Applied Energy 192 (2017) 563–574

Contents lists available at ScienceDirect

Applied Energy
journal homepage: www.elsevier.com/locate/apenergy

Technical and economic assessment of a SOFC-based energy system

for combined cooling, heating and power
Andrea L. Facci a,⇑, Viviana Cigolotti b, Elio Jannelli c, Stefano Ubertini a
a
DEIM – School of Engineering, University of Tuscia, 01100 Viterbo, Italy
b
ENEA – Portici Research Center, 80055 Portici (Na), Italy
c
Department of Engineering, University of Napoli Parthenope, 80143 Napoli, Italy

h i g h l i g h t s

Assessment of the performances of a SOFC based CHCP plant for residential applications.

Optimal plant design as a function of investment cost and control strategy.

Optimized control strategies to minimize the costs and primary energy consumption.

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

Article history: Here we present the technical and economical performances of a small scale trigeneration power plant
Received 18 April 2016 based on solid oxide fuel cells and designed for a small residential cluster (i.e. 10 apartments). The energy
Received in revised form 17 June 2016 system features a natural gas solid oxide fuel cell, a boiler, a refrigerator, and a thermal storage system.
Accepted 18 June 2016
We compare different power plant configurations varying the size of the fuel cell and the refrigeration
Available online 5 July 2016
technology to satisfy the chilling demand (i.e. absorption or mechanical chiller). Given that the ability
to meet the power demand is crucial in this kind of applications, the plant performances are assessed
Keywords:
following an optimal control strategy, as a function of different energy demand profiles and electricity
Trigeneration
Distributed generation
prices, and of rated and part load efficiencies of each energy converter. The optimization of the energy
CHCP system operating strategy is performed through a graph theory-based methodology. Results are provided
Fuel cells in terms of electrical and thermal efficiency, operating strategy, as well as economic saving, primary
SOFC energy consumption reduction, and pay back period, considering different capital costs of the fuel cell.
Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction reducing the energy consumption of buildings by 50% compared

Increasing the energy efficiency and reducing the environmen-
tal impact of buildings is a key aspect for developed countries
[1–3]. In fact, buildings are responsible for 40% of the energy
demand and 36% of the carbon dioxide emissions in the European
Union (EU) [4]. In particular, households cause 25% of the total
greenhouse gases emission related to fossil fuel combustion in
the EU [5]. Similarly, in 2010, buildings used about 41% of the pri-
mary energy consumption of the United States (US) [6,7], 54% of
which only for the residential sector [7]. In this scenario, both US
and EU undertook actions to reduce the buildings energy waste
[3,4]. The US Department of Energy envisages the possibility of
⇑ Corresponding author.
E-mail addresses: andrea.facci@unitus.it (A.L. Facci), viviana.cigolotti@enea.it
(V. Cigolotti), elio.jannelli@uniparthenope.it (E. Jannelli), stefano.ubertini@unitus.it
(S. Ubertini).
to 2010 [3]. The EU requires that all new buildings must be ‘‘near
ly-zero-energy” by 31 December 2020 [4], and all new public
buildings must be ‘‘nearly-zero-energy” by 2018 [8] as a determi-
nant measure to meet the objective of reducing the primary energy
consumption by 20% with respect to the business as usual
projection.
Cogeneration, or combined heat and power (CHP), and trigener-
ation or combined heat cooling and power (CHCP) are reliable
technologies that are already contributing to the global energy
demand [9]. According to the International Energy Agency (IEA),
CHP coupled with district heating and cooling could save
950 Mton/year of carbon dioxide emissions by 2030 [9]. Although
the main driver for large CHP investments is economic [9],
cogeneration and distributed generation (DG) offer several other
opportunities, such as: (i) increased overall efficiency compared
to separate production [10,11]; (ii) reduction of pollutant and
green house gases emissions [11,12]; (iii) deferring expensive

