Energy Simulation of a Heat Pump-driven Liquid Desiccant System Using Dynamic Analysis

Transcription

1 Energy Simulation of a Heat Pump-driven Liquid Desiccant System Usg Dynamic Analysis Jang-Hoon Sh a, Sung-Joon Lee a, Won-Jun Kim a, Jae-Weon Jeong a, * a Department of Architectural Engeerg, College of Engeerg, Hanyang University, Seoul, 04763, Republic of Korea Abstract A heat pump driven liquid desiccant (HPLD) system, a highly energy efficient system terms of aste heat recovery utilizg the coolg capacity of an evaporator and the heatg capacity of a condenser. In this study, e assess and compare the energy savg potential of an HPLD system ith a conventional liquid desiccant (LD) system. A packed-bed toer type LD system ith an aqueous ution of lithium chloride (LiCl) and a ater-toater heat pump ith a refrigerant of R134a ere used this study. he performance of the LD system as terpreted an engeerg equation ver (EES) and the heat pump as simulated through RNSYS 17, a dynamic analysis program. When the total energy consumption of the proposed HPLD system and the conventional LD system are analyzed and compared, the HPLD system shoed better performance than the conventional LD system both aspects of COP and energy savg Stichtg HPC Selection and/or peer-revie under responsibility of the organizers of the 12th IEA Heat Pump Conference Keyords: Underground spaces; liquid desiccant and evaporative coolg-assisted 100% door air system (LD-IDECOAS); ground source heat pump (GSHP) 1. Nomenclature emperature [ C] Humidity ratio [kg/kg] Q hermal load [W] ṁ Mass flo rate [kg/s] P Vapor pressure [kpa] C Concentration [-] h Enthalpy [kj/kg] h fg a0 a2, b0 b2, c0 c2 Heat of vaporization of ater [=2257 kj/kg] Coefficients of vapor pressure equation * Correspondg author. el.: ; fax: address: jjarc@hanyang.ac.kr

2 Jang-Hoon Sh et al / 12th IEA Heat Pump Conference (2017) P A1 A5, B1 B5 Coefficients of ater-to-ater heat pump model for coolg mode Greek symbols ε Effectiveness [-] Subscripts p1 - p8 e moi abs reg load source c Designated pots the liquid desiccant system Equilibrium Moisture Absorber Regenerator Desiccant ution Inlet Outlet Load side Source side Coolg mode Abbreviations COP LD LiCl HP HPLD SHE coefficient of performance liquid desiccant lithium chloride Heat pump Heat-pump-driven liquid desiccant sensible heat exchanger 2. Introduction Accordg to the 2014 report by the Department of Energy (DOE), liquid desiccant (LD) air conditiong system applied HVAC technologies are ranked high stage for highly promisg air-conditiong options for the next generation terms of their energy savg potential [1]. he liquid desiccant system, hich removes the moisture content the door air (i.e. process air) has its on unique feature of requirg heatg and coolg sources at the same time for its effective operation and stable system performance. Regardg this feature, researchers have conducted vibrant studies to determe adequate heat supply sources for the crease the system s performance. Among various proposed systems, the tegration of a vapor compression heat pump system ith an LD system is beg actively discussed and vestigated lately. Previous studies of heat-pump-driven liquid desiccant (HPLD) systems have focused primarily on energy consumption, overall system performance, and capacity matchg beteen the required ution coolg load, heatg load, and heat-pump-generated load. Yamaguchi et al. [2] conducted a performance evaluation and discussed methods for improvg system efficiency via mathematical calculations. hey concluded that the coefficient of performance (COP) could be creased by enhancg the isentropic efficiency of the compressor and ution heat exchanger. Bergero and Chiari [3] evaluated the system performance of HPLD through simulation. In their study, they suggest that the HPLD system is driven ith a hygroscopic ution and a hydrophobic 2

