UNISOL – Solar Combistore Evaluation and Optimization

Available online at www.sciencedirect.com ScienceDirect Energy Procedia 48 (2014) 264 – 272 SHC 2013, International Conference on Solar Heating and Cooling for Buildings and Industry September 23-25, 2013, Freiburg, Germany UNISOL – solar combistore evaluation and optimization. Ricardo Amorim*a, Jorge Facãoa, João C. Rodriguesa, Maria João Carvalhoa a LNEG – Laboratório Nacional de Energia e Geologia, Estrada do Paço do Lumiar,22, 1649-038 Lisboa, Portugal Abstract In the frame of UNISOL project, a test bench was installed to perform tests of a combistore which includes a two-way heat exchanger already submitted to a national patent application. The present work describes the main characteristics of the test bench installed and the tests performed with the objective of optimization of the configuration of the inner storage tank of the combistore (two way heat exchanger), used for DHW pre-heating or as back-up of the space heating. Tests according to EN 12977-3:2012 were performed in order to evaluate distinct configurations of the inner storage tank. Tests according to EN 129774:2012 were also performed for characterization of the complete combistore. Long-term performances of solar thermal systems using these combistore are presented. Long-term performance prediction based on testing results according to EN 12977-3:2012, showed how an increased active volume of the heat exchanger enhances the energy available for space heating, although it may decrease solar collector thermal performance and increase the energy losses of the combistore. Long-term performance prediction based on testing results according to EN 12977-4:2012, considering a lower heat loss coefficient since a better insulation of the combistore is expected in future prototypes, showed fsav values for Davos and Wurzburg of 39.3% and 25.3%, respectively. © 2014The TheAuthors. Authors. Published by Elsevier © 2014 Published by Elsevier Ltd. Ltd. SHCunder 2013responsibility under responsibility Selection andpeer peerreview review scientific conference committee Selection and byby thethe scientific conference committee of SHCof2013 of PSE AGof PSE AG. Keywords: Combistore; Optimization; Domestic Hot Water (DHW); Space Heating (SH) 1. Introduction UNISOL is a national project aiming at the development of an innovative, autonomous and intelligent universal system for management and accumulation of solar heat that can practically use any solar collector in the market [1]. The system will simultaneously pre-heat domestic water (DHW) and space heating (SH). The main component is a * Corresponding author. Tel.: +351 210924766; fax: +351210922641. E-mail address: ricardo.amorim@lneg.pt 1876-6102 © 2014 The Authors. Published by Elsevier Ltd. Selection and peer review by the scientific conference committee of SHC 2013 under responsibility of PSE AG doi:10.1016/j.egypro.2014.02.031 Ricardo Amorim et al. / Energy Procedia 48 (2014) 264 – 272 265 combistore which includes a two-way heat exchanger already submitted to a national patent application (patent n. 10561) [2]. A test bench was installed in order to evaluate and optimize the performance of the combistore developed in the frame of UNISOL project [1]. The present work describes the main characteristics of the test bench installed. Numerical simulations were performed in the software TRNSYS [3] in order to test the selected numeric model Type 340 [4] and optimization algorithm available in GENOPT [5] according to EN 12977-3:2012 [6], respectively, Annex A and C. The errors were calculated according to equations in Annex B of the referred standard and it was possible to conclude that the selected methodology is able to fulfill EN 12977-3:2012 [6] requirements [7]. Two storage prototypes were tested according to EN 12977-3:2012 [6]. The prototypes will be referred as prototype 1 and 2. The objective of these tests was to evaluate the heat transfer capacity between the inner tank and the exterior tank. Only prototype 2 was tested according to EN 12997-4:2012 [8]. Two CFD simulations [9] were also performed to evaluate the heat transfer capacity of prototype 2 considering different inlet and outlet positions. Future work will be testing of prototype 2 with changed positions of inlet and outlet of the heat exchanger since CFD simulations showed that this solution would have a better heat transfer capacity. Nomenclature fsav Q fractional energy savings [%] annual energy [GJ] subscripts aux,net Loss solar sh d hw auxiliary net losses to ambient solar energy space heating demanded domestic hot water 2. Description of test rig 2.1. Test bench, data acquisition and software The test bench was installed following the requirements of EN 12977-3:2012 [6] and EN 12977-4:2012 [8]. In Fig.1 and 2, photos of the test bench are shown. The main features of the data acquisition system are the use of an Agilent 34970A Multimeter with one card 34901A with 20 reading channels, allowing the reading of voltage, current and resistance for reading of temperature sensors and flow meters and two multifunction cards, 34907A, with 16 digital I/O TTL, 2 analog outputs (digital to analog converters) of ±12 Volt and one input counter, for actuation of electromechanical valves, circulation pumps and a motorized valve. Fig. 1. General view of the testing bench (front view) (left) and view of first prototype of the combistore during tests (right). 