http://dx.doi.org/10.1016/j.apenergy.2016.06.105
0306-2619/Ó 2016 Elsevier Ltd. All rights reserved.
564 A.L. Facci et al. / Applied Energy 192 (2017) 563–574
investments on large size plants [11], and on transmission and dis-
tribution system [12]; (iv) reducing losses in the distribution sys-
tem [12]; (v) providing grid support or ancillary services [12];
(vi) promoting the use of alternative technologies and renewable
sources [11,13,14]. Several case studies related to the perfor-
mances of CHP systems can be found in literature [15–21], also
coupled with renewable sources [13,15–17,22–24].
Solid oxide fuel cells (SOFC) are a promising technology for
high-efficiency and sustainable energy conversion [25–27], and
are expected to play an important role in future distributed energy
generation [28]. SOFCs have many advantages over conventional
power plants including: (i) high efficiency [25]; (ii) fuel adaptabil-
ity [29]; (iii) very low NOx ; SOx and particulate matter emissions,
and reduced CO2 emissions [30]; (iv) vibration-free operation
[31]. Theoretically, the electrical efficiency of a SOFC may be larger
than 70% [32]. On the other hand, commercially available SOFC
systems have a net electrical efficiency in the range 40–60% [32].
SOFC systems can produce an electrical power from few W up to
several hundreds of kW [25,32] and their efficiency is only slightly
influenced by the system scale [32]. SOFC plants with an electrical
power of 1 kW and an efficiency of 60% have been demonstrated in
[32]. The high working temperature facilitates the integration of
SOFCs within CHCP plants [32]. Therefore SOFCs, are becoming a
promising prime mover for CHCP applications [25–28,33,34].
After decades of research and development, SOFCs are now
close to commercialization [32], and there are already several SOFC
high-efficiency micro-cogenerators conceived primarily for resi-
dential use or small-scale commercial applications. More than
1000 CHP installations of polymer electrolyte fuel cells (PEMFC)
and SOFC have been realized starting from the late 2014, mainly
within the European project ENE.FIELD [35]. The ENE-FARM is
the most successful project on micro CHP [35]. It has been realized
in Japan from 2008 until the end of 2015, whith more than 150,000
CHP units (both PEMFC and SOFC) installed in residential areas
[35]. Typical applications are those with high power and heat
demands, like office buildings, swimming pools, and small and
medium enterprises [36].
The control strategy determines the performances of DG plants
[10,37] as much as the technological level of the components
[38–40]. Therefore, realistic working conditions must be considered
for their economical and environmental evaluation [38,39,41–43].
Determining the set-point is critical for CHP and CHCP, as each of
the energy vectors simultaneously produced represents a con-
straint for the plant [27,38,44]. Optimized control strategies
emphasize the strengths of DG plants, compared to predetermined
strategies (i.e. thermal tracking or electrical tracking), impacting
the return on investment and the environmental benefit [43].
In this paper we evaluate the potential of a SOFC-based CHCP
plant in terms of economic, energy, and environmental perfor-
mances. We hypothesize that such a plant satisfies the energy
demand of a small residential cluster (i.e. 10 apartments) and we
compare different configurations varying the SOFC electrical power
and the refrigerator machine technology (i.e. absorption or mechan-
ical chiller). Two control strategies are considered for each configu-
ration: an economically optimized strategy that minimizes the total
daily cost, and an energy consumption optimized strategy that min-
imizes the daily primary energy consumption (PEC). Both strategies
are determined through the methodology presented in [45–47] that
accounts for hourly energy demand and costs, and for the efficiency
of the energy system components as a function of their set-point.
The paper is organized as follows. In Section 2 we describe the
methodology utilized to determine the total energy cost. The case
study is presented in Section 3, with particular reference to the
plant configurations, the energy demand, and the energy tariffs.
Results are dissected in Section 4 and conclusions are drawn in
Section 5.
2. Methodology
We dissect the performances of a SOFC-based CHCP plant in a
realistic energy management scenario. In particular, we
concentrate on the annual energy cost and on the primary energy
consumption. We compare different configurations, varying the
equipment size and the refrigerator technology (further details
are given in 3.3).
The operating efficiency of such a plant is largely determined by
its control strategy as highlighted in [10,37,38,45,46,48]. There-
after, we compare the performances of the different configurations
assuming two possible management policies: cost minimization
and PEC minimization. Both control strategies are determined
through the optimization methodology introduced in [45], and fur-
ther developed in [46,47]. Such a methodology minimizes a pre-
scribed objective function on a daily basis accounting for: (i) the
design performances of all the subsystems; (ii) the derating of
the performances at part load; (iii) the effects of environmental
conditions; (iv) energy demand and costs as functions of time;
(v) maintenance, and cold start costs; (vi) constraints related to
the dynamic behavior of the equipment, such as the minimum
time interval between two consecutive starts or shutdowns. All
the energy converters are modeled as black-boxes, through their
efficiency curves as functions of the set-point and environmental
conditions.
The prescribed objective function for cost minimization is:
X
24
GCost ¼ C f ðh; sðhÞÞ þ C m ðh; sðhÞÞ þ C s ðh; sðhÞÞ  Rðh; sðhÞÞ; ð1Þ
h¼1
being h the time interval, C f the cost of fuel, C m the maintenance
cost, C s the cold-start cost, and R the revenue/cost yielding from
the electricity exchanged with the grid. Costs are functions of the
time interval and the plant state (i.e. the set-point of the
subsystems) sðhÞ.
PEC minimization requires the following objective function
reported in Eq. (2)
X
24
GPEC ¼ Ef ðh; sðhÞÞPEFf þ Egrid PEFgrid ; ð2Þ
h¼1
where Ef is the energy content of the fuel, PEFf is the primary
energy factor of the fuel [49], Egrid is the electricity exchanged with
the grid, and PEFgrid is the primary energy factor of electricity [49].
Eqs. (1) and (2) are discretized with respect to the plant state
and in time, and the problem is represented as a weighted and ori-
ented graph. The optimal control strategy is determined seeking
for the shortest path across the graph.
We combine two energy demand profiles, one typical of sum-
mer and one representative of winter (see Section 3.1), and two
electricity cost profiles (see Section 3.2), one for working days
and one for non-working days, to obtain four sample combinations.
The control strategy is optimized for the four days, and the daily
results are projected on the whole year to compare the perfor-
mances of the different plant configurations.
3. Case study
In this paper we evaluate the performances of the CHCP plant
described in Section 3.3 to comply the energy demand of a rela-
tively small residential unit.
The prime mover is a cogenerative SOFC. A natural gas boiler is
included as a back-up and to satisfy peak thermal energy demand.
Both absorption and mechanical chillers are considered for the
production of chilling energy.
 A.L. Facci et al. / Applied Energy 192 (2017) 563–574 565

The CHCP plant is grid connected, but the excess energy fed into Table 1
Electricity prices before VAT expressed in €/kW h for year 2015, as functions of the
the power grid is not remunerated. Thus, the term Rðh; sðhÞÞ in Eq.
consumption quota for single housing unit, the time slot, and the quarter. Slot F1
(1) may only assume negative values. We comment that such an refers to hours between 8 a.m. and 7 p.m. of working days; slot F2 refers to the rest of
assumption facilitates the generalization of the methodology and the week. Data are retrieved from [51].
of the results to different countries. In fact, the remuneration of
Q Slot Consumption quota [kW h/year]
electricity produced by DG plants is subject to stringent national
rules. Moreover, prices are strongly influenced by complex 0–1800 2801–2640 2641–4440 >4440

subsidizing mechanisms that vary on a national basis. Finally, we I F1 0.1279 0.1863 0.2541 0.3006
note that such an assumption is expected to underestimate the F2 0.1215 0.1800 0.2478 0.2942

profitability of SOFC based plants. Thereafter, the design guidelines II F1 0.1257 0.1844 0.2526 0.2990
emerging herein can be considered conservative. F2 0.1820 0.1991 0.2390 0.2816
III F1 0.1243 0.1835 0.2523 0.2988
F2 0.1824 0.1996 0.2394 0.2820
3.1. Energy demand
IV F1 0.1297 0.1895 0.2591 0.3056
F2 0.1894 0.2065 0.2463 0.2890
Fig. 1 reports the energy demand used in the present analysis,
representative of a group of 10 houses. Data relative to the single
housing unit are retrieved from [50]. Significant variations of the Table 1 are intended for singe housing unit. We comment that a
energy demand profiles are not expected at least for industrialized similar mechanism based on the total consumption per year is uti-
countries. In fact, in the residential sector, energy demand is lized on most of the European countries [53]. Hours between 8 a.m.
mainly influenced by climate and lifestyle. Thus, energy demand and 7 p.m of working days fall into the slot F1, the rest of the week
deviations among different countries are expected to compare with is considered into the slot F2. For generality, we averaged over the
regional variations. We consider two different demand profiles, quarter, while retaining the variations with respect to the con-
one representative of a winter day, the other of a summer day. sumption and to the time slot, that may impact the optimal control
Cooling is required only in summer for air conditioning. Thermal strategy. A 10% VAT must be added to the cost reported in Table 1
energy is requested for space heating, only in the winter, and to obtain the final cost, reported in Fig. 2 as a function of the total
domestic hot water, during the whole year. consumption. The increase of the electricity price varying the
During summer, heating load has an average value of 8.6 kW total consumption is significant, in particular, in the range
and a peak of 27 kW. Chilling power request is similar, with an 2–20 MW h/year, and the effective averaged cost of electricity is
average of 7.2 kW, and a peak value of 23 kW. Electricity demand influenced by the plant control strategy.
is slightly higher, being 32 kW and 21 kW the peak and average According to AEEG [52], the natural gas price varies as a func-
power demands respectively. tion of the consumption quota (per single housing unit), and the
During winter, the peak request of thermal energy is 124 kW, quarter. A similar mechanism based on the total consumption
more than 3 times larger compared to electricity demand. Simi- per year is utilized on most of the European countries [54]. The
larly, the average thermal power, whose value is 44 kW, doubles Italian territory is divided into six macro-regions characterized
the electrical one which is just 19 kW. by different prices: North West (NW), North East (NE), Center
(C), Center South West (CSW), Center South East (CSE) and South
3.2. Energy prices (S). For generality we consider an average national and annual cost,
averaging over the quarters and the regions. The dependence on
Prices of electricity and natural gas are retrieved from the the total consumption may impact the optimal control strategy
Italian Authority for Electrical Energy and Gas (AEEG) [51,52], and is retained, as evidenced in Fig. 3. Such a variation is concen-
and reported in Tables 1 and 2. trated in the range 100–500 Sm3/year, while the natural gas cost
The cost of electricity varies as a function of the quarter (Q), of could be considered constant outside this range. 10% VAT also
the time slot, and of the consumption quota. Quotas reported in applies to the natural gas cost.

