Energy Conversion and Management 106 (2015) 1024–1038

Contents lists available at ScienceDirect

Energy Conversion and Management

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

Economic and environmental bi-objective design of an off-grid

photovoltaic–battery–diesel generator hybrid energy system
Marco Bortolini a, Mauro Gamberi b,⇑, Alessandro Graziani a, Francesco Pilati b
a
Department of Industrial Engineering, University of Bologna, Viale del Risorgimento 2, 40136 Bologna, Italy
b
Department of Management and Engineering, University of Padova, Stradella San Nicola, 3, 36100 Vicenza, Italy

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

Article history: Generally, off-grid fossil fuel generators provide energy supply to remote regions. The integration of pho-
Received 22 July 2015 tovoltaic (PV) plants to battery energy storage (BES) systems potentially increases reliability, the system
Accepted 16 October 2015 autonomy and lifetime, reducing the generator working hours and the system environmental impact. PV–
Available online 11 November 2015
BES–Diesel generator hybrid energy systems (HESs) offer technical, economic and environmental benefits
compared to traditional off-grid systems. This paper proposes a bi-objective design model for off-grid PV–
Keywords: BES–Diesel generator HESs. The aim is to identify the PV plant rated power, the BES system capacity and
Hybrid energy system
the technical configuration able to jointly reduce the levelised cost of the electricity ðLCOEÞ and the car-
Photovoltaic plant
Battery energy storage
bon footprint of energy ðCFOEÞ. Furthermore, the comparison of the LCOE and CFOE values of the HES
Diesel generator against a traditional diesel generator allows determining the economic and environmental advantages
Levelised cost of electricity coming from the described system. Despite the proposed model is general and suitable for any installa-
Carbon footprint of electricity tion site and HES configuration, this paper exemplifies its application designing a HES to be installed in a
remote village in Yakutsk, Russia. The model takes into account the hourly energy demand, the irradia-
tion and temperature profiles of the installation location calculating the hourly PV plant yield, the battery
charge–discharge processes and the required generator energy. Results highlight the technical, economic
and environmental feasibility of the system for a context with a medium irradiation level, i.e.
1400 kW h/(m2 year), and relatively low fuel cost, i.e. 0.7 €/l. For the best economic scenario LCOE
and CFOE reductions are of about 8% and 28%, respectively. Finally, the most effective trade-off between
economic and environmental performances leads to a CFOE decrease of about 48% and a slight decrease of
the economic performances (2%).
Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction for electrification of rural areas. A recent review is by Akikur

et al. [3]. Table 1 classifies recent papers on HESs integrating PV
Worldwide, about 1.3 billion people still have no access to plants, BES units and other RESs and traditional sources. Deep
reliable electricity services and rural electrification despite efforts attention is paid on the PV–BES–Diesel generator configurations.
for bringing electrical power to remote regions are encouraged HESs have the purpose to meet the energy demand optimising
by national and international agencies and work programmes [1]. the operation of each energy source and energy storage system.
At the same time, in remote areas, the high investments for grid The identification of the most effective size for each component
connections and the parallel cost increase of fossil fuels make and of the most proper operation control strategy are crucial for
hybrid energy systems (HESs) potentially competitive, creating the reduction of the lifetime cost, the environmental impact, the
favorable conditions for their diffusion within off-grid applications unmet load demand and other key objective functions [41]. Khatib
[2]. HESs integrate renewable energy sources (RESs), e.g. photo- et al. [42] present a review of optimisation techniques for HESs
voltaic (PV) plant, small wind turbines, etc., fossil fuel based integrating PV plants, while Luna-Rubio et al., Erdinc & Uzunoglu
devices, e.g. diesel generators, and energy storage systems to meet and Upadhyay & Sharma [43–45] review the configurations, con-
the energy demand at any time and everywhere. The interest in trol and sizing methodologies for HESs highlighting the technolog-
HESs increases for both grid connected and off-grid applications. ical, economic, socio-political and environmental factors involved
Several scientific contributions face the design of HESs suitable within the HES assessment criteria. Globally, the current literature
highlights the technical feasibility and the long-term economic
⇑ Corresponding author. Tel.: +39 0444 998735; fax: +39 0444 998888. sustainability as the most frequently adopted design metrics. The
E-mail address: [email protected] (M. Gamberi). inclusion of the environmental sustainability perspective is less

http://dx.doi.org/10.1016/j.enconman.2015.10.051
0196-8904/Ó 2015 Elsevier Ltd. All rights reserved.
 M. Bortolini et al. / Energy Conversion and Management 106 (2015) 1024–1038 1025

Nomenclature

Aa PV system area, m2 Plim

B;D BES discharge power limit, kW
Am PV single module area, m2 PD power dissipated through the PV inverter, kW
AH BES autonomy hours, h PG;h generator power for hour h, kW
C AC ðP max Þ AC distribution unit cost (function of P max ), €/kW
Pmax generator maximum net power, kW
cB BES system unitary cost, €/kW h G
Pm PV single module rated power, kWp
C binv bidirectional inverter cost, €
Pmax maximum user load, kW
C cd control devices cost, €
Po PV system nominal power rate, kWp
cf generator fuel cost, €/l
rB total number of BES system replacements
CFOE carbon footprint of electricity, kgCO2eq/kW h r binv total number of bidirectional inverter replacements
C inv ðP o Þ PV inverter cost (function of Po ), € r inv total number of PV inverter replacements
Cj total cost for year j, € SOC B;h BES system state of charge for hour h, kW h
C Bj PV–BES system battery replacement cost for year j, € T a;h ambient temperature for hour h, °C
C binv PV–BES system bidirectional inverter cost for year j, € T c;h PV cell temperature for hour h, °C
j
T c;ref PV cell reference temperature, °C
C Gf
j
diesel generator cost due to fuel consumption for year VB BES system nominal voltage, V
j, € yB lead acid battery life-cycle emissions per kW h,
C Gm
j diesel generator cost due to maintenance activities kgCO2eq/kW h
for year j, € Y B ðK B Þ BES system life-cycle emissions (function of K B ),
C Ij PV–BES system inverter cost for year j, € kgCO2eq
C OM&I
j PV–BES operation and maintenance cost for year j, € Y binv ðPbinv Þ bidirectional inverters life-cycle emissions (function
cm generator hourly maintenance cost, €/h of Pbinv ), kgCO2eq
C Cn BES system maximum charge rate yG unitary diesel generator emissions, kgCO2eq/kW h
C Dn BES system maximum discharge rate Y G ðP o ; K B Þ diesel generator life-cycle emissions (function of Po
C PV ðP o Þ PV module cost including installation cost (function and K B ), kgCO2eq
of Po ), € yinv PV inverter life-cycle emissions per kW, kgCO2eq/kW
C0 PV–BES system turnkey cost, € Y inv ðPo Þ PV inverter life-cycle emissions (function of Po ),
DOD BES system depth of discharge, % kgCO2eq
EA;h energy supplied by PV system for hour h, kW h yPV PV module life-cycle emissions per m2, kgCO2eq/m2
EB;h energy supplied by BES system for hour h, kW h Y PV ðPo Þ PV system life-cycle emissions (function of Po ),
eG unitary diesel generator system energy cost, €/kW h kgCO2eq
EG;h energy supplied by diesel generator for hour h, kW h yPVstr PV supporting structures life-cycle emissions per m2,
Ej PV–BES–Diesel generator system energy production kgCO2eq/m2
for year j, kW h Y PVstr ðPo Þ PV supporting structures life-cycle emissions (function
EL total energy demand for the reference year, kW h of Po ), kgCO2eq
EL;a average hourly energy load, kW h yw connecting wire life-cycle emissions per kg, kgCO2eq/kg
EL;h energy load for hour h, kW h Y W ðP o Þ connecting wires life-cycle emissions (function of
f G;hðPG;h =Pmax Þ generator specific fuel consumption for hour h, kg/kW h P o ), kgCO2eq
fG
G
generator average fuel consumption for the reference zbinv number of bidirectional inverters
year, kg/kW h
g inflation rate Greek letters
HI;h total in-plane irradiation for hour h, kW h/m2 b temperature coefficient of solar cell efficiency, 1/°C
HI;r yearly module reference in-plane irradiation, kW/m2 r BES system hourly self-discharge rate, %
IB BES nominal capacity, A h q wire specific weight per wire length, kg/m
Ilim BES charge current limit, A qf fuel density, kg/l
B;C
gbinv bidirectional inverter efficiency, %
Ilim
B;D BES discharge current limit, A gch BES system charging efficiency, %
KB BES system nominal capacity, kW h gd PV module annual degradation ratio, %
K BAH¼1 BES system hourly storage autonomy, h gdch BES system discharging efficiency, %
K min
B BES system minimum capacity, kW h ge PV electrical efficiency, %
K max BES system maximum capacity, kW h
ginv PV inverter efficiency, %
l
B
connecting wire length per PV system rated power,
gmodule PV module conversion efficiency, %
m/kWp
gPV;h PV system overall efficiency for hour h, %
LCOE levelised cost of electricity, €/kW h
gtemp;h PV system temperature efficiency factor for hour h, %
n PV–BES system lifetime, years
NOCT normal operating cell temperature, °C Indices
OCC opportunity cost of capital h index for hours
Pbinv bidirectional inverter nominal power, kW j index for years
Plim
B;C BES charge power limit, kW

