Testing solar collectors as an energy source for a heat pump

ARTICLE IN PRESS Renewable Energy 33 (2008) 832–838 www.elsevier.com/locate/renene Technical Note Testing solar collectors as an energy source for a heat pump A. Georgiev Department of Mechanics, Technical University of Sofia, branch Plovdiv, P.O. Box 7, 4023 Plovdiv, Bulgaria Available online 2 July 2007 Abstract The article presents the experimental study of a heat pump possessing solar collectors as an energy source. A method to test the combined work of collectors delivering heat to the evaporator of a heat pump was devised. The layout of the test facility is shown and the system construction with the measurement equipment is described. The planning experiment to test the installation was chosen. The _ c and the medium fluid evaporator temperature t̄ev were medium fluid condenser temperature t̄c , the fluid condenser mass flow rate m chosen as experiment factors to determine both objective functions—the coefficient of performance (COP) of the heat pump and the efficiency of the system Zs. The reverberation of both objective functions is shown. r 2007 Elsevier Ltd. All rights reserved. Keywords: Heat pump; Flat plate solar collectors; Experimental design; Study 1. Introduction Freeman [1] has studied some systems for residential space and domestic hot water heating. These are the conventional solar system, the conventional heat pump system, the series system, in which the solar storage is used as the source for the heat pump, the parallel system, in which ambient air is used as the source for the heat pump, and the dual source system, in which the storage or ambient air is used as the source depending on which source yields the lowest work input. The systems for residential space were studied by several authors. O’ Dell et al. [2] presented mathematical models and experimental results of similar systems for residential space in detail. Sakai et al. [3] showed the effects of different system parameters change and delivered some test data. Hatheway et al. [4] carried out detailed economic analysis of different heat pumps for residential space using solar energy. The most comprehensive study was done by General Electric [5] and Bessler [6] et al. Cottingham [7] considered in brief the mentioned systems—he described the system method of work and gave some experimental data. The heat pump systems using the solar energy for residential space and domestic hot water heating were Tel.: +359 32 649785; fax: +359 32 650270. E-mail address: [email protected] 0960-1481/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.renene.2007.05.002 studied, too. MacArthur et al. [8], Chandrashekar et al. [9], Tleimat et al. [10] and Chaturvedy et al. [11] considered this type of systems in detail giving an economic analysis as well. Roeder et al. [12], Speiser et al. [13], Terrell [14] and Ucar et al. [15] presented particular details of the overall installation analysis. The objective of the article is to explain the series heat pump system in detail. A bench consisting of a heat pump ‘‘water–water’’ and flat plate solar collectors was constructed. Methods to investigate the given system were devised [16]. It also discusses the investigation regime of the system, the accuracy of the measured parameters and the processing of the experimental data. A parameter investigation of the mentioned system heat pump ‘‘water– water’’— solar collectors was undertaken. A planning experiment was used to test the system and a description of the main characterizing system parameters, the coefficient of performance (COP) of the heat pump and the system efficiency Zs is shown. An error analysis evaluating the measured parameters was conducted, too. 2. Investigation methods of a system heat pump ‘‘water–water’’-solar collectors The methods are concerned with operational schemes where the additional energy source (AES) is situated after the solar collectors and before the evaporator of the heat ARTICLE IN PRESS A. Georgiev / Renewable Energy 33 (2008) 832–838 A c COP I _ m N p Q_ s t n d Z j area, m2 specific heat capacity, J/kg K coefficient of performance insolation, W/m2 mass flow rate, kg/s electrical power, W pressure, Pa heat flux, W collector tilt from horizontal surface, 1 temperature, 1C velocity, m/s declination, 1 efficiency geographic latitude, 1 Subscripts a AES c col ev g hp i o s w pump. The layout of the test facility heat pump ‘‘water– water’’-solar collectors is presented in Fig. 1 [16]. 