1、270 2016 Her Majesty the Queen in Right of Canada, as represented by the Minister of Natural ResourcesABSTRACTThis paper discusses a technical comparison of a direct-expansion ground-source heat pump (DX-GSHP) and an air-sourceheatpump(ASHP).Tolowerthegroundheatexchangersize for cost reduction purpo
2、ses, the system performance of ahybrid system is also evaluated using a supplementary airevaporator combined with the DX-GSHP. Detailed screeningmodels previously developed for ASHPs and DX-GHSPs arefirst used to compare the seasonal performance of these twooptions for a residential building in the
3、cold-climate city ofMontral, Canada. Then, the model is adopted for perfor-mance evaluation of the hybrid system. Additionally, differentparameters including borehole total length and heat pumpcapacity are varied to evaluate their impact on the seasonalsystemperformance.Theresultsshowthatbyadequates
4、izing,totalandpeakelectricityconsumptionoftheDX-GSHPsystemcanbereducedby50%and40%,respectively,comparedtoanundersizedDX-GSHPsystem.However,systemenergysavingsfrom using a hybrid ground-source heat pump (HGSHP) aremarginal compared to a DX-GSHP (5.5% for a low-capacitysystem), and these savings are h
5、appening during the shouldermonths.Suchresultshighlighttheimportanceoffurtherinves-tigations in the area of DX-GSHPs to reduce the boreholeinstallation cost and increase its performance.INTRODUCTIONGenerally, heat pumps can be divided into two main cate-gories according to the heat sink/source mediu
6、m: air-sourceheat pumps (ASHPs) and ground-source heat pumps(GSHPs). ASHP systems use ambient air as a heat source/sinktoprovidespaceheatingandcooling.TheefficiencyofASHPsvaries significantly with ambient temperature levels, and theircoefficient of performance (COP) falls drastically at lowambient t
7、emperatures. This is a major problem for ASHPapplications in cold climates. On the other hand, ASHPs arecheaper and easier to install when compared to GSHPs. Tosome extent, the new generation of ASHPs has increasedperformances in cold climates (Bertsh et al. 2005). However,cold-climate ASHPs still r
8、emain an ongoing challenge andmore improvements are still needed at low temperatures(Hakkaki-Fard et al. 2014).Alternatively, GSHP systems use the ground as a heatsource/sink medium to overcome the poor cold-climateperformance of ASHPs. However, high installation costshinder their widespread adoptio
9、n. Among the various GSHPsystems available, the vertical-borehole GSHP system hasattracted the greatest interest (Yang et al. 2010). There are twomain categories of vertical-borehole GSHPs: the direct-expansion ground-source heat pump (DX-GSHP) and thesecondary-loop ground-source heat pump (SL-GSHP)
10、. Of thetwo systems, DX-GSHPs are the most efficient for spaceconditioning and domestic hot-water production (Guo et al.2012). However, DX-GSHPs entail more system design andenvironmental complications, including compressor starting,oil return, potential ground pollution, and high refrigerantcharge.
11、 Moreover, general design guidelines for DX-GSHPsystems are not well documented.Modeling and performance analyses of various types ofASHP systems are very well addressed in the literature(Hakkaki-Fard et al. 2015a). Performance and operation char-acteristics of SL-GSHP systems are also very well eva
12、luatedtheoretically in different climates (Michopoulos et al. 2013).However, equivalent information is rather scarce for DX-GSHPs, because of the complexity of two-phase flow model-Technical Assessment of Ground-Source,Air-Source, and Hybrid Heat Pumps forSingle-Family Buildings in Cold ClimatesParh
13、am Eslami-Nejad, PhD Ali Hakkaki-Fard, PhDZine Aidoun, PhD Mohamed Ouzzane, PhDParham Eslami-Nejad, Zine Aidoun, and Mohamed Ouzzane are research scientists at CanmetENERGY-Varennes, Natural ResourcesCanada, Varennes, Canada. Ali Hakkaki-Fard is an assistant professor at Sharif University of Technol
14、ogy, Tehran, Iran.ST-16-027Published in ASHRAE Transactions, Volume 122, Part 2 ASHRAE Transactions 271linginsidetheboreholes.MostexistingstudiesonDX-GSHPsexperimentally evaluated the whole-system performance(Wang et al. 2009; Fannou et al. 2014), while other studieswere performed to establish some
15、guides and improvementsforthedesign(Wangetal.2013).Veryfewworkshavelookedat the whole heat pump cycle (Austin and Sumathy 2011).Among few studies on modeling the DX-GSHP, Beauchampet al. (2013) developed a numerical model to analyze theperformance of a ground heat exchanger. Fannou et al. (2015)perf
16、ormed a short-term (500 s) comparative study on a DXgroundevaporatorusingR-410A,R-407C,andR-22asrefrig-erants.AstudybyRousseauetal.(2015)recentlypresentedanexperimentally validated model for a DX vertical groundevaporator using R-22 as the working fluid. Eslami-Nejad etal. (2014) also performed a nu