frequent and, generally, it consists of the ex-post environmental between the system economic and environmental sustainability,
assessment of the previously designed system. Such a methodol- are rare and no tool, including HOMER software, implements
ogy is behind, also, some existing commercial tools, e.g. HOMER bi-objective approaches. This paper addresses this lack presenting
software [46]. In the system design phase, the adoption of a bi-objective design model for off-grid PV–BES–Diesel generator
bi-objective approaches, able to find good trade-off solutions HESs suitable for applications to a generic installation site. The
1026 M. Bortolini et al. / Energy Conversion and Management 106 (2015) 1024–1038

Table 1 contrary, the proposed model is applied to study effective solutions

Off-grid HESs, literature contribution review. for installations in a remote village in Yakutsk, Russia, a location
Energy plants and storage devices References marked by a medium irradiation level, i.e. 1400 kW h/(m2 year).
PV SWT Fuel Diesel Micro Micro-hydro Battery The model takes into account the hourly energy demand, the solar
cell engine turbine irradiation level and the temperature profiles for the installation
U U [4]
site calculating the hourly PV plant yield, the battery charge–
U U [5–10] discharge processes and the generator energy request, i.e.
U U U [11–16] back-up energy. A multi-scenario analysis, varying the PV plant
U U U [17–20] and BES system capacity, allows discussing the economic and
U U U [21–35]
environmental benefit toward standard fossil fuel based solutions.
U U U [36]
U U U U [37–39] According to the introduced purposes, the reminder of this
U U U U U [40] paper is organised as follows: the next Section 2 introduces the
U U U U U [41] HES architecture, defining the strategy and the logic of control
for energy dispatching and load supply. Section 3 analytically
describes the model to design the HESs and to compute their eco-
nomic and environmental performances, while Section 4 presents
aim is to identify the PV plant rated power, the BES system the introduced case study including the definition of the model
capacity and the technical configuration able to jointly reduce parameters and revising the input data. The obtained results are
the levelised cost of the electricity ðLCOEÞ and the carbon footprint extensively discussed in Section 5 before drawing the paper con-
of energy ðCFOEÞ. Furthermore, the comparison of the LCOE and clusions together with suggestions for further research in the last
CFOE values between HESs and traditional diesel generators Section 6.
allows determining the economic and environmental advantages
introduced by these sustainable systems. 2. Hybrid energy system architecture
Furthermore, the most of the contributions presented in Table 1
evaluates the effectiveness and the best component size of HESs A reference diagram for off-grid HESs, including the energy flow
through their application in high irradiation regions. On the directions and key notations, is in Fig. 1. The system integrates the

(EA,h - EL,h)·ηbinv/∆t > PB,C-lim EA,h > EL,h

Hourly PV plant energy PD = (EA,h - EL,h)·ηbinv/∆t-PB,C-lim SOCB,h = KB-max
production Battery charge power limit exceeded Battery fully charged
EA,h
Energy dissipation

EA,h > 0

PV plant PV Inverter

1. Battery discharge power limit EA,h > 0

exceeded EL,h > 0
EL,h > EA,h
(EL,h - EA,h)/(ηbinv·ηdch·∆t) > PB,D-lim
AC Main
distribution
2. Battery not available unit
EL,h > EA,h
(EL,h - EA,h)/(ηbinv·ηdch) > (SOC B,h - KB-min) User load
Diesel generator
Hourly energy demand
Hourly generator power Generator power to EL,h
production prevent battery
degradation Load supply sources :
EG,h
EA,h < EL,h SOC B,h ≤ KB-min Photovoltaic + Battery
(SOC B,h - KB-min)>(EL,h - EA,h)/(ηbinv·ηdch) Only Generator
(EL,h -EA,h)/(ηbinv·ηdch·∆t)< PB,D-lim Only battery
Battery available
Only Photovoltaic
Photovoltaic + Generator

Bidirectional
BES system inverter
EA,h > EL,h
Battery state of charge SOCB,h < KB-max
SOCB,h Battery not fully charged

Fig. 1. Reference diagram for the off-grid HES.

 M. Bortolini et al. / Energy Conversion and Management 106 (2015) 1024–1038 1027

diesel generator, the PV plant, the battery banks, the AC main and the BES system capacity, allows identifying the economic and
distribution unit, which is connected to the load and the electric environmental performances of each tested configuration.
conversion and control devices. The PV plant supplies the AC load,
through a PV inverter including the maximum power point 3.1. PV system analytic model
tracking (MPPT) device, the electric protections and the reactive
and exceeding power management. The battery charge/discharge Focusing on the PV system, the model quantifies the hourly
processes, the generator start/stop signals and the modulation of yield of the PV arrays, EA;h , according to Eq. (4) and considering
the power coming from the PV inverter are controlled through the effect of the efficiency decrease due to the PV module degrada-
the bidirectional inverter, which also protects the batteries from tion [47]
degradation avoiding the risks of overcharge and excessive
EA;h ¼ HI;h  Aa  gPV;h ð4Þ
discharge. In the case of a three-phase system, each phase is
controlled by a different bidirectional inverter. where Aa ¼ Po =ðgmodule  HI;r Þ, Po refers to the standard test condi-
The BES system stores the energy excesses, best manages the tions, i.e. solar spectrum of AM 1.5, module temperature of 25 °C
diesel generator, e.g. pre-heating, cooling times and transient con- and HI;r equal to 1 kW/m2 [48]. The overall PV system efficiency
ditions, and it guarantees the energy supply during the generator depends on the reference module conversion efficiency, the electri-
failures and low irradiation periods. The diesel generator charges cal energy efficiency, the efficiency decrease due to the cell temper-
the BES system in the case of risk of low charge condition, only, ature and the progressive PV module degradation. Furthermore, the
i.e. the BES system state of charge goes below the minimum stor- DC/AC electric conversion is affected by the PV inverter efficiency
age capacity. This is a mandatory technical condition to prevent (see Eq. (5)).
the battery degradation. The BES system physical constraints are
defined in the following Eq. (1), for each working hour, h. The
gPV;h ¼ gmodule  ge  gtemp;h  ginv  ð1  ðj  1Þ  gd Þ ð5Þ
hourly state of charge, SOC B;h , cannot exceed the maximum capac- The PV module efficiency decreases linearly with the temperature
ity, K max min
B , and it has to be higher than the minimum capacity, K B , respect to the reference condition of 25 °C, as defined in Eq. (6) [49].
defined through the so-called maximum allowable depth of
gtemp;h ¼ 1  b  ðT c;h  T c;ref Þ ð6Þ
discharge, DOD, according to Eq. (2).
where
K min
B 6 SOC B;h 6 K max
B ð1Þ
T c;h ¼ T a;h þ ½ðNOCT  20Þ=800  HI;h ð7Þ
K min
B ¼ ð1  DODÞ  K max
B ð2Þ In Eqs. (6) and (7), b and NOCT depends on the considered PV
module type. The module manufacturers generally provide such
The system aims to satisfy the user load thanks to the PV plant, the
specifications.
energy stored in the BES system and the diesel generator, in case of
the PV–BES system is not able to fully supply the energy demand.
3.2. BES system analytic model
The hourly overall energy balance is in Eq. (3).

