A Solar Thermal Cooling And Heating System.pdf

Available online at www.sciencedirect.com

Solar Energy 84 (2010) 166–182 www.elsevier.com/locate/solener

A solar thermal cooling and heating system for a building: Experimental and model based performance analysis and design Ming Qu a,*, Hongxi Yin b,1, David H. Archer c,2 a

School of Civil Engineering, Purdue University, 550 Stadium Mall Drive, West Lafayette, IN 47907-2051, USA School of Engineering Education, Purdue University, 701 W. Stadium Ave., West Lafayette, IN 47907-2061, USA c Department of Mechanical Engineering, Carnegie Mellon University, 5000 Forbes Ave., Pittsburgh, PA 15213, USA b

Received 29 December 2008; received in revised form 13 October 2009; accepted 14 October 2009 Available online 16 December 2009 Communicated by: Associate Editor Ruzhu Wang

Abstract A solar thermal cooling and heating system at Carnegie Mellon University was studied through its design, installation, modeling, and evaluation to deal with the question of how solar energy might most effectively be used in supplying energy for the operation of a building. This solar cooling and heating system incorporates 52 m2 of linear parabolic trough solar collectors; a 16 kW double effect, water– lithium bromide (LiBr) absorption chiller, and a heat recovery heat exchanger with their circulation pumps and control valves. It generates chilled and heated water, dependent on the season, for space cooling and heating. This system is the smallest high temperature solar cooling system in the world. Till now, only this system of the kind has been successfully operated for more than one year. Performance of the system has been tested and the measured data were used to verify system performance models developed in the TRaNsient SYstem Simulation program (TRNSYS). On the basis of the installed solar system, base case performance models were programmed; and then they were modified and extended to investigate measures for improving system performance. The measures included changes in the area and orientation of the solar collectors, the inclusion of thermal storage in the system, changes in the pipe diameter and length, and various system operational control strategies. It was found that this solar thermal system could potentially supply 39% of cooling and 20% of heating energy for this building space in Pittsburgh, PA, if it included a properly sized storage tank and short, low diameter connecting pipes. Guidelines for the design and operation of an efficient and effective solar cooling and heating system for a given building space have been provided. Ó 2009 Elsevier Ltd. All rights reserved. Keywords: Parabolic trough solar collector; Double effect absorption chiller; Solar cooling and heating; System optimization; Carnegie Mellon University; TRNSYS

1. Introduction At present, the buildings in the United States account for more than one third of the total primary energy consumption.3 Increases in building energy demands and *

Corresponding author. Tel.: +1 765 494 9125; fax: +1 765 269 7903. E-mail addresses: [email protected] (M. Qu), [email protected] (H. Yin), [email protected] (D.H. Archer). 1 Tel.: +1 765 269 7917. 2 Tel.: +1 412 268 2004. 3 US Energy Information Administration, “Energy consumption by section,” http://www.eia.doe.gov/emeu/aer/txt/ptb0201a.html. 0038-092X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.solener.2009.10.010

energy costs have caused people to seek alternate cheaper, renewable energy sources for the operation of buildings. As one of renewable energy sources, solar energy as heat can be used as the energy source for building cooling, heating, and ventilation to conserve energy, as well as to protect the environment by avoiding pollutant and CO2 emissions associated with the generation of electric power and the burning of natural gas. For cooling systems, solar thermal energy can be used in an absorption cycle, a desiccant cycle, or a mechanical process. Compared with solar desiccant cycles and mechanical processes, a solar absorption cycle is more reliable,

M. Qu et al. / Solar Energy 84 (2010) 166–182

feasible, and quiet. Significant research on solar thermal absorption cooling occurred from 1970 to 1980, when a number of demonstration projects were conducted in the United States. However, these systems failed to establish a significant global market for solar thermal absorption cooling due to their high initial cost, lack of commercial hot water driven absorption chillers, and scarcity of demonstrations and impartial assessments by reputable institutions (Kulkarni, 1994). There are couple of successful studies demonstrating the technical feasibility of solar thermal absorption cooling, specifically those that are based on high temperature solar receivers and a double effect absorption chiller. A demonstration solar absorption cooling system (Duff et al., 2004), for instance, was installed for a commercial building in Sacramento, California. The integrated compound parabolic concentrator collectors have been operated over 140 °C to provide hot water for a 70 kW (20 ton) double effect McQuay/Sanyo chiller. Another high temperature solar thermal absorption cooling system was developed by the US Army at Yuma. This system has operated 1245 m2 of Hexel parabolic trough solar collectors driving a 160 ton water–lithium bromide (LiBr) double effect absorption chiller for space cooling nearly 14 years since its installation in 1979 (Hewett, 1995). However, the economics of this system were unattractive because of the high capital and operating costs of solar collectors with absorption chillers, compared to those with electrically driven vapor compression chillers, as reported in 1995. Recently solar thermal absorption cooling has again aroused researchers’ interest in the development of high temperature solar receivers, double effect chillers, and advanced control. The Center for Building Performance and Diagnostics (CBPD) at Carnegie Mellon University has carried out research on solar thermal absorption cooling and heating to assess the feasibility of this technol-