2.1. Theory Some parameters are defined: Efficiency Zcol of the solar collector—it is the ratio of the heat flux extracted by the solar collector Q_ col and the solar collector area Acol multiplied by the global solar insolation in the plane of the solar collectors Icol: Q_ col . Acol :I col (1) ambient additional energy source condenser collector evaporator global heat pump inlet outlet system wind circuit Nc and of the pumps in the solar circuit Ncol) Zs ¼ Zcol ¼ time, s; h; d t Nomenclature 833 Q_ col Q_ col þ Q_ AES . þ Q_ AES þ N hp þ N c þ N col (5) 2.2. Investigation procedure All parameters are measured simultaneously every 3 min during a period of 30 min. The measurements are made under quasi-stationary conditions. The tested collector is orientated towards the sun and tilted to the horizon at an angle (j–d). The heat flux Q_ col is defined as a function of the fluid mass flow rate through the solar collectors and the evaporator _ col , the fluid specific heat capacity in the solar collector m ccol and the inlet and outlet fluid collector temperatures tcol,i,tcol,o: _ col ccol ðtcol;o  tcol;i Þ, Q_ col ¼ m (2) COP of the heat pump—the ratio of the condenser heat flux Q_ and the consumed electrical power Nhp: c COP ¼ Q_ c . N hp (3) The condenser heat flux is defined as a function of the fluid _ c , the fluid specific mass flow rate through the condenser m heat capacity in the condenser cc and the inlet and outlet fluid condenser temperatures tc,i, tc,o: _ c cc ðtc;o  tc;i Þ. Q_ c ¼ m (4) Efficiency of the system heat pump–flat plate solar collectors Zs is defined as the ratio of the gained heat flux Q_ col plus the used power from an AES Q_ AES to the full inlet power (Q_ col ,Q_ AES and the electrical powers of the heat pump compressor Nhp, of the pumps in the condenser Fig. 1. Layout of the test facility heat pump ‘‘water–water’’-flat plate solar collectors with AES: 1—heat pump; 2—flat plate solar collectors; 3— additional energy source (AES); 4—heat exchanger; 5–7—expanding vessels; 8, 9—pumps; 10–12—flow meters; 13—sensor of the solarimeter; 14—sensor of the anemometer; 15–21—thermometers; 22–24—out-off valves; 25—solarimeter; 26—anemometer; 27–30—wattmeters. ARTICLE IN PRESS 834 A. Georgiev / Renewable Energy 33 (2008) 832–838 The following must be observed during a measuring period of 30 min:       _ col and The fluid mass flow rates through the collector m _ c must be kept within precision of the condenser m 75%; The measured temperatures must be kept within precision of 71 1C; The intensity of the global insolation must be kept within precision of 710%; The collector construction error relating to the angle tilted to the horizon must not exceed 721; The change of the measured electrical powers must not exceed 75%; The change of the measured wind velocity must not exceed 70,5 m/s (if nwo2 m/s, it is to be assumed that nw ¼ 0 m/s). 2.3. Processing of the experimental data The semi-integral inlet and outlet temperatures of the collector, evaporator and condenser and the ambient temperature are to be determined during a period of 30 min as follows: ! 10 Dtj t1 þ t11 X t̄ ¼ þ tj , (6) 2 tg j¼2 where t1 and t11 are the temperature values-obtained at the first and the last measurement of the period of 30 min, 1C; tj  temperature value  obtained at the j intermediate measurement;  C; The semi-integral wind velocity is to be determined during a period of 30 min analogically: ! 10 Dtj vw;1 þ vw;11 X þ vw;j . (10) v̄w ¼ 2 tg j¼2 The semi-integral electrical powers of the heat pump, of the pumps and of the AES are to be determined during a period of 30 min analogically: ! 