EL;h ¼ EA;h þ EB;h þ EG;h ð3Þ The BES system capacity is expressed by both the nominal
capacity, K B , and the number of autonomy hours, AH. AH is the
The PV system operates to meet the energy demand, while the number of hours that a fully charged battery is able to supply
energy surplus, if present, flows to the BES system, until it is fully the energy demand considering the average hourly user load, EL;a
charged. The energy excess is dissipated through the PV inverter. (see Eq. (8)).
The battery charging process is limited by the technical battery
limits in the charging power, Plim AH ¼ ðK B  gbinv  gdch  DODÞ=EL;a ð8Þ
B;C . The charging power is controlled
by the bidirectional inverter and eventual energy excess is directly
The temperature has a significant impact on the electrical perfor-
dissipated by the PV inverter. On the contrary, in the case of low
mance of BES systems. Generally, the performances increase with
irradiance and during the night-time hours, BES system supplies
a temperature increase but, at the same time, aging and the self-
the energy deficit until the SOC B;h decreases to its minimum level
discharge processes are accelerated. The BES system available
or the required power overcomes the battery discharging power
capacity is affected, also, by the external temperature. A capacity
limit, P lim
B;D . In such two cases, the bidirectional inverter asks the decrease of 1% per Celsius degree occurs below 20 °C. For these
diesel generator to supply the load. The PV plant works in parallel reasons, BES systems are typically stored in 20–25 °C controlled
to the diesel generator and the BES system, while the battery rooms [50]. During the charging process, SOC B;h increases according
charge/discharge processes are stopped when the diesel generator to Eq. (9).
supplies the load.
The flow-chart in Fig. 2 summarises the management rules to SOC B;h ¼ minfSOC B;h1  ð1  rÞ þ ðEA;h  EL;h Þ  gbinv  gch ; K max
B g
define, for each studied hour, h, the electric energy flows through ð9Þ
the system to supply the energy load. Such logic of control and
management is in accordance with the standard practice and it is Furthermore, the battery charging process is restricted by the
the basis to develop the HES system analytic model. maximum battery power charge limit, Plim
B;C , defined by the battery

technical limitations in the charging current, Ilim

B;C , in order to avoid
3. Analytic model improper charging process, battery degradation and efficiency
decrease. Ilim
B;C is defined through the BES system maximum charge
This section presents the analytic model of the HES system
together with the economic and the environmental models to com- rate, C Cn , and the nominal capacity IB , expressed in A h.
pute the LCOE and the CFOE. A parametric tool, implementing such ðEA;h  EL;h Þ  gbinv =Dt < Plim lim
Dt ¼ 1 h
B;C ¼ I B;C  V B ð10Þ
models according to the introduced energy flow control algorithm,
is developed in Microsoft Visual Basic for ApplicationsTM environ- C
ment. A multi-scenario analysis, varying the PV plant rated power
Ilim
B;C ¼ I B =C n ð11Þ
1028 M. Bortolini et al. / Energy Conversion and Management 106 (2015) 1024–1038

START

Set PV-BES system power

rates
PO - KB

Read hourly data

Irradiation, Temperature,
Load
HI,h - Ta,h - EL,h

Calculate the PV system hourly energy

production
EA,h

No No
(SOCB,h -KB-min)>(EL,h - EA,h)/
EA,h > EL,h?
(ηbinv·ηdch)?

Yes
Yes
Supply the user energy
demand
No
EL,h (EL,h - EA,h)/(ηbinv·ηdch·∆t)
< PB,D-Lim ?
∆t=1h

No
Yes
Charge BES system SOCB,h = KB-max ?

Request energy from the Diesel

Discharge BES system and
Yes generator
supply the energy demand EB,h
EG,h

(EA,h - EL,h)·ηbinv/∆t < PB,C-Lim No

or
SOCB,h < KB-max ?

Yes

The PV inverter dissipates

the
exceeding energy

END

Fig. 2. HES system flow chart for each studied hour, h, to supply the energy load.

IB ¼ K B  1000=V B ð12Þ Plim lim

Dt ¼ 1 h
B;D ¼ IB;D  V B ð14Þ
The charging process ends if the BES system reaches the maximum
capacity or the process to store the available energy is completed. D
Ilim
B;D ¼ I B =C n ð15Þ
On the contrary, if the PV production, EA;h , cannot satisfy the energy
demand, EL;h , the battery starts its discharging process in the cases
of available energy and a power request lower than the power Otherwise, the discharging process continues until the load demand
is completely met (see Eq. (16)).
discharge limit, P lim
B;D .
If the requested power exceeds the power
discharge limit or the hourly energy demand exceeds the remaining
SOC B;h ¼ SOC B;h1  ð1  rÞ  ðEL;h  EA;h Þ=gbinv  gdch ð16Þ
battery capacity, the demand is satisfied through the diesel genera-
tor (see Eq. (13)).
In Eqs. (9), (13) and (16) the initial BES system state of charge is
supposed to be SOC B;h ¼ K max
B . It is also possible to set a discharge
SOC B;h ¼ SOC B;h1  ð1  rÞ if
8
 power limit higher than Plim
B;D for a limited period to handle the
< ðEL;h  EA;h Þ=g
binv  gdch > SOC B;h  K B
min
generator transient conditions, e.g. pre-heating and cooling time,
ð13Þ power source switches, etc., to guarantee the proper system
: ðE  EA;h Þ=ðgbinv  gdch  DtÞ > Plim
L;h B;D operation.
 M. Bortolini et al. / Energy Conversion and Management 106 (2015) 1024–1038 1029

3.3. Economical analytic model C Gm

j , are proportional to the number of working hours and the
maintenance hourly cost, cm . Globally, the annual operating costs
The HES system economic assessment includes the LCOE and are in Eq. (19).
the lifetime system study. The LCOE is a widely adopted index to v
quantify the economic feasibility of energy systems based on RESs C j ¼ C OM&I
j þ C Bj þ C Ij þ C bin
j þ C Gf Gm
j þ Cj ð19Þ
[49]. As presented in Bortolini et al. [51], the LCOE is the ratio The HES system LCOE is compared to the traditional diesel genera-
between the sum of the system costs and the yearly energy tor value ðPo ¼ 0; K B ¼ 0Þ. The LCOE of a traditional diesel generator
demand, both discounted and affected by the inflation rate (see comes from Eq. (20), calculating the unit energy cost for the refer-
Eq. (17)). ence year j ¼ 1, considering maintenance operation for 8760 h
!, and the average specific fuel consumption per produced kW h, f G .
X
n
C j  ð1 þ gÞj1 Xn
EL  ð1 þ gÞj1
LCOE ¼ C0 þ ð17Þ eG ¼ cf  f G =qf þ ðcm  8760Þ=EL ð20Þ
j¼1 ð1 þ OCCÞ j j¼1 ð1 þ OCCÞ j

