine. VSDG, subsequent release of the battery is evaded and the entire
T ¼ power produced by PV panel
PPV net load is drawn from the VSDG and consequently, the operation of
PlT ¼ energy demand for the specific hour. VSDG over its base loading is guaranteed [27].
hbc ¼ Charge efficiency of the battery bank
PRG ¼ PPV þ PWIND þ PBIO
hinverter ¼ Efficiency of the inverter PDEFICIT ¼ PDEMAND  PRG  ðhINVERTER  PBATTERY Þ
(6)
t ¼ The hourly self-release rate
T þ PT T T
c) PPV WIND þ PBio < Pdemand When load demand is higher to The feasible sizing of HRES is given a thought of a VSDG. The
total output power of the hybrid system. At this time battery advantage over regular speed DG it has higher operating value at
unit is in release position and the charge quantity of the battery much lesser loading. In this proposed work according to deficit
bank is given in equation. The storage battery bank set at a load, various strategies for DG operation are considered such as:
nominal capacity and it only allow discharge at that limit [17].
The details of the battery bank is given in Table 1  If PDEFICIT < 20% This condition is known as very low working
" # type
PlT    If PDEFICIT ¼ 20%  50% This condition is known as low working
T
Ebattery T1
 Ebattery $ð1  tÞ þ T
 PWIND T
þ PPV T
þ PBio hbf
hinverter type
 If PDEFICIT ¼ 50%  75% This condition is known as medium
(5) working type
where.
 If PDEFICIT ¼ 75%  100% This condition is known as high work-
hbf ¼ Discharging efficiency of battery bank ing type

In the first condition minimum loading of VSDG is 20% in this

situation net load demand is supply through the DG and the
2.6. Variable speed diesel generator (VSDG) discharge of battery unit is evaded.

Since DG consumes a minimum amount of fuel if it is run at an 2.7. Converter

optimal speed while supplying a particular load therefore variable
speed generator has been used for the design of HRES in this pro-
posed work. Constant speed DG set operates at a static speed
irrespective of the power requested by the load demand to guar-
antee that the yield electrical frequency and voltage are kept up
Electronic converters are needed to balance energy flow be-
tween the AC and DC elements. The converters/inverters transfer
the electric power from one form to another (converter AC to DC,
and inverter DC to AC) by means of the required frequency of the
464 Y. Sawle et al. / Renewable Energy 119 (2018) 459e472
load. The converter efficiency is calculated as [24,52e54]:
P through low cost and leading to job increases and reduce the de-
himv ¼
P þ P0 þ kP2
where, P, P0 and k are determined by using following equations as:
 2
10 1 1
P0 ¼ 1  99  9 ; k¼  P0  1; And P ¼
h10 h100 h100
Where, h10 and h100 are the efficiency of the inverter at 10% and
100% of its insignificant energy correspondingly, which are given by
the makers. The P, Pn and P0 are converter output power with
respect to nominal rated capacity, converter nominal power and
losses from the vacuums (load independent) respectively.
3. System design criteria
3.1. Social effect
3.1.1. Maximization of human development index (HDI)
In general, the human development index (HDIyz) is used as a
statistical tool to scale a nation's general achievement in its social
and financial measures. HDI is the sum of elements “y” and control
approach “z” evaluated based on given equation. In this proposed
paper consider that a division of yearly surplus energy is able to be
used by small workshops, new trades, employment that can reform
the standard of living and consequently the human development
index. The surplus energy which is generated by non-conventional
sources which can't be utilized; i.e. surplus energy is energy
generated during every hour by non-conventional sources such as
biomass generator, wind turbines and PV system which cannot be
consumed by ac load because ac load is less and it is as of now
secured. Such surplus energy can't be stored in the battery bank of
the system since it is completely charged. Part of the surplus energy
can be utilized by AC additional loads which were not considered
when the load was characterized. Thus surplus energy generates
new business and jobs The author developed an equation for log-
arithmic dependence by using 128 nations' data through the
various parameters [31].
h i h  .
HDI ¼ 0:091 ln PLY= þ 0:0724HDIy;z ¼ 0:0978 ln PLOAD þ min fmaxSE $PSE ; fmaxpLoad $PLOAD
C
where.
PLY= ¼ Energy yearly electricity Consumption per capita in kWh/
C
yr/person
fmaxSE ¼ Element of maximum surplus energy which may be
utilized for AC additional load
fmaxpLoad ¼ Element to increase yearly AC load in order that the
maximum surplus energy utilized by the new AC the additional
load may not be superior to that product
PSE ¼ yearly surplus energy of the system in kWh/yr
nHUMAN ¼ number of humans which consume the generated
power through the hybrid system.
3.1.2. Job creation
In India, the development non-conventional energy might
become an appropriate energy policy with exceptional environ-
mental benefits, in addition rising the use of native resources
(7)
pendency on conventional energy sources. The job creation analysis
can calculate by various methods, all method is shown in Fig. 6 [33]
the job creation is calculated as, the electrical energy supplied by
various energy source in a year for the number of employees per
Pout GWh/year. The job creation factor is 0.17 jobs/(GWh/yr) for energy
Pn supplied via DG in India during 1984e1985 [28]. Aforementioned
(8) job creation factor is improved because of technological progress
and better employment output; the improved value is 0.14jobs/
(GWh/yr) for power generated through the DG [29].
Analysis of job creation factor for other renewable energy (PV,
Wind) sources is by using different units. The job factor calculation
for PV panel and wind turbine power station in form of job.year/
MW here MW denotes maximum electricity generated by wind
plant and peak energy production through Photovoltaic plant. The
unit job.year/MW is a concern for single employee attached for a
period of one year, it also called direct employment. Nevertheless,
employees which work in O&M for the whole life of power gen-
eration plants are calculated in form of Jobs/MW units, such type of
jobs covered under the indirect employment [30]. PV power gen-
eration plant, the estimated job creation factor range is
0.41e2.48jobs/MW by taking the three different studies together
with the direct and indirect method. Similarly, for wind power
generation plant employment factor is calculated to the range is
0.39 to 0.8jobs/MW on the basis of five different cases. In the same
way job, creation factor for biomass power generation plant and
battery bank unit is 1.64jobs/MW and 0.01jobs/MWhr respectively
by using two different cases including a combination of direct and
indirect employment factor [31].
In literature, various methods are explained for the calculation
of PV and Wind power plant employment factor by considering
various cases with including and excluding direct and indirect
technique, which is given in Fig. 7 [32].
JCy;z ¼ JCPV $PPV þ JCWIND $PWIND þ JCBIOMASS $PBIOMASS þ JCD $ED
þ JCBATTERY $EBATTERY
(10)
i
hHUMAN  0:0319fmaxSE (9)
JCPV ¼ Number of job/MWP through the PV plant
PPV ¼ Peak power generation of PV plant in MWP
JCWIND ¼ Number of job/MWP through the wind plant
PWIND ¼ Power generated through the wind plant in MW
JCBIOMASS ¼ Number of job/MWP through the biomass plant
PBIOMASS ¼ Power generated through the biomass plant in MW
JCD ¼ Number of created through diesel generator in job/GWh/
yr
ED ¼ Yearly energy generated by diesel generator GWh/yr
JCBATTERY ¼ Job created by battery unit in job/MWh
EBATTERY ¼ Nominal capacity of battery unit in MWh
 Y. Sawle et al. / Renewable Energy 119 (2018) 459e472 465

