- Email: [email protected]
An Energy Efficiency Assessment of the Thermal Comfort in an Office building
Available online at www.sciencedirect.com
ScienceDirect
Available online at www.sciencedirect.com Energy Procedia 00 (2017) 000–000
ScienceDirect
www.elsevier.com/locate/procedia
Energy Procedia 134 (2017) 885–893
9th International Conference on Sustainability in Energy and Buildings, SEB-17, 5-7 July 2017, Chania, Crete, Greece
An Energy Efficiency Assessment of the Thermal Comfort in an Office building Zhidan Zhaoa, Mahdi Houchatia, and AbdlMonem Beitelmala, * 0F
Qatar Environment and Energy Research Institute, Hamad Bin Khalifa University, Qatar Foundation, Doha, Qatar
a
Abstract People spend about 90% of their time indoors, so a comfort indoor thermal environment is essential for the satisfaction, productivity and wellbeing of the building occupants. Assessment of the indoor thermal comfort is the key point for building HVAC system design and operation to provide a comfort indoor environment to building occupants. Predicted Mean Vote (PMV) model is the most widely used tool for the indoor thermal comfort assessment. In this study, the application of the PMV model in Qatar with dry, subtropical desert climate is evaluated. An experiment was conducted in an office building in Doha, Qatar to reveal the occupant perception of indoor thermal comfort. Using collected data, the PMV indexes were calculated using Fanger’s theory and compared with the Actual Thermal Sensation (ATS) of the occupants to assess the applicability of the PMV model to predict the indoor thermal comfort in air conditioned buildings in the climate zone of Qatar. The corresponding occupants’ satisfaction level with the indoor thermal comfort and their adaptive behavior were also assessed. © 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of KES International. Keywords: Thermal comfort; PMV; Occupant bahavior; Building enery efficiency
1. Introduction Indoor thermal comfort is the most important factor that determines the overall indoor environment quality which is the major concern of building occupants due to their long time stay indoors [1]. Building occupants’ perception of indoor thermal comfort normally refers to their feelings of room comfort, for example, the room is hot, cold or neutral etc. and is not a direct sensation of indoor air temperature. Due to the essential role that building occupants play in the building operation and further more in the building energy efficiency enhancement, the assessment of their indoor thermal comfort is quite important for not only building design and operation, but also for building energy efficiency improvement. Predicted Mean Vote (PMV) model which was first developed by Fanger [2] is the most widely accepted tool for the indoor thermal comfort assessment and was adopted by international standards such as ISO 7730 [3] and the ASHRAE Standard 55–92 [4] to evaluate the indoor thermal comfort conditions. In previous studies, this model has been evaluated and validated using data collected from different climate zones and different building types. Although the strength of PMV model has been proved by many studies, it was found that the PMV model didn’t perform very well for assessment of the indoor thermal comfort in naturally ventilated buildings. Therefore, the adaptive model [5] was adopted by AHRAE Standard called Adaptive Comfort Standard (ACS) as an extension tool of PMV method for the thermal comfort evaluation in naturally ventilated buildings [6]. Despite that a lot of studies validated the application of PMV model in air conditioned buildings, some recent studies indicate that there are discrepancies between the * Corresponding author. Tel.: +974-4454-7158; fax: +974-4454-1528. E-mail address: [email protected] 1876-6102 © 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of KES International.
1876-6102 © 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of KES International. 10.1016/j.egypro.2017.09.550
Zhidan Zhao et al. / Energy Procedia 134 (2017) 885–893
886
Zhidan Zhao/ Energy Procedia 00 (2017) 000–000
PMV model prediction and the Actual Mean Vote (AMV) of the building occupants while evaluating the indoor thermal comfort in air conditioned buildings [7-11]. In this study, the application of the PMV model in Qatar with dry, subtropical desert climate is evaluated. An experiment was conducted in an office building served with a central VAV air conditioning system in Doha, Qatar to reveal the occupant perception of indoor thermal comfort. The occupants’ adaptive behavior were also examined in this study. . Nomenclature PMV Ta Tmra RH Va MR CL
Predicted Mean Vote Room air temperature Mean radiant temperature Relative humidity of room air Air velocity Metabolic rate Clothing level
2. Methodology 2.1. Experiment Facility The chosen building for this study is an office building located in College of North Atlantic, Qatar numbered as building 7. It is a two story office building with gross area of about 3,184 m2. Each floor can be divided into East and West Zones. The experiment was conducted in the West zone of the ground floor with a gross area of about 657 m2. There are total 22 personal offices, 1 staff lounge and 1 equipment room in this experimental zone, which is served by three AHUs. Chilled water is provided by the campus central plant and no heating device is in use. The floor plan is shown in Fig. 1. Each office is served by one VAV terminal box controlled with local thermostat. Except office 7110a and 7110b, each office is occupied by one university instructor and 14 of the total 20 occupants agreed to participate in the study.
Fig. 1. Experiment zone floor plan.
2.2. Data Collection The PMV index should be calculated using 6 inputs according to Fanger’s theory. As shown in Equation 1, air temperature, mean radiant temperature, relative humidity, air velocity, metabolic rate and clothing level are the six factors that determine the PMV index. The former first 4 inputs are measureable parameters which can be obtained through sensors and the latter 2 inputs need evaluation through subjective studies like survey feedbacks from the occupants.