3.4. Environmental analytic model

3.3.1. Lifetime HES system cost analysis
The impact of the integration of the PV plant and the BES
The environmental model determines the environmental
system to a traditional system is evaluated through the lifetime
impact of the HES system. A cradle to grave approach allows eval-
cost analysis. In this context, the initial investment for the diesel
uating the CFOE index given the HES system as the functional unit.
generator is not accounted because the purpose is to compare
The CFOE index considers the amount of greenhouse gas (GHG)
the HES to a traditional diesel generator, which is necessary for
emissions, quantified in equivalent carbon dioxide (CO2eq) per
any off-grid applications independently on the adoption of
kW h of produced energy, produced during the entire system
additional RESs and storage solutions. The diesel generator size
lifetime [38,52]. Fig. 3 shows the system boundaries defining the
equals the maximum user load to ensure the energy supply in
considered processes within the environmental impact analysis.
the case of solar absence and fully discharged battery.
Particularly, the manufacturing process considers the material
The system cost includes several contributions due to the PV
mining as well as the production of the system components, i.e.
plant, the inverter, the battery banks, the bidirectional inverters,
PV modules and structures, lead-acid batteries, diesel generator,
the AC main distribution unit, the communication and the remote
connecting wires, charge controller and inverter. All components
control devices. The former two contributions, C PV ðPo Þ and C inv ðPo Þ,
are shipped from the production to the working site, where the
are functions of the PV rated power, Po . The AC main distribution
system is installed. Furthermore, both the utilisation and mainte-
unit cost, C AC ðP max Þ is function of the maximum user load, the
nance processes are considered. The former includes the load elec-
battery bank cost is proportional to the battery capacity, K B , and,
tricity supply; the latter considers the necessary component
finally, the bidirectional inverters and control device costs are
substitutions. Finally, recycling and waste disposal are included
assumed to be constant according to the user load profile level.
through the mandatory end-of-life treatments. As a consequence,
The overall system turnkey cost expression is in Eq. (18).
the system input are the natural resources and the process energy
C 0 ¼ C PV ðPo Þ þ C inv ðPo Þ þ C AC ðPmax Þ þ cB  K B þ C binv þ C cd ð18Þ needs. These input are out of the system boundaries and they are
not considered in the environmental impact evaluation. The
The operating annual costs include the maintenance activities and system outputs, included into the boundaries, are the produced
the component replacement preventing failures and the perfor- electricity and the process GHG emissions, quantified in CO2eq.
mance decrease. Replacement are planned for the inverter and
bidirectional inverter at years 9 and 17 after the plant power up,
3.4.1. Environmental impact assessment
for the batteries at years 7, 13 and 19 after the plant power up,
The following analytic model evaluates the HES system GHG
according to the operative best practice. Furthermore, the diesel
emissions as a function of the PV plant rated power and the BES
generator introduces additional costs due to the fuel consumption,
system nominal capacity.
C Gf
j , proportional to the supplied power, the fuel cost, c f , and the Eq. (21) expresses the CFOE calculated considering the sum of
generator specific consumption. Such a specific consumption the GHG emissions produced by each component and relates them
depends on the generator working power respect to its maximum to the reference energy demand, EL . All the processes belonging to
 
power, f G;h PG;h =Pmax
G . In addition, the maintenance operations, the system boundaries of Fig. 3 are included.

System Life-Cycle

Electricity
Manufacturing Installation Operating Recycling
Raw
& Transportation & & &
Materials
Assembly Start-up Maintenance Waste Disposal Emissions

Model Boundaries

Auxiliary
Process Energy
resources

Fig. 3. Environmental model boundaries.

1030 M. Bortolini et al. / Energy Conversion and Management 106 (2015) 1024–1038

Y PV ðP o Þ þ Y PVstr ðPo Þ þ Y B ðK B Þ þ Y inv ðPo Þ þ Y binv ðP binv Þ þ Y W ðPo Þ þ Y G ðPo ; K B Þ

CFOE ¼ ð21Þ
n  EL

The following equations detail the emissions for each component 4.2. Load profile
according to the introduced notations.
The hourly load profile of the remote village is in Fig. 6. It shows
Po an1 average load of about 16 kW (the red dashed line), a base load of
Y PV ðPo Þ ¼  Am  yPV ð22Þ
Pm 8–10 kW, a maximum load, Pmax , of 30 kW and a yearly energy
request of about 138 MW h.
Po
Y PVstr ðPo Þ ¼  Am  yPVstr ð23Þ
Pm 4.3. Hybrid energy system architecture and parameters

The HES architecture to supply the remote village of Yakutsk,

Y B ðK B Þ ¼ K B  yB  rB ð24Þ
Russia, is a three-phase HES system including the following
functional units:
Y inv ðP o Þ ¼ Po  yinv  r inv ð25Þ
 PV plant (to be configured).
Y binv ðPbinv Þ ¼ zbinv  Pbinv  yinv  rbinv ð26Þ  BES system (to be configured).
 2  33.3 kW Maximum power/31.3 kW Continuous power
Y W ðPo Þ ¼ Po  yW  l  q ð27Þ (COP) diesel generators. The adoption of two generators guaran-
tees the energy supply during the maintenance operations or
X failures. The generators work alternatively to equally distribute
Y G ðP o ; K B Þ ¼ ðEG;h Þ  yG  n ð28Þ
h
the working hours and, therefore, the maintenance activities.
 One or more three phase PV inverters according to the PV rated
The described economic and environmental models are able to plant.
drive the technical, economic and environmental design of HESs  3  6 kW COP bidirectional inverters, one for each phase. The
for any installation site and system configuration. The next Section 4 master bidirectional inverter is the HES system control unit. It
applies the model to design the HES system in a remote village in is synchronised with the two slaves, the AC main distribution
Yakutsk, Russia. unit and the three phase inverters. In addition, it controls the
BES system charging/discharging operations and the generator
power request. The size of the bidirectional inverters is in
4. Case study
accordance with the user load profile and the charging power
limit of a BES system is up to 200 kW h. Particularly, they are
The proposed model is used to design the HES system to meet
able to continuously supply 18 kW, i.e. more than the average
the energy demand of a remote village located in Yakutsk, Russia
user load demand, 24 kW for 30 min, 27 kW for 5 min and
(Latitude 62.02°North, Longitude 129.44°East). The main purpose
33 kW for few seconds and a maximum AC power input of
is to identify a suitable PV and BES system configuration, to be
about 35 kW.
integrated to the traditional diesel generators, jointly reducing
 Control and monitoring devices.
the LCOE and CFOE. All the input data are detailed in this section
considering the so-called standard reference year data. The hourly
In the following, the boundary conditions for the previously
temperature, the irradiation and load profiles and the key technical
mentioned parameters and the input data included within the
aspects are in the following paragraphs.
analytic model of such functional units are fully reported adopting
fix values or ranges according to the real industrial component
4.1. Temperature and irradiation profiles datasheets and referring to the standard literature. The multi-
scenario analysis includes PV plant rated power from 0 to
Figs. 4 and 5 show the temperature and the irradiation profiles 250 kWp and BES system capacity up to 200 kW h, in accordance
for Yakutsk, Russia. The temperature profile is referred to the with the user load profile and the identified electric conversion
monthly average high temperature conditions, which corresponds devices. 735 system configurations occur and they are compared
to the daylight time. The presence of snow and ice on PV modules to determine the economic and environmental performances. For
during the winter period, i.e. from November to February, may lead each scenario, the LCOE and the CFOE are evaluated together with
to PV plant performance reductions. The irradiation levels on such the hourly energy yield, the energy request from the generator,
months cover the 2.9%, 0.7%, 1.4%, 7.0%, respectively, of the yearly the fuel consumption, the generator working hours, the energy
irradiation. The PV module cleaning should be concentrated in surplus, the battery state of charge and cycles. Section 5 compares
February. Anyway, the monthly precipitation during the winter and discusses the key results and outcomes.
period is limited (mean precipitation of about 0.42 mm/month)
with an average number of precipitation days per month equal 4.3.1. PV–BES system parameters
to 3.5 (2.0 in February) [53]. The irradiation profile is referred to The PV–BES module, integrated to the system, includes the PV
the hourly conditions estimated through the PV-Watts calculator plant, the BES system and the electric conversion devices. The PV
of the National Renewable Energy Laboratory [54]. The graph plant input data are in Table A.1 of Appendix A, and refers to high
refers to an optimal inclination angle of the solar collector equal quality multi-crystalline modules manufactured in the European
to 44°. The aggregate global annual irradiation level is of
1437 kW h/(m2 year) and the average daily irradiation level is of 1
For interpretation of color in Figs. 6, 8 and 12, the reader is referred to the web
3.94 kW h/(m2 day). version of this article.
 M. Bortolini et al. / Energy Conversion and Management 106 (2015) 1024–1038 1031

30

25

20

15

10

Temperature [°C]
5

0
January February March April May June July August September October November December
-5

-10

-15

-20

-25

-30

-35

-40

Fig. 4. Temperature profile for Yakutsk, Russia [53].