266 Ricardo Amorim et al. / Energy Procedia 48 (2014) 264 – 272 The software necessary for the data acquisition and control of the test bench was developed in Visual Studio ®. It includes not only the functions of sensors reading but also the functions of command and control of the hydraulic circuit that constitutes the test bench. 2.2. – The Combistore In Fig.2 a schematic presentation of the combistore is shown. In order to identified parameters for the configuration of the inner storage tank, used for DHW pre-heating or as back-up of the space heating, tests according to EN 12977-3:2012 [6] were performed. The work presents the identified parameters for both store configurations. Fig. 2. Schematic representation of the combistore of the type tank in tank where the tank inside works as pre.heat for DHW preparation (Left) or back-up for SH (right). Fig. 2 shows the used ports for space heating, external storage solar heating, inner tank for DHW preparation and backup heat exchanger. Fig.2 also shows the flow direction for each port. The external store has multiple ports for water extraction for space heating at several heights (green circles in Fig.2). In the current work it is only used the connections indicated with arrows in Fig. 2. The external tank has a nominal volume of 600 liters with internal dimensions of 800 x 400 x 2000 mm (width x length x height). The inner tank is totally immersed in water and has a nominal volume of 120 liters for domestic hot water preparation. The exterior tank is not pressurized and it was filled with water until a height of 1900mm. 3. Performed tests To use the data acquired in the laboratory it was necessary to build a CSV file to feed the data into Type 340 [4]. This file compiles the tests necessary to identify the parameters that describe the thermal behavior of the store. This file is defined as a sequence of tests, and it is referred in this text as a sequence. Ricardo Amorim et al. / Energy Procedia 48 (2014) 264 – 272 3.1. Tests according to EN 12977-3:2012 In Table 1 are listed the performed tests for both prototypes and the group qualification based in EN 129773:2012 [6]. For the prototype 2 it was only necessary to perform two more tests since the only change was the length and diameter of the tube that delivers hot water to the inner tank. Since there are two prototypes, two CSV files were built as listed in Table 2. Table 2 lists the tests contained in each sequence and the respective sequence name. Table 1. Performed tests for both prototypes and the group qualification based in EN 12977-3:2012 [4]. Test name Group Prototype Abbreviation Test C 1 1 C1 Test S 1 1 S1 Test L 1 1 L1 Test NiA 2 1 NiA1 Test NB 2 1 NB1 Test NiA 2 2 NiA2 Test NB 2 2 NB2 Table 2. List of tests contained in each sequence based on Table 1. Sequence Name Tests Prototype Sequence1 C1; S1; L1; NiA1; NB1 1 Sequence2 C1; S1; L1; NiA2; NB2 2 3.2. Tests according to EN 12977-4:2012 Tests according to EN 12977-4:2012 [8] were performed for characterization of the complete combistore. Since this combistore can be classified in different ways according to the charge/discharge ports (see Table 1 of reference [8]), Table 3 shows the classification and configurations considered for testing of the combistore. The performed tests based on Table 3 are listed in Table 4. The tests listed in Table 4 were compiled in a CSV file called Sequence3. Table 3. Classification of the combistore according to EN 12977-4:2012 [7]. Group Charge Mode Port Discharge Mode Port 1 Direct Solar Direct SH 2 Indirect Inner tank as backup of SH Direct SH 3 Direct Solar indirect Inner tank as pre-heat of DHW 267 268 Ricardo Amorim et al. / Energy Procedia 48 (2014) 264 – 272 Table 4. Performed tests for the prototype 2 based on the group classification listed in table 3. Test name Abbreviation Test CD CD1 Test DD DD1 Test CI CI1 Used the test NiA2 DI1 Store Fully Charged + low flow rate DI2 Store Fully Charged + high flow rate DI4 Aux. Fully Charged + low flow rate DI5 Aux. Fully Charged + high flow rate Test DI Obs. 4. Long-term performance prediction In order to calculate the thermal performance of the tested combistore a new deck was created in TRNSYS [3]. This new deck has 4 circuits, the solar collector loop, auxiliary loop, space heating loop and DHW loop. The solar loop consists of a 10 m2 flat plate collector area and a flow rate of 300 l/h in the loop. It was considered that the collector has an optical efficiency of 0.8, overall heat losses of 3 W/K and the second order heat losses coefficient of 0.01 W/m2.