Simulation Study of Performance of Active Ceilings with Phase Change Material in Office Buildings under Extreme Climate Conditions.
As the population continues to grow, the energy use of buildings is expected to increase. An increased focus on energy efficiency is needed in order to lower the energy use in buildings. To achieve a lower energy use in buildings, new research is necessary. Studies of the use of Phase Change Material (PCM) have previously shown that PCM has the potential to store energy in building fabrics, because the heat storage capability of PCM is larger than other conventional thermal storage materials (Mehling and Cabeza 2008; Kosny 2015; Dincer and Rosen 2002). There are two types of heat storage, latent and sensible (Dincer and Rosen 2002). PCM has a large specific latent heat capacity and is therefore suitable in building fabrics. Because of PCMs ability to store energy, PCM increases the thermal mass of a building, without increasing the physical mass of the building (Babiak, Olesen and Petras, 2009). Furthermore, studies have shown that adding PCM reduces temperature fluctuations in ventilated rooms, and thereby improves the thermal environment (Richardson and Woods, 2008). An area that lacks research regarding PCM is how PCM performs in extreme climates, which will be the main target of this study. In this study organic paraffin PCM was used, due to organic PCMs capability to melt and freeze repeatedly without phase segregation (Kosny 2015). The melting point of the PCM is 23[degrees]C (73[degrees]F). The PCM used was integrated as microencapsulated particles into gypsum ceiling panels in a two-person office. The ceiling panels had embedded pipes with water flowing in a pipe circuit. The idea behind the active ceiling with PCM is that the PCM absorbs and stores the heat from the office room during the day, while discharging it during nighttime by circulating cold water in the embedded pipes and thereby making it able to store heat again.
Eight different climates were chosen for this study. Some of them are considered extreme climates in terms of hot or cold temperatures, and dry or humid. The tested climates in this study are listed in Table 1 and are catagorized using Koppen-Geiger classification.
Two models were built for each climate; a model with an active ceiling with PCM and one all-air ventilation model. This is to be able to compare the two models and see the effects of using PCM in active ceilings for each climate. Both the indoor climate and the yearly energy use were examined and compared based on the European Standard EN15251(2007) Category II. The main target is to keep operative temperatures within the range of 20[degrees]-24[degrees] (68[degrees]F-75[degrees]F) during winter and 23[degrees]C -26[degrees] (73[degrees]F-79[degrees]F) during summer, the C[O.sub.2] concentration below 900 ppm and the PPD below 10%. Some of the tested climates have summer or winter periods all year round, if compared to European climates, so the specific summer or winter temperature ranges were compared for these locations. Commercially available building simulation software was used for the simulations. Eight different climate files were used in the simulations. The climate files, ASHRAE IWEC2, were downloaded directly from an application in the software.
The model is built as a two-person office located at an intermediate floor with a total of two zones. The occupancy zone from the internal floor to a suspended ceiling, and a ventilated plenum zone from the suspended ceiling to the roof. The office had a total floor area of 22.9 [m.sup.2] (246 [ft.sup.2]) and a height from the floor to the suspended ceiling of 2.7 m (9 ft). The total volume of the occupancy zone was 61.8 [m.sup.3] (2182 [ft.sup.3]). The height of the ventilated plenum was 0.5 m (1.8 ft). The office had three external surfaces, a wall facing south, a wall facing east and a roof. A 5.8 [m.sup.2] (63 [ft.sup.2]) window was located in the south wall. When simulating climates on the southern hemisphere the model was turned 180[degrees], which meant that the window was facing north instead of south. The reason for this was to expose the window facade to direct solar radiation.
The U-values of the different construction parts are listed in Table 2, as well as the requirements according to the current Danish building code, BR15.