gross job net job

direct & indirect &

direct job direct & indirect
induce job
job

analytical models in/out models & computable general

equilibrium models

job factor computable general

in/out models equilibrium models
job factor & multipliers

Fig. 6. Techniques for analysis of job factor HRES.

3.1.3. Particular matter (PM) 8

This work proposed the optimal size calculation with PV, wind,
PV wind
DG, biomass and battery unit amongst all sources DG and Biomass 6
generator unit producing emissions. The emission mainly consists
of carbon dioxide, carbon monoxide, Sulfur Dioxide (SO2), partic- 4
ular matter (PM), Nitrogen Oxides (NOx), Unburned Hydrocarbons
(UHC). Amongst all these elements of emission, particulate matter 2
(A blend of smoke, residue, and fluid beads that can bring about
0
respiratory issues) has an adverse effect on human health [39]
particularly from air pollutants comprising of little particles with
Min Mean max
a diameter less than or equal to 2.5 mm.
Fig. 7. Total job creation for PV and wind.
PM is also put into two categories namely PM2.5 (air pollutants
comprising of little particles with an aerodynamic diameter across
of not exactly or equivalent to 2.5 mm) and PM10 (air pollutants 3.1.4. Renewable factor
comprising of little particles with a diameter less than or equal to The electrical energy supplied to the required load demand
10 mm). Carbon is one of the components in atmospheric particulate through the renewable energy sources is known as a renewable
matter, which involves around 20e60% of PM2.5 fixation and for factor (RF). In proposed hybrid system renewable fraction explains
the most part exists as organic carbon (OC) and element carbon the limit of power supply as compared to non-renewable energy
(EC). Both of thus in PM are enhancing the probability of global source to renewable energy source. An ideal hybrid system which
climate effects, visibility degradation, animal health, ecosystems has hundred percent renewable fractions means total generated
and human health [40]. A study of the temporal behavior of par- power supply to the load from renewable energy source. Renew-
ticulate matter (PM2.5) and determination of the quantity of PM2.5 able fraction is calculated by given equation (16). According to this
of the HRES for remote location barwani in Madhya Pradesh, India equation if the value of renewable fraction is zero means power
has been done in this proposed work. The calculation of PM is based supply through non-renewable source (as PV, wind) is equal to
on linear regression equation given as [41,46]: supply of generator [24,37].
 P 
PM2:5 ¼ ð0:185*carbon contentÞ þ 0:05 with correlation R Pdiesel
Renewable fraction ¼ 1P P  100 (16)
pPV þ pwind
¼ 0:884
(11)

3.2. Technical effects

PM10 ¼ PM2:5 =0:6495 (12)
3.2.1. Loss of power supply probability (LPSP)
As a result of the infrequent nature of supply from renewable
PMtotalyr ¼ ðPM2:5 þ PM10 Þ (13)
energy resource due to variable wind velocity and solar radiation
characteristics, which significantly affect the subsequent power
PMreff 2:5 ¼ 40; PMreff 10 ¼ 60; production, control dependability examination has recently been
(14)
PMtotalreff yr ¼ PMreff 2:5 þ PMreff 10 considered as a crucial level in any design of system configuration.
Reliability is a major concern with load demand and a system which
 .  has enough electrical energy to fulfill the required load demand for
PM factor ¼ PMtotalyr PMtotalreff yr (15) a given time duration or another meaning, has a lesser value of
LPSP. LPSP is characterized as the possibility when necessary load
The reference values of PM2.5, PM10 are taken from national demand is not fulfilled by hybrid renewable energy system because
ambient air quality standards report which is shown in Table 2 [41]. of inadequate electrical power supply [22]. The range of LPSP is
R represents the regression values. between 0 and 1. If the value of LPSP is 1 it means HRES is not
466 Y. Sawle et al. / Renewable Energy 119 (2018) 459e472

Table 2
National ambient air quality standards (2009) [41].

Pollutants Time Weighted Concentration in Ambient Air Methods of

Average Measurement
Industrial, Residential, Rural and Ecologically Sensitive Area (Notified by Central
other Areas Government

Particulate Matter (Size less than 10 mm or Annual 60 100 Gravimetric

PM10, mg/m3) 24 Hours 60 100 TEOM
Beta attenuation
Particulate Matter (Size less than 2.5 mm or Annual 40 40 Gravimetric
PM2.5, mg/m3) 24 Hours 60 60 TEOM
Beta attenuation