ºC2. The auxiliary heater for space heating has a maximum power of 24 kW and a flow rate of 600 l/h. The controller for the auxiliary heater has a set point of 47 ºC and a turn on difference of 5ºC. This means that the auxiliary heater loop stops when controller sensor in the outside tank has a temperature of 47 ºC and turns on when the temperature falls below 42 ºC. This sensor is located at a relative height of 0.5. Also it was considered that the auxiliary heater is activated when the loop of space heating is on. The set-point is lower than the set-point considered in reference conditions of EN 12977-2:2012 [10] considering that the space-heating loop will be preferably a heated floor. Also the fact that the system will work as pre-heat system for DHW, i.e., in the DHW loop the auxiliary heater is in series with inner store. The space heating loop is simulated for each city using the data in the input file (flow temperature, return temperature and mass flow rate) as described in EN 12977-2:2012 [10]. The mains temperature is calculated according to EN 12977-2:2012 [10] and is pre heated in the inner tank. The necessary power is provided to the DHW stream by an auxiliary heater in series with the inner tank. A mixing valve with an objective temperature of 45ºC is used to supply the heat load for DHW and the draw off flow rate is 600 l/h. 5. Results and discussions Based on the tests performed on the laboratory and the sequences described in Tables 2 and 4, three different optimizations were performed. For the sack of simplicity in this section only the best results of the identification of parameters process are presented. 5.1. According to EN 12977-3:2012 Table 5 resumes the parameters identified with Sequence1 and Sequence2, respectively, prototype 1 and prototype 2. A better fitting and a stratified charging was achieved considering that the heat exchanger relative inlet position is 1 and the outlet position is optimized, but physically, as shown in Fig. 2, the heat exchanger has its inlet position almost at the bottom and the outlet at the top. As listed in Table 5 the relative position of the heat exchanger outlet changes from 0.275 to 0.063. This was expected since the length of the new tube that delivers the hot water in the heat exchanger is longer. The other parameters will not be discussed or compared here. To compare the differences between the two inner tanks, in this section, we will present the results from a long-term performance prediction considering only the heat load of space heating. 269 Ricardo Amorim et al. / Energy Procedia 48 (2014) 264 – 272 Table 5. Parameters identified based on Sequence1 and Sequence2 listed in Table 2. Parameter name Sequence1 Sequence2 Unit Description DP1in 0.122 0.153 - S.H. port relative inlet position DP1out 0.950 0.942 - S.H. port relative outlet position DP2in 0.975 0.878 - Solar port relative inlet position Hx1out 0.275 0.063 - Heat exchanger port relative outlet position Hx1b3 0.494 0.39 - Heat exchanger mean temperature exponent coefficient Hx1UA 48.264 92.8 W/ºC Heat exchanger constant heat transfer coefficient Hx1Vol -82.4 -75.2 l Heat exchanger volume stoDeg 0.313 2.292 W/m.ºC Storage degradation coefficient stoLoss 8.274 9.375 W/ºC Storage lateral heat loss coefficient stoLossB 5.083 4.948 W/ºC Storage bottom heat loss coefficient stoLossT 6.380 4.37 W/ºC Storage top heat loss coefficient stoVol 658.5 652.5 l Storage volume nrNodes 65 155 - Storage stratification coefficient 5.1.1. Long-term performance prediction The deck described in section 4 was used to estimate the long-term performance prediction of the two heat exchangers with the input file of Davos (higher solar energy available and higher heat demand). The annual heat demand (Qd,sh) for space heating for Davos city is 42.3 GJ. Table 6 resumes the long-term performance prediction for both heat exchangers with and without the solar collector loop with exception of DHW load. The auxiliary energy (Qaux,net) increases 3.3% with the new heat exchanger without solar collectors and 4.4% considering the solar loop described in section 4. This is justified by the fact that the exterior tank has more volume heated by the auxiliary heater through the inner tank, which also increases the energy losses (QLoss). With the new heat exchanger the solar energy also decreases because the bottom of the tank as a higher temperature, which will impose a higher inlet temperature to the collectors. This does not mean that the new heat exchanger has