Due to differences in the tested climates the models were adapted to the corresponding climates to ensure the validity of the extracted reults. In the cold climates, Moscow, Nuuk and Tromso, the insulation thickness of the exterior walls and the roof were doubled from 150 mm to 300 mm (6 in to 12 in) and from 200 mm to 400 mm (8 in to 16 in) respectively to prevent too many occupancy hours below 20[degrees]C (68[degrees]F). The increase in insulation lowered the U-value of the exterior walls to 0.12 W/[m.sup.2]K (0.02 Btu/[ft.sup.2]hr[degrees]F) and to 0.09 W/[m.sup.2]K (0.016 Btu/[ft.sup.2]hr[degrees]F)) for the roof. Solar shading was installed in all climates to prevent overheating. To lower the amount of energy used for heating in the colder climates the solar shading was scheduled to be operating from 12 PM to 4 PM only. Two persons with a 9-hour working day schedule from 8 AM to 5 PM are set to occupy the room from Monday to Friday throughout the year. The activity level of the occupants is set to 1.2 met. The clothing factor of the occupants varied according to the different climates, ranging from 0.85 [+ or -] 0.25 clo in the colder climates to 0.5 [+ or -] 0.1 clo in the warm climates. Furthermore, the room had two computers and two desk lamps with a total heat load of 200 W (682 Btu/h). Lighting was set to 21 W (72 Btu/h) in total, the total internal heat gain being 17.3 W/[m.sup.2] (5.5 Btu/h*[ft.sup.2]).
The PCM ceiling panels consisted of the 18 mm (0.7 in) PCM layer between two 1 mm (0.04 in) layers of aluminum. In Table 3 the properties of the PCM are listed. Because of numerical problems in the software the thermal conductivity of liquid was changed to the value for solid.
The active ceiling was built as a 2-zone radiant heating and cooling system. The two zones cover a total area of 19.1 m2 (206 [ft.sup.2]) and have a maximum power of 100 W/[m.sup.2] (32 Btu/hr*[ft.sup.2]), which makes the total maximum power of the active ceiling 1911 W (6500 Btu/hr) for both heating and cooling purposes. The water supply temperature for cooling was 15[degrees]C (59[degrees]F). The cooling signal was activated if the operative temperature exceeded 24.5[degrees]C (76[degrees]F) and at the same time the surface temperature of the PCM layer exceeded 22[degrees]C (72[degrees]F). Both had a deadband of 1.5[degrees]C (3[degrees]F). The water supply temperature for heating depended on the ambient temperature. The temperature ranged from 40 to 60[degrees]C (104 to 140[degrees]F). The temperatures were at these levels as it also supplied the coils of the air handling units. A VAV mechanical ventilation system with an air flow varying from 1.7 ACH to 8.7 ACH was applied to both the office zone and the plenum. To avoid overheating the air flow rate would maximize to 8.7 ACH only if the temperature exceeded 24.5[degrees]C (76[degrees]F). The air supply temperature was 17[degrees]C (63[degrees]F) throughout the year for all examined locations.
A model with only mechanical ventilation was built to compare with the PCM model in each climate. The build-up of the all-air models was the same that of the PCM model, except for the radiant ceiling panels. The ventilation system in this model had an air supply temperature for heating that depended on the ambient temperature, as shown in Figure 1. The maximum air supply temperature was 23[degrees]C (73[degrees]F) if the ambient temperature was -20[degrees]C (-4[degrees]F). The lowest air supply temperature was 16[degrees]C (61[degrees]F) if the ambient temperature was 35[degrees]C (95[degrees]F). The maximum air flow of the system was 8.7 ACH.
When calculating the energy use of the models, primary energy factors were used. A PEF for district heating was 0.8 according to the Danish building regulations, BR15. The PEF for district cooling was 0.4, and for electricity 2.0.
Further details regarding the simulation study can be found in (Stefansen and Farhan, 2017).
RESULTS AND DISCUSSION
In order to analyze the performance of the active ceiling with PCM the PCM model was compared to the all-air model for each climate. In the following the models are compared, according to yearly simulations in terms of operative temperature, PPD, and the annual energy use. The total number of occupied hours were 2349. The C[O.sub.2] concentration was 900 ppm or lower, which meets the recommended criteria of Category II of EN15251, stating the C[O.sub.2] concentration should not be more than 500 above outdoor concentration, which was set to 400 ppm for each climate. The C[O.sub.2] concentration was consequently omitted in the results.
Table 4 shows the minimum and maximum operative temperatures as well as the number of occupancy hours with operative temperature below 20[degrees]C (68[degrees]F), and above 26[degrees]C (79[degrees]F).