proficient towards supply for required load demand furthermore if 0.6078kg/kWh, co2w is carbon weight engaged like tons, and ETRC is
the value of LPSP is 0, it means power supply through the HRES is tradable renewable energy certificate cost in $/kWh.
always fulfilling the required load demand. In any stand-alone
HRES, application of LPSP can be applied in two different ways: 3.3.2. Cost of energy
probabilistic method and chronological simulation. The probabi- In this section, HRES design based on Cost of energy (COE) is
listic method is utilized to integrate the intermittent nature of the described, it is a well-known and important factor for obtaining
renewable resource in addition to the load demand, which reduces cost-effective optimal hybrid system. Average cost per kWh of
necessitate for time series record. The need for time series record is Useful electrical power generated by HRES is known as COE [37].
reduced when probabilistic method is used to match the inter- The COE evaluated,
mittent nature of the renewable sources to the load demand.
However, the second method chronological simulation is compu- TNPC
COE ¼  CRF (20)
tationally tough and requires the accessibility of information tra- P
H¼8760
versed in a specific timeframe [34,35]. In this proposed work PL
H¼1
battery is used to store the energy for the feasible performance of
the hybrid system; so that LPSP is applying in the form chrono- Where PL is the hourly power consumption in kWh, TNPC is a
logical method. The LPSP is implemented for a range of time 0 to T total net present cost of $ and CRF is the capital recovery factor
which is explained in equation (18).The surplus energy must be (CRF).The CRF is a proportion utilized to figure current estimation
kept to a base to have a productive system together with the proper of benefits (a progression of equivalent yearly money streams). It is
feasible configuration of a storage unit in favor of a persistent calculated as given below equation [35]:
generation still with no accessible alternative sources towards
guarantee self-governing process [36]. The surplus energy is Ið1 þ IÞN
CRF ¼ (21)
calculated by equation (17). ð1 þ IÞN  1

X
T
PLOAD  PPV  Pwind  PBIoMASS where, I ¼ actual interest.
Surplus Energy ðSEÞ ¼ (17)
0
PPV þ Pwind þ PBIoMASS
N¼ Number of lifetime
PT
0 PLOAD  PPV  Pwind  PBIo þ PSOCM þ PDISEL 4. Multi-objection function
LPSP ¼ PT (18)
0 PLOAD
The multi-objective function contains six separate objectives in
where. SOCM Minimum state of charge which some objectives minimized like as cost of energy, LPSPI, PMI
and also some objectives are maximized like renewable factor, HDI
and JCI.
3.3. Economical effects Minimization of multi-objective function.
   
3.3.1. Penalty cost of emission
Diesel generator (DG) is used in proposed HRES which produces Min FMultiobjective ¼ COEI þ LPSPI þ 1
=RFI þ 1
=HDI þ PMI
a number of harmful gases such as carbon monoxide (CO), Carbon  
Dioxide (CO2), Sulphur dioxide (SO2), nitrogen oxides, particulate þ 1=JCI
matter and unburned hydrocarbons of all harmful gases, carbon
dioxide takes up the largest share of emissions. Because of this (22)
reason, the penalty cost is considered only for carbon dioxide in the Where, COEI, LPSPI, RFI, HDI, PMI, and JCI are COE, LPSP,
proposed case. The calculation of penalty cost of co2 emission renewable factor, human development, and particular matter and
through the tradable renewable energy certificate [38]. job creation indices respectively. This novel multi-objective func-
tion is made up of priority based indices, which has the first priority
C $E
CO2W ¼ C DG of COE, second priority of LPSP, third priority RF, the fourth priority
1016:04
  of HDI, the fifth priority of PM and six priority of JC.
E (19)
CO2 ¼ TRC $1016:04
CC 5. Optimization techniques
PC ¼ CO2W $CO
Hybrid system is designed by four different optimization tech-
where Pc is emission cost, Cc is carbon content engaged like niques as genetic algorithm (GA), particle swarm optimization
 Y. Sawle et al. / Renewable Energy 119 (2018) 459e472
(PSO), butterfly particle swarm optimization (BFPSO) and teacher
learning-based optimization. The fundamental points of interest of
the GA are that it is for the most part powerful in finding global
ideal results, especially in multimodal and multi-target improve-
ment issues. Genetic Algorithm basically works on 3 different op-
erators such as selection, crossover, and mutation for analysis of
results [35]. The PSO is first introduced through Kennedy in the
year 1995 is a famous optimization technique regarding the pop-
ulation of swarms. All through the PSO, the element has a learning
to fly nears a rising region. The PSO procedure introduces the
arbitrary position of particles and keeps informed their particles
location/position in light of the individual and nearby best
involvement [35,42,53]. In this section butterfly, intelligent
network is formed when the sensitivity of butterfly meet the
probability of the nectar so insensitive region minimum sensitivity
of butterfly matches minimum probability of nectar. In this region,
nodes are called into an active state, while in outer region sensi-
tivity of a butterfly is not able to detect the probability of nectar in
this region; all nodes are called in the inactive state [44]. The active
and inactive regions are separated by the boundary which is form
when the minimum sensitivity of butterfly meets to the minimum
probability of nectar. The TLBO algorithm is effective optimization
method [43e45]. The TLBO is inspired by the teaching-learning
process. The TLBO is based on the influence effect of a teacher on
the students (beginners) in a particular group. The main principle of
the method can explain with two parts, one is Teacher part, as well
as second, is Learner part. These two parts are the heart of opti-
mization method. In this process, a class of students is considered
as population with various topics as variables and corresponding
grades of learners be considered as the ‘fitness’ significance of
problem.
6. Results and discussion
This section presents the optimal design and sizing of HRES for
rural electrification of rural area district Barwani, India as a case
study using GA, PSO, BFPSO and TLBO optimization techniques. The
optimal design and sizing of HRES for rural electrification in bar-
wani district of India are presented as a case study using GA, PSO,
BFPSO and TLBO methods. The viable planning of HRES comprises
of power back up of battery bank as well as diesel generator (DG).
MATLAB (2009a) has been used in implementing this work. For GA,
PSO, BFPSO, and TLBO the population size, swarm size and class size
have been taken as 30 and the number of iterations is 50. The
minimization of multi-objective function (Fmulti-objective) is consid-
ered for each case/configuration using PSO, BFPSO, GA and TLBO;
hence the optimal solutions are obtained according to the lowest
value of the multi-objective function (Fmulti-objective). Six separate
objective indices are included in the multi-objective function. To
find the minimum value of the multi-objective function (Fmulti-
objective) is the main objective and also the significance of feasibility
for the objective indices is determined by the minimum value of
(Fmulti-objective). The different optimal configuration of hybrid
renewable energy systems are presented as follows:
 Case-I: Hybrid system with PV, Biomass, Diesel & Battery
 Case-II: Hybrid system with PV, Diesel & Battery
 Case-III: Hybrid system with Wind, Biomass, Diesel & Battery
 Case-IV: Hybrid system with Wind, Diesel & Battery
 Case-V: Hybrid system with PV, Wind, Diesel & Battery
 Case-VI: Hybrid system with PV, Wind, Biomass, Diesel &
Battery
467
6.1. CASE-I (PV, Biomass, Diesel & Battery)
The optimal design of hybrid system including PV, Biomass,
Diesel Generator (DG) and Battery using GA, PSO, BFPSO and TLBO
are given in this section. The optimal results are presented in
Tables 3and4 respectively. Table 3 demonstrates the values of
multi-objective function and indices for case-I using GA, PSO,
BFPSO and TLBO. The minimum values of the multi-objective
function (Fmulti-objective) are 8.310147, 7.902398, 7.894871 and
7.781196 using GA, PSO, BFPSO and TLBO respectively. The optimal
values of different objectives with respect to minimum value of
Fmulti-objective are also given in Table 3 with different techniques. The
optimal solution for different parameters of hybrid system is re-
ported in Table 4. The Npv, Nbatt, Emission, Penalty cost, PM and
Diesel operating hours for GA are 100, 31, 5.901744, 145.65015,
28.99825 & 738; for PSO are 100, 33, 5.612548, 138.513034,
27.63951 & 641; for BFPSO are 123, 30, 5.474288, 135.100895,
26.98992 & 606 and for TLBO are 154, 30, 5.040458, 124.39432,
24.95163 & 460 respectively. Table 3 shows annual power of hybrid
system by different units.
The comparative analysis of various parameters such as emis-
sion, penalty cost, PM, COE, LPSP, RF, HDI and Job creation are given
using GA, PSO, BFPSO and TLBO, which shows that the optimal
results obtained by TLBO are superior to GA as well as PSO & BFPSO.
The values of emission, penalty cost, PM, COE, LPSP are minimum
for TLBO as compared GA, PSO & BFPSO and also the values of RF,
HDI and Job creation are higher for TLBO than GA, PSO & BFPSO. The
effective optimal configuration solution of the hybrid system is
achieved by TLBO, therefore, the various power profile results are
reported for a more optimal solution (i.e. using TLBO).
6.2. CASE-II (PV, Diesel & Battery)
This section presents the optimal design of hybrid system con-
sisting of PV, Diesel Generator (DG) and Battery using GA, PSO,
BFPSO and TLBO. The optimal results for case-II are illustrated in
Tables 3and4 respectively. The optimal values of different objective
indices with respect to minimum value of Fmulti-objective for case-II
using GA, PSO, BFPSO and TLBO are expressed in Table 4. The
minimum values of the multi-objective function (Fmulti-objective) are
7.893357, 7.130776, 7.11042 and 7.085658 using GA, PSO, BFPSO and
TLBO respectively. Table 3 exhibits the optimal solution for different
parameters of the hybrid system. The annual power requirement of
a hybrid system for different units is given in Table 4.
The evaluation of different object parameters such as emission,
penalty cost, PM, COE, LPSP, RF, HDI and Job creation are given using
GA, PSO, BFPSO and TLBO, which clearly shows the superiority of
TLBO. The better solution of the hybrid system is attained by TLBO,
hence the various power profile results are computed for a more
optimal solution.
6.3. CASE-III (Wind, Biomass, Diesel & Battery)
Case-III considers the optimal design of hybrid system consist-
ing of Wind, Biomass, Diesel Generator (DG) and Battery using GA,
PSO, BFPSO and TLBO. The optimal results are presented in
Tables 3and4 respectively. The multi-objective function and indices
values for case-III using GA, PSO, BFPSO and TLBO are given in
Table 3. The minimum values of the multi-objective function
(Fmulti-objective) are 11.44973, 11.28528, 11.19663 and 11.19663
using GA, PSO, BFPSO and TLBO respectively. The optimal solutions
of objective indices regarding minimum value of Fmulti-objective are
specified in Table 3 using different techniques.
The comparative analysis of emission, penalty cost, PM, COE,
LPSP, RF, HDI and Job creation using GA, PSO, BFPSO and TLBO
468 Y. Sawle et al. / Renewable Energy 119 (2018) 459e472