PMV f Ta , Tmra , RH , Va , MR, CL
(1)
The indoor thermal conditions like temperatures and relative humilities and indoor air velocities are monitored using sensors and recorded in the data acquisition system at 15 min intervals. However, due to the experiment conduction limitation, we didn’t get a chance to measure the mean radiant temperature in each office room. Considering the big thermal mass of the building,
Zhidan Zhao et al. / Energy Procedia 134 (2017) 885–893
Zhidan Zhao/ Energy Procedia 00 (2017) 000–000
887
shading wall outside the north external wall for sunblock function, the mean radiant temperature was assumed to be equal to the air temperature according to previous study [12]. Some sensors and online data recording interface are shown in Fig. 2.
Fig. 2. Data collection sensing system
The occupant perception of the indoor thermal comfort was evaluated through the Preference Monitoring Application (PMA). The PMA is a 3-5 minute Web-based survey that collects time-stamped occupant responses and contains about 10 questions with the opportunity to provide open-ended comments. It was created using Qualtrics survey development software accessible on both webpage and smartphones as shown in Fig. 3.
Fig. 3. PMA mobile App example
The survey was designed based on the ASHRAE standard including three major parts regarding occupants’ sensation of indoor thermal comfort on a seven-point scale from cold (-3) to hot (+3), occupants’ personal values like their activity levels and cloth levels and also their adaptive actions. Table 1 shows the rating scale with evaluated parameters. Table 2 and Table 3 provide some examples to illustrate that how to correlate the occupant reported activity levels and clothing levels with the metabolic rate values and clothing insulation values that should be used in the PMV calculation [4]. In our study, the building occupants were asked to fill the survey from time to time to give their feedbacks whenever they feel uncomfortable or very comfortable or take some energy related actions for duration of one year. To evaluate the thermal comfort in an air conditioned building, the experimental zone temperature is pushed to 18˚C as low limit and 26˚C as high limit regularly using intervention plan. From March 2016 until March 2017, total 613 effective occupant feedbacks are collected and in use for the study. Table 1. Occupant evaluated parameters and rating scale Rating scale value
Thermal comfort evaluation
Thermal comfort satisfaction
+3
Hot
Very satisfied
+2
Warm
Satisfied
+1
Slightly Warm
Moderately satisfied
0
Neutral
-
-1
Slightly cool
Moderately unsatisfied
-2
Cool
Unsatisfied
-3
Cold
Very Unsatisfied
Zhidan Zhao et al. / Energy Procedia 134 (2017) 885–893
888
Zhidan Zhao/ Energy Procedia 00 (2017) 000–000
Table 2. Occupant reported activity level and corresponding metabolic rate Reported activity level
Metabolic rate [Met]
Seated, quiet
1.0
Standing, relax
1.2
Typing
1.1
Filing, seated
1.2
Filing, standing
1.4
Walking about
1.7
Lifting/packing
2.1
Table 3. Occupant reported clothing level and corresponding clothing insulation value Reported activity level
Clothing insulation value [Clo]
Short-sleeve shirt, trousers
0.57
Long-sleeve shirt, trousers
0.61
Long-sleeve shirt, trousers, suit jacket
0.96
Short-sleeve shirt, Knee-length skirt
0.54
Long-sleeve shirt, Knee-length skirt, full slip
0.67
Long-sleeve shirt, Knee-length skirt, suit jacket
1.04
Long-sleeve shirt, Ankle-length skirt, suit jacket
1.1
3. Results and Discussion 3.1. PMV application
3
3
2
2
1
1
0
0
-1
-1
-2
-2
-3 0
100
200
300
400
CaseNumber
500
600
PMV
ATS
Based on the timestamps of the 613 occupant feedbacks, corresponding air temperature, humidity, air velocity values are located from the database. The metabolic rate quantified with “met” and clothing level quantified with “clo” were also determined according to ASHRAE standard using the occupant reported activity level such as “seated quietly” and clothing level such as “long sleeve shirt with thin trouser” as inputs. Then the PMV indexes were calculated for these 613 scenarios. The PMV and ATS values for the 613 cases are shown in Fig. 4. It was observed that there are big differences between the reported ATS from the occupants and calculated PMV. The PMV lost track of most of the occupants’ thermal comfort sensation when they evaluate the thermal comfort as neutral or above.