In all cases the minimum operative temperature for the PCM models did not go below 20[degrees]C (68[degrees]F), except for Moscow with 19.8[degrees]C (68[degrees]F), although this temperature drop only occurred for a short period of time, which is why the "hours below 20[degrees]C (68[degrees]F)" is 0 hours for the Moscow PCM model. On the other hand, the all-air system in the cold climates of Moscow, Nuuk and Tromso was not able to maintain all its operative temperatures above the recommend design temperature of 20[degrees]C (68[degrees]F) throughout the year, as the "hours below 20[degrees]C (68[degrees]F)" are 16, 24 and 5 hours respectively. From Table 4 a pattern is also seen in the maximum operative temperatures when comparing the two diffenrent models. In all cases the PCM models had lower peak room temperatures than the models with the all-air system, which could be due to the PCM panels absorbing the heat from the room and thereby providing a more comfortable thermal indoor environment. It should be noted that none of the models are having hours above the recommend design value of 26[degrees]C (79[degrees]F). The PCM models show less temperature fluctutations than the models with the all-air system in the cold and temperate climates. The warm climates Dubai, Salvador and Lima show a higher range in minimum and maximum temperatures, which may affect the thermal comfort of the occupants negatively. The reason for the minimum temperature being lower in the PCM models is due to the temperature set points for cooling of the active ceiling being 24.5[degrees]C (76[degrees]F) for the operative temperature and 22[degrees]C (72[degrees]F) for the surface temperature of the PCM layer both with a deadband of 1.5[degrees]C (3[degrees]F). The set points could be optimized in these climates by increasing them, which could provide a better indoor environment and potentially reduce the energy use for cooling as well.
Predicted Percentage of Dissatisfied (PPD)
Another important factor, when analyzing the quality of indoor environment, is the Predicted Percentage of Dissatisfied (PPD). In Table 5 the maximum PPD levels show that the PPD is lower for all PCM models except for the hottest climates of Dubai and Salvador with respectively 11.7% and 7.0%. Multiple factors affect the PPD such as clothing factor, metabolic rate, operative temperature, radiant temperature, air velocity and relative humidity.
As the maximum operative temperatures were found to be lower for the PCM models, and the active ceiling only being in function during unoccupied hours, this could be caused by the ventilation system maximizing the air volume as the ventilation system will maximize the air volume supplied if operative temperatures above 24.5[degrees]C (76[degrees]F) were reached. Still, when looking at the "hours within Category II" column, it can be seen that the only PCM model with hours outside Category II is Dubai with only 2 hours. Temperate climates like Istanbul and Tokyo have the highest number of hours outside Category II for the all-air system with 161 hours and 441 hours respectively. Table 4 showed that no overheating hours occurred, so this could be due to temperatures above 25[degrees]C (77[degrees]F) together with the clothing factor being too high, as it is set to 0.85 [+ or -] 0.25 clo, which is better suited for colder climates. The colder climates Moscow, Nuuk and Tromso also have hours outside Category II with 49, 28 and 4 hours respectively, which could be linked with too many hours below the temperature of 20[degrees]C (68[degrees]F). All occupied hours were within Category III. Based on the analysis of the indoor environment it is clear that the PCM models are ensuring a more comfortable thermal indoor environment compared to the all-air system, as the PCM models in every case fulfilled the recommended criteria of EN15251. Especially in the colder climates, the indoor environment was found to be better for the PCM models than the all-air system as the PCM increases the thermal mass.
Annual energy use
Table 6, shows the energy use for heating, cooling, HVAC auxiliary and the total energy use for both models for each climate.