3 Jang-Hoon Sh et al / 12th IEA Heat Pump Conference (2017) P membrane. he simulation results dicate that the proposed hybrid system can lead to energy savgs. Zhang et al. [4] focused on methods for effectively removg the heat left over after regeneratg ution. o different methods for exhaustg the extra heat are suggested hich are utilizg either an air-cooled assistant condenser or a ater-cooled assistant condenser, and the results are compared from the perspective of COP. hey found that systems ith air-cooled condenser and ater-cooled condenser exhibit better performance compared to a basic HPLD system ith no assistant condenser ith COP values approximately 18% and 35% higher, respectively. Niu et al. [5] suggested methods for matchg the capacity of four major heat and mass transfer components the HPLD, hich are absorber, regenerator, evaporator, and condenser. he results dicate that the ution flo rate the condenser, revolution of the compressor, and air flo rate the air-cooled condenser should be simultaneously controlled for capacity matchg the HPLD system. In this study, the energy consumption beteen conventional liquid desiccant system and proposed HPLD system have been analyzed and compared quantitatively. he suggested HPLD system this paper is comprised of a counter-flo packed-bed type liquid desiccant system and a vapor compression heat pump. Lithium chloride (LiCl) ution is adopted as the orkg ution the liquid desiccant system and R-134a is chosen as the refrigerant of the vapor compression heat pump. Also the evaporator and the condenser the heat pump is connected ith coolg coil and heatg coil the liquid desiccant system respectively right after sensible heat exchanger to treat the required ution coolg and heatg load. he simulation for performance and required load of the liquid desiccant system is terpreted usg the commercial engeerg equation ver (EES). Based on these results, a ater-to-ater heat pump model the RNSYS17 is used to simulate the proposed system. he thermal properties of the air and LiCl ution embedded the EES program are utilized for the simulation. 3. Proposed system overvie 3.1. Liquid Desiccant (LD) system A liquid desiccant (LD) system is used to treat the latent load of the door air before it is transferred doors. A typical LD system is composed of an absorber, a regenerator, a sensible heat exchanger, a heatg coil, and a coolg coil, as shon Figure 1. In the absorber, a concentrated ution (i.e., strong ution) dehumidifies the process air that passes through the absorber. he diluted ution (i.e., eak ution) leavg the absorber enters the regenerator to be regenerated to strong ution by humidifyg the regeneration air that passes through the regenerator. he to processes of dehumidification and regeneration are repeated the LD system, and the drivg force of moisture transfer the system is the difference vapor pressure beteen the desiccant ution and the air passg through the ution. Namely, the direction of vapor pressure difference is the determant for hich process beteen dehumidification and regeneration ill take place [6]. Solution temperature is one of the ma key factors s for effective operation of the liquid desiccant system (or stable system performance), makg thermal treatment of the ution dispensable. Generally, desiccant ution should be heated to C for the regeneration process, hereas it should be cooled to C for the dehumidification process [7-8]. o meet the required ution temperature, the utions leavg the absorber and regenerator are firstly met the sensible heat exchanger terms of recoverg the heat. Solutions leavg the sensible heat exchanger are heated and cooled the coolg and the heatg coils, hich a gas boiler and chiller are generally used to treat the required loads the coils for a conventional liquid desiccant system. Dehumidification and regeneration units can be sorted to to types based on the heat extraction process: adiabatic and ternally cooled type. In this study, a packed-bed toer type is used; hich is the most commercialized among adiabatic types [9]. 3