Table 3
The multi-objective function and indices value of all six cases.

Cases Optimization Technique F multi-objective COEI LPSPI PMFI RFI HDI JCI

Case I GA 8.310147 0.224285 0.026954 0.289983 0.931305 0.570924 0.219195

PSO 7.902398 0.223854 0.024092 0.276395 0.940188 0.570683 0.219195
BFPSO 7.894871 0.223584 0.026217 0.269899 0.953988 0.575724 0.218771
TLBO 7.781196 0.21352 0.021879 0.249516 0.971603 0.581701 0.219879
Case II GA 7.893357 0.239472 0.024593 0.106549 0.951105 0.57696 0.248925
PSO 7.130776 0.237113 0.024468 0.098045 0.961128 0.580892 0.249425
BFPSO 7.11042 0.236427 0.023416 0.092713 0.961003 0.580142 0.249246
TLBO 7.085658 0.236098 0.023635 0.08698 0.969609 0.584215 0.250256
Case III GA 11.44973 0.311599 0.043633 0.401125 0.897789 0.580757 0.127265
PSO 11.28528 0.309199 0.041292 0.392315 0.895906 0.578875 0.12989
BFPSO 11.27468 0.30892 0.041124 0.392305 0.892546 0.577971 0.131372
TLBO 11.19663 0.309075 0.040292 0.387431 0.895122 0.577905 0.131372
Case IV GA 10.33335 0.344604 0.045702 0.263409 0.883448 0.581697 0.146443
PSO 10.14558 0.342039 0.044339 0.262145 0.873164 0.579081 0.150945
BFPSO 10.13789 0.342037 0.044205 0.261122 0.87582 0.57911 0.150945
TLBO 10.07422 0.342012 0.043328 0.258843 0.870805 0.578106 0.152658
Case V GA 9.680154 0.228802 0.02379 0.104699 0.955273 0.578423 0.152737
PSO 9.446509 0.227777 0.021391 0.090858 0.960831 0.577321 0.157889
BFPSO 9.414946 0.226921 0.021259 0.090848 0.962417 0.578464 0.158019
TLBO 9.161588 0.22662 0.018733 0.085148 0.967823 0.580986 0.164565
Case VI GA 11.83285 0.205789 0.019367 0.255269 0.971298 0.583569 0.116154
PSO 10.49285 0.196498 0.019054 0.255372 0.964019 0.576729 0.137918
BFPSO 10.48324 0.196945 0.017409 0.249694 0.966918 0.576607 0.137918
TLBO 10.18189 0.195166 0.017638 0.251353 0.967126 0.577653 0.143852

Table 4
Output result of each parameter of HRES.