Fig. 1. User energy demand time series: (a) electricity demand; (b) thermal energy demand; (c) cooling energy demand.
566 A.L. Facci et al. / Applied Energy 192 (2017) 563–574

Table 2
Natural gas prices before VAT, expressed in €/Sm3 for year 2015, as functions of the
consumption quota (for single housing unit), the term, and the region. Data from [51].

Q Reg. Total consumption [Sm3/year]

0–120 121–480 481–1560 1561–5000
NW 0.390 0.506 0.483 0.479
NE 0.385 0.483 0.462 0.458
C 0.390 0.512 0.489 0.485
I
CSW 0.386 0.566 0.538 0.534
CSE 0.388 0.540 0.514 0.510
S 0.383 0.622 0.589 0.586
NW 0.362 0.478 0.456 0.452
NE 0.357 0.456 0.435 0.430
C 0.362 0.484 0.461 0.457
II
CSW 0.359 0.538 0.510 0.506
CSE 0.360 0.512 0.486 0.482
S 0.356 0.595 0.562 0.558
NW 0.356 0.472 0.449 0.445
NE 0.350 0.449 0.428 0.424
C 0.356 0.478 0.455 0.451
III
CSW 0.352 0.532 0.504 0.500
CSE 0.354 0.505 0.480 0.476
S 0.349 0.588 0.555 0.551
Fig. 4. Schematic representation of the CHCP plant analyzed throughout the paper.
NW 0.372 0.488 0.465 0.461
NE 0.366 0.465 0.444 0.440
C 0.372 0.494 0.471 0.467
IV
CSW 0.368 0.548 0.520 0.516
CSE 0.370 0.521 0.496 0.492 Table 3
S 0.365 0.604 0.571 0.568 Configuration of the five energy systems considered for the analysis. Ref. case stands
for the reference scenario.

SOFC Absorption chiller Mechanical chiller Boiler (kW)

Ref. case – – 23 kW 125
Case A 25 kW 23 kW – 125
Case B 25 kW – 23 kW 125
Case C 40 kW 23 kW – 125
Case D 40 kW – 23 kW 125