17、merical modeling of a carbondioxide (CO2) filled vertical geothermal borehole underforced circulation. Another study by Eslami-Nejad et al.(2015) focused on a quasitransient modeling of the single-stage CO2transcritical GSHP with a gas bypass. A model wasused to simulate a heat pump under real opera
18、ting conditionsfor annual simulations focusing on space heating and hot-water requirements.Despite many debates on economic competitiveness andenergy efficiency of ASHPs versus GSHPs, no systematiccomparison has been performed and made available in theliterature. Among very few studies, Urchuegua et
19、 al. (2008)evaluated the annual difference between energy efficiency ofa conventional ASHP and a SL-GSHP operating in coolingand heating modes. They showed that for the worst-casescenario, the GSHP consumed 26% and 19% less primaryenergy than the ASHP in heating and cooling modes, respec-tively.Hakk
20、aki-Fardetal.(2015b)performedalife-cyclecostanalysis to account for the difference between initial and 10-year operating costs of DX-GSHP and ASHP systems basedon the current prices in the Qubec province of Canada. Theycalculated that the relative payback period of the DX-GSHPcompared to that of the
21、 ASHP is more than 15 years withcurrentboreholeinstallationprices.Theyconcludedthatiftheborehole installation price is reduced by 50%, the paybackperiod would be reduced to just a few years.However, this study conducted a technical comparisonbetweenacommonlyusedASHP,aDX-GSHP,andaHGSHPin which the ai
22、r evaporator is added to the DX-GSHP. Basedon the definition in the literature, HGSHP systems are GSHPsequipped with any type of supplemental heat rejection orextraction system. Different HGSHPs have recently beenstudied by a number of researchers (Man et al. 2008; Hackelet al. 2008). To the best of
23、 the authors knowledge, no studyhas ever been performed to compare the ASHP, DX-GSHP,and the combination of the two (HGSHP). A detailed numer-ical modeling of these systems is performed to evaluate thesystem performance for satisfying identical space-heatingloads of a single-family detached home loc
24、ated in the cold-climate Canadian city of Montral. The three systems areentirely identical, except that the air-to-refrigerant evaporatoroftheASHPisreplacedbyboreholesintheDX-GSHPandtheHGSHP system is connected to both the evaporator and theboreholes.HEAT PUMP CONFIGURATION AND MODELINGFigure1showss
25、chematicviewsofatypicalASHP,aDX-GSHP,andaHGSHPsystem.TheASHPconsistsoffourmainsystem components, including a compressor, condenser,expansion valve, and evaporator. The DX-GSHP consists ofthesamecomponents,exceptthattheevaporatorisreplacedbyvertical ground heat exchangers (boreholes). In the ASHP, th
26、erefrigerant inside the evaporator extracts the heat from theenvironment.IntheDX-GSHPsystem,therefrigerantisflow-ing directly down in the borehole, changing direction at thebottom (U connection), and coming back up. The refrigerantFigure 1 Schematics of a typical (a) ASHP system, (b) DX-GSHP system,
27、 and (c) HGSHP system.Published in ASHRAE Transactions, Volume 122, Part 2 272 ASHRAE Transactionsevaporates by extracting heat from the ground. The HGSHPsystem is equipped with both the evaporator and the borehole,but only one will work at a time. The HGSHP system will takeadvantage of the warm day
28、s in shoulder/transition seasons,when the outside ambient temperature becomes warmer thanthe borehole wall temperature. Not only will this increase thetemporary COP of the system, but also the ground will alsorecover during this period, which will increase the seasonalperformance of the system. The
29、duration of using air coils inHGSHPsystemsdependsonbothboreholesizeandheatpumpcapacity.The performance of the heat pumps is evaluated usingcombined steady-state numerical models for the boreholes,the air condenser, and evaporator; a transient analytical modelof the ground; as well as other steady-st
30、ate heat transfer andthermodynamic models for the expansion valves and thecompressor. In this work, only the main features of the modelpreviouslydevelopedbyHakkaki-Fardetal.(2014,2015a)areoutlined. This model is also coupled to the ground heatexchangermodelpreviouslydevelopedbyEslami-Nejadetal.(2014
31、, 2015) for DX-GSHP simulations. For validation of theground-heat exchanger and the ASHP models, readers arereferred to the aforementioned publications. However, somegeneral assumptions and simplifications to develop the theo-retical models are as follows:The heat pump is used for heating only (for
32、the heating-dominated climate of Montral).All system components are operating under steady-stateconditions.Heat transfer in the ground is transient, one-dimen-sional, and perpendicular to the boreholes.Grout and ground material are homogeneous.Pressure drop in connecting tubes is neglected.Thermal i
33、nteraction between boreholes is neglected.Flow for the refrigerant inside tubes is one-dimensional.Heat loss to the surroundings is assumed to be negligi-ble.Heat required to defrost the ASHP evaporator isneglected.The expansion valve undergoes an isenthalpic process(HCND exit= HEVAP inlet).The comp
