Embed Size (px)
Citation preview
ARTICLE IN PRESS
0301-4215/$ - se
doi:10.1016/j.en
�CorrespondE-mail addr
Energy Policy 35 (2007) 6132–6144
www.elsevier.com/locate/enpol
Trigeneration primary energy saving evaluation for energy planning andpolicy development
Gianfranco Chicco�, Pierluigi Mancarella
Dipartimento di Ingegneria Elettrica, Politecnico di Torino, Corso Duca degli Abruzzi 24, 10129 Torino, Italy
Received 1 May 2007; accepted 16 July 2007
Available online 31 August 2007
Abstract
Trigeneration or combined heat, cooling and power (CHCP) is becoming an increasingly important energy option, particularly on a
small-scale basis (below 1MWe), with several alternatives nowadays available for the cooling power production and the coupling to
cogeneration systems. This paper deals with the introduction of a suitable framework for assessing the energy saving performance of
trigeneration alternatives, orientated towards energy planning studies and the development of regulatory policies. In particular, a new
generalized performance indicator—the trigeneration primary energy saving (TPES)—is introduced and discussed, with the aim of
effectively evaluating the primary energy savings from different CHCP alternatives. The potential of the TPES indicator is illustrated
through specific analyses run over different combinations of trigeneration equipment, providing numerical examples based on time-
domain simulations to illustrate the dependence of the energy saving characteristics on the CHCP system configurations and equipment,
as well as on the loading levels. In addition, the key aspect of adequately establishing the reference efficiencies for the conventional
separate production of electrical, thermal and cooling power is addressed in detail. This aspect affects both equipment selection and
potential profitability of the considered solutions under the outlook of receiving financial incentives.
r 2007 Elsevier Ltd. All rights reserved.
Keywords: Trigeneration; Energy efficiency; Primary energy savings
1. Introduction
In recent years, the operators of the energy sector haveput an increasingly high focus on issues concerning energysaving and implementation of high-efficiency energysystems, both from the technical and from the regulatorypoint of view (Cardona and Piacentino, 2005). Inparticular, the latest concerns in the energy sector aremainly related to the worldwide increase of energyconsumption, the attempts to reduce the energy depen-dence from some regions of the world holding a relevantshare of fossil primary sources and the emergence ofbinding environmental constraints aimed at limiting theproduction of greenhouse gases (GHGs). In addition, thedevelopment of liberalized energy markets in manycountries has created new interests for analyzing the
e front matter r 2007 Elsevier Ltd. All rights reserved.
pol.2007.07.016
ing author. Tel.: +39 011 090 7141; fax: +39 011 090 7199.
ess: [email protected] (G. Chicco).
possibility of exploiting the equipment available forelectricity production in a more profitable way.Cogeneration (Horlock, 1997) is being extensively used
as an efficient technique to produce heat and electricity,leading to a substantial energy saving with respect to the‘‘conventional’’ separate production (SP) of the sameenergy vectors, respectively, in heat generators and in thepower system. In particular, in the past, mostly because ofeconomy-of-scale reasons, cogeneration was limited tolarge-sized (industrial and district heating) plants. Yet, therecent development of ‘‘thermal’’ distributed generation(DG) technologies, such as microturbines (MTs) andinternal combustion engines (ICEs) (Willis and Scott,2000; Borbely and Kreider, 2001) has enabled the deploy-ment of various small-scale (below 1MWe) applications. Inaddition, DG technologies are being encouraged in severalcountries owing to their high potential for emissionreduction of CO2 and other hazardous pollutants, as, forinstance, discussed by Strachan and Dowlatabadi (2002)and Strachan and Farrell (2006). As a further point, fuel
ARTICLE IN PRESS
Nomenclature
Acronyms
CERG compression electric refrigerator groupCHCP combined heat, cooling and powerCHG combustion heat generatorCHP combined heat and powerCOP coefficient of performanceDG distributed generationEDC engine-driven chillerEHP electric heat pumpFC fuel cellFESR fuel energy saving ratioGARG gas absorption refrigerator groupGHG greenhouse gasesICE internal combustion engine
LHV lower heating valueMT microturbineSP separate productionTPES trigeneration primary energy savingWARG water absorption refrigerator group
Symbols
Subscripts represent energy sources or end use (y ¼cogeneration, z ¼ trigeneration, e ¼ electricity,t ¼ thermal, c ¼ cooling, F ¼ fuel, d ¼ demand)and specify the measuring units. For the energyvectors, the same symbols are used for energy(kWh) or average power (kW): W for electricity,Q for heat, R for cooling (refrigeration), F forfuel thermal content. The Greek letters Z and edenote efficiency.
G. Chicco, P. Mancarella / Energy Policy 35 (2007) 6132–6144 6133
cells (FCs) (Willis and Scott, 2000; Borbely and Kreider,2001) could play an important role in the future, withinalternative high-efficiency energy scenarios based on apotential hydrogen economy (Clark and Rifkin, 2006;McDowall and Eames, 2006).
Several small-scale cogeneration applications, besidesheat and electricity, require cooling power (e.g., for airconditioning purposes). In order to supply this threefoldenergy need, it is possible to set up the so-calledtrigeneration or combined heat, cooling and power (CHCP)plants (EcoGeneration Solutions LLC Companies, 1999;Resource Dynamics Corporation, 2003).