167

ogy through installation, testing, modeling, and evaluation of a new system with an advanced system configuration using recently available parabolic solar receivers and absorption chiller equipment. This system is the smallest high temperature solar cooling system in the world. To date, only this system of the kind has been successfully operated for more than 1 year. This research has proved the feasibility of this technology, collected the detailed operational data, provided techniques and tools for the design and evaluation of such systems, and helped to promote widespread use of this technology. 2. System description The solar thermal absorption cooling and heating system was installed and studied in the Intelligent Workplace (IW) at Carnegie Mellon University (CMU) in Pittsburgh, PA. It consists of 52 m2 of linear parabolic trough solar collectors (PTSC’s); a 16 kW double effect absorption chiller; and a heat recovery heat exchanger, HX-2, indicated on the process and instrumentation diagram of Fig. 1. The PTSC’s were installed in series and oriented perpendicular to the axis of the building space, which is 15° east from true north. They collect and convert solar energy to thermal energy in a heat transfer fluid (HTF), an aqueous solution containing 50% propylene glycol. The absorption chiller or the HX-2 then uses thermal energy converted from sun to generate chilled water in summer or hot water in the winter. If solar radiation is not adequate for the absorption chiller operation, a natural gas burner in the chiller is then used to provide energy for space cooling. The chiller or the HX-2 connect with the heat exchanger, HX-1, serving as a simulated, variable building load. In addition, there are two main circulation pumps, S1 and S5; an expansion tank, TK-1; and a three-way valve. The pump S1 circulates the HTF through the absorption chiller or the bypass and

Fig. 1. Process and instrumentation diagram of the solar absorption cooling and heating system.

168

M. Qu et al. / Solar Energy 84 (2010) 166–182

the pump S5 is used to circulate the HTF through the HX2.The three-way valve regulates the flow rate of the HTF through the bypass and the chiller. 2.1. The parabolic trough solar collector A parabolic trough solar collector is a concentrating solar thermal receiver comprised of a long parabolic trough reflector with a surface treated receiver pipe enveloped by a glass tube at its focal line as shown in Fig. 2. The glass envelope, at temperature much lower than that of the absorber pipe, significantly reduces the convective and radiative heat losses from the absorber pipe to the surroundings. The annular space between the absorber pipe and the glass envelope is evacuated to minimize conduction and convection of heat between them. All these features of the PTSC result in a higher solar efficiency compared with those of various flat plate and evacuated tube collectors, when the solar field operates at a high temperature. The trough rotates on an axis parallel to its focal line to track the altitude of the sun throughout a day. HTF flows through the receiver pipe to absorb the focused solar energy as heat for use outside of the PTSC’s. The installed PTSC’s comprise four modules, each 6 m long by 2.3 m aperture width with a concentration ratio of 19.6.

temperature, to vaporize water from the solution by using thermal energy provided by either the solar collectors or the natural gas burner. The water vapor is condensed by rejecting heat to cooling water in the condenser. Next, the condensate water is passed through an expansion nozzle into the evaporator. The water is vaporized there at a low pressure, absorbing heat transferred from chilled water flow. This cycle then repeats. A single valve shown in Figs. 3 and 4 can be opened to switch the chiller from the cooling mode to heating mode. In the heating mode, the water vapor boiled from the LiBr solution in the regenerator, directly flows into the evaporator passing through the single valve. The evaporator now acts as a condenser and heats the water stream that is used for space heating. 3. System performance test program A test program for the solar thermal cooling and heating system was carried out to characterize devices and system from February to September in 2007. It includes three types of tests: PTSC performance tests, absorption chiller based solar cooling/heating tests, and heat exchanger based solar heating tests. An instrumentation, control and data acquisition systems, provided by the Automated Logical Corporation (ALC), has been used for the tests and the

2.2. The absorption chiller The absorption chiller installed is a dual fired, double effect, LiBr chiller integrated with a cooling tower. It has a natural gas burner in its regenerator to provide heat when solar energy is inadequate. It has both cooling and heating functions. In the cooling cycle, water vapor is absorbed into concentrated LiBr solution in the absorber shown in Fig. 3. A solution pump then circulates the dilute LiBr solution to the regenerator, which operates at a higher pressure and

Fig. 2. The parabolic trough solar collector.

Fig. 3. Absorption chiller in cooling cycle. Refer to the absorption chiller brochure of the courtesy Broad Air Conditioning Co.

M. Qu et al. / Solar Energy 84 (2010) 166–182

169

and its variability was less than ±4% throughout the tests. Since wind velocity greatly impacts the convective heat loss from the PTSC’s, all of performance tests were performed with wind speed less than 4.5 m/s. Turbulent flow was maintained within the absorber pipe to ensure good heat transfer between the fluid and the pipe. Figs. 5 and 6 depict a typical PTSC performance test. Fig. 5 shows the temperatures at the inlet and outlet of the PTSC’s and the two heat exchangers, cold side of HX-2 and cold side of HX1. Fig. 6 is the corresponding calculated heat quantities of total solar radiation availability, solar collectors, and two heat exchangers. The time step for the measurements was 1.0 min. Based on 55 steady state data from the PTSC performance tests, the statistical method was used to identify the PTSC’s performance. The system energy balances were checked by using the data. The results showed that the measurements were in reasonable agreement with the energy balances, taking into account the sensor accuracy and the impacts on energy of other devices such as the pumps. The mean values of experimental data were introduced into the statistical tool to generate the correlation of the solar collector efficiency, g, by using multiple regressions. The PTSC efficiency was determined as indicated in Eq. (1). The optical efficiency is 0.634 and the linear coefficient of thermal losses is 1.4 W/°C m2. g ¼ 0:634  1:4 Fig. 4. Absorption chiller in heating cycle. Refer to the absorption chiller brochure of the courtesy Broad Air Conditioning Co.