900

800

700
Irradiation level [W/m2]

600

500

400

300

200

100

0
0 730 1460 2190 2920 3650 4380 5110 5840 6570 7300 8030 8760
Hours [h]

Fig. 5. Irradiation profile for Yakutsk, Russia [54].

36
34
32
30
28
26
24
Load [kW]

22
20
18
Average load
16
14
12
10
8
6
4
2
0
0 730 1460 2190 2920 3650 4380 5110 5840 6570 7300 8030 8760
Hours [h]

Fig. 6. Hourly load profile for the remote village in Yakutsk, Russia.
1032 M. Bortolini et al. / Energy Conversion and Management 106 (2015) 1024–1038

Table 2 Table 3
PV plant installation costs [€/kWp]. Diesel generator maintenance program and cost.

PV plant Module Labor and Supporting Engineering PV plant Maintenance activity Maintenance Cost [€] Cost
rated installation structure total costs period [h] [€/h]
power
Oil, air and fuel filter replacement 1000 95 0.095
3 690 400 300 250 1640 Blower belt, distribution kit and valve 2500 295 0.118
10 690 400 300 115 1505 cover gasket replacement
35 670 400 300 85 1455 Complete generator revision 12,000 4150 0.346
75 640 400 300 85 1425
Total costs [€/h] 0.559 €/h
150 620 400 300 60 1380
Total costs [€/h], including 0.6 €/h
10% of labor cost

Union area. The PV plant installation cost includes the PV module

cost, the installation and labor costs, the supporting structure costs of the HES system. Finally, Tables A.4 and A.5 of Appendix A list
and the engineering cost. The following Table 2 shows such costs, all the other parameters and input data related to the diesel
obtained through a market survey, for the five considered PV plant generators and to the environmental and economic issues.
rated powers, from 3 kWp to 150 kWp.
The impact of the modules on the total cost is of about 45%,
5. Results and discussions
while the remaining percentage is due to the installation and labor
cost (27%), the supporting structure cost (20%) and the engineering
This section presents the key economic and environmental
cost (8%). The following correlation function between the PV
results for the proposed HES system together with the
overall installation cost and its rated power allows extrapolating
bi-objective analysis integrating and balancing such two design
the following analytic function for C PV ðP o Þ.

 targets, i.e. the economic and environmental performances.
C PV ðPo Þ ¼ 1690:3  P0:041
o  Po ð29Þ
5.1. Economic analysis
The BES system adopts batteries based on the valve regulated lead
acid (VRLA) technology, which is a mature cheap technology suit- According to Eq. (8) and the parameters defined in Table A.2,
able for this kind of applications. The input data of BES system each hour of storage autonomy, i.e. AH ¼ 1, corresponds to,
are in Table A.2 of Appendix A, while Table A.3 presents the input approximately, 25 kW h of the BES system capacity (see Eq. (31)).
data for the electric conversion devices in accordance with the
standard literature and the technical datasheets of the commercial K AH¼1
B ¼ ð16  1 AHÞ=ð0:7  0:9  0:94Þ ffi 25 kW h ð31Þ
units. Tables A.1–A.3 further include the specific CO2eq. emissions
The minimum BES capacity is assumed equal to 25 kW h (1 AH) to
for the PV modules and structure, the BES system, the inverter
guarantee the proper HES system operation. Fig. 8 shows the LCOE
and the wire connections.
values for the most relevant configurations of PV rated power, up
to 120 kWp, and BES capacity, up to 8 AH. The results are compared
4.3.2. Diesel generator and other parameters to the LCOE of a traditional diesel generator system with the
The diesel generators are able to provide a maximum power of hypothesis of adopting the same generator type, represented
39.2 kW, for a limited period, and a continuous power of 37.2 kW. through the red line. Such unitary energy cost, corresponding to
The alternator efficiency is of 85% leading to a maximum net power the LCOE with no PV–BES system integrated, is calculated in
of 33.3 kW and a continuous net power of 31.3 kW making the Eq. (32), according to Eq. (20).
generators able to satisfy the remote village demand peaks
(see Fig. 6) even in the case of solar absence and fully discharged eG ¼ 0:7  0:232=0:835 þ ð0:6  8760Þ=138; 129
battery. The generator investment cost is neglected because it is ¼ 0:232 €=kW h ð32Þ
not differential, while costs for operating and maintenance are
function of the maintenance activities and the fuel consumption. The HES system is economically feasible even considering a low fuel
The required standard maintenance activity data are in Table 3 cost, i.e. 0.7 €/l, and a moderate irradiation level, i.e. about
together with their cost, expressed as a function of each operating 1400 kW h/(m2 year). The most effective PV plant rated power
hour. ranges from 40 to 60 kWp depending on the battery capacity. The
Furthermore, the generator specific fuel consumption, in kg of choice of the BES capacity depends on the installer priorities, e.g.
fuel per produced kW h, depends on the specific supplied power. fuel consumption, maintenance operations, low initial investment
Fig. 7 shows its trend, while Eq. (30) presents the correspondent and payback time, emergency BES autonomy hours, etc. The battery
analytic piecewise function. banks can be extended or reduced during the three programmed

8
3
2

  >
< 0:330  Pmax P G;h PG;h P G;h
P G;h P
G;h
þ 0:808  Pmax  0:605  Pmax þ 0:369 if 0:25 6 Pmax 61
f G;h ¼ G G G G
ð30Þ
P max >
: 0:263 P
G if 0 < Pmax
G;h
< 0:25
G

The transport costs of fuel and spare parts are not accounted in the maintenance replacements according to the HES behaviour experi-
analysis even if the remote position of the village may lead to sig- enced in the previous years and the bidirectional inverter power
nificant logistic costs, increasing the convenience and profitability limit. Further considerations deal with the significant cost and tech-
 M. Bortolini et al. / Energy Conversion and Management 106 (2015) 1024–1038 1033

0.275

[kg/kWh]
0.265

PG,h
PGmax
0.255

Fuel consumption fG,h

0.245

0.235

0.225

0.215
0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00
PG,h
PGmax

Fig. 7. Diesel generator fuel consumption as a function of the supplied power.

0.28

0.27 Battery
capacity
AH [h]
0.26
1.0
1.5
LCOE [€/kWh]

0.25 2.0
2.5
3.0
0.24 LCOE [€/kWh]
Only diesel generator 4.0
5.0
0.23
6.0
7.0
0.22 8.0

0.21
10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115 120
PV Plant rated power [kWp]

Fig. 8. LCOE values for different PV rated power and BES system capacity.

nical difficulties concerning the transportation of fuel and spare incidence, i.e. about the 50% of the maintenance activity costs, con-
parts to such a remote area, not accounted in this analysis and firming the great opportunity of the PV–BES system integration.
strengthening the HES system.
Concerning the economic best scenario, i.e. 0.214 €/kW h for 5.2. Environmental assessment
50 kWp–2.5 AH, the LCOE reduction is of about 8% compared to a
traditional diesel generator. The HES system installation is also Fig. 10 shows the CFOE values for PV rated powers up to
economically sustainable excluding the PV energy production dur- 250 kWp and BES capacity up to 8 AH (200 kW h). The results are
ing the winter period, when the snow and ice may cover the PV compared against the CFOE of a traditional diesel generator sys-
modules. Particularly the LCOE reduction is of about 7%, 6.5%, tem with the hypothesis of adopting the same generator type,
4.5%, 3% excluding the PV energy production in December–January, represented through the red line. The PV–BES module inclusion
November–December, from December to February and from introduces significant environmental benefits within the most of
November to February, respectively. Finally, Fig. 9 shows the the considered scenarios leading to CFOE reductions greater than
lifetime HES system cost distribution, discounted to year 0, in order 20%. The best environmental scenario occurs for a PV plant rated
to satisfy the remote village overall energy demand. power of 95 kWp and a BES system capacity of 200 kW h (8 AH).
The PV–BES turnkey cost covers about the 19% of the total cost, The CFOE is of 0.374 kgCO2eq/kW h (saving of about 50%).
while the cost due to the fuel consumption has the main incidence, Such environmental performances require high BES system
i.e. about 58%. Consequently, better conditions for the PV–BES capacity showing that, from the environmental viewpoint it is
modules, e.g. higher fuel cost or higher irradiation level, increase preferable to store energy in high capacity BES systems respect
the PV plant rated power and BES system capacity. The mainte- to using the backup generator and consuming fossil fuel and oil.
nance activities cause the remaining 24% of the total cost. Among Fig. 11 highlights such an evidence for the best environmental
them, the cost due to the generator operation has the larger scenario. The diesel generator has the largest impact (91%), while
1034 M. Bortolini et al. / Energy Conversion and Management 106 (2015) 1024–1038