Trigeneration can be seen as the simultaneous produc-tion of electricity, heat and cooling power from the samesource of energy (typically gas). From this point of view, atrigeneration plant can be considered as the extension of acogeneration or combined heat and power (CHP) plant.The literature typically refers to trigeneration as thecombination of a traditional CHP prime mover (i.e., athermal machine such as an ICE, a MT or a FC thatcogenerates electricity and heat) with an absorption group,fed by hot water or steam produced by the cogenerationgroup (Colonna and Gabrielli, 2003; Bassols et al., 2002;Maidment and Prosser, 2000; Hwang, 2004). The rationaleof this approach is based on the potential efficiency ofusing the thermal power cogenerated also in the summer-time to fire the absorption machine for cooling production,enabling better and longer exploitation of the prime mover,as shown, for instance, by Havelsky (1999), Heteu andBolle (2002), and Cardona and Piacentino (2003). Thiskind of application may be referred to as ‘‘seasonal’’trigeneration. However, an array of other applications (forinstance, hospitals, department stores, hotels and so forth)require an actual trigeneration production throughout thewhole year, so that the optimal setup of the plant, alsoaccounting for the economic issues, could be different from
the cases of seasonal trigeneration. Thus, in previous works(Chicco and Mancarella, 2005, 2006; Mancarella, 2006),the authors have considered a generalized concept oftrigeneration, considering a set of different optionaltechnologies and sizes for the cooling side coupled to theCHP side.As a consequence of the increasing diffusion of various
types of plants, the evaluation of a trigeneration system isbecoming a crucial issue and requires the adoption ofadequate performance indicators. From this perspective,the energy savings attributable to adopting one plantconfiguration compared with another could be a suitableindicator for evaluating and comparing the effectiveness ofeach alternative. However, the definition of ‘‘energysaving’’ in a trigeneration system also needs to be discussedand clarified. In fact, as pointed out by Chicco andMancarella (2006), classical tools for evaluating CHPplants, such as the fuel energy saving ratio (FESR)indicator (Horlock, 1997), are not always adequate forCHCP plant assessment. Thus, other approaches may benecessary, such as the ones taken up by Havelsky (1999)and Heteu and Bolle (2002), that assess trigenerationsystems by explicitly accounting for the SP of coolingpower, besides heat and electricity. In addition, it is notalways clear how to evaluate specific energy savings andwhat reference situation to apply (Boonekamp, 2006). As afurther fundamental point, to date and to the authors’knowledge, there is no official regulatory frameworkdealing with the issue of evaluating the performance ofCHCP systems. Differently, CHP plants, whose energysaving are officially recognized and expressed throughsuitable indicators, receive financial incentives in manycountries. The details are discussed by Cardona andPiacentino (2005), with practical applications provided,for instance, in Italy by Deliberation no. 42/02 ofthe Italian AEEG (2002), and in the European Directive
ARTICLE IN PRESS
cogeneration side
(�W, �Q)
F
W
Q
cooling side (COPc, �t)
F R
Q
Fig. 1. Separate cooling power generation mode with heat recovery.
G. Chicco, P. Mancarella / Energy Policy 35 (2007) 6132–61446134
2004/8/EC (2004) on the establishment of a commonframework for regulating cogeneration at a continental level.
On these premises, it is clear how the proper evaluationof the performance of a trigeneration system couldrepresent a key point for promoting, through regulatedincentives, the diffusion of high-efficiency CHCP plants ona wider basis.
In this light, in analogy with the cogeneration FESR,and according to the lines drawn by Havelsky (1999) andHeteu and Bolle (2002) for evaluating the performance of‘‘classical’’ CHCP systems with absorption chillers, thispaper introduces a generalized indicator called trigenera-tion primary energy saving (TPES), aimed at assessing thepotential energy saving from any kind of trigenerationplant. After discussing the main issues relevant to theformulation of the new indicator, the effectiveness ofadopting the TPES is shown in a comprehensive case studythat highlights the manifold key variables and parametersfor plant performance assessment. The benefit of this kindof modeling is that it is possible, starting from the threefolduser’s energy needs, to track back the energy inputs to thewhole system, typically gas from a gas distribution networkand electricity from the electrical grid. In this way, seeingthe whole plant as a black box with fuel and electricity asinput and electricity, heat and cooling as output, it is easyto calculate and compare the TPES for every differenttrigeneration solution.
The paper is structured as follows: Section 2 contains ageneral introduction to the main characteristics andcomponents of trigeneration plants; Section 3 describesthe CHCP equipment performance models and defines anddiscusses the TPES indicator; Section 4 presents acomprehensive case study, based on time-domain simula-tions of different types of CHCP plants and loads, pointingout the potential of the proposed indicator; and Section 5contains the concluding remarks.
2. Trigeneration systems: configurations, characteristics,
and components
2.1. General structure of a CHCP plant
A CHCP plant can be seen as composed of a‘‘cogeneration side’’ and a ‘‘cooling side’’, made up of thefollowing components:
W
�cogeneration side
(�W, �Q)Q
W
F
A cogenerative prime mover, that for small-scaleapplications is typically an ICE or an MT (or in casemore units in parallel), both gas-fed (Willis and Scott,2000; Borbely and Kreider, 2001), while FCs could playa relevant role in the future (Clark and Rifkin, 2006;McDowall and Eames, 2006).
�cooling side
(COPc)
R
Fig. 2. Electrical bottoming cooling power generation mode.
A combustion heat generator (CHG) group, typicallyindustrial boilers (Kreider, 2001; ASHRAE, 2000;Danny Harvey, 2006), needed for both back-up andthermal peak-saving operation. Prime mover and CHGtogether compose the ‘‘cogeneration side’’.
�
The ‘‘cooling side’’, which can be composed of differentequipment (Chicco and Mancarella, 2006; Mancarella,2006), as detailed in Section 2.3.2.2. Possible trigeneration configurations
In general, it is possible to categorize the CHCPconfigurations into two models, according to the equip-ment used for the cooling generation and to the waythis equipment is linked to the CHP side: the separate
(or parallel) cooling generation mode and the bottoming
(or series) cooling generation mode.In the separate cooling generation mode (Fig. 1), the
cooling side is ‘‘decoupled’’ from the cogeneration side, i.e.,it is fed by means of energy vectors not coming from theCHP unit. The driving energy is typically gas. The relevantenergy efficiencies and energy flows shown in Fig. 1 arediscussed in Section 3.1.In the bottoming cooling generation mode, the cooling
side is cascaded to the topping cogeneration unit. In thiscase, depending on the cooling equipment adopted, andthus on the energy vector from the CHP plant used to drivethe chiller, it is possible to consider an electrical bottoming
cycle (Fig. 2) or a thermal bottoming cycle (Fig. 3).The relevant energy efficiencies and energy flows shownin Figs. 2 and 3 are discussed in Section 3.1.In general, although the possible schemes refer basically
to separate or bottoming models for the cooling genera-tion, there may be a far tighter interconnection between thesingle equipment of the energy system (for instance,because of heat recovery within the plant), which could
ARTICLE IN PRESS
cogeneration side
(�W, �Q)
F
W
Q
R
Q
cooling side
(COPc)
Fig. 3. Thermal bottoming cooling power generation mode.
G. Chicco, P. Mancarella / Energy Policy 35 (2007) 6132–6144 6135
enhance the energy saving potential of a trigeneration system(Mancarella, 2006). The heat-recovery possibility is high-lighted in Fig. 1 with explicit indication of heat recoveredfrom the separate cooling generation plant (see alsoSection 2.3), but this possibility could be in general envisagedalso for bottoming cooling generation solutions (although notexplicitly shown in Figs. 2 and 3). In addition, the varioussystems can be completed by including a thermal storagesystem (Hasnain, 1998a), as well as a cooling storage system(Hasnain, 1998b), in order to improve plant design andmanagement by enabling the creation of an energy buffer tobe profitably used for thermal load control purposes.