34、ression process is assumed to be polytropicwith constant efficiency.Saturated liquid and saturated vapor conditions areassumed at the condenser outlet and compressor inlet,respectively.A scroll compressor is considered in simulations. Therefrigerant mass flow rate is calculated by the fol-lowing rel
35、ation:(1)The volumetric efficiency vis extracted from the com-mercial compressor performance curves.It is assumed that the geothermal borehole is made of along stainless steel U-tube (two pipes) embedded in asolid material (grout). A schematic cross section of asingle U-tube borehole is presented in
36、 Figure 2. Formodeling, the borehole is divided into a number of con-trol volume elements in the vertical direction. Heat flowper unit length through each pipe, down and up, is calcu-lated based on the approach proposed by Hellstrm(1991) and elaborated by Zeng et al. (2003). Thisapproach has been us
37、ed previously to evaluate the per-formance of conventional secondary-loop geothermalboreholes (Zeng et al. 2003). However, Eslami-Nejad etal. (2014) adapted it to solve the two-phase fluid circula-tion problem in geothermal applications. To take intoaccount the latent heat of evaporation, they repla
38、ced thefirst-order derivatives of temperature by the first-orderderivatives of the fluid enthalpy as a function of bore-hole length.The ground model is based on the classical infinitecylindrical source (ICS) model with a variable heatpulse. The solution is usually found by superposing theresponse of
39、 the heat pulses. This form of solution isbasically the discrete convolution of the variable heatload with the heat pulse response. The discrete convolu-tion calculation is known to be very time consuming.Lamarche and Beauchamp (2007) proposed a nonhis-tory scheme, valid for the ICS solution, which
40、can solvethe problem rapidly. It assumes that the heat flux isknown at the borehole. Here the fluid temperature at theinlet or outlet of the borehole is imposed, so the initialscheme is modified.The undisturbed ground temperature (Tg) in Montral is10C (50F), calculated using an experimentally vali-d
41、ated correlation (Ouzzane et al. 2015). The groundthermal properties are obtained from a thermal responsetest report. The ground and the grout thermal propertiesand borehole characteristics are listed in Table 1.Cross-flow heat exchange is assumed in both the con-denser and air-source evaporator (Ha
42、kkaki-Fard et al.2015a). The air-source heat exchangers are assumed tobe finned-tube type. The overall heat transfer coefficientUA is obtained by the following relation:mrmrvVcpsuc=Figure 2 Schematic cross section of borehole.Published in ASHRAE Transactions, Volume 122, Part 2 ASHRAE Transactions 2
43、73(2)To find the overall heat transfer coefficient using Equa-tion 2, the following correlations and coefficients arerequired: air-side heat transfer coefficient, one-phaserefrigerant heat transfer coefficients, and condensationand boiling correlations for refrigerant phase change.These coefficients
44、 are obtained using the correlationsavailable in the literature (Hakkaki-Fard et al. 2014).For two-phase flow calculation, refrigerant pressurechange is linked to temperature change; therefore cou-pled momentum and energy equations have to be solvedsimultaneously. For refrigerant pressure calculatio
45、n,readers are referred to the study by Eslami-Nejad et al.(2014) for two-phase flow calculation in vertical geo-thermal boreholes and Hakkaki-Fard et al. (2014) for aircondensers and evaporators.Each system component is calculated separately. Heatexchangers are divided into a number of control volum
46、eelements. Fundamental conservation equations of mass,momentum, and energy, based on the appropriate cor-relations, are applied to each control volume element.The system of equations obtained is nonlinear and itsparameters are strongly linked. An iterative method istherefore applied to solve the set
47、 of equations. To speedup the convergence of the program, a gradient-basedoptimization method is used to update the iterations.The code is developed in FORTRAN, to which REF-PROP Version 9.1 subroutines (Lemmon et al. 2013) arelinked, to calculate the thermodynamic properties of airand the refrigera
48、nt.SYSTEM SIMULATIONThe objective of this study is to assess the performance ofthe three heat pump options selected with different capacitiesfor satisfying an identical space-heating need of a residentialhouse for which the cooling and hot-water loads are ignored atthis stage. In this case, heat pum
49、p operation supplemented byanelectricheaterasabackupisconsidered.Thesystemenergyconsumption is defined here as the sum of the heat pumpenergy consumption (compressor and fans) plus the energyconsumption of the backup electric heater.Thesystemsimulationwasperformedusingacodedevel-oped in FORTRAN for this purpose. The logic flow chart ofthe