4 Jang-Hoon Sh et al / 12th IEA Heat Pump Conference (2017) P Heat pump (HP) system Fig. 1. Schematic diagram of the liquid desiccant (LD) system A heat pump, a vapor compression refrigeration system, utilizes the reverse flo of the basic existg heat transfer la (i.e., the second la of thermodynamics) hich is that heat cannot be transferred from a cooler to a hotter body spontaneously [10]. In other ords, the heat pump transports the heat from a loer temperature heat source to a higher temperature heat sk. he system is comprised of an evaporator, a compressor, a condenser, and an expansion valve. In the evaporator, the refrigerant flog the heat pump absorbs the heat as it changes its phase from liquid to gas (i.e., evaporation), hereas the compressor the reverse phenomenon (i.e., condensation) takes place. Additionally, the refrigerant is pressurized a compressor to a higher temperature and high pressure state order to crease the energy level to a higher one. o drive this thermodynamic refrigeration cycle a heat pump, poer put a compressor from an external energy source is required [11] Heat pump driven liquid desiccant (HPLD) system Figure 2 shos a schematic of the proposed HPLD system, hich is comprised of a conventional liquid desiccant system and a vapor compression heat pump. he condenser and evaporator of the heat pump are directly tegrated ith the heatg and coolg coils, respectively, to utilize the heatg and coolg load generated the heat pump. Namely, the heat release the condenser and the coolg effect from the heat absorption the evaporator are used simultaneously to treat the ution heatg load the regenerator and ution coolg load the absorber, respectively. In terms of energy consumption, the energy requirement an air-cooled chiller and a boiler a conventional liquid desiccant system can be replaced ith the poer put the compressor of a heat pump an HPLD system. Furthermore, the feature of usg aste heat the compression heat pump can enhance the efficiency of the HPLD system. 4

5 Jang-Hoon Sh et al / 12th IEA Heat Pump Conference (2017) P Energy Simulation Overvie Fig. 2. Schematic diagram of the heat pump driven liquid desiccant (HPLD) system; (a) LD, (b) HP, (c) HPLD ith auxiliary devices In the energy simulation, the energy consumption needed to treat the required load the coolg and heatg coils to match the set temperature of the ution enterg the absorber (25 C) and the regenerator (60 C) is compared beteen the conventional system and the proposed HPLD system. he basic assumptions for the simulation are as follos: the air flo rate the absorber and the regenerator are set at a constant 2000 and 4000 cubic meters per hour, respectively. Additionally, liquid to gas ratio is set at 1 the absorber, hich means that the ution flo rate is designed to be 0.67 kg/s from the constant air flo rate. Moreover, the ution let temperature the absorber and the regenerator are mataed at 25 C and 60 C, respectively, and the concentration of the ution enterg the absorber is set at 38%. Along ith the above conditions, the ternational eather for energy calculations (IWEC) eather data of Seoul, Sh Korea, for 744 hours from August 1st to August 31st from ASHRAE [12], is applied to predict the performance of the liquid desiccant system and evaluate the operatg energy comparison beteen the conventional and suggested systems Liquid desiccant (LD) system able 1. Coefficients of vapor pressure equation Coefficients of vapor pressure equation Dehumidification process Regeneration process a 0 a 1 a b 0 b 1 b c 0 c 1 c a 0 a 1 a b 0 b 1 b c 0 c 1 c

6 Jang-Hoon Sh et al / 12th IEA Heat Pump Conference (2017) P he temperature of the utions leavg the absorber and the regenerator must be found order to compute the required coolg and heatg loads [13]. In the calculation process, three ution conditions: the ution mass flo rate, ution concentration, and ution temperature for every pot from 1 to 8 as marked the Figure 2 and three air conditions: the air flo rate, air temperature, and air humidity ratio before enterg and after leavg the absorber and regenerator are needed. he entire process of the condition changes the ution and the air are based on the mass balance and energy balance equations. o