2.3. Cooling side components
Within the separate cooling generation mode, possibleequipment includes:
�
Gas absorption refrigerator group (GARG): the coolingplant is composed of one or more direct-fired absorptionchillers, directly fed by gas, whose thermal power F isdirectly transformed into cooling power R (Kreider,2001; ASHRAE, 2000; Danny Harvey, 2006; ResourceDynamics Corporation, 2003; US DOE, 2004). � Engine-driven chiller (EDC): the cooling side is com-posed of one or more engine-driven chillers, also directlyfed by fuel. In this case, a ‘‘conventional’’ compressionchiller is driven, instead of electrically, by a mechanicalcompressor directly connected to the shaft of an ICE(ASHRAE, 2000; Danny Harvey, 2006; ResourceDynamics Corporation, 2003).
In general, both GARG and EDC solutions allow forrecovery of waste heat (Fig. 1). For instance, the heatusually discharged from a GARG through a cooling towercould be recovered by means of a heat pump that increasesthe thermal level of the water circulating in the coolingtower (at around 30 1C) to a thermal level more suitable forthe specific user’s applications (for instance, for producinghot air at 50 1C for ambient heating). Likewise, an EDCallows for even higher heat recovery from the drivingengine exhaust gases (ASHRAE, 2000). Heat recovery canfurther increase the overall energy efficiency of thetrigeneration plant (Mancarella, 2006).
Considering the bottoming cooling production mode,among the possible equipment to adopt it is possible tomention (Kreider, 2001; ASHRAE, 2000; Danny Harvey,2006; US DOE, 2004):
�
Water absorption refrigerator group (WARG): it is the‘‘classical’’ case of trigeneration in the literature and inmost installations. The machine is fired by heat Qproduced by the cogenerator (thermal bottoming).
� Compression electric refrigerator group (CERG) andelectrical heat pump (EHP), both representing theclassical solution to produce cooling power R fromelectricity W, which in the specific case can be producedby the CHP unit (electrical bottoming). The EHP isusually a reversible machine that can operate in bothheating and cooling modes.
Also, for the bottoming cases, heat recovery from thecooling generation plant is in general possible. Forinstance, Havelsky (1999) reports a scheme where anelectric heat pump is used to enhance the thermal level ofthe heat recovered from a WARG, in analogy with what ismentioned above for the GARG.
3. Trigeneration equipment, performance evaluation and
indicators
3.1. Performance indicators for CHCP equipment
The most suitable way to synthetically evaluate anenergy system component is to resort to its efficiency,defined, energetically, as the ratio of the useful effect to theenergy spent to obtain that effect, i.e., the output-to-inputenergy ratio (Horlock, 1997; ASHRAE, 2000; DannyHarvey, 2006).The energy performance of a CHP prime mover can be
described by means of the electrical efficiency ZW and thethermal efficiency ZQ, ratio of, respectively, the electricaloutput W and the thermal output Q to the fuel thermalinput F (Horlock, 1997; Borbely and Kreider, 2001):
ZW ¼W
F; ZQ ¼
Q
F. (1)
The efficiencies in (1) depend in general on thetechnology, the loading level, the heat-recovery systemand the application (Horlock, 1997; Borbely and Kreider,2001; EDUCOGEN, 2001).The performance of the CHG units, typically boilers, is
usually described through the thermal efficiency Zt, ratio ofthe thermal output Q to the fuel thermal input F
(ASHRAE, 2000; Danny Harvey, 2006):
Zt ¼Q
F. (2)
For the cooling plant equipment, the most usedperformance indicator is the coefficient of performance(Kreider, 2001; ASHRAE, 2000; Danny Harvey, 2006; US
ARTICLE IN PRESS
Fz
Wz
Qz
Rz
Trigeneration system
Fig. 4. Trigeneration system input–output model (black box).
G. Chicco, P. Mancarella / Energy Policy 35 (2007) 6132–61446136
DOE, 2004), ratio of the desired output (cooling power R
for chillers, heating power Q for EHPs in heating mode) tothe input (electrical power W for the CERG and the EHP,thermal power Q in the form of steam or hot water for theWARG, thermal power F from the fuel for the GARG andthe EDC):
COPc ¼R
WðCERG; EHP cooling modeÞ; (3)
COPt ¼Q
WðEHP heating modeÞ; (4)
COPc ¼R
QðWARGÞ; (5)
COPc ¼R
FðGARG; EDCÞ: (6)
In addition, relevant effectiveness indicators can bedefined for the heat recovery from the cooling plant(Wulfinghoff, 1999; Mancarella, 2006). For instance, inFig. 1a heat-recovery efficiency for the separate coolinggeneration plant is indicated, defined as �t ¼ Q=F , where Q
represents the recovered heat.Depending on the details of the specific study, all the
above performance indicators for thermal and coolinggeneration equipment can be modeled as a function ofseveral variables, such as the loading level, the outdoorconditions, which is crucial for the cooling side equipment,and so forth (ASHRAE, 2000; Danny Harvey, 2006).
3.2. Evaluation of cogeneration plants: the FESR indicator
The traditional approach to CHP plant evaluationresorts to different types of performance indicators(Horlock, 1997). The fuel energy saving ratio (FESR)indicator is widely used, also from a regulatory standpoint,as discussed by Cardona and Piacentino (2005), to evaluatethe fuel energy saving obtained in a cogeneration plantwith respect to the SP of heat (in conventional boilers) andelectricity (in the centralized power system). The FESR isdefined in Horlock (1997) as
FESR ¼FSP � F y
FSP¼ 1�
F y
W y=ZSPe þQy=ZSP
t
. (7)
In Eq. (7), Fy is the fuel thermal energy input (in generalcalculated on an LHV basis) to the CHP plant; FSP is thefuel thermal energy input required for the SP of the samecogenerated energy vectors Wy and Qy; and the entries ZSP
e
and ZSPt are the reference efficiencies for the conventional
SP of electricity and heat, assumed to occur in anequivalent power plant and an equivalent boiler, respec-tively (Horlock, 1997; Martens, 1998).
In principle, the same indicator could be adopted forCHCP plant evaluation. However, where this kind ofapproach is applied to ‘‘classical’’ trigeneration systemswith absorption chillers fired by cogenerated heat, itseffectiveness could be arguable. In fact, the evaluation of a
WARG-based CHCP system would depend on the energysavings brought by the cogenerated heat used to fire the chiller(Cardona and Piacentino, 2005; European Directive 2004/8/EC, 2004; Deliberation 42/02 of the Italian AEEG, 2002), andwould not explicitly take into account the role and theefficiency of the cooling generation (in both the CHCP systemand the SP). Thus, the FESR, as shown, for instance, byChicco and Mancarella (2006), is inadequate for comprehen-sive evaluation of different trigeneration alternatives.