10 Dtj N 1 þ N 11 X þ . (11) Nj N̄ ¼ tg 2 j¼2 2.4. Accuracy of the measured parameters The directly measured parameters must be measured with the following precision:       Intensity of the global solar insolation—72%; Temperatures—72%; Flow rate—72%; Refrigerant pressures—72%; Wind velocity—76%; Electrical powers—75%. The indirectly measured parameters must be measured with the following precision:      Condenser heat flux of the heat pump—76%; Solar collector heat flux—76%; Collector efficiency—76%; Heat pump coefficient of performance (COP)—76%; System efficiency—76%. Dtj ¼ 3 min time step between two consecutive measurements; tg ¼ 3 min time period of the measurement: The semi-integral intensity of the global solar insolation on the collector plane is to be determined during a period of 30 min analogically: ! 10 Dtj I col;1 þ I col;11 X þ . (7) Ī col ¼ I col;j tg 2 j¼2 The semi-integral fluid flow rates through the condenser, evaporator and collector are to be determined during a period of 30 min analogically: ! 10 _1 þm _ 11 X Dtj m ¯_ ¼ _j . (8) þ m m tg 2 j¼2 The semi-integral refrigerant pressures of condensation and evaporation are to be determined during a period of 30 min analogically: ! 10 Dtj p1 þ p11 X p̄ ¼ þ pj . (9) 2 tg j¼2 3. System design 3.1. Construction An experimental installation consisting of a heat pump ‘‘water–water’’ and two flat plate solar collectors (Figs. 2 and 3) was constructed. There are two main components— the heat pump 1 and the solar collectors 2 (Fig. 1). The flat plate solar collectors are of the type KWM (Table 1). They are orientated to the South and tilted at an angle s ¼ 301. The heat pump is of the type ‘‘water–water’’ (Table 2). It is constructed and produced at the ‘‘Institute of Refrigerator Technique’’ in Sofia, Bulgaria. The evaporator consists of a tube with a diameter +30  1.5 and a length of 2816 mm. It is shaped as a cylinder with a diameter of +640 mm. Three small tubes with diameter +10  1 are situated therein (Fig. 4). The heating fluid is flowing through the small tubes and the refrigerant is flowing between the small tubes and the big tube. The condenser is shaped as a horizontal cylinder with a length of 500 mm and a diameter of +108 mm, where a tube serpentine is situated (Fig. 5). The cooling fluid flows ARTICLE IN PRESS A. Georgiev / Renewable Energy 33 (2008) 832–838 835 Table 2 Technical data of a heat pump with compressor of the type KK-6,6 Nominal compressor power Refrigerant Cooling power Fig. 2. View of the heat pump and the measuring instruments in the test installation. 1500 W R22 4710 W Fig. 4. Cross section of the evaporator. Fig. 5. Cross section of the condenser. Fig. 3. View of the solar collectors in the test installation. Table 1 Technical data of flat plate solar collector KVM Absorber plate length Absorber plate width Insulation thickness Distance between absorber and glass Collector number 1.9 m 0.92 m 0.05 m 0.025 m 2 through the serpentine and the refrigerant enters the upper condenser part, condenses round the serpentine and flows out through its lower part. Two circuits are available in the installation. The working fluid enters the solar collectors as part of the solar circuit, leaves them and enters an electrical heat exchanger 3 (maximum power of 4000 W), which heats the fluid in the solar circuit when the solar radiation is not enough. After that the fluid flows through the evaporator and gets cooled by means of the pumps 9 to the expanding vessel 5, mounted at the height of 3.04 m. Next, it flows through the flow meters 11 and enters the solar collectors. The valves 24 are used to draw out the fluid from the solar circuit. Both pumps 9 have the following correspondent parameters: nominal power—70 and 100 W; volume flow rate— 300 and 300 l/h; head—20 and 30 kPa. The fluid flows through the condenser and the heat exchanger 4 in the second circuit and reaches the expanding vessel 7 with raised temperature. Tap water (with a temperature of about 12 1C) is mixed in this vessel with the heated fluid in the condenser. The cooled water flows by means of the pumps 8 to the expanding vessel 6 and enters the condenser through the flow meters 10 and 12. The valve 23 regulates the flow rate in the condenser circuit. The two pumps 8 have the following respective parameters: nominal power—70 and 100 W; volume flow rate—420 and 420 l/h; head—20 and 30 kPa. All tubes between the installation parts are well