When analyzing the energy used for heating in the "Heating" column in Table 6, it is seen that the difference of the amount of energy used is large for the climates that had higher heating demands. The highest difference in energy use for heating, when comparing the two models, is found in the coldest climates Moscow, Nuuk and Tromso. One of the reasons for the differences in energy use for heating is that the thermal mass of the room in the PCM models was increased by having PCM. Another important cause, is that the PCM stores the heat during the day and releases it during nighttime when the occupants have left the office. A lesser need for district heating is consequently needed for the PCM models and turns out to be more energy efficient as a heating system because of the combination of convection and radiation from the radiant ceiling system. From Table 6 it can also be seen that the hot climates like Dubai and Salvador use large amounts of energy for cooling to keep temperatures below 26[degrees]C (79[degrees]F) for both the PCM models and the all-air models, while nothing was used for heating. For Dubai the energy use for cooling was lower for the PCM model with 44 kWh/[m.sup.2] (13900 Btu/[ft.sup.2]), against 50 kWh/[m.sup.2] (15900 Btu/[ft.sup.2]) for the all-air model. For Salvador the energy use for cooling was higher for the PCM model than the all-air model with 57.3 kWh/[m.sup.2] (18200 Btu/[ft.sup.2]) versus 56.1 kWh/[m.sup.2] (17800 Btu/[ft.sup.2]) for the all-air model. The differences between the energy uses of the two models on cooling was low in both climates. The low difference in energy use for cooling could be due to the PCM being fully charged early in the day, as it has reached its heat storage capacity because of high ambient temperatures. In the more temperate climates, Istanbul, Lima and Tokyo it is noticed that the energy used for cooling was also higher for the PCM models than the all-air models, although less energy is used on HVAC auxiliary (the energy used by operating pumps and fans) for all PCM models except for Lima. The energy use of cooling was higher for the PCM models in all climates except for Dubai. This indicates that the controls of the active ceiling system should be optimized to improve the performance of the PCM models. In Lima the water-based system in the PCM model used more energy on HVAC auxiliary than the all-air model. This is also an indication that the water-based system needs modification of the controls to improve the performance. The differences between the energy used on heating for the two models are of a significant level and the active ceiling with PCM could for this cause also be beneficial in buildings located in temperate and cold climates. The energy use for lighting was 2.2 kWh/[m.sup.2] (700 Btu/[ft.sup.2]) for all simulated models.
The differences in energy use are shown in Table 7. By comparing the two types of models percentage-wise it can easily be seen in which climates the PCM models are most efficient and thereby best suited. The highest differences in total energy use by the two types of models are found in the cold climates, Nuuk, Tromso and Moscow where also the highest need for heating were found.
The highest difference was found in Nuuk with a total energy use for the all-air model of 75.8 kWh/[m.sup.2] (24000 Btu/[ft.sup.2]), and 43.8 kWh/[m.sup.2] (13900 Btu/[ft.sup.2]) for the PCM model. The Nuuk PCM model had a total energy use of 32 kWh/[m.sup.2] (10100 Btu/[ft.sup.2]) lower than the all-air model, which means that the PCM model was 42% more energy efficient than its counterpart. Second is Tromso with 39% and Moscow third with 30%. The PCM model of the temperate climate Istanbul, was 24% more energy efficient caused by a significant difference in energy use of heating between the PCM model and the all-air model. In all other cases except for Lima, the PCM models were also proving to be more energy efficient when comparing total energy use. The cause of the all-air system performing better in Lima could be due to the lack of heating demand in combination to the cooling demand being too low, however the HVAC auxiliary was also higher for the PCM model.
In a future study it could be interesting to compare the PCM model with a totally similar model without the PCM panels. Furthermore, the design and operation of the current system will greatly benefit from an exergy analysis. An exergy analysis of this system will be carried out in the future and will be reported separately in further studies.
In a simulation study of eight different climates, a model with active ceiling with PCM was compared to an all-air system in terms of the created thermal indoor environment and annual energy use. The study showed that the PCM model provided a more comfortable thermal indoor environment caused by lower maximum operative temperatures compared to the all-air system, disregarding the climate, which also led to a lower PPD level. The PCM models fulfilled the recommended criteria of Category II of temperature, C[O.sub.2] concentration and PPD of EN15251 in all tested climates. This was achieved while also lowering the annual energy use in all climates except for Lima. The biggest difference in total energy use was found in the cold climates Nuuk, Tromso and Moscow as the differences were 30% or higher. Although PCM are mostly used for cooling applications in offices, this study showed that PCM could also be beneficial for heating applications.
Babiak, J., Olesen, B. W., & Petras, D. (2009). Low temperature heating and high temperature cooling. Rehva.
CEN, E. (2007). 15251, Indoor environmental input parameters for design and assessment of energy performance of buildings addressing indoor air quality, thermal environment, lighting and acoustics. European Committee for Standardization, Brussels, Belgium.
Dincer, I., & Rosen, M. (2002). Thermal energy storage: systems and applications. John Wiley & Sons.