3.3. Trigeneration evaluation: general definition of the
TPES indicator
As broadly discussed by Cardona and Piacentino (2005)for cogeneration, a suitable indicator to assess theperformance of a combined energy system should be basedon a comparison with the SP ‘‘conventional’’ means.Therefore, the trigenerated cooling power should explicitly
appear in the definition of the indicator, as well as therelevant reference for comparison with the relevantseparate cooling production means. In addition, the energyflows and interactions within a trigeneration system couldbe quite complicated, and a suitable performance indicatorshould be able to synthetically describe the energy savingcharacteristics of the system even without considering thedetails of the plant structure.On these premises, describing the trigeneration plant as a
black box with only the relevant input–output energy flows(Fig. 4), it is possible to generally define the TPES indicatorfor trigeneration system assessment as
TPES ¼FSP � Fz
FSP¼ 1�
Fz
W z=ZSPe þQz=ZSP
t þ Rz=ZSPc
.
(8)
It is worthwhile to clarify the role of the entriescontained in expression (8):
�
Fz is the overall fuel thermal energy input (in generalcalculated on an LHV basis) to the trigeneration system(including the fuel feeding a GARG or an EDC, as wellas the auxiliary CHG); � FSP is the total fuel thermal energy input required for theSP of the same energy vectors Wz, Qz and Rz producedin the trigeneration system;
� Wz is the net trigenerated electricity output (including theelectricity sold to the grid, and excluding the possibleenergy needed to feed electric equipment such as chillersor heat pumps);
ARTICLE IN PRESSG. Chicco, P. Mancarella / Energy Policy 35 (2007) 6132–6144 6137
�
Qz is the net useful trigenerated heat output (produced bythe cogenerator, boilers, heat pumps or some wayrecovered within the plant, and excluding the possiblethermal energy needed to feed absorption chillers); � Rz is the net trigenerated cooling energy output(excluding the energy in cases used for cooling purposeswithin the plant, such as pre-cooling of the turbine inletair, as, for instance, in Hwang (2004));
� the efficiencies ZSPe , ZSPt and ZSP
c are the SP referenceefficiencies, referred to the primary energy (fuel thermalenergy content) as input, for electrical, thermal andcooling power, respectively.
Clearly, expression (8) is an extension of the FESRindicator (7). Its formulation is the most general possibleand enables one to evaluate any type of trigenerationsystem. As such, it is a powerful and synthetic tool forassessing energy savings from trigeneration systems. Infact, no matter what the content of the plant black box inFig 4, that is, no matter how complex the energy flowswithin the plant are, nor the specific plant layout (forinstance, bottoming or parallel cooling generation, with orwithout heat recovery), the only relevant variables are theinput–output energy flows contained in (8). Thus, theTPES can be used for general trigeneration evaluation withdifferent types of equipment, being fed by and generatingdifferent types of energy, accounting for internal thermalrecovery and so forth. In addition, the separate generationreference efficiencies can be selected in the most suitablefashion according to the study purposes. From thisstandpoint, it is up to the energy designer, or to the policymaker, to select the most suitable numerical referencevalues. The implications of these choices are illustrated inthe sequel.
The overall calculation of the energy entries in (8) shouldbe performed by considering a specified observation period(for instance, 1 year). However, it is important to use,possibly, the values resulting from time-domain simulations(for instance, hourly based) accounting for the out-of-
design behavior of all the equipment involved, in order tocarry out an unbiased and comprehensive evaluation of theenergy system performance while evaluating differentpotential alternatives (Voorspools and D’haeseleer, 2003;Chicco and Mancarella, 2006; Mancarella, 2006). Thedifferent alternatives to envisage could include districtheating and district cooling systems typically involvingpowers higher than the ones relevant to the DG solutionsconsidered in this work.
3.4. Models and numerical values for the TPES evaluation
Likewise, for cogeneration system primary energy savingevaluation (Cardona and Piacentino, 2005; Directive2004/8/EC, 2004; Deliberation no. 42/02 of the ItalianAEEG, 2002), a crucial point in the numerical assessmentof the TPES is the merit of assigning suitable values to theSP reference efficiencies, which, in general, will depend on
the technologies replaced. As such, it is necessary to definethe most suitable models for separate generation of thedifferent energy vectors. For cogeneration systems, the SPof electricity and heat is modeled through an equivalentthermal power plant (including transmission and distribu-tion efficiency) and through an equivalent boiler, respec-tively (Horlock, 1997; Martens, 1998), as the moststraightforward solution. For the cooling production, thebasic alternatives would be a mechanical compression cycleor a thermally fed cycle (Kreider, 2001; ASHRAE, 2000;US DOE, 2004; Danny Harvey, 2006). Undoubtedly, themost widespread equipment for cooling power generationis based on vapor compression cycles, where the compres-sor is activated by electrical energy. Thus, from (8) theTPES can be rewritten in a more practical form as
TPES ¼FSP � Fz
F SP¼ 1�
Fz
W z=ZSPe þQz=ZSP
t þ Rz= ZSPe COPSP
� � ,
(9)
where COPSP is one of the equivalent CERGs chosen as amodel for the cooling power generation.Again, on the basis of the considerations drawn above
for (8) and with reference to Fig. 4, expression (9) can beadopted for comparatively evaluating any kind of trigen-eration system, in line with the approach to classicaltrigeneration outlined by Havelsky (1999) and Heteu andBolle (2004).With respect to assigning numerical values to the
reference efficiencies, there are no official guidelines forthis kind of evaluation. However, two approaches may beconsidered in more detail.For a straightforward comparison, it is possible to
assume average efficiency values, clearly dependent on thetype of production units operating in a specific country.For instance, ZSP
e could be set equal to 0.4 (about theaverage power system production efficiency in Italy,including line losses for transmission and distribution),while ZSP
t could be set to a value in the range from 0.8 to 0.9(average boiler efficiencies). Similarly, an average valuecould be chosen for CERG performance. Yet, since theperformance of a chiller strongly depends on the outdoortemperature, besides the temperature at which to providethe cooling effect (Kreider, 2001; ASHRAE, 2000; DannyHarvey, 2006), it would be better to adopt differentreference values for different seasons; values from 3 to 4are reasonable references for COPSP in (9).The other limit-case approach would be to compare the
combined energy system with the best available technolo-gies for SP. Thus, possible values could be ZSP
e ¼ 0:55(combined cycles), ZSP
t ffi 1 (condensing boilers, withreference to the fuel LHV) and COPSP
¼ 6–7 (althoughagain to be evaluated with respect to the operatingconditions).In between the illustrated approaches, several other
approaches would be possible, with variation in specificdetails such as, for instance, the type of input fuel adopted,
ARTICLE IN PRESSG. Chicco, P. Mancarella / Energy Policy 35 (2007) 6132–61446138
the plant size, the application (industrial or residential) andso on. These were discussed by COGEN EUROPE (2005),addressing the European Directive 2004/8/EC (2004) onthe regulation of cogeneration, and were considered inItaly through Deliberation no. 42/02 of the Italian AEEG(2002). Setting the reference values for the conventional SPrepresents a key specific task for regulators.