-3
Fig. 4. Final PMV and ATS results
The crossover comparison between the PMV and ATS is shown in Fig. 5. If the PMV could be a good estimation for the ATS in this case study, these data marks should collapse along the red line, which is not observed from the plot. It is shown that when
Zhidan Zhao et al. / Energy Procedia 134 (2017) 885–893
889
Zhidan Zhao/ Energy Procedia 00 (2017) 000–000
the occupant actual thermal sensation went beyond slight cool (-1), the PMV indexes overestimated the comfort level while underestimated the thermal comfort level when the actual occupant thermal sensation is slightly warm or even hotter. 3 2
PMV
1 0 -1 -2 -3
-3
-2
-1
0
ATS
1
2
3
4
Fig. 5. Crossover comparison between ATS and PMV
To further assess the validity of PMV application, under each occupant actual thermal comfort sensation (ATS) category, from cold (-3) to hot (3), the calculated PMV indexes are averaged to see how well the PMV can capture the real thermal comfort sensation under different thermal comfort levels. The comparison between the averaged PMV values and the ATS is shown in Fig. 6. It is observed that, when the occupant actual thermal sensation between “Neutral” and “Cool”, the PMV indexes could be a good guess of the AMV, but when the occupants’ evaluation of the thermal comfort are out of this range, the PMV estimation has big errors as marked inside the red boxes in Fig. 6. Especially when the actual thermal comfort sensation is “slightly warm” or hotter, the PMV gives an opposite guess that the thermal comfort level should be “Neutral” or cooler. The PMV model seems lost its validity when the actual occupant reported thermal sensations went to extremes like “Hot” or “Cold” based on the observation. One possible reason for this result could be the experiment limitation. Due to the operative facility, it is hard to push the whole zone environment into very cold or hot conditions, therefore, the amount of test cases reported as “Hot” and “Cold” for occupant thermal sensation is not as much as other cases, which may cause bigger uncertainty of the PMV results and furthermore reduce the validity of the PMV model under the extreme conditions. On the other hand, it is also should be considered that it may be more difficult for some occupants to articulate their feelings when they going through thermal discomfort. They may tend to use strong words like “Hot” or “Cold” to describe the discomfort rather than words like “Warm” or “Cool” etc., which are more neutral description in semantic representation. It is also observed that the error is bigger in the hot region compared with that in the cold region. It is probably because the occupants’ tolerances of the hot and cold conditions are different. Since the experiment was conducted in Qatar with hot arid climate, the occupants who live in such climate zone may be more sensitive to hot conditions and have less tolerance of these conditions. 3 2
Avg.PMV
1 0 -1 -2 -3 -3
-2
-1
0
ATS
1
2
Fig. 6. Average PMV versus ATS under each thermal sensation category
3
890
Zhidan Zhao et al. / Energy Procedia 134 (2017) 885–893
Zhidan Zhao/ Energy Procedia 00 (2017) 000–000
The linear regression between the PMV/ATS and the operative temperature are normally performed in previous studies to find out the neutral (optimal) temperature for building occupants in specified climate zone. In our study, clear linear relationship between the PMV/ATS value and the operative temperature wasn’t observed as shown in Fig. 7. So as shown in Fig. 8, under different operative temperature, the average PMV and ATS value were calculated to further investigate this relationship. It is observed that the average PMV values are linearly correlated with the operative temperatures and the predicted neutral temperature is 24.6˚C. The average ATS values weren’t perfectly fit into the linear relationship with the operative temperature. When the operative temperature is lower than 20˚C and higher than 24˚C, the average ATS are quit scattered across the operative temperature. These may also due to the limited amount of the test cases under extreme indoor conditions. The actual neutral temperature will definitely be lower than predicted neutral temperature 24.6˚C and there is a shift between the ATS and the PMV based on the observation.
Fig. 7. PMV, ATS versus the operative temperature
Fig. 8. Average PMV, ATS versus the operative temperature
Zhidan Zhao et al. / Energy Procedia 134 (2017) 885–893
891
Zhidan Zhao/ Energy Procedia 00 (2017) 000–000
According to above analysis, the PMV model couldn’t provide a good estimation for the actual occupant thermal comfort sensation in the air conditioned building in the hot arid climate area. One reason could be that in our calculation of PMV, the mean radiant temperature was estimated using the air temperature instead of real measurement. This approach may reduce the accuracy of the PMV calculation, but with the big gap between the calculated PMV and the ATS observed from above analysis, the improvement will not change the current results too much. It is reasonable to infer that the outdoor environment conditions should be taken into account while evaluating the human indoor thermal comfort. 3.2. Occupant satisfaction level with the thermal comfort Through the PMA application, the occupant satisfaction level with their indoor thermal comfort was also examined. As shown in Table 1, this parameter was evaluated using 6-points scale from very unsatisfied (-3) to very satisfied (3). As shown in Fig. 8, about 43% of the 613 scenarios were evaluated as satisfied with the thermal comfort by the occupants while only 27% of these scenarios were rated as neutral regarding the thermal comfort.
Fig. 8. Satisfaction level with indoor thermal comfort
The occupants’ sensation of the indoor thermal comfort doesn’t necessarily comply with their perceived satisfaction level with the thermal comfort. This is revealed by the crossover comparison between the two parameters shown in Fig. 9. When the occupants evaluated that they feel cold or hot inside the office, they still perceived that they are satisfied with the thermal comfort. Similarly, when the occupants sensed the thermal comfort as neutral, they may also feel unhappy about this in the meantime. In the PMV-PPD model, when the PMV number is close to 0 (neutral), the PPD number will be close to 5%. In our case study, when the AMV is 0, only 10% of these feedbacks gave unsatisfied evaluation of the thermal comfort, which is reasonable. 4 3
ATS
2 1 0 -1 -2 -3 -4 -4
-3
-2
-1
0
1
Satisfaction Level
2
3
4
Fig. 9. Crossover comparison between the ATS and satisfaction level with thermal comfort
3.3. Occupant adaptive behavior The occupant adaptive behavior to the thermal discomfort was summarized in Fig. 10. In general, adjusting the thermostat and adjusting cloth are the two major ways that building occupants would take to improve their comfort level, which were found to
892
Zhidan Zhao et al. / Energy Procedia 134 (2017) 885–893
Zhidan Zhao/ Energy Procedia 00 (2017) 000–000
be 47% and 35% separately among all the adaption responses. Opening the internal office door to introduce the cold or warm corridor air is third common used approach for comfort enhancement and accounts for about 12% of all the reported adaptive actions. All the other actions are less the 5%. Some occupants even took more than two approaches at once hoping to improve the comfort level sooner. Among the three mostly reported adaptive behaviors, adjusting the thermostat will directly impact the building energy system and therefore impact the building energy consumption. Opening the internal door may also affect the performance of the building energy system indirectly. Adjusting cloth is normally a self-adaptive way for building occupant to improve their thermal comfort but may also impact the building energy efficiency in the long run.