Fig. 2. Average electricity cost, after VAT, as a function of the total consumption per
year.
Fig. 3. Average natural gas cost, after VAT, as a function of the total consumption
per year.
3.3. Power plant
Fig. 4 schematically represents the CHCP plant studied in this
paper. It features a cogenerative SOFC, a natural gas fuel boiler, a
refrigeration machine, and a heat storage [55,56].
We consider four scenarios varying the size of the SOFC and the
refrigeration technology, as reported in Table 3. The performance
of the power plant in the four configurations are compared to a
reference scenario that represents the business as usual: electricity
is only acquired from the grid, thermal energy is produced through
the boiler, and cooling energy is obtained from a mechanical refrig-
erator (see Table 3).
The technical characteristics of the single fuel cell are taken
from the commercial unit EnGENTM  2500 module produced by
SolidPOWER [57,58]. Such a system is operated at 700–750 °C
and each module has a nominal electrical output of 2.5 kW, a ther-
mal power of 2 kW and a net electrical efficiency of 50% [59] at
maximum power output. Power can be regulated in the range
30–100%. Electrical and thermal efficiencies are displayed in
Fig. 5(a) as functions of the set-point, and are estimated according
to the data available in [60,61]. Cases A and B utilize 10 modules to
obtain 25 kW of electrical power and 20 kW of thermal power,
while cases C and D consider 16 modules, yielding 40 kW and
32 kW of electricity and heat, respectively.
The nominal power of the boiler is 125 kW for all the configura-
tions (see Table 3), to entirely satisfy the peak thermal demand.
The efficiency curve is represented in Fig. 5(b) and is estimated
according to the model presented in [37].
A 23 kW double effect absorption chiller [62] is utilized to pro-
duce cooling energy in cases A and C. Given the high temperature
operation of a SOFC, without loosing generality, we assume here
that the fuel cell is able to provide heat at the temperature required
by the chiller [63–65]. We note that this assumption would require
an optimization of the thermal flows within the fuel cell energy
system with respect to the state of the art [62]. Fig. 5 reports the
COP as a function of the thermal load [62].
 A.L. Facci et al. / Applied Energy 192 (2017) 563–574
Fig. 5. Efficiency of each of the energy converter in the CHCP plant as functions of the set-point: (a) SOFC; (b) boiler; (c) absorption chiller; (d) mechanical chiller.
A 23 kW mechanical chiller is utilized to satisfy the cooling
demand for cases B and D. Its design coefficient of performance
is 4.3 (Daikin models RYYQ/RXYQ) [66]. The COP varies as a func-
tion of the set-point according to the model in [37], as reported
in Fig. 5(d).
Following the results in [47] we sized the heat storage to accu-
mulate thermal energy to satisfy the peak demand for three hours.
We here assume a 90% round trip efficiency for the heat storage
system [67].
The plant is grid-connected such that electricity can be acquired
from the grid when the energy produced by the SOFC is lower than
the one requested by the user. Electricity can also be fed into the
grid in case of over-production. Nevertheless, the excess energy
fed into the power grid does not produce any income.
We comment that the proposed configuration maximizes the
flexibility of the plant. In fact, the thermal storage and the grid con-
nection relax the constraints imposed by the energy demand on
the control strategy. Thermal energy can be produced and con-
sumed at different times, mitigating the non-simultaneity of ther-
mal and electrical demand profiles. Electricity can be acquired or
fed into the power grid, thus releasing the SOFC from electrical
tracking.
The PEF of the electricity exchanged with the grid is set to 2.174
[46], and the PEF of the natural gas is set to 1.1 [46,49].
4. Results and discussion
4.1. Minimum cost control strategy
Figs. 6 and 7 compare the global performances obtained
through the minimum cost control strategy.
Fig. 6. Global performances of the power plant with the minimum cost control strategy: (a) total cost for the energy supply; (b) primary energy consumption; (c) total fuel
consumption; (d) electricity acquired from the grid.
567
All the considered configurations significantly reduce the yearly
energy cost with respect to the reference scenario, as evidenced in
Fig. 6(a). In particular, from 58% to 63% of the energy expenditure
is saved with respect to the reference case. The lowest cost,
26 k€/year, pertains to the SOFC-based power plant with electrical
chiller (Case D). The SOFC size has a limited impact on the yearly
energy cost: cost is reduced by about 3.5–3.7% by increasing the
nominal power from 25 kW to 40 kW. Similar results are obtained
switching from the absorption to the mechanical refrigerator.
Trigeneration also reduces the PEC (see Fig. 6(b)), despite the
CHCP plants are controlled with the only objective of minimizing
energy costs. PEC reduction is in the range 21–27% with respect
to the reference case. Case D yields the minimum PEC. From the
comparisons between case A and case C and between case B and
case D shown in Fig. 6(b), it is evidenced that PEC is reduced by
about 5% by increasing the nominal power of the SOFC from
25 kW to 40 kW. Also the utilization of the mechanical chiller
reduces the PEC by about 5% with respect to the absorption chiller
(i.e. Case D vs Case C).
Electricity exchanged with the grid is drastically reduced to
below 10% of the amount required by the reference case, as shown
in Fig. 6(d). The electricity price is significantly reduced, as a con-
sequence of its dependence on the total consumption exemplified
in Fig. 2. Electricity unit price is 0:293 €/kWh for slot F2, and
0:297 €/kW h for slot F1, in the reference case. The electricity
acquired from the grid is below 1800 kW h/year per single housing
unit for all the other plant configurations. Thereafter the unit price,
is 0.140 €/kW h for slot F1 and 0.133 €/kW h for slot F2.
Trigeneration increases the natural gas consumption by 54–66%
with respect to separate production (see Fig. 6(c)). However, the
effect of such a variation on the natural gas unit price is negligible.
568 A.L. Facci et al. / Applied Energy 192 (2017) 563–574
