J. Cent. South Univ. (2012) 19: 783-790 

DOI: 10.1007/s11771-012-1072-8

Energy consumption, indoor environmental quality, and benchmark for

office buildings in Hainan Province of China

KONG Xiang-fei(孔祥飞)¹, L? Shi-lei(吕石磊)¹, XIN Ya-juan(辛亚娟)¹, WU Wei(吴薇)²

1. School of Environmental Science and Technology, Tianjin University, Tianjin 300072, China;

2. Department of Housing and Urban-Rural Development of Hainan, Haikou 570100, China

? Central South University Press and Springer-Verlag Berlin Heidelberg 2012

Abstract: With rapid economy growth, building energy consumption in China has been gradually increased. The energy consumption and indoor environmental quality of 51 office buildings in Hainan Province, a hot and humid area, were studied through collection of verified data in site visits and field tests. The result revealed that, electricity accounted for 99.79% of the total energy consumption, natural gas 0.17%, and diesel 0.04%. The air conditioning dominated the energy use with a share of 43.18%, equipment in the particular areas 26.90%, equipment in the office rooms 11.95%, lighting system 8.67%, general service system 7.57%, and miscellaneous items 1.73%. Statistical method including six indicators obtained the energy consumption benchmark with upper limit of 98.31 kW·h/m² and lower limit of 55.26 kW·h/m². According to ASHRAE standard (comfortable standard) and GB/T 18883—2002 (acceptable standard), the indoor environmental quality of 51 sampled office buildings was classified into three ranks: good, normal and bad. With benchmark of building energy consumption combined with indoor environmental quality, it was found that only 3.92% of sampled buildings can be identified as the best performance buildings with low energy consumption and advanced indoor environmental quality, and the buildings classified into normal level accounted for the maximum ratio.

Key words: building energy consumption; benchmark; indoor environmental quality; energy use index; Hainan Province

1 Introduction

The rapid economy growth in China has a positive relation with the increase of energy consumption [1-3]. Population growth, enhancement of building services and comfort levels, together with more time spent inside buildings, have increased building energy consumption (BEC) [4]. The building energy consumption doubled from 1998 to 2009 in China [5], and according to the developing law of developed countries, the share of building in total energy consumption will reach 20%-40% [6-8]. The current demographic, economic and urbanization growth indicates nearly 60% of Chinese population is expected to live in urban areas in 2030, while the current level is about 45% [9]. This rapid urbanization process in China means a large potential of increase in demand for energy services in buildings for the next 20-30 years [10]. The growth of energy consumption will have no doubt to be accompanied by an increase in greenhouse gas emissions, with an exhaustion of energy resources and environmental pollution.

Hainan Province, the only tropical island province of China, is located between 18°10′-20°10′ north latitude and 108°37′-111°05′ east longitude. Without heating in winter, the building energy performance of Hainan Province maintains its own characteristic. According to the national energy statistics [11], the building energy consumption in Hainan Province showed an upward trend in recent years. Besides, with related policies developed, such as “Offshore Duty-free Policy”, Hainan Province has launched the plan to construct the “International Traveling Island”, to increase the numbers of tourists and stimulate the economy growth. Hainan Province is nearly in the same latitude with Hawaii, where the energy consumption was found to be closely related to the tourism economy growth [12]. Therefore, energy benchmark system is necessary to restrain the growth of building energy consumption in Hainan Province.

Office building is one of the buildings with high energy consumption [13]. Energy consumption in office buildings is about 70-300 kW·h/m² per year, 10-20 times that of residential buildings [14]. Over the years in the new century, energy consumption in office buildings has increased significantly due to the modernization of office buildings, the installation of HVAC equipment and the widespread use of office equipment [15]. There are two analysis methods to study the office building energy consumption: actual data analysis and virtual data analysis. Obtained through site surveys and field tests in actual buildings, actual data consist of energy consumption data, characteristics of buildings, period, etc. Energy use index (EUI) is calculated on the analysis of available data. ROULET et al [16] collected related data from 56 office buildings in nine European countries, and revealed that 25% of the buildings had an annual EUI <222 kW·h/m² and <389 kW·h/m² for 75% of buildings, with an average value of 278 kW·h/m². Through collecting data in 39 bank branches over a period of 6 years, SPYROPOULOS and BALARAS [17] reported the average annual EUI of 345 kW·h/m² in Greece, and revealed the breakdown of the different end-uses with averaged 48% of the final energy consumption for air-conditioning, 35% for lighting, and 17% for the other electronic equipment. The method of virtual data analysis, based on realistic input-parameters, is used to calculate EUIs in office buildings by mathematical formulations or numerical software. Collecting basic information on 68 office buildings, SAIDUR [4] deduced the annual EUI of 130 kW·h/m² for the office buildings in Malaysia by related mathematical formulations, and the ratios of electricity consumption for air conditioning, lighting and general equipment were 57%, 19%, and 24%, respectively. Besides, CHIRARATTANANON and TAWEEKUN [13] obtained annual EUI of 161 kW·h/m² for office buildings in Thailand by implementing DOE-2 simulation software.