Fig. 10. Occupant adaptive behavior
4. Conclusion The applicability of the PMV model in Qatar with dry, subtropical desert climate is evaluated through an experiment that was conducted in an air conditioned office building in Doha, Qatar to assess the indoor thermal comfort. 14 occupants participated in this study and provided continuous feedbacks in 12 months duration and total 613 effective feedbacks were used for the analysis. The results showed that there was a big gap between the PMV and the ATS. Not only were the PMV index values far from the ATS values, the PMV also failed to capture the trend of the ATS under the same indoor scenarios. The linear regression used to find out the neutral temperature also failed due to the quit scattered distribution of the ATS and PMV along the operative temperatures. Therefore the average PMV and average ATS versus the operative temperature correlation was performed to further assess the neutral temperature. It was observed that the average PMV showed clear linear correlation with the operative temperature and the neutral temperature was found to be 24.6 ˚C while there is no obvious linear correlation between the average ATS and the operative temperatures. However, it could found out that the actual natural temperature should be lower than 24.6 ˚C and there is an shift between the PMV estimation and the ATS. The occupants’ satisfaction level with the indoor thermal comfort and their adaptive behavior were also examined in the study. Among all the feedbacks, about 43% of them rated the thermal comfort as satisfied, which is higher than the 27% that rated the room thermal comfort as neutral comfort. Therefore, the occupants can still feel satisfied with the indoor thermal comfort while they perceived the room as cold or hot at the same time. Adjusting thermostat and adjusting cloth are the two major approaches that building occupants took to improve the indoor thermal comfort, both of which have impact on the building energy efficiency. In the future, a new thermal comfort prediction model or an improved PMV model should be developed to assess the indoor thermal comfort in air conditioned buildings in the hot arid climate area. More parameters such as outside weather conditions may need to be taken into account for a better estimation of the occupant indoor thermal comfort sensation in air conditioned buildings.
Zhidan Zhao et al. / Energy Procedia 134 (2017) 885–893
Zhidan Zhao/ Energy Procedia 00 (2017) 000–000
893
Acknowledgements This publication was made possible by NPRP grant # 6-1370-2-552 from the Qatar National Research Fund (a member of Qatar Foundation). The findings achieved herein are solely the responsibility of the authors.
References [1] Frontczak, M., R.V. Andersen, and P. Wargocki, Questionnaire survey on factors influencing comfort with indoor environmental quality in Danish housing. Building and Environment, 2012. 50: p. 56-64. [2] Fanger, P.O., Thermal comfort. Analysis and applications in environmental engineering. Thermal comfort. Analysis and applications in environmental engineering. 1970. [3] 7730, I., Ergonomics of the thermal environment — Analytical determination and interpretation of thermal comfort using calculation of the PMV and PPD indices and local thermal comfort criteria, I.O.f. Standardization, Editor. 2005. [4] ANSI/ASHRAE, Standard 55-2013 Thermal Environmental Conditions for Human Occupancy. 2013. [5] De Dear, R. and G. Schiller Brager, The adaptive model of thermal comfort and energy conservation in the built environment. International journal of biometeorology, 2001. 45(2): p. 100-108. [6] Hoof, J.v., Thermal comfort: research and practice. Frontiers in Bioscience, 2010. 15(1): p. 765. [7] Azizpour, F., S. Moghimi, S. Mat, C. Lim, and K. Sopian. Objective and subjective assessment of thermal comfort in hot-humid region. in Proceedings of 5th WSEAS international conferences on Recent Researches in Chemistry, Biology, Environment and Culture, Montreux, Switzerland. 2011. [8] Shaari, N.A., S.A. Zaki, M.S. Mat Ali, and A. Abd Razak, Investigation of the PMV and TSV Models of Thermal Comfort in Air-Conditioned University Classrooms in Malaysia. Applied Mechanics and Materials, 2016. 819: p. 207-211. [9] Kim, J.T., J.H. Lim, S.H. Cho, and G.Y. Yun, Development of the adaptive PMV model for improving prediction performances. Energy and Buildings, 2015. 98: p. 100-105. [10] Gilani, S.I.u.H., M.H. Khan, and W. Pao, Thermal Comfort Analysis of PMV Model Prediction in Air Conditioned and Naturally Ventilated Buildings. Energy Procedia, 2015. 75: p. 1373-1379. [11] Al-ajmi, F.F. and D.L. Loveday, Indoor thermal conditions and thermal comfort in air-conditioned domestic buildings in the dry-desert climate of Kuwait. Building and Environment, 2010. 45(3): p. 704-710. [12] Walikewitz, N., B. Jänicke, M. Langner, F. Meier, and W. Endlicher, The difference between the mean radiant temperature and the air temperature within indoor environments: A case study during summer conditions. Building and Environment, 2015. 84: p. 151-161.