Fig. 7. Comparison between the different plant configurations in term of economic saving, PEC reduction, and pay back period obtained thorough the minimum cost control
strategy: the area of the circles is proportional to the pay back period and the center of the circles defines relative cost and PEC reduction. The three plots correspond to
different SOFC costs: (a) SOFC cost 2000 €/kW; (b) SOFC cost 10,000 €/kW; (c) SOFC cost 20,000 €/kW.
In fact, the gas acquired by the single housing unit, being in the
range 3000–5000 Sm3/year, is located in the plateau depicted in
Fig. 3.
The size of the SOFC is expected to significantly impact the
system cost. SOFCs are at a pre-commercial stage, thus a precise
estimation of the SOFC capital cost per installed kW is still difficult.
According to [68], the cost of a 1 kW SOFC system is between
12,000 €/kW and 23,000 €/kW, and between 3000 €/kW and
5500 €/kW for a 5 kW system. The IEA [69] reports a cost of
12,000–15,000 €/kW for stationary systems with a power in the
range 200–300 kW, and forecasts a reduction to 1500–1600 €/kW
with increasing production volumes. According to the model in
[70], the cost of a 50 kW SOFC energy system is in the range
3000–5500 €/kW. In [71] the installed cost of a SOFC plant is
expressed as a function of the installed power and of the number
of units produced per year and varies between 12,000 €/kW, for a
production of 100 systems per year with an average power of
1 kW, and 400 €/kW for a production 50,000 units per year and
an average power of 100 kW. According to [72] the cost of a
270 kW system decreases from 3500 €/kW to 670 €/kW by increas-
ing the production volume from 50 units/year to 10,000 units/year.
Fig. 7, summarizes the performances of the analyzed scenarios
in terms of relative cost and PEC reduction with respect to the
reference scenario, and of the pay back period (PBP). We hypothe-
size to update the reference scenario to a CHCP plant. Thereafter,
the PBP is evaluated considering only the costs of the equipment
that are not present in the reference scenario (i.e. the absorption
chiller and the SOFC). For the absorption chiller we assumed a cap-
ital cost of 35 k€ in line with the prices of double effect systems
available in the market. Given the large variability of the capital
cost for the SOFC, we considered three different scenarios: (i) SOFC
cost = 2000 €/kW; (ii) SOFC cost = 10,000 €/kW; (iii) SOFC
cost = 20,000 €/kW. If we assume that the cost of the SOFC energy
system is 2000 €/kW (see Fig. 7(a)), for case C the investment
is paid during year 2 by utilizing the minimum cost control
strategy, while PBP ¼ 3 for the other cases. Thereafter, in this
scenario, Case D is characterized by the largest primary energy
saving and cash flow and thus it should be considered the
optimal solution.
When the cost of the SOFC is increased up to 10,000 €/kW, only
cases A and B have a PBP lower than 8 years and could still be
considered feasible investments for the residential sector. On the
contrary, cases C and D are characterized by PBP ¼ 11 years that
may be too long for a profitable investment into the housing sector.
Case B has a lower PBP (7 years) compared to case A (8 years) and
higher cash flow and energy saving, and can be considered the
optimal solution in this scenario.
None of the proposed scenarios appear to be competitive, from
an economical perspective, when the SOFC energy system cost is
equal or above 20,000 €/kW, given that more than 10 years are
required to pay the investment for all the analyzed cases.
Notably, configurations utilizing the mechanical refrigerator
(B and D) perform better with respect to those featuring the
absorption chiller (A and C). This is related to the fact that, if we
compare the energy demand, reported in Fig. 1, and the SOFC per-
formance curves, displayed in Fig. 5(a), we could notice that the
thermal efficiency of the SOFC is always lower than the electrical
one. On average, the producible thermal energy is about 1/2 of
the electrical energy. The energy demand in the summer is equal
to 454 kW h/day of electricity and 205 kW h/day of heat. Thus,
for configuration B and D, heat can be completely produced
through cogeneration. On the other hand, utilizing the absorption
chiller, 172 kW h/day of chilling demand are converted into a fur-
ther thermal demand. As a consequence, in configurations A and C
heat cannot be completely cogenerated.
Fig. 8 displays the set-point time series that minimizes the
energy cost for the SOFC. In all the configurations and energy
demand profiles, the fuel cell is never turned off. For cases A and
B, that utilize the 25 kW SOFC, the prime mover is operated at full
load during evening hours (from 6 p.m. to 11 p.m), that are charac-
terized by large electricity and heat demand. During the rest of day
the SOFC power is modulated to match the energy demand profile
and to increase the electrical efficiency of the fuel cell. The 40 kW
SOFC utilized for cases C and D always operates at part load,
because the electricity sold to the grid is not remunerated and
the electrical efficiency of the fuel cell increases by reducing the
set-point.
The control strategy of the boiler, shown in Fig. 9, is not directly
related to the thermal demand. During winter, the boiler is always
utilized to complement the cogenerative thermal production, but
heat demand and production need not to be strictly simultaneous.
The thermal storage, whose state of charge (SOC) is reported in
Fig. 10, allows deferred thermal production. In particular, in win-
ter, the SOC is gradually reduced during central hours, and is then
suddenly recharged during evening, when the SOFC operates at full
load, and the boiler is turned on at high load. We comment that,
having hypothesized a cyclic behavior of the CHCP plant, we con-
strained the SOC to be 50% both at the first and last time intervals.
Only cases A and C utilize the boiler to satisfy the summer thermal
demand, confirming that the configurations that utilize the
mechanical refrigerator are better tailored for the energy demand
in study. Finally, the thermal storage allows the boiler to operate
almost always at high efficiency (set-point larger than 25%)
irrespective of the heat demand.
 A.L. Facci et al. / Applied Energy 192 (2017) 563–574 569