Some dedicated applications are shown in the nextsection, addressing the usage of the TPES indicator forenergy system assessment and also as a tool for assistingthe possible development of regulatory frameworks fortrigeneration.
4. Case study applications
This section presents a case study in which differenttrigeneration solutions are evaluated and comparedthrough the TPES calculated on the basis of results fromtime-domain simulations. In particular, different loadpatterns (scenarios) are considered, to highlight the impactof different loading levels on the trigeneration plantperformance. In addition, different reference values forthe SP efficiencies are considered, so as to highlight theimpact of their selection on the plant evaluation.
4.1. Description of load scenarios
Let us consider different load scenarios, each of themrepresented by a demand ‘‘loading vector’’ d composed ofa triplet of the form d ¼ (d1,d2,d3), where the componentd1 indicates the electrical load level, the component d2represents the thermal load level and the component d3represents the cooling load level. The different load levelsare referred to a base load scenario (Scenario D1) indicatedby the loading vector d1 ¼ (1,1,1) and characterized by anassigned hourly load pattern for electrical (Fig. 5), thermaland cooling power. In particular, the hourly patterns of thethermal load (Fig. 6) and of the cooling load (Fig. 7) are
0
50
100
150
200
250
300
1 2 3 4 5 6 7 8 9 10 11
[kW
e]
Fig. 5. Base hourly electrical l
specified for three different reference seasons (with 150days for winter, 125 days for midseason and 90 days forsummer), and the electrical load (Fig. 5) is assigned for aunique reference period. These load pattern models reflectthe typical seasonal variability encountered in outdoortemperature-related loads in most non-industrial applica-tions. For instance, such load patterns (at differentabsolute load levels) could be applicable to a hospital ora hotel. More specifically, the presence of winter coolingloads refers to the general possibility of dealing withspecific applications requiring ambient conditioning evenin winter (e.g., special hospital rooms, food storage and soforth), above all in warm climates. Furthermore, under thehypothesis that the electrical load does not exhibitsignificant seasonal variations for instance due to daylightduration cycles, for the specific case the major seasonalvariations in the electrical demand could be typicallyattributable to cooling power generation through electricchillers. Since the focus of the paper is indeed to addressthe potential of adopting different alternatives for coolingproduction, considering a unique electrical load patternhelps highlight the impact of different cooling solutions onoverall plant performance.As a further step in the analysis, in order to point out the
impact of the loading level on the plant performance, let usconsider the following scenarios, corresponding to varia-tion in the ‘‘absolute’’ level of some of the loads withrespect to the base Scenario D1, although holding the sameload pattern models:
�
12
hou
oad
Scenario D2, loading vector d2 ¼ (1,0.5,0.5): The thermaland electrical power load patterns are uniformly halvedwith respect to the values specified in Scenario D1, whilethe electrical load remains the same;
� Scenario D3, loading vector d3 ¼ (1,0.5,1): The thermalload pattern is uniformly halved with respect to thevalues specified in Scenario D1, while the electrical andcooling loads remain the same;
13 14 15 16 17 18 19 20 21 22 23 24
rs
pattern (all scenarios).
ARTICLE IN PRESS
0
50
100
150
200
250
300
350
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 21 22 23 24
summer
midseason
winter
[kW
t]
hours
20
Fig. 6. Base hourly thermal load for three seasons (Scenario D1).
0
50
100
150
200
250
300
350
400
450
500
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
summer
midseason
winter
[kW
c]
hours
Fig. 7. Base hourly cooling load pattern for three seasons (Scenario D1).
Table 1
G. Chicco, P. Mancarella / Energy Policy 35 (2007) 6132–6144 6139
�
Reference efficiency scenariosEfficiency scenario ZSPe ZSP
tCOPSP
Scenario D4, loading vector d4 ¼ (1,2,1): the thermalload pattern is uniformly doubled with respect to thevalues specified in Scenario D1, while the electrical andcooling loads remain the same;
Low efficiency 0.4 0.8 3
� Average 0.4 0.9 4Intermediate 0.45 0.95 5
State-of-the-art 0.55 0.98 6
Scenario D5, loading vector d5 ¼ (1,3,1): the thermalload pattern is uniformly tripled with respect to thevalues specified in Scenario D1, while the electrical andcooling loads remain the same.
4.2. Description of the reference efficiency scenarios
In order to highlight the impact of the selection of thenumerical values for the SP references in expression (9),let us consider the reference efficiency scenarios reportedin Table 1.
The low-efficiency reference values could be used for atypical comparison with equipment used in non-centralizedsystems for residential or tertiary applications. The average
reference values could be applied for dealing with the
ARTICLE IN PRESS
Table 2
Equipment used in the case study
Equipment Number of units Unit rated
capacity
Unit rated
efficiency
Prime mover (gas-
fed Otto ICE)
1 330 kWe ZW ¼ 0.36
ZQ ¼ 0.43
CHG (gas boilers) 1 (3)*, 2 (5)**, 4
(7)***, 6 (9)****
200 kWt Zt ¼ 0.9
CERG 2 280 kWc COP ¼ 3.5
WARG 1 515 kWc COP ¼ 0.65
GARG 2 250 kWc COP ¼ 1.1
*Scenario D2 and Scenario D3; **Scenario D1; ***Scenario D2; ****
Scenario D3. The entries between brackets refer to the WARG-based
plants.
Table 3
TPES (%) results from time-domain simulations: low-efficiency separate
production references (ZSPe ¼ 0:4, ZSP
t ¼ 0:8, COPSP¼ 3)
Cooling
generation
plant
Operation
strategy
Loading level scenarios
D1 D2 D3 D4 D5
CERG A 11.4 �0.1 0.4 16.6 16.6
B 8.9 1.6 1.3 20.2 21.6
C 11.1 1.3 1.1 19.8 21.4
WARG A 8.4 4.2 1.6 10.6 10.8
B 11.8 4.2 5.9 17.6 19.1
C 11.6 5.4 4.9 17.0 18.2
GARG A 8.7 �1.7 �2.6 13.6 14.2
B 7.9 1.3 0.1 19.0 20.6
C 9.8 1.1 0.3 18.4 20.1
G. Chicco, P. Mancarella / Energy Policy 35 (2007) 6132–61446140
equipment used in centralized systems for residentialor tertiary or some industrial applications. The intermedi-
ate reference values could be used for dealing withindustrial equipment. Finally, the state-of-the-art referencevalues point to the best technologies that can normallybe encountered today (ruling out specific higher-efficiency equipment and prototypes available but not yetcommercialized).