Kosny, J. (2015). PCM-enhanced building components: An application of phase change materials in building envelopes and internal structures. Springer.
Mehling, H., & Cabeza, L. F. (2008). Heat and cold storage with PCM. Berlin: Springer.
Richardson, M.J. & Woods, A. W. (2008). An analysis of phase change material as thermal mass. London: Physical and engineering sciences.
Stefansen, C & Farhan, H.R. (2017). Performance of active ceilings with PCM in buildings. Technical University of Denmark
Student Member ASHRAE
Student Member ASHRAE
Bjarne W. Olesen, Ph.D.
Student Member ASHRAE
Ongun B. Kazanci, Ph.D.
Associate Member ASHRAE
Casper Stefansen a B.Eng graduate, Hajan Farhan MSc. Student, Eleftherios Bourdakis Ph.D student, Ongun B. Kazanci post doc and Bjarne W. Olesen a professor, Technical University of Denmark, Lyngby
Table 1. Overview of the tested climates City Country Climate Dubai UAE Desert, dry climate Istanbul Turkey Diverse climate Lima Peru Dry carid and semiaridal climate Moscow Russia Humid continental climate Nuuk Greenland Tundra climate Salvador Brazil Tropical rainforest climate Tokyo Japan Humid subtropical climate Tromso Norway Subarctic climate Table 2. U-values for the simulation model Construction part Two- person office [W/[m.sup.2]K (Btu/([ft.sup.2] hr [degrees]F))] Roof 0.18 (0.032) External walls 0.22 (0.039) Internal walls 0.4 (0.070) Window 1 (0.176) Construction part BR15 requirements [W/[m.sup.2]K] (Btu/([ft.sup.2] hr [degrees]F))] Roof 0.2 (0.035) External walls 0.3 (0.053) Internal walls 0.4 (0.070) Window 1.8 (0.317) Table 3. Properties of simulated PCM Material properties PCM SI Peak melting point 23.4 [+ or -] 0.2[degrees]C Density, solid, [??] 910 kg/[m.sup.3] Latent heat capacity 201 kJ/kg Specific heat, solid, [C.sub.p] 1840 J/kgK Specific heat, liquid, [C.sub.p] 1990 J/kgK Thermal conductivity, solid, [lambda] 0.25 W/mK Thermal conductivity, liquid, [lambda] 0.15 W/mK Material properties PCM IP Peak melting point 74 [+ or -] 32[degrees]F Density, solid, [??] 56.8 lb/[ft.sup.3] Latent heat capacity 86.4 Btu/lb Specific heat, solid, [C.sub.p] 0.439 Btu/lb[degrees]F Specific heat, liquid, [C.sub.p] 0.475 Btu/lb[degrees]F Thermal conductivity, solid, [lambda] 0.145 Btu/hr*ft*[degrees]F Thermal conductivity, liquid, [lambda] 0.087 Btu/hrft*[degrees]F Table 4. Operative temperatures for each climate Operative Min. operative Max. operative Hours below temperature temperature temperature 20[degrees]C [[degrees]C [[degrees]C (68[degrees]F) ([degrees]F)] ([degrees]F)] Dubai PCM 22.5 (73) 25.3 (78) 0 All-air 24.5 (76) 25.6 (78) 0 Istanbul PCM 20.5 (69) 25.1 (77) 0 All-air 20.1 (68) 25.4 (78) 0 Lima PCM 20.7 (69) 25.4 (78) 0 All-air 22.6 (73) 25.6 (78) 0 Moscow PCM 19.8 (68) 24.9 (77) 0 All-air 19.7 (68) 25.4 (78) 16 Nuuk PCM 20.8 (69) 23.7 (75) 0 All-air 19.6 (67) 25.2 (77) 24 Salvador PCM 22.9 (73) 25.5 (78) 0 All-air 25.1 (77) 25.6 (78) 0 Tokyo PCM 20.8 (69) 25.0 (77) 0 All-air 20.0 (68) 25.5 (78) 0 Tromso PCM 