Cases Technique Parameters

Npv Nwt Nbatt Emission(t) Penalty cost ($) PM Diesel yr (hr) PV(kW) Wind (kW) Biomass (kW) E DCH yr Diesel (kW)

Case I GA 100 - 31 5.901744 145.65015 28.99825 738 48356.04 - 5209.129 20332 3679.643
PSO 100 - 33 5.612548 138.513034 27.63951 641 48356.04 - 5209.129 20332 3203.835
BFPSO 123 - 30 5.474288 135.100895 26.98992 606 59477.93 - 5209.129 19946 2976.359
TLBO 154 - 30 5.040458 124.39432 24.95163 460 74468.3 - 5209.129 19595 2262.588
Case II GA 139 - 35 1.997518 49.297092 10.65485 598 67214.89 - - 22820.77 3286.473
PSO 159 - 34 1.816526 44.830352 9.804488 551 76886.1 - - 22572.75 2988.69
BFPSO 167 34 1.754148 42.02962 9.271294 532 80754.58 22486.42 2801.975
TLBO 177 - 34 1.581015 39.018135 8.697976 477 85590.19 - - 22387.57 2601.209
Case III GA - 32 9 8.267307 204.030286 40.11248 1397 - 68869.81 5209.129 11764.14 7571.7
PSO - 30 12 8.079795 199.402651 39.23149 1344 - 64565.44 5209.129 12080.75 7263.1
BFPSO e 29 14 8.0717 199.0079 39.221 1342 e 62413.26 5209.129 12255.93 7206.305
TLBO - 29 14 7.975856 196.837514 38.74315 1309 - 62413.26 5209.129 12255.93 7092.1
Case IV GA e 35 11 5.33616 131.691994 26.34094 1498 - 75326.35 - 13698.4 8779.5
PSO e 32 14 5.309256 131.02803 26.21454 1494 - 68869.81 - 14229.71 8735.2
BFPSO e 32 15 5.28074 130.2842 26.20092 1487 e 68869.81 e 14423.71 8692.28
TLBO e 31 16 5.238974 129.293526 25.88433 1479 - 66717.63 - 14428.06 8619.6
Case V GA 100 11 22 1.958161 48.32579 10.46994 621 48356.04 23674 - 13485.99 3221.7
PSO 100 10 25 1.663561 41.0553 9.085805 530 48356.04 21521.81 - 14028.18 2737
BFPSO 100 10 27 1.6526 40.77139 9.085147 521 48356.04 21521.81 - 14182.18 2688.1
TLBO 123 9 25 1.542035 38.05614 8.514833 484 59477.93 19369.63 - 14380.04 2537.1
Case VI GA 100 15 16 5.769044 142.3752 28.37478 521 48356 32282.72 5209.129 9622.04 2467.652
PSO 91 9 21 5.165096 127.4703 25.53723 517 44004 19369.63 5209.129 12094.74 2464.022
BFPSO 91 9 22 5.044248 124.4879 24.96944 482 44004 19369.63 5209.129 12094.74 2368.823
TLBO 100 8 22 5.07954 125.3588 25.13525 491 48356 17217.45 5209.129 12581.02 2326.888

respectively is presented, which shows the performance of TLBO is
better than GA and PSO, BFPSO. The efficient optimal solution of the
hybrid system is attained using TLBO, so all power profile results of
case-III are represented for an optimal solution using TLBO. The
results analysis clearly indicates that the better solution is achieved
by using TLBO.
6.4. CASE-IV (Wind, Diesel & Battery)
The optimal design of hybrid system including Wind, Diesel
Generator (DG) and Battery using GA, PSO and TLBO are given in
this section. The optimal results are presented in Tables below.
Table 3 demonstrates the values of multi-objective function and
indices for case-IV using GA, PSO, BFPSO and TLBO. The minimum
values of the multi-objective function (Fmulti-objective) are 10.33335,
10.14558, 10.13789 and 10.07422 using GA, PSO, BFPSO and TLBO
respectively. The optimal values of different objectives with respect
to the minimum value of Fmulti-objective are also given in Table 3 with
different techniques.
The comparative analysis of various parameters such as emis-
sion, penalty cost, PM, COE, LPSP, RF, HDI and Job creation are given
using GA, PSO, BFPSO and TLBO, which shows that the optimal
results obtained by TLBO are superior to GA as well as PSO & BFPSO.
The values of emission, penalty cost, PM, COE, LPSP are minimum
 Y. Sawle et al. / Renewable Energy 119 (2018) 459e472 469

25
Spring Summer PV
Wind
Biomass
20 Battery
Diesel
Load demand
15 ESurplus
Power (kW)

Winter
Fall
10

0
1 24 48 72 96
Hours

Fig. 8. Seasonal power profile of each unit for hybrid system of case-VI.

30

25

20
PV Power (kW)

15

10

0
0 1000 2000 3000 4000 5000 6000 7000 8000 9000
Time (hr)

Fig. 9. PV power profile for hybrid system of case-VI.