4.3. Description of trigeneration equipment
For trigeneration equipment selection, since the specificfocus is meant to be on the cooling power generation, let usfix the CHP prime mover and select alternative coolinggeneration equipment. In particular, looking up commer-cial catalogs, a prime mover suitable for the given loadscould be, for instance, a natural gas-fed cogenerative ICE.The engine is completely backed up by a CHG composed,for the different loading scenarios, of a proper number ofnatural gas-fuelled boilers in parallel. In addition, whenadopting a WARG, in order to ensure full coolingproduction capacity also when the CHP is switched off(for maintenance reasons or according to specific operationstrategies), the number of boilers is increased so that theabsorption chiller could also be fed by hot water producedby the boilers.
On the electrical side, the CHP system is assumed to begrid-connected, so that electricity generated in excess withrespect to the user’s demand can be sold to the powernetwork. Conversely, the heat produced in excess of theuser’s requirements is wasted. In addition, when operatedunder (thermal or electrical) load following mode, theengine is automatically switched off when its electrical loaddrops below 50% of the rated one, and the requiredelectricity is withdrawn from the grid.
As far as the cooling side is concerned, three differentalternatives are considered, selected from commercialcatalogs, namely:
�
a CERG, composed of two reciprocating units inparallel; � a WARG, composed of one single-effect unit fed by hotwater at 90 1C; and
� a GARG, composed of two direct-fired double-effectunits in parallel.
For the sake of simplicity, no heat recovery from thecooling generation plant is envisaged. The main character-istics of the alternative CHCP plant components aresummarized in Table 2.
4.4. Time-domain simulations and operation strategies
Hourly based time-domain simulations have been runover a 1-year time span for the different trigeneration plantalternatives and the different load scenarios considered. In
particular, three different operation strategies have beensimulated:
A.
electrical load following; B. thermal load following; and C. engine always on at full load.In addition, out-of-design models of the equipment, as inChicco and Mancarella (2006), have been adopted in thesimulations.
4.5. Results of simulations in terms of TPES
The values of the TPES index, computed on theaggregated hourly energy over 1 year for the different loadscenarios, plant alternatives and operation strategiesconsidered, are shown in Tables 3–6 corresponding to thedifferent SP reference efficiency scenarios. For each case,the best values obtained for each loading level scenario arehighlighted in bold characters.
4.6. Comments to TPES results
On the basis of the results shown in Tables 3–6, it ispossible to draw several conclusions from differentperspectives.
ARTICLE IN PRESS
Table 4
TPES (%) results from time-domain simulations: average separate
production references (ZSPe ¼ 0:4, ZSP
t ¼ 0:9, COPSP¼ 4)
Cooling
generation
plant
Operation
strategy
Loading level scenarios
D1 D2 D3 D4 D5
CERG A 6.2 �3.6 �4.5 10.6 9.9
B 3.6 �1.9 �3.6 15.0 16.1
C 7.3 �1.0 �2.2 15.4 16.3
WARG A 3.0 0.8 �3.2 4.2 3.5
B 7.1 0.9 1.6 12.6 13.6
C 8.1 3.3 1.9 12.6 13.1
GARG A 3.4 �5.2 �7.6 7.4 7.2
B 2.6 �2.2 �4.7 13.9 15.2
C 6.2 �1.2 �2.8 14.1 15.2
Table 5
TPES (%) results from time-domain simulations: intermediate separate
production references (ZSPe ¼ 0:45, ZSP
t ¼ 0:95, COPSP¼ 5)
Cooling
generation
plant
Operation
strategy
Loading level scenarios
D1 D2 D3 D4 D5
CERG A �2.9 �13.9 �15.6 2.8 2.5
B �1.0 �3.7 �6.3 8.5 9.2
C �3.9 �13.5 �15.3 6.1 7.8
WARG A �6.9 �8.9 �14.6 �4.6 �4.7
B �1.2 �4.5 �6.1 4.1 5.4
C �3.3 �8.8 �11.0 2.7 4.0
GARG A �6.4 �15.7 �19.6 �1.0 �0.7
B �3.5 �5.0 �9.2 6.6 7.7
C �5.4 �13.8 �16.3 4.4 6.3
Table 6
TPES (%) results from time-domain simulations: state-of-the-art separate
production references (ZSPe ¼ 0:55, ZSP
t ¼ 0:98, COPSP¼ 6)
Cooling
generation
plant
Operation
strategy
Loading level scenarios
D1 D2 D3 D4 D5
CERG A �16.6 �31.3 �33.8 �7.5 �6.5
B �6.1 �5.1 �8.2 0.0 0.2
C �22.9 �36.5 �38.8 �8.4 �4.7
WARG A �21.9 �25.5 �33.3 �16.5 �15.0
B �13.9 �12.5 �18.1 �8.2 �6.3
C �22.9 �30.9 �34.2 �13.0 �9.6
GARG A �21.4 �33.6 �39.4 �12.4 �10.4
B �11.4 �8.0 �14.5 �3.6 �2.6
C �11.4 �36.9 �40.6 �11.1 �7.0
G. Chicco, P. Mancarella / Energy Policy 35 (2007) 6132–6144 6141
First of all, for all the SP efficiency scenarios, there is asort of ‘‘saving of scale’’ for the thermal load, meaning thatthe higher the thermal load, the higher the plant energy
savings. This is essentially due to the fact that less heat iswasted (in operation strategies A and C), and that thecogeneration potential is fully exploited for longer time(in operation strategies B and C). In addition, withrelatively lower SP reference efficiencies, the WARG-basedplant is more efficient for lower loads, while the CERG-based plant is more efficient for higher loads. In fact, aWARG adds thermal load to the ‘‘base’’ heating load, sothat the cogenerator can efficiently operate at higher loads,satisfying the user’s needs for a longer duration than withthe other cooling generators. In Scenarios D4 and D5,however, the higher thermal load already enables thecogenerator to run at high loads, so that the additionalthermal load from WARG needs to be produced in boilers.Therefore, it is more efficient to produce cooling energy bymeans of a CERG or a GARG (composed of double-effectchillers, in this case), which intrinsically exhibit higherefficiency than the WARG (composed of a single-effectchiller, in this specific case). This is confirmed by the factthat for the ‘‘intermediate’’ and ‘‘state-of-the-art’’ referenceefficiency scenarios the plants with CERG and GARGexhibit higher TPES than the plant with WARG. Ofcourse, the situation changes when it is possible to sellcogenerated heat to a district heating network, making itmore profitable to run the cogenerator at full load, with noor lesser heat wasted.As far as the cooling load is concerned, passing from
Scenarios D2 to D3 does not always lead to higher energysavings. Indeed, the cooling loads impact the overallelectrical load (CERG) or the thermal load (WARG) asseen from the cogenerator, and thus the final outcomedepends on the operation strategy. In particular, additionalcooling load has a positive effect for the CERG-basedplant under Strategy A (electrical load following), since theengine can operate at higher loads and thus at higherefficiency. There is a similar effect for the WARG-basedplant, but under Strategy B (thermal load following), as itcould be expected. However, this is true only for thescenario with lower SP reference efficiencies, since increas-ing the cooling load has a negative impact on the TPESwhen the COPSP is higher (if the relevant referenceefficiency is very high, it is better to produce less). It isalso interesting to note that passing from the lower coolingload level to the higher cooling load level has a negativeimpact on the WARG-based plant under Strategy C (fullload at all times). In fact, passing to the higher cooling loadimplies that the engine must be backed up by the auxiliaryboilers to supply the absorption chiller, which leads to aloss of the benefits brought about by cogenerativeproduction. Thus, when there is higher cooling demandthe engine must be backed up, while when there is lowercooling demand and a lower thermal load, the enginewastes cogenerated heat. Both these facts result in anegative impact on the TPES values. The situationimproves consistently with a higher thermal load (fromScenarios D3 to D1), with higher energy savings arisingfrom reduced thermal losses from the ICE.