operation. This control system is a web-based control and data display system using BACnet as protocol. There were a total of 21 sensors installed including electromagnetic flow rate sensors (accuracy: ±0.2% at 100% flow, ±0.1% at 40% flow), RTD temperature sensors (accuracy: ±0.3 °C at 0 °C; ±0.8 °C at 100 °C; ±1.3 °C at 200 °C; transmitter: ±0.1% of span), pressure sensors (non-linearity accuracy: ±0.25% full span), and pyrheliometer for direct normal solar radiation (accuracy: ±0.5% of full span). The measurements from the system performance tests were collected and used to validate the system performance model. 3.1. PTSC performance test data In PTSC performance tests, building loads were simulated by the HX-1 and were adjusted to maintain the PTSC’s operated at a quasi-steady state, which refers to the condition of the collector when the flow rate and inlet fluid temperature are constant, but the exit temperature changes slightly due to the normal variations in solar irradiance that occur with time for clear sky. The PTSC performance tests were performed in 17 clear days when the direct normal solar radiation was greater than 630 W/m2,

ðT in þ T out Þ=2  T am I DN  cosðhÞ

ð1Þ

where, g: solar collector efficiency (no unit); Tin: inlet temperature of the solar collector (°C); Tout: outlet temperature of the solar collector (°C); Tam: ambient air temperature (°C); IDN: direct normal solar radiation (W/m2); H: incident angle of solar collector (°). 3.2. Solar absorption cooling daily test data In the absorption chiller based solar cooling tests, the HTF was circulated through bypass until the temperature required by the absorption chiller was reached in the morning of a sunny day; and then the HTF was diverted through absorption chiller to produce chilled water for space cooling. When the amount of solar energy was no longer adequate to provide heat for the absorption chiller, the HTF was circulated in the collection loop through the bypass until the sun set. The measured data from solar absorption cooling tests on July 16 and July 31, 2007 are selected to present the analyses of system performance. Figs. 7 and 9 show throughout a day, the measured temperatures of the HTF at the exit of the PTSC’s and at the inlet and outlet of the chiller, and also of the chilled water at the inlet and outlet of the chiller. The rapid rise of the temperatures of the HTF at the chiller inlet and outlet at noon was the

170

M. Qu et al. / Solar Energy 84 (2010) 166–182 110.00 100.00 90.00 T_SC_out

80.00

Temperature in °C

T_HX-2_cs_out

70.00 T_SC_in

60.00

T_HX-2_cs_in

50.00 T_HX-1_cs_out

40.00 30.00 20.00

T_HX-1_cs_in

10.00 0.00

Local time on Apr.22, 2007

Fig. 5. Operating temperatures of the PTSC test on 22 April 07.

55.00 50.00 45.00

Idn*Aa*Cos(θ)

40.00

Power rate in kW

35.00 30.00

Q_useful_solar

25.00

Q_HX-2_hs

20.00

Q_HX-2_cs

15.00 10.00 5.00 0.00

Fig. 6. Energy flows of the PTSC test on 22 April 07.

result of the three-way valve opening to admit the HTF to the chiller when its temperature exceeds the temperature required by chiller to produce chilled water. Prior to this time, the chiller was driven by natural gas. Figs. 8 and 10 show the corresponding calculated heat quantities: (1) the solar input marked as IDN  Aa  cos(h) in figures: the product of direct normal solar irradiation from pyrheliometer measurements, actual aperture surface area (considering shadows), and the cosine of the incident angle; (2) the cooling capacity provided by the chiller marked as Q_chiller_cooling: the product of the chilled water flow and the temperature difference over the chiller; (3) the thermal energy to the chiller, marked as Q_useful_sc: the product of the HTF flow and the temperature difference over the solar collector; (4) the thermal energy from solar field to the chiller marked as Q_chiller_solarinput: the product

of the HTF flow and the temperature difference over the chiller. As indicated by these figures, energy delivered by solar receiver was larger than the energy used by chiller at the beginning of the chiller operation while the relation between them was reversed in the later afternoon. When the HTF was operated at 150–160C, the overall solar efficiency of the PTSC’s was approximately 33–40%. The COP of the installed absorption chiller was in the range 1.0–1.1. The solar COP of the overall installed solar cooling system, the product of the COP of absorption chiller and the solar collector efficiency, was therefore about 0.33–0.44. The maximum output of the absorption chiller was 12 kW. The reasons for this capacity, lower than the chillers design capacity of 16 kW, were related to the intensity of direct solar radiation and rate of heat transfer between HTF

M. Qu et al. / Solar Energy 84 (2010) 166–182

171

180 T_chiller_HTF_in

160

Temperature in °C

140

T_chiller_HTF_out

120 100

T_sc_out

80 60 40 T_chiller_CHW_return

20 T_chiller_CHW_supply

0

Local time on Jul.31, 2007 Fig. 7. Operating temperatures of solar cooling test on 31 July 07.

55 50 45

Power rate in kW

40

I DN *Aa*cos(θ)

35 30 25 20 15

Q_chiller_cooling

10

Q_useful_sc

5 0

Q_chiller_bySolar

Local time on Jul.31, 2007 Fig. 8. Cooling capacity of solar cooling system on 31 July 07.

180.00 T_chiller__HTF_in

160.00

Temperature in °C

140.00 T_chiller_HTF_out

120.00 100.00 80.00

T_sc_out

60.00 40.00 T_chiller_CHW_return

20.00 0.00

T_chiller_CHW_supply

Local time on Jul.16, 2007 Fig. 9. Operating temperatures of solar cooling test on 16 July 07.