insulated and covered with aluminium folio. The visible parts of the thermometers are insulated with aluminium folio, too. ARTICLE IN PRESS 836 A. Georgiev / Renewable Energy 33 (2008) 832–838 3.2. Measurement equipment The list of the measured parameters is presented below. The inlet and outlet fluid collector temperatures, the inlet and outlet fluid evaporator temperature and the inlet and outlet fluid condenser temperatures are measured with the corresponding precise mercury thermometers 18, 16, 20, 21, 17 and 19. The thermometers have the following properties—type TGL, range 0–100 1C, absolute measurement error 0.1 1C. The ambient temperature is measured in a small house with a precise mercury thermometer 15. It has the following properties—type TGL, range 0–50 1C, absolute measurement error 0.1 1C. The wind velocity is measured by means of the anemometer 14, 26, which possesses the following properties—type N188, range 1.5–30 m/s, absolute measurement error 0.5 m/s. The intensity of the global solar insolation in the plane of the solar collectors is measured by means of the integrating solarimeter 13, 25. It has the following properties—type IS2-85, range 0.05–1.2 kW/m2, summary error of the apparatus 72%. The electrical power of the heat pump, of the heaters and of the pumps of the solar and condenser circuits is measured by means of the correspondent wattmeters 29, 27, 28 and 30, which possesses the following properties— range 0–3600 W, precision class 0.5. The refrigerant pressure in the condenser and the evaporator is measured by laboratory manometers with the following properties—type MZM, range 0–25 kg/cm2, precision class 0.4. The fluid flow rate in the solar circuit is measured by the flow meters 11, which have the following properties—type SW 16,1, range 0.5–3 l/min, and precision class 2.5. The fluid flow rate in the condenser circuit is measured by the flow meters 10 and 12, which possess the following properties—type GG, range 30–250 l/h, and precision class 2. The equipment was calibrated prior to the test. 4. Parameter investigation of a system heat pump–solar collectors An experiment was conducted on the installation consisting of heat pump and solar collectors (Fig. 1). The methods described in Chapter 2 and an experimental design were used during the investigation. 4.1. Choice of experimental design It was decided to test the installation using a planning experiment. The number of the measured parameters is 15. One of them, the fluid flow rate through the solar collectors and the evaporator was kept constant (4.7 l/min). Three parameters were chosen as experiment factors:
the medium condenser fluid temperature t̄c ¼ tc;i þ tc;o =2, the fluid _ c and the medium evaporator condenser mass flow rate m
fluid temperature t̄ev ¼ tev;i þ tev;o =2. They respond to the following basic conditions:      The measurement of the factors is of high enough precision; The factors are essential; All factors can be regulated directly during the experiment; Every factor has quantitative value and can be controlled in its whole change interval; The realization of all possible combinations of the factor values in their variation region limits is possible. Preliminary tests were done with the aim to specify the technical realization of the factor values combinations. The following variation limits of the basic factors during the experiment were chosen: t̄c 2 ½25; 35  C; _ c 2 ½0:077; 0:111 kg=s; m t̄ev 2 ½15; 25  C. The experiment is conducted at three factor levels (lower, basic and upper) as follows: t̄c 2 f25; 30; 35g  C; _ c 2 f0:077; 0:094; 0:111g kg=s; m t̄ev 2 f15; 20; 25g  C. The change steps of the factors are 5 1C for the temperatures and 0.017 kg/s for the mass flow rate. Two parameters were chosen as objective functions— first the COP of the heat pump and second the whole efficiency of the system Zs. The experimental values of COP and Zs are calculated according to Eqs. (3) and (5) respectively. The heat specific coefficients cc and ccol are determined as tabular values based on the average temperatures and pressures of the water. The COP of the heat pump as objective function meets the basic requirements. It describes the amount of heat delivered to the consumer at a specific input electrical energy of the compressor. The common view of the mathematical model describing the COP is as follows: _ c þ c3 t̄ev þ c12 t̄c m _ c þ c13 t̄c t̄ev COP ¼ co þ c1 t̄c þ c2 m _ 2c þ c33 t̄2ev . _ c t̄ev þ c11 t̄2c þ c22 m þ c23 m ð12Þ The coefficients c0,c1yc33 are obtained after statistical data processing. Two coefficients fall off after the evaluation of the coefficient significance. The final view of Eq. (12) is as follows: _c COP ¼ 3:391957  0; 37132t̄c þ 0:09989m þ 0:31236t̄ev þ 0:045987t̄c t̄ev  0:0200939t̄2c _ 2c  0:1117061t̄2ev .  