terpret the effectiveness of the absorber, the model created by Park [14] is adopted among the several existg models. he effectiveness of the regenerator is defed ith the humidity ratio and the temperature of the regeneration and the temperature of the ution the regenerator. Furthermore, the efficiency of the sensible heat exchanger this simulation is assumed to be 80 % Absorber he dehumidification process occurs as the process air is brought to contact ith the strong ution of 38 % and 25 C side the packed-bed type absorber. Several models have been established current literature to terpret the performance of the absorber and predict the dehumidification effectiveness (ε abs ). In this research, a parameter-estimation-based lear regression model developed by Park is selected. he empirical model is a function of six operatg parameters: air mass flo rate (m OA,state1), ambient air temperature ( OA,state1 ), ambient air humidity ratio ( OA,state1 ), ution mass flo rate (m p2,), ution let temperature ( p2, ), and ution concentration (C p2, ). Additionally, the effectiveness of the absorber can be defed ith the temperature change or humidity ratio change of the process air enterg and leavg the absorber (Eq. (1), Eq. (2)). Once the empirical model is combed ith the defitions, one can estimate the air conditions leavg the absorber. he equilibrium humidity ratio of the ution enterg the absorber ( e ) can be obtaed ith the function of the vapor pressure (p s ) at the saturation condition of the desiccant ution the absorber (Eq. (3)), hich a second-order polynomial model suggested by Fumo and Gosami [15] as used (Eq. (4)). he constants needed the model are presented table 1. he equilibrium temperature of the ution the regenerator is assumed to be the same as the temperature of the ution enterg the regenerator, hich is 25 C. p abs abs eq s eq eq ps p s ( a0 a1 L a2 L ) ( b0 b1 L b2 L ) CL ( c0 c1 L c2 L ) CL he mass and energy balance around the absorber can be expressed via Eq. (5) to Eq. (10), through hich ution conditions leavg the absorber can be obtaed. he dehumidification rate (m p3,) the absorber can be acquired usg the moisture bass balance equation (Eq. (5)). he ution conditions (mass flo rate and concentration) before enterg the absorber are the same p1, p2, and p8 (Eq. (6), Eq. (7)), regardless of the presence of a sensible heat exchanger or a coolg coil. Also, those of the ution leavg the absorber can be obtaed usg Eq. (8) and Eq. (9). (1) (2) (3) (4) m p 3, moi m ( air, air, m p1, m p2, m p8, C p1, C p2, C p8, m p4, m p1, m p3, moi m p1, C p1, m p4, C p4, ) (5) (6) (7) (8) (9) 6

7 Jang-Hoon Sh et al / 12th IEA Heat Pump Conference (2017) P he ution temperature leavg the absorber can be acquired through the LiCl properties embedded the EES ith the ution concentration from Eq. (9), and the ution enthalpy, hich can be determed from the energy balance equation (Eq. (10)). m p4, hp4, m p2, hp2, m p3, moi h fgp3, moi (10) Regenerator he concentration of the ution leavg the absorber becomes loer compared to that of the ution the enterg state hich has to be regenerated the regenerator for the performance of the liquid desiccant system. For the balance the overall system, the regeneration rate the regenerator is assumed to be the same as the dehumidification rate the absorber, through hich the humidity ratio of the air leavg the regenerator can be obtaed (Eq. (11) and Eq. (12)). m p3, moi m p7, moi p 7, moi m ( air, air, m ) (11) (12) he regeneration effectiveness (ε reg ) can be defed ith the change humidity ratio or the change temperature of the regeneration air (Eq. (13), Eq. (14)). he regeneration rate from Eq. (12) can be used Eq. (13) to calculate the regeneration effectiveness (ε reg ), hich can be used Eq. (14) to obta the temperature of the regeneration air leavg the regenerator ( SA,state2 ). he equilibrium ratio the regenerator ( OA,state1 ) can be determed usg Eq. (3) and Eq. (4). he equilibrium temperature of the ution the regenerator is assumed to be the same as the temperature of the