5
Wind Power (kW)

0
0 1000 2000 3000 4000 5000 6000 7000 8000 9000
Time (hr)

Fig. 10. Wind power profile for hybrid system of case-VI.

for TLBO as compared GA, PSO & BFPSO and also the values of RF,
HDI and Job creation are higher for TLBO than GA, PSO & BFPSO. The
effective optimal configuration solution of the hybrid system is
achieved by TLBO, therefore, the various power profile results are
reported for a more optimal solution (i.e. using TLBO).
6.5. CASE-V (PV Wind Diesel & Battery)
The optimal design of hybrid system including PV, Wind, Diesel
Generator (DG) and Battery using GA, PSO and TLBO are given in
this section. The optimal results are presented in Tables. Table 3
demonstrates the values of multi-objective function and indices
for case-V using GA, PSO, BFPSO and TLBO. The minimum values of
the multi-objective function (Fmulti-objective) are 9.680154, 9.446509,
9.414946 and 9.161588 using GA, PSO, BFPSO and TLBO respectively.
The optimal values of different objectives with respect to minimum
value of Fmulti-objective are also given in Table 3 with different tech-
niques. The optimal solution for different parameters of hybrid
system is reported in Table 4. The Npv, Nwind, Nbatt, Emission,
Penalty cost, PM and Diesel operating hours for GA are 100, 11, 22,
1.958161, 48.32579, 10.46994 & 621; for PSO are 100, 10, 25,
1.663561, 41.0553, 9.085805 & 530; for BFPSO are 100, 10, 27,
1.6526, 40.77139, 9.085147& 521 and for TLBO are 123, 9, 25,
1.542035, 38.05614, 8.514833, 484 respectively. Table 4 shows
470 Y. Sawle et al. / Renewable Energy 119 (2018) 459e472

6
Diesel Generator Power (kW)

0
0 1000 2000 3000 4000 5000 6000 7000 8000 9000
Time (hr)

Fig. 11. DG power profile for hybrid system of case-VI.

2.5

2
Fuel consumtion

1.5

0.5

0
0 1000 2000 3000 4000 5000 6000 7000 8000 9000
Time (hr)

Fig. 12. Fuel consumption profile for hybrid system of case-VI.

5
E batt (kWh)

0
0 1000 2000 3000 4000 5000 6000 7000 8000 9000
Time (hr)

Fig. 13. Battery power profile for hybrid system of case-VI.