PV Maintenance; 2.44%

Battery replacement;
Fuel ; 57.70% 3.79%
PV inverter
replacement; 3,09%
Bidirectional inverter
Maintenance replacements; 2.06%
activities; 23.68%

Generator maintenance;
12.30%

PV Plant; 12.78%
Bidirectional inverters;
1.33%

Control devices; 0.22%

Battery; 1.66%
AC Distribution unit;
0.62% PV Inverter; 2.00%

Fig. 9. HES system cost distribution for the best economic scenario.

0.8

0.75
CFOE [kgCO2eq/kWh] Battery
Only diesel generator capacity
0.7 AH [h]
1
CFOE [kgCO2eq/kWh]

0.65
1.5

0.6 2
2.5
0.55 3
4
0.5 5
6
0.45 7
8
0.4

0.35
10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 230 240 250
PV plant rated power [kWp]

Fig. 10. CFOE values for different PV rated power and BES system capacity.

1.31% 0.97%
the remaining percentage is due to the PV module, the structures
5.23% 0.49% and wires (7.5%), the BES system (0.51%) and the electric conver-
0.51%
sion devices (0.49%).

PV Modules

PV Supporting structures 5.3. Trade-off bi-objective analysis

Connecting wires
Considering the introduced LCOE and CFOE functions, leading
PV and bidirectional Inverters to the economic and environmental assessment of the HESs,
Lead-acid batteries Fig. 12 shows the LCOE and CFOE values for each tested scenario.
Each point corresponds to a system configuration in terms of PV
Diesel generator (Fuel and oil)
module size and BES system capacity. The red and yellow squared
91.49%
dots are the economic and environmental best cases defined
before. Among all scenarios, a subset of them is non-dominated
by the others. This means that to improve one goal, i.e. the
Fig. 11. HES system emission distribution for the best environmental scenario.
economic performance or the environmental impact, the other
 M. Bortolini et al. / Energy Conversion and Management 106 (2015) 1024–1038 1035

0.8

0.75

0.7

CFOE [kgCO2eq/kWh]
0.65
50 kWp - 2.5 AH
(0.214 €/kWh;0.546 kgCO2eq/kWh)
0.6

0.55

0.5

0.45

0.4

0.35 95 kWp - 8.0 AH 160 kWp - 8.0 AH

(0.228 €/kWh;0.394 kgCO2eq/kWh) (0.268 €/kWh;0.374 kgCO2eq/kWh)

0.3
0.2 0.21 0.22 0.23 0.24 0.25 0.26 0.27 0.28 0.29 0.3 0.31 0.32 0.33 0.34 0.35 0.36 0.37 0.38
LCOE [€/kWh]
MO Solutions Pareto frontier min LCOE Solution min CFOE Solution ψ solution

Fig. 12. Pareto Frontier for the economic and environmental trade-off bi-objective analysis.

one worsens. Such points compose the so-called Pareto frontier LCOEðPo ; K B Þ CFOEðPo ; K B Þ
wðPo ; K B Þ ¼  ð33Þ
(in light brown in Fig. 12). Possible trade-off scenarios fall on min LCOEðPo ; K B Þ min CFOEðPo ; K B Þ
the Pareto frontier.
To drive the trade-off scenario selection, the Authors propose The scenario that minimises the introduced w function is a candi-
the following practical rule-of-thumb, based on the following date good trade-off between the economic and the environmental
wðPo ; K B Þ function, calculated for each Pareto point. performances. Fig. 13 depicts the trend of w highlighting the

1.5

1.45

1.4

1.35

1.3
ψ function

1.25

1.2

1.15

1.1
ψ Optimal solution
95 kWp - 8.0 AH
1.05 (0.228 €/kWh;0.394 kgCO2eq/kWh)

1
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

Parte frontier solution

Fig. 13. Trade-off w function between economic and environmental performances.

Table 4
HES best scenarios, trade-off and benchmark.

Parameters and KPIs Design target

Benchmark diesel generator Economic best scenario Environmental best scenario Effective trade-off
PV rated power [kWp] 0 50 160 95
BES system capacity [kW h] 0 62.5 (2.5 AH) 200 (8 AH) 200 (8 AH)
Yearly energy demand [kW h] 138,129 138,129 138,129 138,129
Yearly PV energy production [kW h] 0 61,068 195,417 116,029
Yearly PV energy stored [kW h] 0 8444 41,497 34,919
Yearly PV energy lost [kW h] 0 6812 92,775 23,642
Yearly diesel gen. production [kW h] 138,129 86,681 49,337 57,385
Yearly fuel consumption [kg] 32,023 20,317 11,748 13,608
Energy demand met (PV–BES) 0 37.25 64.28 58.46
LCOE [€/kW h] 0.232 (+0%) 0.214 (7.8%) 0.268 (+15.5%) 0.228 (1.7%)
CFOE [kgCO2eq/kW h] 0.763 (+0%) 0.546 (28.4%) 0.374 (51.0%) 0.394 (48.4%)
1036 M. Bortolini et al. / Energy Conversion and Management 106 (2015) 1024–1038

trade-off system parameters further highlighted in green in environmental targets. The levelised cost of the electricity, LCOE,
previous Fig. 12. and the carbon footprint of electricity, CFOE, quantify the
Globally, the economic and environmental best scenarios economic and the environmental impact of the HES configuration.
present significant differences in the PV–BES system parameters. The proposed model is general and able to design HESs for any
The latter model suggests bigger PV and BES systems. PV rated installation site. In this paper, a case study applies the model to
power doubles (from 350 to 700 kWp) and BES capacity is almost supply the off-grid village of Yakutsk, Russia. Results highlight
the triple (from 487.5 kW h to 1200 kW h). Such a difference the technical, economic and environmental feasibility of HESs,
affects the system energy flow. In the LCOE best configuration, respect to fossil fuel solutions, for a context with a medium irradi-
energy production is two times the CFOE best configuration energy ation level, i.e. 1400 kW h/(m2 year), and a relatively low fuel
production. Furthermore, in the environmental model, the PV–BES cost, i.e. 0.7 €/l. Relevant savings in both the fuel and the oil cost
modules supply almost all the energy required by the load (94.7%). (7.8%) and emissions (51.0%) occur for the best economic and
Considering the system performance indices, the economic design environmental scenarios, compared to the benchmark case using
leads to a 43% increase of the CFOE respect to its optimal value, fossil fuel, only. Furthermore, a trade-off analysis allows balancing
whereas environmental model determines a 23% LCOE increase. the divergent trends of the economic and environmental functions
The proposed trade-off analysis looks for effective global scenario. leading to final savings of 1.7% and 48.4% for LCOE and CFOE,
Table 4 summarises the case study results and the identified respectively.
effective scenarios, together with the benchmark configuration Further research mainly deals with the inclusion to the model of
adopting fossil fuel, only. fuel and spare part shipment cost and impact, the application of
the model to multiple locations with different environmental
conditions and a sensitivity analysis on the load profiles to identify
6. Conclusions and further research
economic and environmental lower and upper feasibility limits on
the system size.
This paper presents the study of photovoltaic (PV) plant and
battery energy storage (BES) system integrated to traditional diesel
generators for off-grid applications. The development of the ana-
lytic technical, economic and environmental models for the design Appendix A
of hybrid energy systems (HESs) allows determining the PV system
rated power and the BES system capacity matching economic and See Tables A.1–A.5.