172

M. Qu et al. / Solar Energy 84 (2010) 166–182 55.00 50.00 45.00

IDN *Aa*cos(θ)

Power rate in kW

40.00 35.00 30.00 25.00 20.00 15.00

Q_chiller_cooling

10.00 5.00 0.00

Q_solar_delivered Q_chiller_bySolar

Local time on Jul.16, 2007 Fig. 10. Cooling capacity of solar cooling system on 16 July 07.

and LiBr solution. Due to the relative high humidity of Pittsburgh in summer, the direct solar radiation was relative low with typical values of 600–850 W/m2. Rate of heat transfer depends on the temperature of HTF and the area and heat transfer coefficient of the exchange surface in the regenerator. 3.3. Solar heating daily test data In heat exchanger based solar heating tests, the HTF was circulated and heated in the solar collection loop without operating the load pump, marked S5 in Fig. 1, until the desired operating temperature was reached in the morning of a sunny winter day. Then pump S5 circulated water in the load loop to deliver the heat from HX-2 to HX-1, which simulates the building load by regulating chilled water flow on its cold side. When the amount of solar energy was no longer adequate for the building load, the system operation was halted. In the absorption chiller based solar heating tests, the HTF was circulated through bypass until the temperature required by the absorption chiller was reached in the morning of a sunny day; and then the HTF was diverted through absorption chiller to produce hot water for space cooling. Experimental data have been obtained to define the system performance in solar heating using either the absorption chiller with the HTF at 140 °C, or the heat recovery exchanger with the HTF at 68 °C. A comparison of solar heating using an absorption chiller or a heat recovery exchanger, based on the tests, shows clearly that a heat recovery exchanger is more effective. Figs. 11 and 12 depict the heat exchanger based solar heating testing data on 2 March 07. Use of the exchanger avoids the large temperature difference between the HTF and the heated water in the absorption chiller. And it allows the collectors to be operated at a lower temperature, thus reducing heat losses from the system. In solar heating system tests, the overall solar efficiency of the PTSC’s was

approximately 50–55% when they were operated at 60–70C. The heating generated by the solar system was at range of 16–29 kW with changed direct solar radiation. 4. System performance models System performance models have been developed for the design, evaluation, and optimization of the solar cooling and heating system. A base case model was programmed in accord with the configuration of the solar cooling and heating system installed in the IW at CMU. After verification by experimental data from the system performance tests, the base case model was extended to develop system optimization models. Finally, guidelines for the design and operation of the solar cooling and heating system were formulated on the basis of the results of system optimization by exploring the effects of solar collector area and orientation, the provision of thermal storage, system pipe length and diameter, and system operational control strategies. 4.1. Approach of the system performance model The software selected for modeling system performance is TRaNsient SYstem Simulation program (TRNSYS), developed by Solar Energy Laboratory of the University Wisconsin. The system model developed includes building model and energy supply system model. The building model considers the configuration of the building; weather conditions; the schedules for occupancy, lighting, equipment; and set points for temperature and humidity and for conditioned fresh air flow. The output of the building model quantifies building thermal condition and building sensible loads. The energy supply system model consists of all major system components, most of which are available in the TRNSYS component library. The linear parabolic concentrator component of the library of TRNSYS

M. Qu et al. / Solar Energy 84 (2010) 166–182

173

100.00 90.00

Temperature in °C

80.00 70.00 60.00 50.00 40.00 30.00 20.00 10.00 0.00

Local time on Mar.02, 2007 Fig. 11. Operating temperatures of HX based solar heating system on 2 March 07.

55.00 50.00 45.00

Power rate in kW

40.00 35.00 30.00 25.00 20.00 15.00 10.00 5.00 0.00

Local time on Mar.02,2007 Fig. 12. Heating capacity of HX based solar heating system on 2 March 07.

was modified to represent the installed PTSC. In addition, the overall system control was programmed as a new component to integrate the operation of the PTSC, absorption chiller, pumps, and other devices. The system performance model utilizes two indicators to assess the system performance: solar fraction and saved energy per unit solar receiver area. The solar fraction is the ratio of the building cooling/heating provided by solar energy; and the saved energy per unit solar receiver area is energy provided by solar for cooling/heating divided by the total area of solar receivers. The solar building cooling and heating system models and their details have been published (Qu, 2008). The system performance models use the typical meteorological year (TMY2) data to obtain the weather conditions and geographic information for the location of the solar system installed. TMY2 data sets were typical values of solar radiation and meteorological elements for a 1-year

period derived from the 1961–1990 National Solar Radiation Data Base (NSRDB). TMY2 data for Pittsburgh, PA, indicate that the daily peak direct normal solar radiation intensity in the summer is lower than that in winter. The very humid air in summer of Pittsburgh diffuses a large amount of direct normal solar radiation before it reaches the earth. On the contrary, due to dry and chilled weather in the winter, the daily peak direct normal solar radiation is relatively high, sometimes above 1000 W/m2. The daily average direct normal solar radiation intensity throughout a year in Pittsburgh is lower than 420 W/m2. The solar cooling system performance models use two absorption chiller components, a hot water driven absorption chiller and a direct fired absorption chiller, to simulate the installed dual fire absorption chiller since the installed chiller can be driven either by the solar energy or by the natural gas, not simultaneously by both energy sources.

174

M. Qu et al. / Solar Energy 84 (2010) 166–182

The experimental data from the system performance tests indicates that the heat exchanger based solar system provides more useful heat than the absorption chiller based heating system, so solar heating system performance models are based on the heat exchanger based solar heating system.