0:003156m The verification proves the model adequacy. ð13Þ ARTICLE IN PRESS A. Georgiev / Renewable Energy 33 (2008) 832–838 Fig. 7. Reverberation of the efficiency Zs. Fig. 6. COP reverberation. Fig. 6 shows the COP reverberation, which was obtained from Eq. (13). The areas built present the COP as a _ c and t̄ev . Every area is built at function of three factors t̄c , m _ c –m _ c ¼ 0.077 kg/s, m _ c ¼ 0.094 kg/s a constant flow rate m _ c ¼ 0.111 kg/s. The values of the factors t̄c and t̄ev and m situated on the two axes of the coordinate system and the values of COP are situated on the third axis. The system efficiency Zs was chosen as an objective function in the second case because of its richest value of the investigated system. The basic factors influencing Zs and COP are the same and are in the same variation limits. Zs responds to the basic requirements. The mathematical model describing Zs has the following common view: ð14Þ The coefficients d0,d1yd33 are obtained on the basis of the same tests after statistical data processing. The evaluation of the coefficient significance is made analogically. The model looks in the following way after rejecting the insignificant coefficients: _c Zs ¼ 0:66945  0:02716t̄c þ 0:01723m þ 0:03869t̄ev þ 0:011375t̄c t̄ev _ c t̄ev  0:00165t̄2c  0:0094m _ 2c  0:0162t̄2ev .  0:0045m on the two axes of the coordinate system and the values of Zs are situated on the third axis. 4.2. Error analysis Three types of errors exist, which can be made during an experiment—systematic, occasional and dynamic errors. There is a lack of systematic errors because all the measuring instruments were calibrated recently before starting the measurements. Dynamic errors during the experiment are not present for the following reasons:   _ c þ d 3 t̄ev þ d 12 t̄c m _ c þ d 13 t̄c t̄ev Zs ¼ d 0 þ d 1 t̄c þ d 2 m _ 2c þ d 33 t̄2ev . _ c t̄ev þ d 11 t̄2c þ d 22 m þ d 23 m 837 ð15Þ The verification proves the model adequacy. Fig. 7 shows the Zs reverberation, which was obtained from Eq. (15). The areas built present the Zs as function of _ c and t̄ev . Every area is built at a constant three factors t̄c , m _ c –m _ c ¼ 0.077 kg/s, m _ c ¼ 0.094 kg/s and m _c ¼ value of m 0.111 kg/s. The values of the factors t̄c and t̄ev are situated The measuring equipment is used within the acceptable working condition limits of the instruments; The installation operates for more than an hour before starting the reading. The occasional errors are evaluated by means of the regression analysis. Two linear models are created (one for the COP and another one for the efficiency of the system Zs) as a function of 15 factors (inlet condenser temperature, tc,i; ; outlet condenser temperature, tc,o ; inlet evaporator temperature, tev,i ; outlet evaporator temperature, tev,o ; inlet collector temperature, tcol,i ; outlet collector temperature, tcol,o; global solar insolation in the plane of the solar collectors, Icol; ambient temperature, ta; flow rate through _ col ; flow rate through the condenser the collector loop, m _ c ; refrigerant pressure in the condenser, pc; loop, m refrigerant pressure in the evaporator, pev; electrical power of the heat pump, Nhp; electrical power of the pumps in the condenser loop, Nc and electrical power of the pumps in the collector loop, Ncol) [17]. The relative error of the measuring instruments is presented in Table 3. The coefficient of multiple correlation R and the ^ P̄ and Z̄  Z̄^ are maximum absolute error COP  C̄ Ō s s ARTICLE IN PRESS A. Georgiev / Renewable Energy 33 (2008) 832–838 838 Table 3 Relative