ution enterg the regenerator, hich is 60 C. reg reg e e he ution mass flo rate and the ution concentration before enterg the regenerator are the same p4, p5, and p6 (Eq. (15) and Eq. (16)), and those of the ution leavg the regenerator can be expressed as Eq. (17) and Eq. (18). (13) (14) m p4, m p5, m p6, C p4, C p5, C p6, (15) (16) m p8, m p6, m p7, moi m p6, C p6, m p8, C p8, he let ution temperature the regenerator can be estimated ith the ution concentration from Eq. (18) and the ution enthalpy, hich can be determed from the energy balance equation (Eq. (19)). m p6, hp6, m p8, hp8, m p7, moi h fgp7, moi (17) (18) (19) Solution coolg and heatg load he ution coolg and heatg load the coolg and heatg coil can be estimated ith the temperature of the utions leavg the sensible heat exchanger. he temperature of the utions leavg the absorber and the 7

8 Jang-Hoon Sh et al / 12th IEA Heat Pump Conference (2017) P regenerator ( p4,, p8, ) from Eq. (1) to Eq. (19) can be used to estimate the coolg and heatg loads the coolg and heatg coils. he required loads dicate the poer needed to meet the ution let set-pot temperature the absorber and the regenerator ( p2, = 25 C, p6, = 60 C). he utions leavg the absorber and the regenerator first exchange the sensible heat terms of heat recovery before enterg the coolg and heatg coils. In this simulation, the efficiency of the sensible heat exchanger (ε SHE ) is 80 %. he temperature of the utions leavg the heat exchanger ( p4,, p8, ) can be determed from Eq. (20) and Eq. (21). he required load the coolg and heatg coils can be estimated usg Eq. (22) and Eq. (23). m reg p8, C e 4.2. Conventional liquid desiccant system ith air-cooled chiller and gas boiler In the conventional liquid desiccant system, among various types of heatg and coolg sources, air cooled chillers and boilers are commonly used to treat the required coolg and heatg loads. In this paper, a RANE chiller CGAM35 model and gas boiler model furnished EnergyPlus [16] ere adopted to compute the energy consumptions to treat the required loads Heat pump driven liquid desiccant (HPLD) system In the HPLD system, both the coolg capacity from the evaporator and the heatg capacity from the condenser a heat pump are used, as opposed to usg separate devices for treatg the required loads the conventional LD system. he heat pump performance as simulated ith a parameter estimation based ater-to-ater curvefit model implanted RNSYS 17, a transient analysis program. In the simulation, type927, a sgle stage aterto-ater heat pump model as used. he model treats the ma liquid stream le (the ution that flos through the evaporator this stimulation) by rejectg energy to (coolg mode) or absorbg energy from (heatg mode) a second liquid stream (the ution that flos through the condenser). It is operated an on/off control ith temperature level changes much like an actual heat pump does; once the user defed control signal put is given to be ON either coolg or heatg mode, the system is operated at its capacity level until there s a change the signal value. he model can be used to different modes: coolg and heatg mode. In each mode, the capacity the source and load sides and the poer put the compressor can be computed ith the follog four put parameters: source side flo rate, source side let temperature, load side flo rate, and load side let temperature. In this research, the heat pump as sized for a rated coolg capacity of 16 kw and as operated the coolg mode to focus on the demanded ution coolg load to meet the temperature of the ution enterg the absorber, here the purpose of the LD system occurs. herefore, this simulation, the load side and source side capacities are the load generated the evaporator and condenser, respectively. Also, the estimated poer put dicates the energy needed to operate the heat pump system. 5. Energy Simulation Results ( p8, p 1, ) m p4, C p, ( p5, p4, pp8, p84 Qcoolg m p1, C p, ( 1 p 1, p p2, Qheatg m p5, C p, ( 5 p6, p p5, ) ) Air and ution condition leavg the absorber are determed by the heat and mass transfer beteen the air and the ution enterg the absorber, and the heat generated durg the dehumidification process (i.e., exothermic reaction) hich are fluenced by the properties of the air and the ution. In this simulation, due to the relatively loer temperature of the ution enterg the absorber (25 C) compared to the temperature of the process air enterg the absorber, the ution temperature leavg the absorber rises from the heat exchange ith the process air and the absorption of the generated heat the absorber. Also, the ution flo rate, hich is mataed at a ) (20) (21) (22) (23) 8