operation when there is a cooling demand and direct normal solar radiation is greater than 300 W/m2 which is required to balance the heat loss from the system during its operation. The highest operating temperature was set at 185C. When this temperature is reached, the PTSC’s are defocused to prevent overheating of the system. This control feature is included in PTSC model based on on/ off with hysteresis. The HTF is continuously circulated and heated through the bypass in the solar collection loop until it reaches the temperature required by hot water absorption chiller. Then the HTF is switched from the bypass to the regenerator of hot water driven absorption chiller. The chiller controls the flow of the HTF by adjusting the three-way valve and also the temperature of the HTF by the defocus of the PTSC’s, assuming solar input meets or exceeds the need. The system model uses hysteresis control to switch between the two energy sources, natural gas and solar energy. The chiller uses solar energy when the HTF is at 155 °C or above, and it switches natural gas when the HTF is lower than 135 °C. The base case model for the solar heating system includes the PTSC’s, pumps, HX-2, an electric heater (or natural gas burner), and controls. Similar control for the PTSC’s in solar cooling system is applied to solar heating system. The highest operating temperature of the solar receivers is set at 95C for solar heating. Whenever heating is required, hot water is delivered either from solar field or the auxiliary heater for space heating. When the HTF temperature from PTSC’s is 3 °C higher than the temperature of water entering HX-2 from building and there is a heating demand, the HX-2 transfers the heat from the solar field to the building. The electrical heater (or natural gas burner) is triggered by the temperature of water entering the heater. If it is less than the temperature required by building heating device, the heater is turned on providing hot water for space heating.

4.2. Building model results The IW is a single floor building with 245 m2 of net floor area and 3.1 m of average height. It is an open plan and subdivided by partition walls and furniture in nine offices and one conference space. The building has horizontal shading on the east and west facades. Fig. 13 shows the building cooling and heating loads estimated by the building model. 4.3. Model based performance analysis of the installed solar cooling and heating system In order to ensure the accuracy of the system model, system performance models started with the system performance evaluation models. These evaluation models were programmed based on the installed system configuration and the daily system performance test conditions including the ambient temperature, wind velocity, direct normal solar radiation, working condition, and operational procedures. The measured data from the tests verified the evaluation models. And there was good agreement between model results and experimental data. On the basis of system evaluation models, the base case system models were developed according to the installed solar cooling and heating system. The components in the base case model included PTSCs, hot water driven double effect absorption chiller (HW chiller), direct fired double effect absorption chiller (DF chiller), circulation pumps, space cooling load and controllers. The system starts its

5500 5000

Q_sensible_heating Q_dehumidification Q_sensible_heating Q_humidification

4500

Load in kWh

4000 3500 3000 2500 2000 1500 1000 500 0 1

2

3

4

5

6

7

8

9

10

Month Fig. 13. Building heating and cooling loads estimated by building model.

11

12

M. Qu et al. / Solar Energy 84 (2010) 166–182

4.3.1. Results of the base case model of solar cooling The results of the base case model for solar cooling show that the installed solar cooling system is able to save 36 kW h/m2, which could provide 12.7% (solar fraction) of the building space cooling in a cooling reason. Although the solar availability is in phase with building cooling demands, 76% of useful solar energy collected is lost to the surroundings through the piping and solar collectors due to the system design and operation. The direct normal solar radiation with relatively low intensity in the summer of Pittsburgh remarkably reduces the opportunities for solar cooling. According to the model results, solar cooling can operate only 69 days over 183 days from April 15th to October 15th. A significant portion of useful solar energy collected, as shown in Fig. 14, is used for system preheating due to its large system thermal heat capacity and the loss of heat from the system at night. The remaining, useful solar energy, can only supply the absorption chiller for 3–4 h over 8–10 h required for cooling in a summer day of Pittsburgh. In addition, the high variability of the direct normal solar radiation makes the chiller operate intermittently because chiller controller is sensitive to the temperature of the HTF from solar field entering the chiller. A small buffer storage tank should be used to stabilize chiller operation and to improve system performance. 4.3.2. Results of the base case model solar heating The results of the base case model for solar heating indicated that the installed solar thermal heating system could provide 6.7 kW h/m2, which provide 3.8% of building heating from October 15 to April 15. Although solar receivers potentially provide approximately half amount of energy for building heating, as shown in Fig. 15, solar energy availability is out of phase with building heating demand. Therefore, about 16% of available solar energy is rejected by defocusing and 94% of solar energy collected is lost to the surroundings from the system as illustrated in Fig. 16

175

of a typical operational scenario of the installed solar heating. Building requires heating when solar energy is not available; and when solar energy is available and it exceeds the requirement of space heating during day time, the PTSC’s must be defocused to moderate the solar energy collected and useful solar energy is lost. Therefore, heat storage is needed to overcome this difficulty. Much heat is lost from the system during its operation in the day and at night. The heat lost at night must be replaced at the beginning of the day to rapidly return the solar heating system to its operating conditions, so reducing the surface area, volume, and heat capacity of the system can reduce this problem. 4.4. System optimization of the solar cooling and heating system System optimization was carried out to improve system performance by modifications in the system design and operation, including changes in the system configuration, equipment design details, operating conditions, and control strategies. 4.4.1. Orientation of the PTSC’s The tracking drive system of a PTSC rotates its collector about its axis to track the sun until the sun central ray and the normal direction of the aperture surface of the reflector are coplanar. The angle between the sun rays and the aperture normal is called incident angle. The intensity of solar radiation on the surface is reduced by a factor equal to the cosine of the incident angle, so the incident angle significant influences the efficiency of the PTSC. The orientation of the PTSC refers to the position of its tracking axis. North–south oriented PTSC actually tracks solar azimuth and east–west oriented PTSC tracks solar altitude. Since the orientation of the PTSC directly impacts on the incident angle, it significantly influences the system performance of solar system. To

20.00 Q_usef ul_solar (kW) Q_Load (kW)

16.00

Energy rate in kW

Q_HWchiller_CHW (kW)

12.00

8.00

4.00

0.00 0:00

3:00

6:00

9:00

12:00

15:00

18:00

21:00 Solar time

Fig. 14. Useful solar energy, cooling load, and cooling provided by chiller on 9th August.