ARTICLE IN PRESS
CE
RG
A
CE
RG
B
CE
RG
C
WA
RG
A
WA
RG
B
WA
RG
C
GA
RG
A
GA
RG
B
GA
RG
C
D3D
2D1D
4D5
-0.10
-0.05
0.00
0.05
0.10
0.15
0.20
TPES
load
scenarios
plant and operation strategy
Fig. 8. TPES for the ‘‘average’’ reference efficiency scenario.
G. Chicco, P. Mancarella / Energy Policy 35 (2007) 6132–61446142
With respect to the operation strategies, it can be seenthat the loading level and reference efficiencies determinewhether the WARG-based plant or the CERG-based plantresults in the greatest TPES values. This does not apply toStrategy A.
Reference efficiency scenarios affect both the energysavings and the most effective plant and operation strategy.Generally speaking, increasing SP reference efficienciesleads to smaller energy savings for the trigenerationsystems, which become in practice zero for the state-of-the-art scenario. This confirms the importance of properlychoosing the reference scenarios for the SP. In fact, if thegoal is to promote the diffusion of high-efficiencytrigeneration technologies, for instance by enabling theuse of solutions with energy saving performance betterthan a specified level, the reference efficiencies need not betoo high. Conversely, if the goal is to set up additionalpremium incentives, the reference efficiencies should be setto relatively high values. Consideration of the referencetechnologies needs to take into account that upgrading thesystems for the separate generation to the state of the artcould be a relatively slow process.
From a practical standpoint, assuming that the referenceefficiencies for SP are fixed, the selection of the mostsuitable trigeneration plant configuration should bebased on looking up the relevant results. For instance, letus consider the case with ‘‘average’’ efficiencies for SP(Table 4), whose relevant results are represented in Fig. 8,in which the scenarios have been reordered in such a way asto highlight the trend in dependence of the different loads,as discussed above. The most suitable CHCP plantconfiguration, in terms of energy savings, depends on the
expected loading level. In particular, for the scenarios withlower thermal loads there are small or negative TPES forthe various plants. In Scenario D1 the WARG-based plantexhibits higher TPES, while for higher thermal loads theCERG-based plant, closely followed by the GARG-basedplant, exhibits higher TPES.
5. Concluding remarks
In this paper, the trigeneration primary energy saving(TPES) indicator has been introduced to synthetically
assess, from the most general point of view, the primaryenergy savings brought about by the adoption of any typeof trigeneration system. In particular, after deriving andillustrating the general model, a more specific formulationhas been considered for evaluating the trigenerated energyvectors with respect to the conventional separate produc-tion (SP) of electricity, heat and cooling power.Through a comprehensive case study based on time-
domain simulations, the proposed indicator has proven tobe effective at highlighting the role of the different relevantvariables and parameters involved in determining theefficiency of trigeneration. Of course, the numerical resultsobtained in the case study cannot be generalized to othertrigeneration plants. However, the conceptual explorationof the results has provided a useful indication of the type oftrigeneration solution relevant to different loading levels,operation strategies and reference efficiencies for SP. Inparticular, the numerical results emphasize the importanceof properly selecting the models and the numericalcharacteristics for the SP references in order to accuratelydetermine the most effective trigeneration solution.
ARTICLE IN PRESSG. Chicco, P. Mancarella / Energy Policy 35 (2007) 6132–6144 6143
In the future, regulation will have to explicitly take intoaccount trigeneration because of the increasing importancethat this technique, in different forms and with differentplant schemes, is assuming worldwide. In particular, owingto the enhanced energy generation efficiency of combinedheat, cooling and power (CHCP) plants, proper perfor-mance evaluation of trigeneration will become morerelevant within potential regulatory frameworks envisagingfinancial incentives aimed at improving the energy sectorefficiency, in line with what already occurs for cogenerationsystems (Directive 2004/8/EC, 2004). In this respect, theproposed TPES indicator, as a straightforward extensionof the fuel energy saving ratio (FESR) widely adopted forcogeneration technical and policy assessment in severalcountries, could represent an effective tool for evaluatingthe efficiency characteristics of different types of combinedsystems for electrical, thermal and cooling power genera-tion. For instance, using the TPES indicator in regulatoryanalyses, potential financial incentives for trigenerationenergy savings could be devised. Potentially, this couldboost the spread of high-efficiency energy systems, withbenefits in terms of primary energy savings as well asconsequent CO2 emission reduction. In addition, thediffusion of absorption technologies for cooling generationwould positively impact the electrical power grid manage-ment and structure, by decreasing the electrical loadrequired of the share needed for conventional electricalair conditioners in the summertime peak periods.
However, besides the energy saving evaluation, econom-ic analysis will play a fundamental role in the final selectionof a CHCP plant. Hence, assuming roughly the samefixed cost for electric and absorption chillers, comparativeelectricity and gas prices would determine the selectionof CERG or GARG for the cooling generation (Foleyet al., 2000; Mancarella, 2006). In general, as shown byMancarella (2006) and Chicco and Mancarella (2006),different trigeneration solutions and the relevant energysavings could bring about important economic benefitswith respect to the conventional SP (with pay-back timeseven lower than 3 years), which however strongly dependon the specific market framework and on electricity and gasprices. In particular, if the TPES had regulatory relevancefor delivering financial incentives, as in the case of theFESR for cogeneration, the economic balance could bemoved towards solutions that, although slightly moreexpensive, would ensure higher energy and environmentalbenefits. However, from this perspective, and as illustratedin this paper, a suitable discussion on the selection of theSP reference values is necessary, and could mark theturning point for encouraging the diffusion of higher-efficiency and lower-emission trigeneration systems. In thisrespect, setting average reference values for the SPevaluation, to be determined within the specific regulatoryframework, could represent an effective approach to boostthe diffusion of trigeneration systems. This is in line withconsidering that switching the bulk of current SPtechnologies to the state of the art requires time (of the
order of one decade); meanwhile, trigeneration cancontribute to improve energy efficiency and reduce GHGemissions from energy generation.Further work is in progress to set up a comprehensive
energy, environmental and economic evaluation frameworkfor trigeneration plants, including the assessment of thepotential reduction of CO2 emissions attributable to differenttypes of trigeneration systems, as well as of the economicbenefits relevant to different TPES values for differentCHCP solutions within different energy market scenarios.