Fig. 8. Set-point of the SOFC as a function of time according to the minimum cost control strategy, for all the configurations. Each plot represents a different daily demand
profile: (a) winter working day; (b) winter holiday; (c) summer working day; (c) summer holiday.

Fig. 9. Set-point of the boiler as a function of time according to the minimum cost control strategy, for all the configurations. Each plot represents a different daily demand
profile: (a) winter working day; (b) winter holiday; (c) summer working day; (c) summer holiday.
570 A.L. Facci et al. / Applied Energy 192 (2017) 563–574

Fig. 10. Thermal storage SOC as a function of time according to the minimum cost control strategy, for all the configurations. Each plot represents a different daily demand
profile: (a) winter working day; (b) winter holiday; (c) summer working day; (c) summer holiday.

The variation of the electricity price has a negligible impact on
the power plant control strategy, as both the SOFC and the boiler
are controlled in a similar way for working and non-working days.
In fact, the cost difference between the time slots F1 and F2 is very
limited.
4.2. Minimum PEC control strategy
It is worth to note that the minimization of the daily energy cost
does not guarantee the maximum energy efficiency [46,47].
Therefore, in this section, we study the behavior of the proposed
CHCP solutions when the energy system is controlled following
the management policy that minimizes the PEC. As evidenced in
Eq. (2), the PEC is calculated accounting for the electricity
exchanged with the power grid. As a consequence, PEC needs to
be minimized considering the relevant interactions with the
encompassing energy system. When electricity is acquired from
the grid, the relative PEC is added to the contribution of the fuel
locally consumed. Similarly, when electricity is fed to the power
grid, the corresponding primary energy is subtracted.
Fig. 11 summarizes the results obtained by the different energy
system configurations when operated through the minimum PEC
control strategy. With this management policy, trigeneration
reduces the PEC of an amount comprised between 26% and 39%
with respect to the reference case. Compared to the minimum cost
control strategy, the minimum PEC policy yields a further PEC
reduction in a range between 13 and 15 percentage points. PEC
is minimized by configuration C, that utilizes the 40 kW SOFC

Fig. 11. Global performances of the power plant with the minimum PEC control strategy: (a) total cost for the energy supply; (b) primary energy consumption; (c) total fuel
consumption; (d) electricity acquired from the grid.
 A.L. Facci et al. / Applied Energy 192 (2017) 563–574
and the absorption refrigerator. This is largely related to the fact
that a larger fuel cell allows to cogenerate substantially all the
electrical and thermal power, and the absorption chiller utilizes
the waste energy of the prime mover.
The energy cost is reduced in all the configurations, also
through the minimum PEC control strategy (see Fig. 11(a)).
Specifically, savings are comprised between 41% and 56% com-
pared to reference scenario. Nevertheless, PEC minimization dete-
riorates the economic performances with respect to cost
minimization, as evidenced comparing Fig. 6 with Fig. 11. For con-
figuration A such a variation is relatively limited (about 4%). The
daily cost is moderately incremented by 8% for cases B and C,
switching from economical to PEC based optimization. Configura-
tion D experiences the most important variation of the operating
cost, with a 36% increment compared to cost minimization.
The fuel consumption of each single housing unit (Fig. 11(c))
varies between 3000 Sm3/year and 6100 Sm3/year. Thereafter, the
price of natural gas can be considered constant for all the
configurations (Fig. 3), including the reference case, and equal to
0.533 €/Sm3. We comment that PEC minimization does not
necessarily reduce the local fuel consumption with respect to cost
optimization, as noted by comparing Fig. 6(c) with Fig. 11(c).
Rather, it reduces the global PEC, considering also the real or
avoided consumptions of the surrounding electrical system.
The electricity bought from the grid by each housing unit varies
from 300 kW h/year for case C, to 4800 kW h/year for case B,
significantly impacting the average price of the electrical energy
(see Fig. 2). The resulting cost of electricity is reported in Table 4.
Notably, the electrical energy acquired from the grid is minimized
by the plant configuration C that also minimizes the PEC. Then,
such a configuration takes also advantage from a reduced cost of
electricity, with respect the other cases, as evidenced in Table 4.
Configuration C has also the largest local fuel consumption. The
high electrical efficiency of the SOFC, compared to conventional
power plants utilized for centralized electricity production (i.e. to
the efficiency of the grid) explains this behavior. PEC optimization
maximizes the distributed energy production also when the energy
demand is low. Excess electricity is fed to the grid thus reducing
the demand for relatively inefficient conventional plants. As a
Table 4
Average electricity cost expressed in €/kW h, for the minimum PEC control strategy.
Case A Case B Case C
F1 0.1558 0.2183 0.1396
F2 0.1493 0.2118 0.1331
Fig. 12. Comparison between the different plant configurations in term of economic saving, PEC reduction, and pay back period obtained thorough the minimum PEC
strategy: the area of the circles is proportional to the pay back period and the center of the circles defines relative cost and PEC reduction. The three plots correspond to
different SOFC costs: (a) SOFC cost 2000 €/kW; (b) SOFC cost 10,000 €/kW; (c) SOFC cost 20,000 €/kW.
571
consequence, the most effective configuration utilizes the large
SOFC, to maximize electricity production, and the absorption chil-
ler, to increase heat recovery during summer.
Fig. 12 compares different plant configurations, accounting for
both economical and energy efficiency related aspects. With a cap-
ital cost of the SOFC of 2000 €/kW the PBP is below 5 years for all
the selected cases. In general, the most effective solution could
be configuration C, that has a reasonable PBP (4 years) and the lar-
gest relative PEC reduction (40% with respect to the reference
case). In fact, even if configuration A yields a larger cost reduction,
which is not the primary objective in this case, it produces a signif-
icantly lower PEC (26% with respect to the reference case). If the
capital cost of the fuel cell is increased to 10,000 €/kW, the PBP
becomes too long for cases C and D, assuming that 10 years is a
reasonable time horizon for an investment into the housing field.
Cases A and B have both PBP ¼ 8 years, and are characterized by
a similar primary energy consumption. Case A takes advantage
from a larger cost reduction, 56% with respect to the reference
case, and can be considered the optimal control strategy with a
SOFC cost of 10,000 €/kW. In the last scenario, the large capital cost
of the SOFC (20,000 €/kW) hinders the investment into the CHCP
plant. In fact, the PBP becomes higher than 10 years for all the
configurations.
The SOFC follows the control strategy reported in Fig. 13 to min-
imize the PEC. In case of PEC minimization the optimal control
strategy is not a function of the energy prices. Thereafter, only
two scenarios are simulated: the winter and the summer energy
demand. Comparing Fig. 8 to Fig. 13 we note that, the set-point
of the SOFC is generally larger for PEC minimization, compared
to economical optimization, except for the case D in summer. In
the latter case, the electrical energy produced by the fuel cell is
reduced as a compromise between electricity production and elec-
trical efficiency, that, for the SOFC, is maximized at minimum load
(see Fig. 5). We comment that, despite the electrical efficiency of
the SOFC is higher than that of the grid, the prime mover does
not run at constant full load, because the efficiency of the SOFC
increases by decreasing the power output (see Fig. 5).
The control strategy of the boiler is described in Fig. 14. During
winter, the thermal demand is larger than the heat produced
through cogeneration, and the boiler is utilized to complement
the thermal production of the SOFC. As already evidenced in
Section 4.1, the boiler control strategy is not directly related to
the heat demand. In fact, thanks to the thermal storage, thermal
Case D demand and production do not have to be simultaneous. As a con-
sequence, the boiler operates only in the set-point range of higher
0.1628
0.1563 efficiency. During summer, thermal energy is entirely produced by
cogeneration, except for case A, which is characterized by the
572 A.L. Facci et al. / Applied Energy 192 (2017) 563–574