176

M. Qu et al. / Solar Energy 84 (2010) 166–182

25 Q_usef ul_solar (kW) Q_Load(kW)

Energy rate in kW

20

15

10

5

0 Date Fig. 15. Useful solar energy and building sensible heating load in January.

20.0

Q_usef ul_solar (kW)

16.0

Energy rate in kW

Q_Load(kW) Q_HX-2 (kW)

12.0

8.0

4.0

0.0 0:00

3:00

6:00

9:00

12:00

15:00

18:00

21:00

0:00

Solar time Fig. 16. Useful solar energy, building sensible heating load, and heating provided by HX-2 on 30th December.

explore the effects of PTSC’s orientation on system performance, system optimization calculations compared the solar radiation throughout a day on a PTSC oriented NS and EW in Pittsburgh. The calculations show that the incident solar radiation has two peaks throughout a day for a NS oriented PTSC and one peak for an EW oriented PTSC. The NS oriented PTSC can potentially deliver one third more useful solar energy than an EW oriented PTSC throughout a summer day. On a winter day, however, the NS oriented PTSC can only deliver half the solar energy of that provided by an EW oriented PTSC because daily average incident angle on NS oriented PTSC is smaller than EW oriented PTSC in winter days, as illustrated in Fig. 17. In Pittsburgh, a NS oriented PTSC would receive 25% more solar radiation in the cooling season and 5% less in the heating season, compared to an EW oriented PTSC.

For a combined solar cooling and heating system, the calculations show that a NS oriented PTSC could collect 927.7 kW h/m2 per year, which is 14% more solar energy than an EW oriented PTSC could collect, 815.3 kW h/m2, as indicated in Fig. 18. Based on the similar system optimization calculations in other cities like Seattle, Albuquerque, Madison, San Francisco, generally NS is a better orientation than EW for solar cooling, and conversely EW is better than NS for solar heating although the latitude of a location influence the incident angle. The system performance models also quantified the impacts from the PTSC’s orientation on the overall system performance. If the installed PTSCs were oriented in NS rather than EW, the system would have an improved solar fraction of 18.7% for cooling and a reduced solar fraction of 2% for heating.

Solar irradiations on aperture in W/m2

M. Qu et al. / Solar Energy 84 (2010) 166–182

177

1000 900

EW_Jun10

800

NS_Jun10

700

EW_Dec2 NS_Dec2

600 500 400 300 200 100

Time of day

0 1

2

3

4

5

6

7

8

9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

Fig. 17. Orientation and solar beam irradiation on aperture in Pittsburgh.

Monthly solar irradiations (kWh/m2)

1200 1000 800 NS_kWh/m2

600

EW_kWh/m2 Direct normal_kWh/m2

400 200 0

1

2

3

4

5

6

NS_kWh/m2

36.0

41.3

75.4

93.7

113.6

129.2

EW_kWh/m2

49.8

47.4

69.5

73.8

85.1

Direct normal_kWh/m2

58.4

57.9

90.8

100.7 117.3

7

8

117.6 111.6

97.7

87.8

132.4

120.7 118.3

87.4

9

10

11

12

Total

91.9

69.7

30.5

17.3

927.7

76.1

74.0

40.0

26.8

815.3

104.5

92.3

47.0

30.3 1070.4

Fig. 18. Monthly solar irradiations on aperture throughout a year.

4.4.2. Control for the solar collection loop The flow rate of the HTF in the solar collection loop is a significant operating condition for solar cooling and heating system. There are two alternative control strategies: constant HTF flow and constant outlet temperature of HTF from the PTSC’s. Constant flow control circulates HTF throughout the solar loop at a constant flow. It is common in flat plate solar collector systems to minimize the costs of equipment used in water heating systems. Constant temperature control maintains the HTF temperature constant at the outlet of the PTSC by varying HTF flow rate. This control is typically used in high temperature concentrating solar systems that require a given minimum HTF temperature for use and where excessive temperatures can degrade the HTF or cause high system pressure resulting in damage (Stine, 1985). The system performance model was used to explore the impacts of the two control alternatives on system performance. The

system models used 7.2 gpm (1.96 m3/h) as flow rate set point in the constant flow control system. In the constant temperature control system, 75 and 156C were used as the temperature set points at the outlet of the PTSC’s for heating and cooling, respectively. The HTF flow was adjusted between the maximum value of 80 gpm, determined by the pump capacity, and the minimum value, calculated from a Reynolds number of 4000, the required limit for well developed turbulent flow in the PTSC’s. The performance model calculations indicate that the two alternatives result in little difference in system performance for heating, but for cooling the constant temperature control improves the solar fraction attained by a factor of two compared with the constant flow control. As shown in Fig. 19, the constant temperature control reduces the system preheat time prior to chiller by approximately one hour and extends the operating time of chiller driven by solar energy.

178

M. Qu et al. / Solar Energy 84 (2010) 166–182 20 Q_usef ul_solar_ConsFR

Energy rate in kW

16

Q_usef ul_solar_ConsT Q_HWchiller_chw_ConsFR Q_HWchiller_chw_Cons T

12

Q_load

8

4

Time on Aug.09 0

Fig. 19. System performance in cooling with two alternate controls.