ScienceDirect
Available online at www.sciencedirect.com Energy Procedia 00 (2017) 000–000
ScienceDirect
www.elsevier.com/locate/procedia
Energy Procedia 134 (2017) 885–893
9th International Conference on Sustainability in Energy and Buildings, SEB-17, 5-7 July 2017, Chania, Crete, Greece
An Energy Efficiency Assessment of the Thermal Comfort in an Office building Zhidan Zhaoa, Mahdi Houchatia, and AbdlMonem Beitelmala, * 0F
Qatar Environment and Energy Research Institute, Hamad Bin Khalifa University, Qatar Foundation, Doha, Qatar
a
Abstract People spend about 90% of their time indoors, so a comfort indoor thermal environment is essential for the satisfaction, productivity and wellbeing of the building occupants. Assessment of the indoor thermal comfort is the key point for building HVAC system design and operation to provide a comfort indoor environment to building occupants. Predicted Mean Vote (PMV) model is the most widely used tool for the indoor thermal comfort assessment. In this study, the application of the PMV model in Qatar with dry, subtropical desert climate is evaluated. An experiment was conducted in an office building in Doha, Qatar to reveal the occupant perception of indoor thermal comfort. Using collected data, the PMV indexes were calculated using Fanger’s theory and compared with the Actual Thermal Sensation (ATS) of the occupants to assess the applicability of the PMV model to predict the indoor thermal comfort in air conditioned buildings in the climate zone of Qatar. The corresponding occupants’ satisfaction level with the indoor thermal comfort and their adaptive behavior were also assessed. © 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of KES International. Keywords: Thermal comfort; PMV; Occupant bahavior; Building enery efficiency
1. Introduction Indoor thermal comfort is the most important factor that determines the overall indoor environment quality which is the major concern of building occupants due to their long time stay indoors [1]. Building occupants’ perception of indoor thermal comfort normally refers to their feelings of room comfort, for example, the room is hot, cold or neutral etc. and is not a direct sensation of indoor air temperature. Due to the essential role that building occupants play in the building operation and further more in the building energy efficiency enhancement, the assessment of their indoor thermal comfort is quite important for not only building design and operation, but also for building energy efficiency improvement. Predicted Mean Vote (PMV) model which was first developed by Fanger [2] is the most widely accepted tool for the indoor thermal comfort assessment and was adopted by international standards such as ISO 7730 [3] and the ASHRAE Standard 55–92 [4] to evaluate the indoor thermal comfort conditions. In previous studies, this model has been evaluated and validated using data collected from different climate zones and different building types. Although the strength of PMV model has been proved by many studies, it was found that the PMV model didn’t perform very well for assessment of the indoor thermal comfort in naturally ventilated buildings. Therefore, the adaptive model [5] was adopted by AHRAE Standard called Adaptive Comfort Standard (ACS) as an extension tool of PMV method for the thermal comfort evaluation in naturally ventilated buildings [6]. Despite that a lot of studies validated the application of PMV model in air conditioned buildings, some recent studies indicate that there are discrepancies between the * Corresponding author. Tel.: +974-4454-7158; fax: +974-4454-1528. E-mail address: [email protected] 1876-6102 © 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of KES International.
1876-6102 © 2017 The Authors. Published by Elsevier Ltd. Peer-review under responsibility of KES International. 10.1016/j.egypro.2017.09.550
Zhidan Zhao et al. / Energy Procedia 134 (2017) 885–893
886
Zhidan Zhao/ Energy Procedia 00 (2017) 000–000
PMV model prediction and the Actual Mean Vote (AMV) of the building occupants while evaluating the indoor thermal comfort in air conditioned buildings [7-11]. In this study, the application of the PMV model in Qatar with dry, subtropical desert climate is evaluated. An experiment was conducted in an office building served with a central VAV air conditioning system in Doha, Qatar to reveal the occupant perception of indoor thermal comfort. The occupants’ adaptive behavior were also examined in this study. . Nomenclature PMV Ta Tmra RH Va MR CL
Predicted Mean Vote Room air temperature Mean radiant temperature Relative humidity of room air Air velocity Metabolic rate Clothing level
2. Methodology 2.1. Experiment Facility The chosen building for this study is an office building located in College of North Atlantic, Qatar numbered as building 7. It is a two story office building with gross area of about 3,184 m2. Each floor can be divided into East and West Zones. The experiment was conducted in the West zone of the ground floor with a gross area of about 657 m2. There are total 22 personal offices, 1 staff lounge and 1 equipment room in this experimental zone, which is served by three AHUs. Chilled water is provided by the campus central plant and no heating device is in use. The floor plan is shown in Fig. 1. Each office is served by one VAV terminal box controlled with local thermostat. Except office 7110a and 7110b, each office is occupied by one university instructor and 14 of the total 20 occupants agreed to participate in the study.
Fig. 1. Experiment zone floor plan.
2.2. Data Collection The PMV index should be calculated using 6 inputs according to Fanger’s theory. As shown in Equation 1, air temperature, mean radiant temperature, relative humidity, air velocity, metabolic rate and clothing level are the six factors that determine the PMV index. The former first 4 inputs are measureable parameters which can be obtained through sensors and the latter 2 inputs need evaluation through subjective studies like survey feedbacks from the occupants.