20.6 (69) 24.3 (76) 0 All-air 19.8 (68) 25.4 (78) 5 Operative Hours above temperature 26[degrees]C (79[degrees]F) Dubai 0 0 Istanbul 0 0 Lima 0 0 Moscow 0 0 Nuuk 0 0 Salvador 0 0 Tokyo 0 0 Tromso 0 0 Table 5. PPD for each climate Predicted percentage Maximum Hours within Hours outside of dissatisfied PPD [%] Category II Category II (PPD) < 10% Dubai PCM 11.7 2347 2 All-air 6.7 2349 0 Istanbul PCM 9.7 2349 0 All-air 11.2 2188 161 Lima PCM 8.0 2349 0 All-air 8.4 2349 0 Moscow PCM 9.7 2349 0 All-air 12.0 2300 49 Nuuk PCM 8.4 2349 0 All-air 12.3 2321 28 Salvador PCM 7.0 2349 0 All-air 6.4 2349 0 Tokyo PCM 9.7 2349 0 All-air 11.3 1908 441 Tromso PCM 8.0 2349 0 All-air 11.1 2345 4 Table 6. Energy use for simulation models Heating Cooling Energy use [kWh/[m.sup.2] [kWh/[m.sup.2] (Btu/[m.sup.2])] (Btu/[m.sup.2])] Dubai PCM 0.0 (0) 44.0 (13900) All-air 0.0 (0) 50.0 (15900) Istanbul PCM 13.5 (4300) 9.9 (3100) All-air 23.5 (7400) 8.0 (2500) Lima PCM 0.0 (0) 20.7 (6600) All-air 0.0 (0) 18.2 (5800) Moscow PCM 22.9 (7300) 4.7 (1500) All-air 37.3 (11800) 3.3 (1000) Nuuk PCM 37.1 (11800) 0.5 (200) All-air 65.4 (20700) 0.0 (0) Salvador PCM 0.0 (0) 57.3 (18200) All-air 0.0 (0) 56.1 (17800) Tokyo PCM 8.0 (2500) 15.3 (4900) All-air 12.6 (4000) 11.9 (3800) Tromso PCM 26.3 (8300) 1.0 (300) All-air 45.5 (14400) 0.2 (100) HVAC aux. Total Energy use [kWh/[m.sup.2] [kWh/[m.sup.2] (Btu/[m.sup.2])] (Btu/[m.sup.2])] Dubai 0.8 (300) (*) 47.0 (14900) 2.1 (700) (*) 54.3 (17200) Istanbul 4.7 (1500) 30.3 (9600) 6.5 (2100) 40.1 (12700) Lima 5.1 (1600) 28.0 (8900) 2.9 (900) 23.3 (7400) Moscow 4.6 (1500) 34.4 (10900) 6.4 (2000) 49.1 (15600) Nuuk 4.1 (1300) 43.8 (13900) 8.2 (2600) 75.8 (24000) Salvador -0.9 (-300) (*) 58.5 (18500) 1.2 (500) (*) 59.4 (18800) Tokyo 4.0 (1300) 29.4 (9300) 5.0 (1600) 31.6 (10000) Tromso 4.2 (1300) 33.7 (10700) 7.7 (2400) 55.6 (17600) (*) When simulating for Salvador the HVAC auxiliary shows a negative energy use, which is considered an error in the software. The same goes for the low energy use for HVAC auxiliary for Dubai, as some of the simulated months show a negative value almost covering the energy use for operating the fans. Table 7. Difference of PCM models total energy use vs. All-air Total energy use Difference Difference [kWh/[m.sup.2] (Btu/[ft.sup.2])] [%] Dubai -7.3 / (-2700) 13 Istanbul -9.7 /(-3100) 24 Lima 4.7 / (1500) -20 Moscow -14.7 / (-4700) 30 Nuuk -32 / (-10100) 42 Salvador -0.9 / (-300) 2 Tokyo -2.2 / (-700) 7 Tromso -21.9 / (-6900) 39
|Printer friendly Cite/link Email Feedback|
|Author:||Stefansen, Casper; Bourdakis, Eleftherios; Olesen, Bjarne W.; Farhan, Hajan; Kazanci, Ongun B.|
|Publication:||ASHRAE Conference Papers|
|Date:||Jan 1, 2018|
|Previous Article:||Simulation Study of Active Ceilings with Phase Change Material in Office Buildings for Different National Building Regulations.|
|Next Article:||The Influence of a Radiant Panel System with Integrated Phase Change Material on Energy Use and Thermal Indoor Environment.|