error of measured parameters Parameter Relative error (%) Inlet condenser temperature tc,I Outlet condenser temperature tc,o Inlet evaporator temperature tev,I Outlet evaporator temperature tev,o Inlet collector temperature tcol,I Outlet collector temperature tcol,o Global solar insolation in the plane of the solar collectors Icol Ambient temperature ta _ col Flow rate through the collector loop m _c Flow rate through the condenser loop m Refrigerant pressure in the condenser pc Refrigerant pressure in the evaporator pev Electrical power of the heat pump Nhp Electrical power of the pumps in the condenser loop Nc Electrical power of the pumps in the collector loop Ncol 1.9 0.9 2.0 1.9 1.5 1.2 2.0 2.0 1.1 0.25 0.9 0.9 1.0 2.3 2.1 used as adequacy criterion of the model evaluation. The following values are obtained for both parameters: n o  R ¼ 0.981 and max COP  C̄ Ō^ P̄ ¼ 0:036 for i i COP; n o  R ¼ 0.989 and max Z̄s  Z̄^ s i ¼ 0:0055 for Zs. i 5. Conclusions A system heat pump–flat plate solar collectors was investigated experimentally. A planning experiment was used during the tests for the evaluation of the COP and of the system efficiency Zs. The following conclusions can be drawn:     COP and Zs are higher at lower condenser temperature t̄c ; _ c leads to The increase of the mass condenser flow rate m an increase of COP and Zs; COP and Zs are higher at higher temperature t̄ev ; The same approach can be used by similar heat pump systems for describing different thermal parameters. The presented experiment is a part of the work, which aims at the theoretical and experimental investigation of a series heat pump system. The publication of further results is foreseen. References [1] Freeman TL, Mitchell JW, Audit TE. Performance of combined solar–heat pump systems. Solar Energy 1979;22(2):125–35. [2] O’ Dell MP, Mitchell JW, Beckman WA. Design method and performance of heat pumps with refrigerant-filled solar collectors. Trans ASME, J Sol Energy Eng 1984;106(2):159–64. [3] Sakai I, Takagi M, Terakawa K, Ohue J. Solar space heating and cooling with bi-heat source heat pump and hot water supply system. Solar Energy 1976;18(6):525–32. [4] Hatheway FM, Converse AO. Economic comparison of solarassisted heat pumps. Solar Energy 1981;27(6):561–9. [5] Solar-assisted heat pump for the heating and cooling of buildings. Six month technical report, 6.11.1978-05.05.1979. General electric company corporate research and development. Schenectady: New York 12301; May 1979. [6] Bessler WF, Hwang BC. Solar assisted heat pumps for residential use. ASHRAE J 1980;22(9):59–63. [7] Cottingham JG. Heat pump design: Cost-effectiveness in the collection, storage and distribution of solar energy. ASHRAE J 1979(4):35–8. [8] MacArthur JW, Palm WJ, Lessmann RC. Performance analysis and cost optimization of a solar-assisted heat pump system. Solar Energy 1978;21(1):1–9. [9] Chandrashekar M, Le NT, Sullivan HF, Hollands KGT. A comparative study of solar assisted heat pump systems for Canadian locations. Solar Energy 1982;28(3):217–26. [10] Tleimat BW, Howe ED. A solar-assisted heat pump system for heating and cooling residences. Solar Energy 1978;21(1):45–54. [11] Chaturvedy SK, Murphy LM. Energy conservation potential of large capacity Solar-assisted heat pumps for low temperature IPH applications. Trans ASME, J Sol Energy Eng 1985;107(4):286–92. [12] Roeder DJ, Reid RL. A TRNSYS/GROCS simulation of the TECH House I ground-coupled series solar-assisted heat pump system. Trans ASME, J Sol Energy Eng 1983;105(4):446–53. [13] Speiser K, Diamond E. The energy house. ASHRAE J 1975(11):68–71. [14] Terrell RE. Performance and analysis of a ‘‘series’’ heat pumpassisted solar heated residence in Madison, Wisconsin. Solar Energy 1979;23(5):451–3. [15] Ucar M, Mittur S. Weather sensitivity analysis of solar-assisted heat pump systems. Trans ASME, J Sol Energy Eng 1984;106(2):182–6. [16] Georgiev A, Kartelov Y. Methods for investigation of a system heat pump ‘‘water–water’’—water solar collectors. Thermo techniques for the mode of life 2004;12:15–8. [17] Bozhanov E, Vuchkov I. Statistical methods for modeling and optimization of multifactor objects. Sofia: Tehnika; 1983.