PMV f Ta , Tmra , RH , Va , MR, CL
(1)
The indoor thermal conditions like temperatures and relative humilities and indoor air velocities are monitored using sensors and recorded in the data acquisition system at 15 min intervals. However, due to the experiment conduction limitation, we didn’t get a chance to measure the mean radiant temperature in each office room. Considering the big thermal mass of the building,
Zhidan Zhao et al. / Energy Procedia 134 (2017) 885–893
Zhidan Zhao/ Energy Procedia 00 (2017) 000–000
887
shading wall outside the north external wall for sunblock function, the mean radiant temperature was assumed to be equal to the air temperature according to previous study [12]. Some sensors and online data recording interface are shown in Fig. 2.
Fig. 2. Data collection sensing system
The occupant perception of the indoor thermal comfort was evaluated through the Preference Monitoring Application (PMA). The PMA is a 3-5 minute Web-based survey that collects time-stamped occupant responses and contains about 10 questions with the opportunity to provide open-ended comments. It was created using Qualtrics survey development software accessible on both webpage and smartphones as shown in Fig. 3.
Fig. 3. PMA mobile App example
The survey was designed based on the ASHRAE standard including three major parts regarding occupants’ sensation of indoor thermal comfort on a seven-point scale from cold (-3) to hot (+3), occupants’ personal values like their activity levels and cloth levels and also their adaptive actions. Table 1 shows the rating scale with evaluated parameters. Table 2 and Table 3 provide some examples to illustrate that how to correlate the occupant reported activity levels and clothing levels with the metabolic rate values and clothing insulation values that should be used in the PMV calculation [4]. In our study, the building occupants were asked to fill the survey from time to time to give their feedbacks whenever they feel uncomfortable or very comfortable or take some energy related actions for duration of one year. To evaluate the thermal comfort in an air conditioned building, the experimental zone temperature is pushed to 18˚C as low limit and 26˚C as high limit regularly using intervention plan. From March 2016 until March 2017, total 613 effective occupant feedbacks are collected and in use for the study. Table 1. Occupant evaluated parameters and rating scale Rating scale value
Thermal comfort evaluation
Thermal comfort satisfaction
+3
Hot
Very satisfied
+2
Warm
Satisfied
+1
Slightly Warm
Moderately satisfied
0
Neutral
-
-1
Slightly cool
Moderately unsatisfied
-2
Cool
Unsatisfied
-3
Cold
Very Unsatisfied
Zhidan Zhao et al. / Energy Procedia 134 (2017) 885–893
888
Zhidan Zhao/ Energy Procedia 00 (2017) 000–000
Table 2. Occupant reported activity level and corresponding metabolic rate Reported activity level
Metabolic rate [Met]
Seated, quiet
1.0
Standing, relax
1.2
Typing
1.1
Filing, seated
1.2
Filing, standing
1.4
Walking about
1.7
Lifting/packing
2.1
Table 3. Occupant reported clothing level and corresponding clothing insulation value Reported activity level
Clothing insulation value [Clo]
Short-sleeve shirt, trousers
0.57
Long-sleeve shirt, trousers
0.61
Long-sleeve shirt, trousers, suit jacket
0.96
Short-sleeve shirt, Knee-length skirt
0.54
Long-sleeve shirt, Knee-length skirt, full slip
0.67
Long-sleeve shirt, Knee-length skirt, suit jacket
1.04
Long-sleeve shirt, Ankle-length skirt, suit jacket
1.1
3. Results and Discussion 3.1. PMV application
3
3
2
2
1
1
0
0
-1
-1
-2
-2
-3 0
100
200
300
400
CaseNumber
500
600
PMV
ATS
Based on the timestamps of the 613 occupant feedbacks, corresponding air temperature, humidity, air velocity values are located from the database. The metabolic rate quantified with “met” and clothing level quantified with “clo” were also determined according to ASHRAE standard using the occupant reported activity level such as “seated quietly” and clothing level such as “long sleeve shirt with thin trouser” as inputs. Then the PMV indexes were calculated for these 613 scenarios. The PMV and ATS values for the 613 cases are shown in Fig. 4. It was observed that there are big differences between the reported ATS from the occupants and calculated PMV. The PMV lost track of most of the occupants’ thermal comfort sensation when they evaluate the thermal comfort as neutral or above.