Table A.1
PV plant main parameters.

Parameter Description Value–range Reference

Am PV single module area 1.75 m2 Technical datasheet
C OM&I
j
PV operation and maintenance cost for year j 0:01  P o for j ¼ 1; . . . ; n [55]
C PV ðP o Þ PV plant installation cost See Table 2 and Eq. (29) Real market data
HI;r Yearly module reference in-plane irradiance 1 kW/m2 [48]
n PV plant lifetime 25 years [32]
NOCT Normal operating cell temperature 47 °C [56]
Pm PV single module rated power 0.260 kWp Technical datasheet
Po PV plant rated power 10–250 kWp, step 5 Multi-scenario analysis
T c;ref PV cell reference temperature 25 °C [48]
yPV PV module life-cycle emissions per m2 160 kgCO2eq/m2 [57]
yPVstr PV supp. structures life-cycle emiss. per m2 40.2 kgCO2eq/m2 [57]
gd PV module annual degradation ratio 0.5%/year [58]
gmodule PV module efficiency 15% Technical datasheet
b Temp. coefficient of solar cell efficiency 0.005/°C [56]

Table A.2
BES system parameters.

Parameter Description Value–range Reference

CB BES system unitary cost 150 €/kW h Real market data
C Bj BES system battery replacement cost 150 €/kW h [59]
For year j For j = 7, 13, 19 only Technical data sheet
C Cn BES system charge rate 6 Technical data sheet
CD
n
BES system discharge rate 4 Technical data sheet
DOD Depth of discharge 70% [59]
Technical data sheet
KB BES system nominal capacity 1–50 AH, step 0.5 AH Multi-scenario analysis
rB Total number of BES system replacements 3 Technical data sheet
VB BES system voltage 48 V Standard practice
yB Lead acid battery life-cycle emiss. per kW h 24.25 kgCO2eq/kW h [57]
r BES system hourly self-discharge rate 0.067%/day Technical data sheet
gch BES system charging efficiency 90% [59]
gdch BES system discharging efficiency 90% [59]
 M. Bortolini et al. / Energy Conversion and Management 106 (2015) 1024–1038 1037

Table A.3
PV–BES system electric conversion device parameters.

Parameter Description Value–range Reference

C AC ðP max Þ AC distribution unit cost (function of P max ) 100 €/kW Real market data
C binv Bidirectional inverters cost 7500 € Real market data
C cd Control devices cost 1250 € Real market data
C inv ðP o Þ PV inverter cost (function of P o ) 225 €/kW Real market data
v PV–BES system bidirectional inverter cost 7500 € Real market data
C bin
j
For year j For j = 9, 17 only Technical data sheet
C Ij PV–BES system inverter cost 225 €/kWp Real market data
For year j For j = 9, 17 only Technical data sheet
l Connecting wire length 20 m/kWp Standard practice
P binv Single bidirectional inverter nominal power 6 kW User demand profile
P max User peak demand 35 kW User demand profile
rinv , r binv Total number of inverter replacements 2 Technical data sheet
yinv Inverter life-cycle emissions per kW 26.3 kgCO2eq/kW [57]
yw Connecting wire life-cycle emissions per kg 0.789 kgCO2eq/kg [57]
zbinv Total number of bidirectional inverters 3 Three phase system
ge PV electrical efficiency 95% Technical data sheet
gbinv Bidirectional inverter efficiency 94% Technical data sheet
ginv PV inverter efficiency 96% Technical data sheet
q Wire specific weight per PV rated power and wire length 0.55 kg/m Technical data sheet

Table A.4
Diesel generator input parameters.

Parameter Description Value–range Reference

Diesel generator parameters
cf Generator fuel cost 0.7 €/l Real market data
cm Generator maintenance cost 0.6 €/h (see Table 3) Technical data sheet


PG;h
f G;h Pmax Generator specific fuel consumption for hour h See Fig. 7 and Eq. (30) Technical data sheet
G

P max
G
Generator maximum net power 33.3 kW See Fig. 6
yG Emissions to generate a kW h by diesel gen. 0.763 kgCO2eq/kW h [57]
qf Fuel density 0.835 kg/l

Table A.5 [11] Protogeropoulos C, Brinkworth BJ, Marshall RH. Sizing and techno-economical
Environmental and economic input parameters. optimization for hybrid solar photovoltaic/wind power systems with battery
storage. Int J Energy Res 1997;21(6):465–79.
Parameter Description Value–range [12] Yang H, Lu L, Zhou W. A novel optimization sizing model for hybrid solar–wind
EL;h Energy load for hour h See Fig. 6 power generation system. Sol Energy 2007;81(1):76–84.
HI;h Total in-plane irradiation for hour h See Fig. 4 [13] Yang H, Zhou W, Lu L, Fang Z. Optimal sizing method for stand-alone hybrid
solar–wind system with LPSP technology by using genetic algorithm. Sol
g Inflation 3%
Energy 2008;82(4):354–67.
OCC Opportunity cost of capital 5%
[14] Diaf S, Belhamel M, Haddadi M, Louche A. Technical and economic assessment
T a;h Ambient temperature for hour h See Fig. 5
of hybrid photovoltaic/wind system with battery storage in Corsica Island.
Energy Policy 2008;36(2):743–54.
[15] Kaabeche A, Belhamel M, Ibtiouen R. Techno-economic valuation and
optimization of integrated photovoltaic/wind energy conversion system. Sol
References
Energy 2011;85(10):2407–20.
[16] Askarzadeh A, dos Santos Coelho L. A novel framework for optimization of a
[1] International Energy Agency (IEA). World Energy Outlook 2013; 2013. grid independent hybrid renewable energy system: a case study of Iran. Sol
[2] Mohammed YS, Mustafa MW, Bashir N. Hybrid renewable energy systems for
Energy 2015;112:383–96.
off-grid electric power: review of substantial issues. Renew Sustain Energy Rev
[17] Li CH, Zhu XJ, Cao GY, Sui S, Hu MR. Dynamic modeling and sizing optimization
2014;35:527–39.
of stand-alone photovoltaic power systems using hybrid energy storage
[3] Akikur RK, Saidur R, Ping HW, Ullah KR. Comparative study of stand-alone and
technology. Renew Energy 2009;34(3):815–26.
hybrid solar energy systems suitable for off-grid rural electrification: a review.
[18] Avril S, Arnaud G, Florentin A, Vinard M. Multi-objective optimization of
Renew Sustain Energy Rev 2013;27:738–52.
batteries and hydrogen storage technologies for remote photovoltaic systems.
[4] Celik AN. Optimisation and techno-economic analysis of autonomous
photovoltaic–wind hybrid energy systems in comparison to single photovoltaic Energy 2010;35(12):5300–8.
and wind systems. Energy Convers Manage 2012;43(18):2453–68. [19] Jallouli R, Krichen L. Sizing, techno-economic and generation management
[5] Tan CW, Green TC, Hernandez-Aramburo CA. A stochastic method for battery analysis of a standalone photovoltaic power unit including storage devices.
sizing with uninterruptible-power and demand shift capabilities in PV Energy 2012;40(1):196–209.
(photovoltaic) systems. Energy 2010;35(12):5082–92. [20] Silva SB, Severino MM, de Oliveira MAG. A stand-alone hybrid photovoltaic,
[6] Wissem Z, Gueorgui K, Hédi K. Modeling and technical–economic optimization fuel cell and battery system: a case study of Tocantins, Brazil. Renew Energy
of an autonomous photovoltaic system. Energy 2012;37(1):263–72. 2013;57:384–9.
[7] Glavin ME, Hurley WG. Optimisation of a photovoltaic battery ultracapacitor [21] Muselli M, Notton G, Louche A. Design of hybrid-photovoltaic power
hybrid energy storage system. Sol Energy 2012;86(10):3009–20. generator, with optimization of energy management. Sol Energy 1999;65
[8] Kazem HA, Khatib T, Sopian K. Sizing of a standalone photovoltaic/battery (3):143–57.
system at minimum cost for remote housing electrification in Sohar, Oman. [22] Muselli M, Notton G, Poggi P, Louche A. PV-hybrid power systems sizing
Energy Build 2013;61:108–15. incorporating battery storage: an analysis via simulation calculations. Renew
[9] Rezk H, El-Sayed AHM. Sizing of a standalone concentrated photovoltaic Energy 2000;20(1):1–7.
system in Egyptian site. Int J Electr Power Energy Syst 2013;45(1):325–30. [23] Shaahid SM, Elhadidy MA. Technical and economic assessment of grid-
[10] Semaoui S, Arab AH, Bacha S, Azoui B. Optimal sizing of a stand-alone independent hybrid photovoltaic–diesel–battery power systems for
photovoltaic system with energy management in isolated areas. Energy Proc commercial loads in desert environments. Renew Sustain Energy Rev
2013;36:358–68. 2007;11(8):1794–810.
1038 M. Bortolini et al. / Energy Conversion and Management 106 (2015) 1024–1038