a lower performance in meeting the demanded energy for space heating. For Davos, the new heat exchanger provides 99.7% (42.2/42.3) of the demanded annual energy for space heating against the 99.1% of the old heat exchanger (41.9/42.3). More simulations were performed with a different set-point (40 ºC) as listed in Table 6. With the new set-point the new heat exchanger supplies the heat demand for space heating in 94.3% and with the old heat exchanger the value drops to 89.8% with solar collectors and, respectively, 93.8% and 88.7% without solar collectors. Table 6. Predicted long thermal performance energies, Qaux,net, Qsolar and QLoss for different auxiliary set-point temperatures, considering the reference weather data and space heating load for the Davos city. Aux. setpoint = 47ºC Aux. setpoint = 40ºC Ql,sh [GJ] Qaux,net [GJ] Qsolar [GJ] QLoss [GJ] Ql,sh [GJ] Qaux,net [GJ] Qsolar [GJ] QLoss [GJ] Old Hx without solar collectors 41.9 52.4 0.0 10.5 37.5 45.6 0.0 8.0 New Hx without solar collectors 42.2 54.2 0.0 11.9 39.7 49.0 0.0 9.2 Old Hx with solar collectors 41.9 35.7 28.3 22.1 38.0 29.2 29.1 2.0 New Hx with solar collectors 42.2 37.3 27.2 22.2 39.9 32.0 28.3 2.0 270 Ricardo Amorim et al. / Energy Procedia 48 (2014) 264 – 272 5.2. According to EN 12977-4:2012 Table 7 lists the parameters identified with Sequence3. In this optimization it was fixed the inlet and outlet positions, with exception of the positions listed in table 7. The relative positions of the heat exchanger outlet, space heating inlet and solar double port inlet were fixed with the values, respectively 0.97, 0.974 and 0.158. Table 7. Parameters identified based on Sequence3 based tests listed in Table 4. Parameter name Value Unit Description DP1out 0.9 - S.H. port relative outlet position Hx1In 0.105 - Heat exchanger relative inlet position Hx1b1 0.008 - Heat exchanger mass flow exponent coefficient Hx1b3 0.386 - Heat exchanger mean temperature exponent coefficient Hx1UA 102.8 W/ºC Heat exchanger constant heat transfer coefficient Hx1Vol -84.1 l Heat exchanger volume stoDeg 1.528 W/m.ºC Storage degradation coefficient stoLoss 5.503 W/ºC Storage lateral heat loss coefficient stoLossB 5.177 W/ºC Storage bottom heat loss coefficient stoLossT 8.328 W/ºC Storage top heat loss coefficient stoVol 671.9 l Storage volume nrNodes 82 - Storage stratification coefficient 5.2.1. Long-term performance prediction The same deck as described in section 4 was used to estimate the annual energy performance of the new inner tank. The parameters of the store used in the simulations are listed in table 7. The parameters that describe the heat losses in tables 5 and 7 are high due to the fact that the insulation of the external tank has several imperfections in this prototype. The top insulation plate has several holes to facilitate the heat exchanger position and the lateral insulation plates are not completely fitted to the external storage tank. In order to study the difference between a good and a bad insulation in the fractional energy savings the parameters stoLoss, stoLossB and stoLossT were changed to the values listed in table 8. The values were calculated taking into account the conductivity coefficient of the insulation of 0.04 W/m.ºC and 100 mm of thickness. The fractional energy savings (fsav) was predicted for two European cities, Davos and Würzburg [11], considering a daily load of 110 liters per day for DHW according to EN 12977-2:2012 [10]. Table 9 shows the predicted fractional energy savings for the combisystem with the parameters that characterize the store listed in table 7. Table 10 lists the fractional energy savings (fsav) considering a better insulation and using the parameters listed in table 8. Ricardo Amorim et al. / Energy Procedia 48 (2014) 264 – 272 Table 8. Parameters used to predict the energy saving of a system with better insulation. Parameter name Value Unit Description DP1out 0.9 - S.H. port relative outlet position Hx1In 0.105 - Heat exchanger relative inlet position Hx1b1 0.008 - Heat exchanger mass flow exponent coefficient Hx1b3 0.386 - Heat exchanger mean temperature exponent coefficient Hx1UA 102.8 W/ºC Heat exchanger constant heat transfer coefficient Hx1Vol -84.1 l Heat exchanger volume stoDeg 1.528 W/m.