-3
Fig. 4. Final PMV and ATS results
The crossover comparison between the PMV and ATS is shown in Fig. 5. If the PMV could be a good estimation for the ATS in this case study, these data marks should collapse along the red line, which is not observed from the plot. It is shown that when
Zhidan Zhao et al. / Energy Procedia 134 (2017) 885–893
889
Zhidan Zhao/ Energy Procedia 00 (2017) 000–000
the occupant actual thermal sensation went beyond slight cool (-1), the PMV indexes overestimated the comfort level while underestimated the thermal comfort level when the actual occupant thermal sensation is slightly warm or even hotter. 3 2
PMV
1 0 -1 -2 -3
-3
-2
-1
0
ATS
1
2
3
4
Fig. 5. Crossover comparison between ATS and PMV
To further assess the validity of PMV application, under each occupant actual thermal comfort sensation (ATS) category, from cold (-3) to hot (3), the calculated PMV indexes are averaged to see how well the PMV can capture the real thermal comfort sensation under different thermal comfort levels. The comparison between the averaged PMV values and the ATS is shown in Fig. 6. It is observed that, when the occupant actual thermal sensation between “Neutral” and “Cool”, the PMV indexes could be a good guess of the AMV, but when the occupants’ evaluation of the thermal comfort are out of this range, the PMV estimation has big errors as marked inside the red boxes in Fig. 6. Especially when the actual thermal comfort sensation is “slightly warm” or hotter, the PMV gives an opposite guess that the thermal comfort level should be “Neutral” or cooler. The PMV model seems lost its validity when the actual occupant reported thermal sensations went to extremes like “Hot” or “Cold” based on the observation. One possible reason for this result could be the experiment limitation. Due to the operative facility, it is hard to push the whole zone environment into very cold or hot conditions, therefore, the amount of test cases reported as “Hot” and “Cold” for occupant thermal sensation is not as much as other cases, which may cause bigger uncertainty of the PMV results and furthermore reduce the validity of the PMV model under the extreme conditions. On the other hand, it is also should be considered that it may be more difficult for some occupants to articulate their feelings when they going through thermal discomfort. They may tend to use strong words like “Hot” or “Cold” to describe the discomfort rather than words like “Warm” or “Cool” etc., which are more neutral description in semantic representation. It is also observed that the error is bigger in the hot region compared with that in the cold region. It is probably because the occupants’ tolerances of the hot and cold conditions are different. Since the experiment was conducted in Qatar with hot arid climate, the occupants who live in such climate zone may be more sensitive to hot conditions and have less tolerance of these conditions. 3 2
Avg.PMV
1 0 -1 -2 -3 -3
-2
-1
0
ATS
1
2
Fig. 6. Average PMV versus ATS under each thermal sensation category
3
890
Zhidan Zhao et al. / Energy Procedia 134 (2017) 885–893
Zhidan Zhao/ Energy Procedia 00 (2017) 000–000
The linear regression between the PMV/ATS and the operative temperature are normally performed in previous studies to find out the neutral (optimal) temperature for building occupants in specified climate zone. In our study, clear linear relationship between the PMV/ATS value and the operative temperature wasn’t observed as shown in Fig. 7. So as shown in Fig. 8, under different operative temperature, the average PMV and ATS value were calculated to further investigate this relationship. It is observed that the average PMV values are linearly correlated with the operative temperatures and the predicted neutral temperature is 24.6˚C. The average ATS values weren’t perfectly fit into the linear relationship with the operative temperature. When the operative temperature is lower than 20˚C and higher than 24˚C, the average ATS are quit scattered across the operative temperature. These may also due to the limited amount of the test cases under extreme indoor conditions. The actual neutral temperature will definitely be lower than predicted neutral temperature 24.6˚C and there is a shift between the ATS and the PMV based on the observation.
Fig. 7. PMV, ATS versus the operative temperature
Fig. 8. Average PMV, ATS versus the operative temperature
Zhidan Zhao et al. / Energy Procedia 134 (2017) 885–893
891
Zhidan Zhao/ Energy Procedia 00 (2017) 000–000
According to above analysis, the PMV model couldn’t provide a good estimation for the actual occupant thermal comfort sensation in the air conditioned building in the hot arid climate area. One reason could be that in our calculation of PMV, the mean radiant temperature was estimated using the air temperature instead of real measurement. This approach may reduce the accuracy of the PMV calculation, but with the big gap between the calculated PMV and the ATS observed from above analysis, the improvement will not change the current results too much. It is reasonable to infer that the outdoor environment conditions should be taken into account while evaluating the human indoor thermal comfort. 3.2. Occupant satisfaction level with the thermal comfort Through the PMA application, the occupant satisfaction level with their indoor thermal comfort was also examined. As shown in Table 1, this parameter was evaluated using 6-points scale from very unsatisfied (-3) to very satisfied (3). As shown in Fig. 8, about 43% of the 613 scenarios were evaluated as satisfied with the thermal comfort by the occupants while only 27% of these scenarios were rated as neutral regarding the thermal comfort.
Fig. 8. Satisfaction level with indoor thermal comfort
The occupants’ sensation of the indoor thermal comfort doesn’t necessarily comply with their perceived satisfaction level with the thermal comfort. This is revealed by the crossover comparison between the two parameters shown in Fig. 9. When the occupants evaluated that they feel cold or hot inside the office, they still perceived that they are satisfied with the thermal comfort. Similarly, when the occupants sensed the thermal comfort as neutral, they may also feel unhappy about this in the meantime. In the PMV-PPD model, when the PMV number is close to 0 (neutral), the PPD number will be close to 5%. In our case study, when the AMV is 0, only 10% of these feedbacks gave unsatisfied evaluation of the thermal comfort, which is reasonable. 4 3
ATS
2 1 0 -1 -2 -3 -4 -4
-3
-2
-1
0
1
Satisfaction Level
2
3
4
Fig. 9. Crossover comparison between the ATS and satisfaction level with thermal comfort
3.3. Occupant adaptive behavior The occupant adaptive behavior to the thermal discomfort was summarized in Fig. 10. In general, adjusting the thermostat and adjusting cloth are the two major ways that building occupants would take to improve their comfort level, which were found to
892
Zhidan Zhao et al. / Energy Procedia 134 (2017) 885–893
Zhidan Zhao/ Energy Procedia 00 (2017) 000–000
be 47% and 35% separately among all the adaption responses. Opening the internal office door to introduce the cold or warm corridor air is third common used approach for comfort enhancement and accounts for about 12% of all the reported adaptive actions. All the other actions are less the 5%. Some occupants even took more than two approaches at once hoping to improve the comfort level sooner. Among the three mostly reported adaptive behaviors, adjusting the thermostat will directly impact the building energy system and therefore impact the building energy consumption. Opening the internal door may also affect the performance of the building energy system indirectly. Adjusting cloth is normally a self-adaptive way for building occupant to improve their thermal comfort but may also impact the building energy efficiency in the long run.