Fig. 13. Set-point of the SOFC as a function of time according to the minimum PEC control strategy, for all the configurations. Each plot represent a different daily demand
profile: (a) winter working day; (b) winter holiday; (c) summer working day; (c) summer holiday.

Fig. 14. Set-point of the boiler as a function of time according to the minimum cost control strategy, for all the configurations. Each plot represent a different daily demand
profile: (a) winter working day; (b) winter holiday; (c) summer working day; (c) summer holiday.

smaller fuel cell (25 kW) and utilizes the heat as an input to the
absorption refrigerator to satisfy the chilling demand. In this case,
as already shown in Section 4.1, the thermal power of the SOFC is
not sufficient to cover the whole heat demand, that includes both
domestic hot water as well as the input energy for the refrigerator.
5. Conclusion
This paper presents the technical and economical evaluation of
a small scale trigeneration power plant based on SOFCs for a resi-
dential application. We compare different configurations varying
the SOFC power and the refrigerator machine technology. Two
control strategies are considered for each configuration: an
economically optimized strategy that minimizes the total daily
cost, and an energy consumption optimized strategy that mini-
mizes the daily primary energy consumption.
Results presented in Section 4, highlight that the chosen objec-
tive function determines both the control strategy and the optimal
system design. Thereafter, the principal purpose of the investment
must be established a priori to correctly configure the CHCP.
Notably, cost minimization fosters the production of chilling
energy through the mechanical refrigerator. Such configurations
are not, in general, the most efficient. Rather, these are the better
tailored on the analyzed energy demand and cost profiles, from
an economical perspective. On the other hand, PEC is minimized
utilizing the absorption chiller, in order to maximize the total effi-
ciency of the power plant.
All the considered configurations reduce both the PEC and the
energy cost with respect to separate production. Specifically, fol-
lowing the minimum cost control strategy, 58–63% of the energy
cost, and 21–27% of the PEC are saved with respect to the reference
scenario. PEC minimization yields a cost reduction in the range
41–56% and a decrease of the primary energy consumption
comprised between 26% and 39% with respect to the reference
scenario.
The capital cost of the SOFC is still an open issue, and signifi-
cantly impacts the optimal configuration of the CHCP plant. The
sensitivity analysis performed herein suggests that investments
on SOFC based DG plants are still feasible for capital costs of the
SOFC as high as 10 k€/kW. At least one configuration with PBP
lower than 10 years is present in such a scenario, for both the con-
trol strategies. A further increase of the initial cost hampers the
investments. PBP is always longer than 10 years for a capital cost
of 20 k€/kW for the SOFC. Conversely, when such a cost is reduced
to 2000 €/kW or below, as envisaged in [69,72], all the combina-
tions of plant design and control strategy yield PBP below 5 year.
In this scenario the economical potential of SOFC based DG plants
is huge.
Finally, we comment that PEC minimization generates positive
economical results, and guarantees further 15 percentage points
reduction of the PEC with respect to economical minimization.
Thereafter, designing and controlling the CHCP plant following an
energy efficiency perspective rather than a pure economical objec-
tive, may be reasonable.
The present analysis is performed with reference to the Italian
energy market. The general framework, in terms of regulatory
aspects and of energy prices, is similar also for other European
countries. Thereafter, the general results of this paper can be
extended also to different energy markets. On the other hand, dif-
ferences on the energy prices are expected to impact the specific
numerical results presented herein. Such a sensitivity analysis will
be performed in future works.
 A.L. Facci et al. / Applied Energy 192 (2017) 563–574
Acknowledgments
This research is supported by the Ministry of Education, Univer-
sities and Research Grant PON_03_PE_00109_1 ‘‘Fuel Cell Lab”. The
authors would also like to express their gratitude to Eng. Lorenzo
Lipardi for his profitable support.
References
[1] The European Parliament and the Council of the European Union. Directive
2012/27/eu of the european parliament and of the council of 25 october 2012
on energy efficiency, amending directives 2009/125/ec and 2010/30/eu and
repealing directives 2004/8/ec and 2006/32/ec. Official Journal of the European
Union.
[2] Concerned Action Energy Performance of Buildings. Homepage; 2014. <http://
www.epbd-ca.eu/> [accessed March 11, 2016].
[3] Building Technologies Office. Buildings; 2016. <http://energy.gov/
eere/efficiency/buildings> [accessed March 11, 2016].
[4] European Commission. Buildings; 2016. <https://ec.europa.eu/energy/
en/topics/energy-efficiency/buildings> [accessed March 11, 2016].
[5] European Environmental Agency. Households and industry responsible for
half of eu greenhouse gas emissions from fossil fuels; 2012. <http://www.eea.
europa.eu/highlights/households-and-industry-responsible-for?&utm_campaign=
households-and-industry-responsible-for&utm_medium=email&utm_source=
EEASubscriptions> [accessed March 11, 2016].
[6] Afram A, Janabi-Sharifi F. Theory and applications of HVAC control systems – a
review of model predictive control (MPC). Build Environ 2014;72:343–55.
[7] US Department of Energy. Buildings energy data book; 2012. <http://
buildingsdatabook.eren.doe.gov/ChapterIntro2.aspx> [accessed March 11,
2016].
[8] Europen-Commission. Nearly zero-energy buildings; 2016. <https://ec.europa.
eu/energy/en/topics/energy-efficiency/buildings/nearly-zero-energy-
buildings> [accessed March 11, 2016].
[9] Agency IE. Combined heat and power evaluating the benefits of greater global
investment; 2008.
[10] Onovwiona H, Ugursal V. Residential cogeneration systems: review of the
current technology. Renew Sust Energy Rev 2006;10(5):389–431.
[11] Chicco G, Mancarella P. Distributed multi-generation: a comprehensive view.
Renew Sust Energy Rev 2009;13(3):535–51.
[12] International Energy Agengy. Distributed generation in liberalised electricity
markets; 2002.
[13] Lau K, Yousof M, Arshad S, Anwari M, Yatim A. Performance analysis of hybrid
photovoltaic/diesel energy system under malaysian conditions. Energy
2010;35(8):3245–55.
[14] Franco A, Salza P. Strategies for optimal penetration of intermittent
renewables in complex energy systems based on techno-operational
objectives. Renew Energy 2011;36(2):743–53.
[15] Cardona E, Piacentino A. A measurement methodology for monitoring a CHCP
pilot plant for an office building. Energy Build 2003;35(9):919–25.
[16] Bizzarri G, Morini GL. Greenhouse gas reduction and primary energy savings
via adoption of a fuel cell hybrid plant in a hospital. Appl Therm Eng 2004;24
(23):383–400.
[17] Merše S, Petelin Visočnik B, Riddoch F, Craenen S, Gardiner P, Rotheray T, et al.
Cogeneration case studies handbook. Jožef Stefan Institute Energy Efficiency
Centre; 2011.
[18] Jannelli E, Minutillo M, Cozzolino R, Falcucci G. Thermodynamic performance
assessment of a small size CCHP (combined cooling heating and power)
system with numerical models. Energy 2014;65:240–9.
[19] Hawkes AD, Brett DJL, Brandon N. Fuel cell micro-CHP techno-economics: Part
1 – Model concept and formulation. Int J Hydrogen Energy 2009;34
(23):9545–57.
[20] Hawkes AD, Brett DJL, Brandon NP. Fuel cell micro-CHP techno-economics:
Part 2 – Model application to consider the economic and environmental
impact of stack degradation. Int J Hydrogen Energy 2009;34(23):9558–69.
[21] Fubara TC, Cecelja F, Yang A. Modelling and selection of micro-CHP systems for
domestic energy supply: the dimension of network-wide primary energy
consumption. Appl Energy 2014;114:327–34.
[22] Valente LCG, de Almeida SCA. Economic analysis of a diesel/photovoltaic