Fig. 20 shows the effect of the two control strategies on the inlet and outlet temperatures of the PTSC’s supplying the absorption chiller. When the control is based on constant outlet temperature from the PTSC’s, there is a significant temperature difference between the HTF inlet and outlet temperatures during the morning of the day; these HTF temperatures in the solar loop show that less solar energy is used in system preheating and more for chiller operation. And the duration of the chiller driven by solar energy is in this way extended. The improvement in system performance due to the constant temperature control will not be as significant if the system capacity is reduced, because the energy used for system preheating will be small and the duration of system preheating will be short. In solar heating, however, since the building load is out of phase with the availability of solar energy, the reduction of system preheating period does not significantly improve system performance of solar heating. In general, constant temperature control is recommended for high temperature concentrating solar cooling with a large

system heat capacity. This approach improves the system performance in solar cooling by reducing energy used for system preheating. 4.4.3. Thermal storage requirement Since the daily profiles of solar radiation with time may not coincide with the profiles of the building cooling and heating loads, some of solar energy available from the solar receivers may have to be discarded. Thermal energy storage, therefore, should be used in solar systems to shift excess solar energy recovered during periods of high solar availability to periods of low solar availability. System optimization calculations for thermal storage have estimated the impacts on system performance, from thermal storage as heated HTF with the volumes ranging from 0.5 to 6 m3. 4.4.3.1. Solar heating. On the basis of the base case system models, the system optimization models added a storage tank in the solar collection loop and a pump for discharg-

180 160

Temperature in °C

140 120

T_sc_in_ConsFR T_sc_in_ConsT T_sc_out_ConsFR T_sc_out_ConsT T_ambient

100 80 60 40 20 0

Time on Aug.09

Fig. 20. Operation temperatures of solar receiver in cooling with two alternate controls.

M. Qu et al. / Solar Energy 84 (2010) 166–182

ing the storage tank to HX-2. If the HTF temperature at the outlet of the PTSC’s is higher than the HTF temperature on the top of storage tank, the HTF flows to the storage tank charging it; otherwise, the HTF circulates through bypass. If there is a heating demand, the storage tank is discharged until the stored energy is exhausted. The calculations show that the storage tank significantly improves system performance of solar heating by shifting the excess solar energy. As illustrated in Fig. 21, there is a time lag between heating load and the availability of the useful solar energy. The excess solar energy is collected and stored in storage tank during the day and it is used in the evening and the next morning. In this way, the solar fraction of the base case system is potentially improved by factor six when 4 m3 is used as the optimized volume of storage. Additionally, the calculations indicated that incrementally larger storage volumes lose more heat than they store. 4.4.3.2. Solar cooling. Solar cooling optimization model for evaluating thermal storage placed a storage tank parallel to the bypass in the solar collection loop. The PTSC’s are operated by the constant temperature control. A variable frequency pump was added to discharge the storage tank. The HTF is circulated through the bypass and heated until the operating temperature required by chiller is reached in the morning. The HTF is then diverted through the regenerator of the absorption chiller if there are cooling demands. If the flow through the PTSCs is greater than the flow required by the chiller and the HTF leaving from the PTSCs is at the set temperature, the storage tank is charged. If the flow through the PTSC’s is not adequate for the chiller and the temperature of the HTF at the top of the storage tank is above the minimum operating temperature required by the chiller, the storage tank is discharged until it is exhausted. According to the simulation

179

results, the storage tank did not improve the system performance of solar cooling. The profiles of the building cooling load and the solar radiation are basically in phase. The installed large volume of the solar collection loop is equivalent to a small storage and already provides the storage for cooling; no additional storage is needed. And the area of solar collector is not adequate for both chiller and storage. It merely results in additional heat loss. None the less a small amount of storage might smooth irregularities in the solar supply and building loads. 4.4.4. Auxiliary heater and drain-back tank A system with a large thermal capacity might use auxiliary heat for system preheating in order to improve system performance. A gas fired heater was added in the solar collection loop prior to the PTSC’s in the system performance model to explore the impact of the auxiliary heat on system performance. The results showed that the auxiliary heater significantly improves the system performance as expected. The solar fraction of solar heating is improved by 7.1% when the target temperature of the heater was set to 50 °C; the solar fraction of solar cooling is increased to 30.4% when the target temperature of heater was set to 120 °C. The use natural gas, however, as a substitute for solar energy is contrary to objective of the system. A drain-back tank for the HTF in the solar collection loop might well be an energy saving feature in the design and operation of the system. Drain back might significantly reduce the loss of heat over night from the HTF contained in the system. A drain-back system would drain all the HTF, at about 120 °C in cooling or 50 °C in heating, from the solar collection loop to a well insulated drain-back tank in the late afternoon and pump this hot fluid back into the system the next morning. Calculations have determined that drain back will improve solar fraction of the system by about 4–9% in both heating and cooling.

Energy rate in kW

20

15

Q_usef ul_solar Q_Load Q_HX-2

10

5

0

Solar time Fig. 21. Useful solar energy, heating load, and solar energy used for heating on 14 and 15 November 2007.

180

M. Qu et al. / Solar Energy 84 (2010) 166–182

4.4.5. Length and size of the pipe in the solar collection loop The length and diameter size of the pipe in the solar collection loop play important roles in system performance. Greater pipe lengths and diameters increase the heat loss from the system and increase also the overall heat capacity of the system and of the HTF it contains, so that the system requires more energy for preheating. Greater pipe lengths also increase the pressure loss in circulating the HTF, while greater diameters reduce these losses. The system optimization calculations show that reducing the length and diameter of the loop pipes is an effective means to improve system performance at least until pressure loss and pump energy for HTF circulation in the collection loop become significant. Since the installed solar system has a very long pipe line, around 100 m long. If the solar thermal system

reduces its pipe volume one fourth by reducing the pipe diameter and length, the solar fraction for cooling and heating can be improved by a factor of 1.7. The system uses solar energy to operate the absorption chiller one hour earlier than the base case system, as indicated in Fig. 22, and achieves a solar fraction of 22%, employing constant temperature control. Therefore small pipe lengths and diameters of pipe should be applied in solar absorption cooling and heating system to utilize more useful solar energy for building cooling and heating. 4.4.6. Solar receiver area and storage tank capacity Solar receiver area and thermal storage capacity significantly impact on system performance of solar cooling and heating. The hour by hour delivery of thermal energy by

20 Q_usef ul_ConsT

16

Q_usef ul_ConsT_1/4V

Energy rate in kW

Q_Hchiller_chw_ConsT Q_Hchiller_chw_ConsT_1/4V

12

Q_load

8

4

0

Time on Aug.09 Fig. 22. System energy performance and pipe size on 9th August.