annual power of hybrid system by different units.
The comparative analysis of various parameters such as emis-
sion, penalty cost, PM, COE, LPSP, RF, HDI and Job creation as il-
lustrates in tables using GA, PSO, BFPSO and TLBO, which shows
that the optimal results obtained by TLBO are superior to GA as well
as PSO & BFPSO. The values of emission, penalty cost, PM, COE, LPSP
are minimum for TLBO as compared GA, PSO & BFPSO and also the
values of RF, HDI and Job creation are higher for TLBO than GA, PSO
& BFPSO. The effective optimal configuration solution of the hybrid
system is achieved by TLBO, therefore, the various power profile
results are reported for a more optimal solution (i.e. using TLBO).
6.6. CASE-VI (PV, Wind, Biomass, Diesel & Battery)
The optimal design of hybrid system including PV, Wind,
Biomass, Diesel Generator (DG) and Battery using GA, PSO, BFPSO
and TLBO are given in this section. The optimal results are pre-
sented in Tables. Table 3 demonstrates the values of multi-objective
function and indices for case-VI using GA, PSO, BFPSO and TLBO.
The minimum values of the multi-objective function (Fmulti-objective)
are 11.83285, 10.49285, 10.48324 and 10.18189 using GA, PSO,
BFPSO and TLBO respectively. The optimal values of different ob-
jectives indices with respect to minimum value of Fmulti-objective are
also given in Table 3 with different techniques. The optimal solution
for different parameters of hybrid system is reported in Table 4. The
Npv, Nwind, Nbatt, Emission, Penalty cost, PM and Diesel operating
hours for GA are 100, 15, 16, 5.769044, 142.3752, 28.37478 & 521;
for PSO are 91, 9, 21, 5.165096, 127.4703, 25.53723 & 517; for BFPSO
are 91, 9, 22, 5.044248, 124.4879, 24.96944 & 482 and for TLBO are
100, 8, 22, 5.07954, 125.3588, 25.13525 & 491 respectively. Table 4
shows annual power of hybrid system by different units.
The comparative analysis of various parameters such as emis-
sion, penalty cost, PM, COE, LPSP, RF, HDI and Job creation as given
in tables using GA, PSO and TLBO, which shows that the optimal
results obtained by TLBO are superior to GA as well as PSO & BFPSO.
The values of emission, penalty cost, PM, COE, LPSP are minimum
 Y. Sawle et al. / Renewable Energy 119 (2018) 459e472
16
15
14
F multi-objective
13
12
11
10
5 10 15 20
No. of Iterations
Fig. 14. Multi-objective function (Fmulti-objective) convergence using GA, PSO, BFPSO & TLBO for case-VI.
for TLBO as compared GA, PSO & BFPSO and also the values of RF,
HDI and Job creation are higher for TLBO than GA, PSO & BFPSO. The
effective optimal configuration solution of the hybrid system is
achieved by TLBO, therefore the various power profile results are
reported for a more optimal solution (i.e. using TLBO). The seasonal
power profile for each unit (PV, Biomass, DG & Battery) to manage
the power requirement optimally of the hybrid system is shown in
Fig. 8. The annual profile of various units such as PV power, DG
power, fuel consumption and battery power profile for hybrid
system of case-VI are demonstrated in Figs. 9e13 respectively.
Annual profile of each unit of hybrid system includes the seasonal
variation. The convergence of optimization process to minimize
multi-objective function (Fmulti-objective) using GA, PSO, BFPSO &
TLBO for case-VI is given in Fig. 14. The results exhibit the ascending
order of fine convergence is GA, PSO and TLBO. Hence better
convergence is achieved by using TLBO.
7. Conclusion
The optimal planning of diverse HRES with a multi-objective
function applying GA, BFPSO, PSO and TLBO is presented for a
remote location of district barwani. A summary of results obtained
by considering hybrid system parameters and multiple objectives
involving different cases I to VI are discussed including a new
multi-objective function. Optimal configurations of parameters
have been obtained using GA, BFPSO, PSO and TLBO for the new
multi-objective function involving six separate objectives. The
various parameters such as technical (LPSP, Renewable factor),
economical (COE, Penalty & Fuel consumption) and social (Job
creation, HDI & PM) attributes are explored in the optimal design of
hybrid system. The results indicate that in all cases from I to VI the
best solution is obtained using TLBO, then BFPSO, PSO and GA
respectively. The comparative results analysis of hybrid system
configuration parameters and objective parameters for different
cases from I to VI are presented. The optimal design of a different
hybrid system using GA, PSO, BFPSO and TLBO are given on the
basis of the minimum value of the multi-objective function (Fmulti-
objective) and optimal values of objective indices or parameters. The
optimal values of COE for different hybrid system configuration
cases I, II, III, IV, V & VI are 0.21352, 0.236098, 0.309075, 0.342012,
0.22662 and 0.195166 respectively. Also, the optimal values of LPSP
471
GA
PSO
BFPSO
TLBO
25 30 35 40 45 50
for different hybrid system configuration cases from I to VI are
0.021879, 0.023635, 0.040292, 0.043328, 0.018733 and 0.017638
respectively. Similarly, the comparative analysis can be given for
PM, HDI, Job creation, Renewable factor, Emission, Penalty cost and
Diesel operating hours etc. This investigation indicates that each
case has the best solution using TLBO and case VI is the more
optimal solution using TLBO technique as compared to other cases
with respect to all objective (COE, LPSP, Renewable factor, PM, HDI,
JCI & Emission) parameters consideration. In the case e VI for the
hybrid system configuration COE is 0.195166 and LPSP are 0.017638
which is the minimum solution obtained among all the cases
considered. Similarly for Renewable factor, HDI, Emission, PM, job
creation, Fuel cost, Penalty cost and Diesel operating hours etc. a
comparative analysis can be presented. Finally, it can be concluded
that the TLBO has the superior performance for each of the cases
considered. The technical, economic and social parameters of
hybrid system are successfully improved by using the proposed
methodology.
References
[1] Sonali Goel, Renu Sharma, Performance evaluation of stand-alone, grid con-
nected and hybrid renewable energy systems for rural application: a
comparative review, Renew. Sustain. Energy Rev. 78 (2017) 1378e1389.
[2] J.L. Bernal-Agustín, R. Dufo-L_opez, Simulation and optimization of stand-
alone hybrid renewable energy systems, Renew. Sustain. Energy Rev. 13
(2009) 2111e2118.
[3] P. Nema, R.K. Nema, S. Rangnekar, A current and future state of art develop-
ment of hybrid energy system using wind and PV-solar: a review, Renew.
Sustain. Energy Rev. 13 (2009) 2096e2103.
[4] Vivek Tamrakar, S.C. Gupta, Yashwant Sawle, Study of characteristics of single
and double diode electrical equivalent circuit models of solar PV module, in:
Energy Systems and Applications, 2015 International Conference on. IEEE,
2015.
[5] R. Dufo-L_opez, J.L. Bernal-Agustín, Multi-objective design of PV-wind-diesel
hydrogen-battery systems, Renew. Energy 33 (2008) 2559e2572.
[6] P. Hajela, C.Y. Lin, Genetic search strategies in multi-criterion optimal design,
Struct. Optim. 4 (1992) 99e107.
[7] C. Shyu, Ensuring access to electricity and minimum basic electricity needs as
a goal for the post-MDG development agenda after 2015, Energy Sustain Dev.
19 (2014) 29e38.
[8] L.R. Valer, A. Mocelin, R. Zilles, E. Moura, C.S. Nascimento, Assessment of so-
cioeconomic impacts of access to electricity in Brazilian Amazon: case study in
two communities in Mamiraua  Reserve, Energy Sustain Dev. 20 (2014) 58e65.
[9] Human Development Report 2014, Sustaining Human Progress: Reducing