ºC Storage degradation coefficient stoLoss 1.408 W/ºC Storage lateral heat loss coefficient stoLossB 0.128 W/ºC Storage bottom heat loss coefficient stoLossT 0.128 W/ºC Storage top heat loss coefficient stoVol 671.9 l Storage volume nrNodes 82 - Storage stratification coefficient Table 9. Predicted fractional energy savings for the system with the parameters described in table 7 considering a daily load of 110 l/day and two European cities Davos and Würzburg. Qd,hw [GJ] Qd,sh [GJ] Qaux,net [GJ] fsav [%] Davos 6.6 42.3 40.8 18.6 Würzburg 5.9 32.7 37.2 6.3 Table 10. Predicted fractional energy savings for the system with the parameters described in table 9 considering a daily load of 110 l/day and two European cities Davos and Würzburg. Qd,hw [GJ] Qd,sh [GJ] Qaux,net [GJ] fsav [%] Davos 6.6 42.3 30.5 39.3 Würzburg 5.9 32.7 29.7 25.3 As expected the thermal performance indicator fsav improves significantly with better insulation of the outside tank. The auxiliary energy reduces since less energy is lost to the ambient surrounding the outside storage. The fsav increases more than 4x for the Würzburg city and more than 2x for Davos. These values indicate that improving the insulation of the outside tank improves the annual savings. 6. Conclusions Two combistore prototypes were tested according to EN12977-3:2012 [6] and EN12977-4:2012 [8]. The main difference between the two prototypes is the tube height and diameter at which the auxiliary energy is delivered to the inner store when this inner store works as auxiliary heat exchanger. For prototype 1, auxiliary energy is delivered at almost half height of the store, while for prototype 2, auxiliary energy is delivered at the bottom of the inner tank. Using the testing results according to EN 12977-3:2012 [6] and long-term performance prediction for the city of Davos it was possible to conclude that the heat exchanger to backup the space heating load 271 272 Ricardo Amorim et al. / Energy Procedia 48 (2014) 264 – 272 interferes with the solar collector thermal performance and also with the energy losses of the combistore. Although prototype 2 reduces the solar energy delivered to the system and also imposes higher losses, it delivers more energy to the space heating. The controller set-point also influences the performance of the combisystem. Lower set points reduce the energy lost and the interference with the solar loop. Prototype 2, when tested according to EN 12977-3:2012 [6] fulfilled better the energy demanded for space heating. It was then tested as combistore according to EN 12977-4:2012 [8]. Since the prototype has imperfections in the way the insulation is applied to the store walls, it shows high heat losses coefficients (Top, Bottom and side losses). Simulations using TRNSYS [3] were performed, for a system using this combistore and a solar field of 10 m2 and delivering energy to space heating and preheating DHW. The performance indicator used is fsav (fractional energy savings) and the results for Davos and Würzburg show values of fsav, respectively, of 18.6% and 6.3%. Simulations using lower heat losses coefficients, considering a heat conduction value for the insulation of 0.04 W/m.ºC, showed large improvement in the fsav values, respectively 39.3% and 25.3%. These results show that the performance of the combisystem increases substantially with a better insulation. Future work will consist of the testing of the combistore considering the inverted positions of the inlet and outlet of the inner tank, following the conclusions of two CFD simulations [9] performed to evaluate the heat transfer capacity of prototype 2. Acknowledgements The work presented was developed in the frame of project n. 21507, UNISOL- Sistema Universal de Energia Solar, financed in the frame of QREN – Quadro de Referência Estratégico Nacional, Programa Operacional Factores de Competitividade. References [1] UNISOL- Sistema Universal de Energia Solar, project n. 21507, financed in the frame of QREN – Quadro de Referência Estratégico Nacional, Programa Operacional Factores de Competitividade. [2] Patent INPI n. 105061 - Circuito Auxiliar para Aquecimento de Acumuladores Térmicos. [3] TRNSYS, A Transient System Simulation Program, version 17.00.0019, Solar Energy laboratory, Univ. Of Wisconsin-Madison. [4] Harald Drück, MULTIPORT Store – Model for TRNSYS Stratified fluid storage tank with four internal heat exchangers,ten connections for direct charge and discharge and an internal electrical heater, Type 340, Version 1.99F, March 2006 [5] GENOPT®, Generic Optimization Program, version 3.1.0, Lawrence Berkeley National Laboratory [6] EN 12977-3:2012 (Ed. 2) - Thermal solar systems and components. Custom built systems. Part 3: Performance test methods for solar water heater stores. [7] Ricardo Amorim, 2013, “Projeto Unisol - Determinação de parâmetros de acordo com a norma EN 12977 – 3 utilizando os softwares Trnsys 17 e GenOpt 3.1.0”. LES Nº1/2013, Internal report. [8] EN 12977-4:2012 (Ed. 1) - Thermal solar systems and components. Custom built systems. Part 4: Performance test methods for solar combistores. [9] Jorge Facão, 2013, Estudo CFD da carga Indireta do depósito UNISOL, LES Nº4/2013, Internal report. [10] EN 12977-2:2012 (Ed. 2) - Thermal solar systems and components. Custom built systems. Part 2: Performance test methods for solar water heater stores. [11] http://www.estif.org/solarkeymarknew/ View publication stats