45% Solar cooling (N/S, 52 m2, 1/4 V)

40%

Solar fraction

35% Solar heating (E/W, 52 m2, 1/4 V)

30%

Solar cooling (E/W, 52 m2, 1/4 V) 25% Solar heating (E/W, 52 m2)

Solar cooling (E/W, 52 m2)

20% Solar heating (N/S, 52 m2, 1/4 V) 15% 10% 5%

Volume of tank (m3) 0% 0

1

2

3

4

5

6

7

8

9

Fig. 23. Solar absorption cooling/heating system optimization.

10

11

12

M. Qu et al. / Solar Energy 84 (2010) 166–182

the collectors should be studied to examine the effects of solar collector area and storage capacity on the utilization of the collected solar energy. This energy supply profile should then be compared with the application’s demand profile and the storage capacity available for storing excess solar thermal energy. A larger area of collectors and greater capacity of the storage will increase the quantity of solar heat made available for cooling and heating, but also increase the cost of the installation. The selection of solar receiver area and storage tank capacity is an economic decision. The system optimization models and calculations explored the effects of collector area and storage capacity on the system performance for cooling and heating. The model results indicate that a larger solar field could improve system performance of solar cooling and solar heating. 4.4.7. Optimized system performance System optimization models and calculations have characterized the performance of an optimal solar cooling and heating system for the IW s by taking into account all the factors addressed above. According to the model results, this optimal system would have a 4 m3 storage tank, one fourth of the installed pipe volume of the current system, and constant temperature control. If the PTSC’s were oriented in NS, the system could potentially provide 39% of the cooling and 20% of the heating for the IW in Pittsburgh, PA, and correspondingly save annually 100 kW h/ m2 and 34 kW h/m2, respectively. The PTSC’s are oriented in EW; this system could provide 37% of cooling and 29% of heating by solar energy and correspondingly save annually 74 kW h/m2 and 61 kW h/m2, respectively, as indicated in Fig. 23. In summary, the improved solar cooling and heating system could totally save annually 135 kW h/m2 for both NS and EW orientation. 5. Recommendations Based on the system optimization analyses, guidelines are generated for the design and operation of a solar absorption cooling and heating system, primarily composed of high temperature PTSC’s and a double effect absorption chiller.  The orientation of PTSC’s significantly impacts on the performance of solar cooling and heating system. A NS orientation is recommended for solar cooling dominated system and an EW orientation should be applied to solar heating dominated system to derive the greatest amount of energy from the sun for building operation.  Solar receiver area is reasonably determined by the solar energy collected and building cooling and heating load in the design day. Although solar receivers of larger area with adequate storage tank capacity improve the system performance for both cooling and heating in days when the solar incidence is intermittent or low, the selection of solar receiver area and thermal storage capacity is properly determined by an economic analysis.

181

 A storage tank should be placed in solar collection loop, so that it can switch the excess thermal energy as much as possible from the periods of high solar availabilities and low loads to the periods of low solar availabilities and high loads. The storage, in addition, can also be used to minimize the effects of short term fluctuations in the solar radiation or in the cooling and heating loads. Since the solar availability and the cooling demands are nearly coincident, only a small storage tank is required to smooth operation. Solar availability and the heating loads, during the evening, night, and early morning, are far from coincident. A sizable storage tank can significantly improve the system performance of solar heating by storing the excess solar energy during the day and making it available in the evening and early morning.  A drain-back tank should be considered for a solar cooling and heating system, which has a large heat capacity, to achieve a higher system performance. The system can drain all the hot HTF from the solar collection loop back to the well insulated drain-back tank in the late afternoon and fill the HTF in the system in the next morning in order to avoid energy lost from HTF in system pipes to surroundings at night.  Decreasing the length and diameter of the pipe in the solar collection loop reduces the system heat capacity, the heat loss, and the preheating time and energy required by the system. And these dimensions also affect the pressure loss and the energy required for circulating the HTF in the solar collection loop. The length and diameter of the pipe line should be designed as small as feasible within the limitations placed by the pressure loss and pumping energy for the loop in order to optimize the system performance in both solar cooling and heating.  The constant temperature control should be applied in a high temperature solar cooling and heating system to achieve higher system performances since it effectively reduces the time and energy for system preheating of a solar thermal system with a large system heat capacity. Its advantage, however, is dependent on the heat capacity of the system. In addition, the design and operation of a solar absorption cooling and heating system should consider the building and its load profiles, the climate conditions and the incident solar radiation profiles, the physical limitations of the situation, and also the economics of the energy supply. References Duff, W.S., Winston, R., O’Gallagher, J.J., Bergquam, J., Henkel, T., 2004. Performance of the Sacramento demonstration ICPC collector and double effect chiller. Solar Energy 76, 175–180. Hewett, R., 1995. Solar absorption cooling: an innovative use of solar energy. AIChE Symposium Series 91 (306), 291–299. Kulkarni, P.P., 1994. Solar absorption cooling for demand-side management. Energy Engineering 91 (5), 29–39.

182

M. Qu et al. / Solar Energy 84 (2010) 166–182

Qu, M., 2008. Model Based Design and Performance Analysis of Solar Absorption Cooling and Heating System. Carnegie Mellon University, Pittsburgh.

Stine, W.B., 1985. Energy Fundamentals and Design: With Computer Applications. John Wiley and Sons, New York.