Fig. 10. Occupant adaptive behavior
4. Conclusion The applicability of the PMV model in Qatar with dry, subtropical desert climate is evaluated through an experiment that was conducted in an air conditioned office building in Doha, Qatar to assess the indoor thermal comfort. 14 occupants participated in this study and provided continuous feedbacks in 12 months duration and total 613 effective feedbacks were used for the analysis. The results showed that there was a big gap between the PMV and the ATS. Not only were the PMV index values far from the ATS values, the PMV also failed to capture the trend of the ATS under the same indoor scenarios. The linear regression used to find out the neutral temperature also failed due to the quit scattered distribution of the ATS and PMV along the operative temperatures. Therefore the average PMV and average ATS versus the operative temperature correlation was performed to further assess the neutral temperature. It was observed that the average PMV showed clear linear correlation with the operative temperature and the neutral temperature was found to be 24.6 ˚C while there is no obvious linear correlation between the average ATS and the operative temperatures. However, it could found out that the actual natural temperature should be lower than 24.6 ˚C and there is an shift between the PMV estimation and the ATS. The occupants’ satisfaction level with the indoor thermal comfort and their adaptive behavior were also examined in the study. Among all the feedbacks, about 43% of them rated the thermal comfort as satisfied, which is higher than the 27% that rated the room thermal comfort as neutral comfort. Therefore, the occupants can still feel satisfied with the indoor thermal comfort while they perceived the room as cold or hot at the same time. Adjusting thermostat and adjusting cloth are the two major approaches that building occupants took to improve the indoor thermal comfort, both of which have impact on the building energy efficiency. In the future, a new thermal comfort prediction model or an improved PMV model should be developed to assess the indoor thermal comfort in air conditioned buildings in the hot arid climate area. More parameters such as outside weather conditions may need to be taken into account for a better estimation of the occupant indoor thermal comfort sensation in air conditioned buildings.
Zhidan Zhao et al. / Energy Procedia 134 (2017) 885–893
Zhidan Zhao/ Energy Procedia 00 (2017) 000–000
893
Acknowledgements This publication was made possible by NPRP grant # 6-1370-2-552 from the Qatar National Research Fund (a member of Qatar Foundation). The findings achieved herein are solely the responsibility of the authors.
References [1] Frontczak, M., R.V. Andersen, and P. Wargocki, Questionnaire survey on factors influencing comfort with indoor environmental quality in Danish housing. Building and Environment, 2012. 50: p. 56-64. [2] Fanger, P.O., Thermal comfort. Analysis and applications in environmental engineering. Thermal comfort. Analysis and applications in environmental engineering. 1970. [3] 7730, I., Ergonomics of the thermal environment — Analytical determination and interpretation of thermal comfort using calculation of the PMV and PPD indices and local thermal comfort criteria, I.O.f. Standardization, Editor. 2005. [4] ANSI/ASHRAE, Standard 55-2013 Thermal Environmental Conditions for Human Occupancy. 2013. [5] De Dear, R. and G. Schiller Brager, The adaptive model of thermal comfort and energy conservation in the built environment. International journal of biometeorology, 2001. 45(2): p. 100-108. [6] Hoof, J.v., Thermal comfort: research and practice. Frontiers in Bioscience, 2010. 15(1): p. 765. [7] Azizpour, F., S. Moghimi, S. Mat, C. Lim, and K. Sopian. Objective and subjective assessment of thermal comfort in hot-humid region. in Proceedings of 5th WSEAS international conferences on Recent Researches in Chemistry, Biology, Environment and Culture, Montreux, Switzerland. 2011. [8] Shaari, N.A., S.A. Zaki, M.S. Mat Ali, and A. Abd Razak, Investigation of the PMV and TSV Models of Thermal Comfort in Air-Conditioned University Classrooms in Malaysia. Applied Mechanics and Materials, 2016. 819: p. 207-211. [9] Kim, J.T., J.H. Lim, S.H. Cho, and G.Y. Yun, Development of the adaptive PMV model for improving prediction performances. Energy and Buildings, 2015. 98: p. 100-105. [10] Gilani, S.I.u.H., M.H. Khan, and W. Pao, Thermal Comfort Analysis of PMV Model Prediction in Air Conditioned and Naturally Ventilated Buildings. Energy Procedia, 2015. 75: p. 1373-1379. [11] Al-ajmi, F.F. and D.L. Loveday, Indoor thermal conditions and thermal comfort in air-conditioned domestic buildings in the dry-desert climate of Kuwait. Building and Environment, 2010. 45(3): p. 704-710. [12] Walikewitz, N., B. Jänicke, M. Langner, F. Meier, and W. Endlicher, The difference between the mean radiant temperature and the air temperature within indoor environments: A case study during summer conditions. Building and Environment, 2015. 84: p. 151-161.