9 Jang-Hoon Sh et al / 12th IEA Heat Pump Conference (2017) P constant 0.67 kg/s hen enterg the absorber, creased by the amount of the dehumidification rate due to the moisture absorption from the process air Energy consumption the conventional liquid desiccant system Figure 3 shos the required coolg and heatg load and the energy consumption of the air-cooled chiller and gas boiler the conventional LD system accordance ith the required ution coolg and heatg load. One can see that the heatg load is generally higher than the coolg load. Fig. 3. Required load and energy consumption the conventional LD system 5.2. Required heatg and coolg load for LD operation and heat pump generated capacity Figure 4 shos the required coolg and heatg load and heat pump generated load. One can see that there s quite a difference beteen the demanded load and the generated load hich as caused due to the on/off control the heat pump. In the proposed system, a control strategy that treats most of the required load ith a primary system (i.e., the heat pump) ith full speed control and the rest of the load ith supplementary systems (i.e., auxiliary devices) as used: the heat pump as operated ith an on/off control at full speed and the sufficient coolg and heatg load ere treated ith an auxiliary air-cooled chiller and an auxiliary gas boiler, respectively. Fig. 4. Required load and heat pump generated load the HPLD system 5.3. otal energy consumption comparison beteen conventional LD system and HPLD system Figure 5 shos a comparison of the energy consumption beteen the conventional LD system and the proposed HPLD system hen treatg the required ution coolg and heatg loads. When comparg the total energy consumption, the primary energy conversion factors, for hich 2.75 for electricity and 1.1 for gas ere used [17]. In the conventional system, electricity and gas ere used to operate the air-cooled chiller and the gas boiler, respectively. Whereas the HPLD system, electricity and gas ere used the heat pump the auxiliary chiller 9

10 Energy consumption [kwh] Jang-Hoon Sh et al / 12th IEA Heat Pump Conference (2017) P and auxiliary gas boiler, respectively. he HPLD system shoed 34 % less electricity consumption than the conventional LD system, of hich 68 % as used for the heat pump and 32 % as used for the auxiliary aircooled chiller. Furthermore, the HPLD system shoed 68 % reduction gas energy consumption hen compared to the conventional system. Due to the on/off control the heat pump, capacity matchg could not be made beteen the heat pump generated load and the required load, makg the need for auxiliary device operation evitable. As a result, the HPLD system shoed 58 % reduction the total energy consumption hen compared to the conventional LD system. he significant energy savg is due to the reduction energy from the gas boiler, hich takes the majority of the total energy consumption the liquid desiccant system. In the HPLD system only the sufficient heatg load, or that left after the ution is treated ith the heatg capacity from the condenser the heat pump, is processed ith the auxiliary gas boiler. Whereas the conventional LD system the total required heatg load is treated ith the gas boiler Conventional LD HPLD ith auxiliary devices Electricity 9885; Gas 26665;4 8639;7 otal primary energy 36550; ;7 Electricity Gas otal primary energy Fig. 5. Energy consumption comparison beteen conventional LD and HPLD systems 6. Energy Simulation Results In this paper, energy