Vulnerabilities and Building Resilience, United Nations Development Pro-
gramme, 2014, 978e992-1-126340-4.
472 Y. Sawle et al. / Renewable Energy 119 (2018) 459e472
[10] M. Gomez, S. Silveira, Rural electrification of the Brazilian Amazon d
Achievements and lessons, Energy Policy 38 (2010) 6251e6260.
[11] I. Oosterlaken, Design for development; a capability approach, Des. Issues 25
(4) (2009) 91e102.
[12] ul Haq M, Reflections on Human Development, Oxford University Press, 1995.
[13] J.A. Momoh, M.E. El-Hawary, R. Adapa, A revie of selected optimal power flow
literature to 1993. Part II: Newton, linear programming, and interior point
methods, IEEE Trans. Pow. Syst. 14 (1) (Feb. 1999), 105111.
[14] J.F. Manwell, et al., An investigation of variable speed operation of diesel
generators in hybrid energy systems, Renew. energy 2.6 (1992) 563e571.
[15] Amanda Northcross, et al., Estimating personal PM2. 5 exposures using CO
measurements in Guatemalan households cooking with wood fuel, J. Environ.
Monit. 12.4 (2010) 873e878.
[16] D. McKain, S. Wayne, N. Clark, Relationship between Carbon Monoxide and
Particulate Matter Levels across a Range of Engine Technologies, SAE Tech-
nical Paper 2012011346, 2012.
[17] Y.S. Zhao, et al., The optimal capacity configuration of an independent wind/
PV hybrid power supply system based on improved PSO algorithm, in: Ad-
vances in Power System Control, Operation and Management (APSCOM 2009),
8th International Conference on. IET, 2009.
[18] Yashwant Sawle, S.C. Gupta, A novel system optimization of a grid indepen-
dent hybrid renewable energy system for telecom base station, Int. J. Soft
Comput. Math. Control (IJSCMC) 4 (2) (May 2015) 49e57.
[19] Yashwant Sawle, S.C. Gupta, Aashish Kumar Bohre, Optimal sizing of stand-
alone PV/Wind/Biomass hybrid energy system using GA and PSO optimization
technique, Energy Procedia 117 (2017) 690e698.
[20] Yashwant Sawle, S.C. Gupta, Optimal sizing of photo voltaic/wind hybrid
energy system for rural electrification, in: Power India International Confer-
ence (PIICON), 2014 6th IEEE, IEEE, 2014, pp. 1e4.
[21] J. Park, Jeongje, et al., A probabilistic reliability evaluation of a power system
including solar/photovoltaic cell generator, in: 2009 IEEE Power & Energy
Society General Meeting, IEEE, 2009.
[22] B. Ghosh, Towards Modernized Bio-energy e A First Hand Book for the
Stakeholders, School of Energy Studies, Jadavpur University, Kolkata, India,
2004. Technical Backup Unit on Biomass Gasifier Promotion (A Unit of Min-
istry of Non-Conventional Energy Sources, Govt. of India).
[23] A. Gupta, Modelling of Hybrid Energy System, Ph.D. thesis, Alternate Hydro
Energy Centre, Indian Institute of Technology, Roorkee, 2010.
[24] Maleki Akbar, Ameri Mehran, Keynia Farshid, Scrutiny of multifarious particle
swarm optimization for finding the optimal size of a PV/wind/battery hybrid
system, Renew. Energy 80 (2015) 552e563.
[25] C.G. Justus, Wind energy statistics for large arrays of wind turbines (New
England and Central US Regions), Sol. Energy 20.5 (1978) pp.379e386.
[26] www.int-elec.com/pdf/varispeed.pdf.
[27] V. Prema, Swagatha Datta, Uma Rao, An effective dispatch strategy for hybrid
power management, in: TENCON 2015-2015 IEEE Region 10 Conference, IEEE,
2015.
[28] R. Ramanathan, L.S. Ganesh, Energy resource allocation incorporating quali-
tative and quantitative criteria: an integrated model using goal programming
and AHP, Socio-Economic Plan. Sci. 29.3 (1995) 197e218.
[29] Tesis doctoral J.C. Rojas-Zerpa, Planificaci_on del suministro el_ectrico en
_areas rurales de los países en vías de desarrollo: un marco de referencia para
la toma de decisiones, Univ. Zaragoza, 2012.
[30] M. Wei, S. Patadia, D.M. Kammen, Putting renewables and energy efficiency to
work: how many jobs can the clean energy industry generate in the US?
Energy Policy 38 (2010), 919e931.
[31] Rodolfo Dufo-Lo pez, Iv  bal-Monreal, Jose
an R. Cristo  M. Yusta, Optimisation of
PV-wind-diesel-battery stand-alone systems to minimise cost and maximise
human development index and job creation, Renew. Energy 94 (2016)
View publication stats
280e293.
[32] L. Cameron, B. van der Zwaan, Employment factors for wind and solar energy
technologies: a literature review, Renew. Sustain. Energy Rev. 45 (2015),
160e172.
[33] Margarita Ortega, et al., Employment effects of renewable electricity
deployment. A novel methodology, Energy 91 (2015) 940e951.
[34] Hongxing Yang, Wei Zhou, Lin Lu, Zhaohong Fang, Optimal sizing method for
stand-alone hybrid solarewind system with LPSP technology by using genetic
algorithm, Sol. Energy 82 (4) (April 2008) 354e367.
[35] Sawle, Y., Gupta, S.C. and Bohre, A.K. Techno-economic scrutiny of HRES
through GA and PSO technique, Int. J. Renew. Energy Technol., Vol. X, No. Y,
pp.xxxexxx.
[36] R.K. Rajkumar, V.K. Ramachandaramurthy, B.L. Yong, D.B. Chia, Techno-
economical optimization of hybrid pv/wind/battery system using Neuro-
Fuzzy, Energy 36 (8) (August 2011) 5148e5153.
[37] http://www.homerenergy.com.
[38] S.C. Gupta, Y. Kumar, Gayatri Agnihotri, REAST: Renewable energy analysis
and sizing tool, J. Electr. Syst. 7e2 (2011) 206e224.
[39] http://www.homerenergy.com/.
[40] Wilbes Mbiya, The Behaviour of Particulate Matter and Carbon Dioxide Con-
centrations in Harare. Diss, UNIVERSITY OF ZIMBABWE, 2007.
[41] http://cpcb.nic.in/upload/NewItems/NewItem_197_NAAQMS_Volume-II.pdf.
[42] A.K. Bohre, G. Agnihotri, M. Dubey, Optimal sizing and sitting of DG with load
models using soft computing techniques in practical distribution system, IET
Gener. Transm. Distrib. 10 (11) (8 4 2016) 2606e2621.
[43] R.V. Rao, V.J. Savsani, D.P. Vakharia, Teachingelearning-based optimization:
an optimization method for continuous non-linear large scale problems, Inf.
Sci. 183 (1) (2012) 1e15.
[44] Aashish Kumar Bohre, et al., A novel method to find optimal solution based on
modified butterfly particle swarm optimization, Int. J. Soft Comput. Math.
Control (IJSCMC) 3.4 (2014) 1e14.
[45] Aashish Kumar Bohre, et al., Impacts of the load models on optimal planning
of distributed generation in distribution system, Adv. Artif. Intell. 2015 (2015)
6.
[46] Jing Duan, Yong Chen, Wen Fang, Zhengjun Su, Characteristics and Relation-
ship of PM, PM10, PM2.5 Concentration in a Polluted City in Northern China,
Procedia Eng. 102 (2015) 1150e1155.
[47] Yashwant Sawle, S.C. Gupta, Aashish Kumar Bohre, PV-wind hybrid system: A
review with case study, Cogent Eng. 3.1 (2016) 1189305.
[48] Yashwant Sawle, S.C. Gupta, Aashish Kumar Bohre, Review of hybrid renew-
able energy systems with comparative analysis of off-grid hybrid system,
Renew. Sustain. Energy Rev. 81 (2) (2018) 2217e2235.
[49] Vivek Tamrakar, S.C. Gupta, Yashwant Sawle, Single-diode and two-diode pv
cell modeling using matlab for studying characteristics of solar cell under
varying conditions, Electr. Comput. Eng. An Int. J. (ECIJ) 4 (2) (June 2015)
67e77.
[50] Vivek Tamrakar, S.C. Gupta, Yashwant Sawle, Single-diode pv cell modeling
and study of characteristics of single and two-diode equivalent circuit, Electr.
Electron. Eng. An Int. J. (ELELIJ) 4 (3) (August 2015).
[51] Rangan Banerjee, Comparison of options for distributed generation in India,
Energy Policy 34 (1) (2006) 101e111.
[52] Said Diaf, et al., A methodology for optimal sizing of autonomous hybrid PV/
wind system, Energy Policy 35.11 (2007) 5708e5718.
[53] C. Darras, et al., Sizing of photovoltaic system coupled with hydrogen/oxygen
storage based on the ORIENTE model, Int. J. Hydrogen Energy 35.8 (2010)
3322e3332.
[54] Hanieh Borhanazad, et al., Optimization of micro-grid system using MOPSO,
Renew. Energy 71 (2014) 295e306.