[24] Shaahid SM, Elhadidy MA. Economic analysis of hybrid photovoltaic–diesel– [40] Dufo-López R, Bernal-Agustín JL. Multi-objective design of PV–wind–diesel–
battery power systems for residential loads in hot regions—a step to clean hydrogen–battery systems. Renew Energy 2008;33(12):2559–72.
future. Renew Sustain Energy Rev 2008;12(2):488–503. [41] Kusakana K, Vermaak HJ. Hybrid diesel generator/renewable energy system
[25] Shaahid SM, El-Amin I. Techno-economic evaluation of off-grid hybrid performance modeling. Renew Energy 2014;67:97–102.
photovoltaic–diesel–battery power systems for rural electrification in Saudi [42] Khatib T, Mohamed A, Sopian K. A review of photovoltaic systems size
Arabia-a way forward for sustainable development. Renew Sustain Energy Rev optimization techniques. Renew Sustain Energy Rev 2013;22:454–65.
2009;13(3):625–33. [43] Luna-Rubio R, Trejo-Perea M, Vargas-Vázquez D, Ríos-Moreno GJ. Optimal
[26] Hrayshat ES. Techno-economic analysis of autonomous hybrid photovoltaic– sizing of renewable hybrids energy systems: a review of methodologies. Sol
diesel–battery system. Energy Sustain Develop 2009;13(3):143–50. Energy 2012;86(4):1077–88.
[27] Saheb-Koussa D, Haddadi M, Belhamel M. Economic and technical study of a [44] Erdinc O, Uzunoglu M. Optimum design of hybrid renewable energy systems:
hybrid system (wind–photovoltaic–diesel) for rural electrification in Algeria. overview of different approaches. Renew Sustain Energy Rev 2012;16(3):
Appl Energy 2009;86(7–8):1024–30. 1412–25.
[28] Rehman S, Al-Hadhrami LM. Study of a solar PV–diesel–battery hybrid power [45] Upadhyay S, Sharma MP. A review on configurations, control and sizing
system for a remotely located population near Rafha, Saudi Arabia. Energy methodologies of hybrid energy systems. Renew Sustain Energy Rev 2014;38:
2010;35(12):4986–95. 47–63.
[29] Khatib T, Mohamed A, Sopian K, Mahmoud M. Optimal sizing of building [46] Lilienthal P, Gilman P, Lambert T. HOMERÒ micropower optimization
integrated hybrid PV/diesel generator system for zero load rejection for model. USA: Department of Energy; 2005.
Malaysia. Energy Build 2011;43(12):3430–5. [47] Markvart T. Solar electricity. 2nd ed. USA: Wiley; 2000.
[30] Kaldellis J, Zafirakis D, Kavadias K, Kondili E. Optimum PV–diesel hybrid [48] International Standard IEC 61724:1998. Photovoltaic system performance
systems for remote consumers of the Greek territory. Appl Energy 2012;97: monitoring – guidelines for measurement, data exchange and analysis;
61–7. 1998.
[31] Khelif A, Talha A, Belhamel M, Hadj Arab A. Feasibility study of hybrid diesel– [49] Hernández-Moro J, Martínez-Duart JM. Analytical model for solar PV and CSP
PV power plants in the southern of Algeria: case study on AFRA power plant. electricity costs: present LCOE values and their future evolution. Renew
Int J Electr Power Energy Syst 2012;43(1):546–53. Sustain Energy Rev 2013;20:119–32.
[32] Ismail MS, Moghavvemi M, Mahlia TMI. Techno-economic analysis of an [50] Wenham SR, Green MA, Watt ME, Corkish R. Applied photovoltaics. United
optimized photovoltaic and diesel generator hybrid power system for remote Kingdom: Earthscan; 2007.
houses in a tropical climate. Energy Convers Manage 2013;69:163–73. [51] Bortolini M, Gamberi M, Graziani A. Technical and economic design of
[33] Tazvinga H, Xia X, Zhang J. Minimum cost solution of photovoltaic–diesel– photovoltaic and battery energy storage system. Energy Convers Manage
battery hybrid power systems for remote consumers. Sol Energy 2013;96: 2014;86:81–92.
292–9. [52] Wiedmann T, Minx JA. Definition of carbon footprint. Chapter in ecological
[34] Tazvinga H, Zhu B, Xia X. Energy dispatch strategy for a photovoltaic–wind– economics research trends. USA: Nova Science Publishers; 2008. p. 1–11.
diesel–battery hybrid power system. Sol Energy 2014;108:412–20. [53] Pogodaiklimat; 2014. <http://www.pogodaiklimat.ru/>.
[35] Suresh Kumar U, Manoharan PS. Economic analysis of hybrid power systems [54] National Renewable Energy Laboratory (NREL). PV Watts tool; 2014. <http://
(PV/diesel) in different climatic zones of Tamil Nadu. Energy Convers Manage rredc.nrel.gov/solar/calculators/pvwatts>.
2014;80:469–76. [55] Bortolini M, Gamberi M, Graziani A, Mora C, Regattieri A. Multi-parameter
[36] Ismail MS, Moghavvemi M, Mahlia TMI. Design of an optimized photovoltaic analysis for the technical and economic assessment of photovoltaic systems
and microturbine hybrid power system for a remote small community: case in the main European Union countries. Energy Convers Manage 2013;74:
study of Palestine. Energy Convers Manage 2013;75:271–81. 117–28.
[37] Elhadidy MA, Shaahid SM. Promoting applications of hybrid (wind [56] Tina GM, Scandura PF. Case study of a grid connected with a battery
+photovoltaic+diesel+battery) power systems in hot regions. Renew Energy photovoltaic system: V-trough concentration vs. single-axis tracking. Energy
2004;29(4):517–28. Convers Manage 2012;64:569–78.
[38] Dufo-López R, Bernal-Agustín JL, Yusta-Loyo JM, Domínguez-Navarro JA, [57] Ecoinvent centre. Ecoinvent data v. 2.0: ecoinvent reports no. 1–25. Swiss
Ramírez-Rosado IJ, Lujano J, et al. Multi-objective optimization minimizing centre for life cycle inventorie; 2007.
cost and life cycle emissions of stand-alone PV–wind–diesel systems with [58] Thevenard D, Pelland S. Estimating the uncertainty in long-term photovoltaic
batteries storage. Appl Energy 2011;88(11):4033–41. yield predictions. Sol Energy 2013;91:432–45.
[39] Kaabeche A, Ibtiouen R. Techno-economic optimization of hybrid [59] Battke B, Schmidt TS, Grosspietsch D, Hoffmann VH. A review and probabilistic
photovoltaic/wind/diesel/battery generation in a stand-alone power system. model of lifecycle costs of stationary batteries in multiple applications. Renew
Sol Energy 2014;103:171–82. Sustain Energy Rev 2013;25:240–50.