hybrid system for decentralized power generation in northern brazil. Energy
1998;23(4):317–23.
[23] McGowan J, Manwell J. Hybrid wind/PV/diesel system experiences. Renew
Energy 1999;16(14):928–33. Renewable energy energy efficiency, policy and
the environment.
[24] Cozzolino R, Tribioli L, Bella G. Power management of a hybrid renewable
system for artificial islands: A case study. Energy 2016;106:774–89.
[25] Sammes N, Boersma R. Small-scale fuel cells for residential applications. J
Power Sources 2000;86(12):98–110.
[26] Farhad S, Hamdullahpur F, Yoo Y. Performance evaluation of different
configurations of biogas-fuelled {SOFC} micro-CHP systems for residential
applications. Int J Hydrogen Energy 2010;35(8):3758–68.
[27] Chiappini D, Facci AL, Tribioli L, Ubertini S. SOFC management in distributed
energy systems. J Fuel Cell Sci Technol 2011;8(3).
573
[28] Kazempoor P, Dorer V, Ommi F. Evaluation of hydrogen and methane-fuelled
solid oxide fuel cell systems for residential applications: system design
alternative and parameter study. Int J Hydrogen Energy 2009;34(20):8630–44.
[29] Singhal SC, Kendall K. High-temperature solid oxide fuel cells: fundamentals,
design and applications. Elsevier; 2003.
[30] Baratto F, Diwekar UM, Manca D. Impacts assessment and tradeoffs of fuel cell
based auxiliary power units: Part ii. Environmental and health impacts, LCA,
and multi-objective optimization. J Power Sources 2005;139(1):214–22.
[31] Singhal S. Advances in solid oxide fuel cell technology. Solid State Ion
2000;135(1):305–13.
[32] McPhail SJ, Leto L, Boigues-Muñoz C. The yellow pages of SOFC technology
international status of SOFC deployment 2012–2013. ENEA; 2013.
[33] Park SK, Kim TS, Sohn JL, Lee YD. An integrated power generation system
combining solid oxide fuel cell and oxy-fuel combustion for high performance
and CO2 capture. Appl Energy 2011;88(4):1187–96.
[34] Papurello D, Borchiellini R, Bareschino P, Chiodo V, Freni S, Lanzini A, et al.
Performance of a solid oxide fuel cell short-stack with biogas feeding. Appl
Energy 2014;125:254–63.
[35] Annex 33. Fuel cells for stationary applications. Tech rep. International Energy
Agency; 2014. <http://www.ieafuelcell.com/annexdescriptions.php>.
[36] Enefield. Fuel cells x combined heat and power; 2016. <http://enefield.eu/>
[accessed March 11, 2016].
[37] Fabrizio E, Filippi M, Virgone J. An hourly modelling framework for the
assessment of energy sources exploitation and energy converters selection and
sizing in buildings. Energy Build 2009;41(10):1037–50.
[38] Andreassi L, Ciminelli M, Feola M, Ubertini S. Innovative method for energy
management: modelling and optimal operation of energy systems. Energy
Build 2009;41(4):436–44.
[39] Andreassi L, Ubertini S. Optimal management of power systems. In: Energy
management. Francisco Maria Perez; 2010. intech ISBN: 978-953-307-065-0..
[40] Wakui T, Kawayoshi H, Yokoyama R. Optimal structural design of residential
power and heat supply devices in consideration of operational and capital
recovery constraints. Appl Energy 2016;163:118–33.
[41] Doering R, Lin B. Optimum operation of a total energy plant. Comput Oper Res
1979;6(1):33–8.
[42] Cardona E, Piacentino A. Optimal design of CHCP plants in the civil sector by
thermoeconomics. Appl Energy 2007;84(7–8):729–48.
[43] Cappa F, Facci AL, Ubertini S. Proton exchange membrane fuel cell for
cooperating households: a convenient combined heat and power solution for
residential applications. Energy 2015;90:1229–38.
[44] Kong X, Wang R, Li Y, Huang X. Optimal operation of a micro-combined
cooling, heating and power system driven by a gas engine. Energy Convers
Manage 2009;50(3):530–8.
[45] Facci AL, Andreassi L, Ubertini S. Optimization of CHCP (combined heat power
and cooling) systems operation strategy using dynamic programming. Energy
2014;66:387–400.
[46] Facci AL, Andreassi L, Martini F, Ubertini S. Comparing energy and cost
optimization in distributed energy systems management. J Energy Resour
Technol 2014;136(3):032001.
[47] Facci AL, Andreassi L, Ubertini S, Sciubba E. Analysis of the influence of thermal
energy storage on the optimal management of a trigeneration plant. Energy
Proc 2014;45:1295–304.
[48] Hawkes A, Leach M. Cost-effective operating strategy for residential micro-
combined heat and power. Energy 2007;32(5):711–23.
[49] Schicktanz M, Wapler J, Henning H-M. Primary energy and economic
analysis of combined heating, cooling and power systems. Energy 2011;36
(1):575–85.
[50] Ren H, Gao W. Economic and environmental evaluation of micro CHP systems
with different operating modes for residential buildings in japan. Energy Build
2010;42(6):853–61.
[51] AEEG. Condizioni economiche per i clienti del mercato tutelato, energia
elettrica – servizio di maggior tutela, anno 2015, clienti domestici e non
domestici; 2015. <http://www.autorita.energia.it/it/dati/condec.htm]>
[accessed March 11, 2016].
[52] AEEG. Condizioni economiche per i clienti del mercato tutelato, gas – servizio
di tutela, anno 2015, clienti domestici e condomini con uso domestico; 2015.
<http://www.autorita.energia.it/it/dati/condec.htm]> [accessed March 11,
2016].
[53] Eurostat. Electricity prices – price systems 2014; 2015.
[54] Eurostat. Gas prices – price systems 2014; 2015.
[55] Haeseldonckx D, Peeters L, Helsen L, Dâhaeseleer W. The impact of thermal
storage on the operational behaviour of residential CHP facilities and the
overall CO2 emissions. Renew Sust Energy Rev 2007;11(6):1227–43.
[56] Barbieri ES, Melino F, Morini M. Influence of the thermal energy storage on the
profitability of micro-CHP systems for residential building applications. Appl
Energy 2012;97:714–22.
[57] SolidPOWER. EngenTM2500; 2016. <http://www.solidpower.com/en/engen-
2500/]engen2500> [accessed March 11, 2016].
[58] Bucheli O, Bertoldi M, Modena S, Ravagni A. Development and manufacturing
of SOFC-based products at SOFCpower SpA. ECS Trans 2013;57(1):81–8.
[59] SolidPOWER. EnGENTM2500 technical specifications; 2016. <http://
www.solidpower.com/wp-content/uploads/2014/03/Data_Sheet_Engen2500_
eng.pdf>.
[60] Wuillemin Z, Ceschini S, Antonetti Y, Beetschen C, Modena S, Montinaro D,
et al. High-performance SOFC stacks tested under different reformate
compositions. In: Proc 11th Eur SOFC SOE Forum.
574 A.L. Facci et al. / Applied Energy 192 (2017) 563–574
[61] Modena S, Ceschini S, Contino AR, Bini R, Tognana L, Bertoldi M, et al. Evolution
of sofcpowerâstack performances. ECS Trans 2013;57(1):359–66.
[62] Systema s.p.a. SYBCT assorbitori monoblocco per esterno e interno; 2016.
<http://www.systema.it/scheda-sybct> [accessed March 11, 2016].
[63] Burer M, Tanaka K, Favrat D, Yamada K. Multi-criteria optimization of a district
cogeneration plant integrating a solid oxide fuel cell–gas turbine combined
cycle, heat pumps and chillers. Energy 2003;28(6):497–518.
[64] Yu Z, Han J, Cao X, Chen W, Zhang B. Analysis of total energy system based on
solid oxide fuel cell for combined cooling and power applications. Int J
Hydrogen Energy 2010;35(7):2703–7.
[65] Al-Sulaiman FA, Dincer I, Hamdullahpur F. Energy analysis of a trigeneration
plant based on solid oxide fuel cell and organic rankine cycle. Int J Hydrogen
Energy 2010;35(10):5104–13.
[66] Daikin. VRV Catalogo Prodotti; 2015. <http://www.daikin.it/docs/Catalogo%
20VRV%202015-tcm478-349195.pdf> [accessed March 11, 2016].
[67] IEA-IRENA. Thermal energy storage technology brief; 2013. <www.irena.org/
Pubblications>.
[68] Manufacturing cost analysis of 1 kw and 5 kw solid oxide fuel cell (SOFC) for
auxilliary power applications. Tech rep. DE-EE0005250, Battelle; 2014.
[69] International Energy Agency. Fuel cells. In: IEA energy technology essentials;
2007.
[70] Palazzi F, Autissier N, Marechal FM, Favrat D. A methodology for thermo-
economic modeling and optimization of solid oxide fuel cell systems. Appl
Therm Eng 2007;27(16):2703–12.
[71] James BD, Spisak AB, Colella WG. Manufacturing cost analysis of stationary
fuel cell systems. Tech rep. Strategic Analysis; 2012.
[72] Weimar MR, Chick LA, G DW, Whyatt GA. Cost study for manufacturing of solid
oxide fuel cell power systems. Tech rep. PNNL-22732. US Department of
Energy.