Acknowledgment
This work has been supported by the Regione Piemonte,Torino, Italy, under the research Grant C65/2004.
References
American Society of Heating, Refrigerating and Air-conditioning
Engineers (ASHRAE), 2000. ASHRAE HVAC Systems and Equip-
ment Handbook, SI ed. ASHRAE, USA.
Bassols, J., Kuckelkorn, B., Langreck, J., Schneider, R., Veelken, H.,
2002. Trigeneration in food industry. Applied Thermal Engineering 22,
595–602.
Boonekamp, P.G.M., 2006. Evaluation of methods used to determine
realized energy savings. Energy Policy 34, 3977–3992.
Borbely, A.M., Kreider, J.F. (Eds.), 2001. Distributed Generation—The
Power Paradigm of the NewMillennium. CRC Press, Boca Raton, FL.
Cardona, E., Piacentino, A., 2003. A method for sizing a trigeneration
plant in Mediterranean areas. Applied Thermal Engineering 23,
1665–1680.
Cardona, E., Piacentino, A., 2005. Cogeneration: a regulatory framework
toward growth. Energy Policy 33, 2100–2111.
Chicco, G., Mancarella, P., 2005. Planning aspects and performance
indicators for small-scale trigeneration plants. In: Proceedings
of Future Power Systems (FPS) Conference, Amsterdam,
The Netherlands, Nov. 16–18, 2005.
Chicco, G., Mancarella, P., 2006. From cogeneration to trigeneration:
profitable alternatives in a competitive market. IEEE Transactions on
Energy Conversion 21, 265–272.
Clark II, W.W., Rifkin, J., 2006. A green hydrogen economy. Energy
Policy 34, 2630–2639.
COGEN EUROPE, Position statement on the definition of harmonised
efficiency reference values, 28 February 2005, av /http://www.cogen.
org/Downloadables/Publications/Position_Paper_CHP_Reference_
Values.pdfS.
Colonna, P., Gabrielli, S., 2003. Industrial trigeneration using ammonia-
water absorption refrigeration systems (AAR). Applied Thermal
Engineering 23, 381–396.
Danny Harvey, L.D., 2006. A Handbook on Low Energy Buildings and
District Energy Systems. James and James, UK.
Deliberation n. 42/02 of the Italian AEEG (Electricity and Gas Authority,
Autorita per l’Energia Elettrica e il Gas), 2002. Available at
/www.autorita.energia.itS.
Directive 2004/8/EC of the European Parliament and of the Council of 11
February 2004 on the promotion of cogeneration based on a useful
heat demand in the internal energy market and amending Directive
92/42/EEC, 2004, Official Journal of the EU, 21.2.2004, L 52, 50–60.
EcoGeneration Solutions LLC Companies, 1999. Renewable Energy
Technologies. Available at /www.trigeneration.comS.
EDUCOGEN, 2001. The European Educational Tool on Cogeneration.
Available at /www.cogen.org/projects/educogen.htmS.
Foley, G., De Vault, R., Sweetser, R., 2000. Future of Absorption
Technology in America: a Critical Look at the Impact of BCHP and
ARTICLE IN PRESSG. Chicco, P. Mancarella / Energy Policy 35 (2007) 6132–61446144
Innovation. Oak Ridge National Laboratory, June 2000. Available at
/http://www.eere.energy.gov/de/publications.htmlS.
Hasnain, S.M., 1998a. Review on sustainable thermal energy storage
technologies, Part I: heat storage materials and techniques. Energy
Conversion and Management 39, 1127–1138.
Hasnain, S.M., 1998b. Review on sustainable thermal energy storage
technologies, Part II: cool thermal storage. Energy Conversion and
Management 39, 1139–1153.
Havelsky, V., 1999. Energetic efficiency of cogeneration systems for
combined heat, cold and power production. International Journal of
Refrigeration 22, 479–485.
Heteu, P.M.T., Bolle, L., 2002. Economie d’energie en trigeneration.
International Journal of Thermal Sciences 41, 1151–1159.
Horlock, J.H., 1997. Cogeneration—Combined Heat and Power (CHP).
Krieger, Malabar, FL.
Hwang, Y., 2004. Potential energy benefits of integrated refrigeration
system with microturbine and absorption chiller. International Journal
of Refrigeration 27, 816–829.
Kreider, J.F. (Ed.), 2001. Handbook of Heating, Ventilation and Air
Conditioning. CRC Press, Boca Raton, FL.
Maidment, G.G., Prosser, G., 2000. The use of CHP and absorption
cooling in cold storage. Applied Thermal Engineering 20, 1059–1073.
Mancarella, P., 2006. From cogeneration to trigeneration: energy planning
and evaluation in a competitive market framework. Ph.D. Disserta-
tion, Politecnico di Torino, Torino, Italy.
Martens, A., 1998. The energetic feasibility of CHP compared to the
separate production of heat and power. Applied Thermal Engineering
18, 935–946.
McDowall, W., Eames, M., 2006. Forecasts, scenarios, visions, backcasts
and roadmaps to the hydrogen economy: a review of the hydrogen
futures literature. Energy Policy 34, 1236–1250.
Resource Dynamics Corporation, 2003. Cooling, heating, and power for
industry: a market assessment. Final Report prepared for US
Department of Energy and Oak Ridge National Laboratory, Vienna,
VA. Available at /www.cogeneration.netS.
Strachan, N., Dowlatabadi, H., 2002. Distributed generation and
distributed utilities. Energy Policy 30, 649–661.
Strachan, N., Farrell, A., 2006. Emission from distributed vs. centralized
generation: the importance of system performance. Energy Policy 34,
2677–2689.
US Department Of Energy (DOE), 2004. Review of thermally activated
technologies. Distributed energy program report. Available at /http://
www.eere.energy.gov/de/publications.htmlS.
Voorspools, K.R., D’haeseleer, W.D., 2003. The impact of the imple-
mentation of cogeneration in a given energetic context. IEEE
Transactions on Energy Conversion 18, 135–141.
Willis, H.L., Scott, W.G., 2000. Distributed Power Generation: Planning
and Evaluation. Dekker, New York.
Wulfinghoff, D.R., 1999. Energy Efficiency Manual. Energy Institute
Press, Wheaton, MD, USA.