consumption to treat the required ution coolg and heatg loads a conventional LD system usg a chiller and a gas boiler and an HPLD system usg a heat pump and auxiliary devices ere compared. he proposed HPLD system shoed significant energy savgs over the conventional LD system; ith a 34 % reduction electric energy, a 68 % reduction gas energy, and a 68 % reduction the total primary energy. In this simulation, sce the heat pump as operated usg an on/off control, the heat pump generated capacity ere not able to match the required loads causg the needs for the use of auxiliary devices. In terms of usg auxiliary devices, further researches on capacity matchg beteen the heat pump generated loads and the required loads ould be needed. Once capacity matchg is achieved, the proposed HPLD system ill be beneficial not only the aspect of energy savg but also terms of system compactness. Acknoledgements his ork as supported by a National Research Foundation (NRF) of Korea (No.2015R1A2A1A ), the Korea Agency for Infrastructure echnology Advancement (KAIA) grant (16CAP-C ), and the Korea Institute of Energy echnology Evaluation and Planng (KEEP) (No ). 10

11 Jang-Hoon Sh et al / 12th IEA Heat Pump Conference (2017) P References [1] Kishorkumar L Makana and Ketankumar N Gohel. Revie: Potential applications of liquid desiccant technology dehumidification and coolg process air. Engeerg Science and Futurisic echonology. Vol. 1, no.1, 2015 [2] Seiichi Yamaguchi, Jongsoo Jeong, Kiyoshi Saito, Hikoo Miyauchi, Masatoshi Harada.Hybrid liquid desiccant airconditiong system: Experiments and simulations Appl. herm. Eng., vol.31, 2011, p [3] Stefano Bergero, Anna Chiari. Performance analysis of a liquid desiccant and membrane contactor hybrid air-conditiong system. Energy Buil., vol. 42, no. 11, 2010, p [4] ao Zhang, Xiaohua Liu, Yi Jiang. Performance optimization of heat pump driven liquid desiccant dehumidification system. Energy Build., Vol. 52, 2012, p [5] Xiaofeng Niu, Fu Xiao, Zhenjun Ma. Investigation on capacity matchg liquid desiccant and heat pump hybrid air-conditiong systems. Int. J. Refrig. Vol. 35, no. 1, 2012, p [6] Ahmed A.M., Abdalla K.N., Experimental study of heat and mass transfer of Solar poered liquid desiccant regeneration system. Annual conference of postgraduate studies and scientific research, Khartoum: Friendship Hall; 2012, p [7] Jag S., raipathi S., Das R.S., Experimental performance of a liquid desiccant dehumidification system under tropical climates. Energy Convers. Manag., vol. 52, no. 6, 2011, p [8] Jag S., raipathi S., Das R.S., Experimental performance of a liquid desiccant dehumidification system under tropical climates. Energy Convers. Manag., vol. 52, no. 6, 2011, p [9] Loenste A., Slayzak S., Ryan J., Peasaran A., Advanced commercial liquid desiccant technology development sutdy. Natl. Rene. Energy Lab., bol. NREL/P-55, 1998 [10] Alasdiar C. M., Paul., Assessg the performance of a reservoir-based ater source heat pump, Department of Mechanical and Aerospace Engeerg, 2013 [11] Natural Resource Cananda s Office of Energy Efficiency, Heatg and coolg ith a heat pump [12] ASHRAE International eather for energy calculations, Version 2.0 User s manual. Atlanta: American Society of Heatg, Refrigeratg and Air-Conditiong Engeers, Inc. [13] Kim M.H., Park J.S., Jeong J.W., Energy savg potential of liquid desiccant evaporative-coolg-assisted 100% door air system. Energy, vol. 59, 2013, p [14] Park J.Y., Yoon D.S., and Jeong J.W. Empirical model for predictg the dehumidification effectiveness of a liquid desiccant system. Energy and Buildgs, vol. 126, 2016, p [15] Fumo N., Goasami D.Y. Study of an aqueous lithium chloride desiccant system: Air dehumidification and desiccant regeneration. Sol. Energy, vol. 72, no , p [16] US Deparment of Energy. EnergyPlus Engeerg Reference: he Reference to EnergyPlus Calculation., US Dep. Energy, 2010 [17] Park D.J., Yu K.H., Yoon Y.S., Kim K.H., Kim s.s., Analysis of a buildg energy efficiency certification system Korea., Sustaability, vol. 7, no.12, 2015, p