Benchmarking results usually are issued to the public, bringing public pressure to the building owners of high energy consumption. So, the benchmark system can be used to encourage building owners to enhance their building energy efficiency. Study on energy consumption benchmark for office buildings generally includes two methods: simulation methods and statistical methods. In simulation methods, simulation software is used to create virtual model of buildings, then the building energy consumption of virtual model is computed under a series of operating conditions, and the benchmark is ascertained on the basis of estimated results. NIKOLAOU et al [18] created a virtual building dataset (inclusive of 30 000 buildings) for office buildings in Greece, taking into account the Greek constructional and operational characteristics of office buildings and Greek legislation. Based on virtual building dataset, the energy and indoor thermal comfort benchmarks for office building were assessed. With numerical software, FEDERSPIEL et al [19] created a benchmark that represents the minimum amount of energy required to meet a set of basic functional requirements of the laboratory buildings. Although simulation is one of the most popular methods for analyzing and assessing the energy consumption in buildings, it is deficient for benchmarking the energy consumption of existing buildings because of the possible difference between simulation result and actual energy use. Therefore, the statistical method is generally used for benchmark of building energy consumption. It is always based on the actual data analysis mentioned above, to ascertain reasonable indicators to represent the general level of energy consumption through available statistical analysis. Under the research of the relationship between EUI and the explanatory factors, CHUNG et al [20] developed a benchmarking process for energy e?ciency by means of multiple regression analysis. With actual data analysis, ZHANG [21] set a benchmark to ensure EUI of 60% buildings to be lower than it. Pursuit of high building energy efficiency needs to take into account the environmental quality which can influence the health and productivity of office staff. This imposes the requirement to statistical analysis on benchmark system to focus on both building energy consumption and indoor environmental quality (IEQ).

2 Office building characteristics

With the assistance from Department of Housing and Urban-Rural Development of Hainan Province, China, a comprehensive data collection about 51 office buildings and their energy consumption was conducted in Hainan Province through the field work. Related forms were filled by the building authorities in site meetings, and the process for verification on the collected data was completed through the site visits. Data were collected from the records of sub-meters, while for the buildings with no sub-meters, detailed field tests were carried out, which included collecting operating records and rated power input for the major equipment, and measuring the power for building lighting and lift systems, etc. Besides, IEQ field tests were completed so as to verify the actual indoor environmental quality. Finally, the complete data for the energy consumption in 2010 year, indoor environment quality, and background information such as construction year, and total gross floor area (GFA), were obtained from 51 office buildings in Hainan Province.

The sampled 51 buildings are located in 6 cities: 30 in Haikou (provincial capital of Hainan Province), 16 in Sanya, 2 in Zhanzhou, 1 in Baoting, 1 in Wanning, and 1 in Wenchang. With the total GFA of 774 071 m², 51 samples cover about 21% of the office buildings in Hainan Province, and have the following characteristics:

1) Year of construction between 1982 and 2009, with 33 buildings constructed in last century and 18 buildings in this century;

2) Quantity of floor varying from 5 to 38;

3) Heterogeneous GFA from 1 800 to 123 183 m²;

4) Number of office workers ranging from 31 to   1 950;

5) Different energy performances among the sampled buildings, because of their different operation schedules and functions.

Hainan Province lies in the tropical area. The hot and humid tropical climate makes almost year round air-conditioning maintain a comfortable thermal environment for the office workers. In the sampled office buildings, 20 buildings were centrally air-conditioned, and the remaining 31 buildings used split units. All the centrally air-conditioned buildings operated and maintained their own chiller plants. In 4 buildings, air cooled central chiller plants were used, and central chiller plants in the remaining 16 buildings were water cooled.

3 Results and discussion

3.1 Breakdown of energy consumption by fuel type

The sampled office buildings mainly used three types of fuels: electricity, natural gas and diesel. The electricity was used in all of the buildings, while the natural gas and diesel were only used in few buildings. The electricity was used for the office equipment such as computers and printers, lighting systems, air- conditioning and lift systems. The studied buildings with staff canteens used the natural gas to cook. Diesel was only for generating power during power outages. The share of total energy consumption for the electricity, natural gas and diesel were 99.79%, 0.17% and 0.04%, respectively.

The electricity absolutely dominated in the energy use in Hainan Province, which is close to the situation of Dalian (in north China with heating in winter), Suzhou (in eastern China with heating in winter), Xi’an (in western China with heating in winter) and Changsha (in central China with no heating in winter) [22-25]. However, the electricity only accounted for 59.37%, 60%, 76% and 80.1%, respectively, in the above four cities. The comparison among Hainan and the four cities reveals that because of no heating in winter in Hainan, the energy consumption of the cooling is significant, considering the hot and humid tropical climate.

3.2 Breakdown of energy consumption based on major building services

Based on the type of service systems, the energy consumption mainly consists of six parts: air-conditioning system, lighting system, equipment in the office rooms, general service system, professional equipment in the particular areas, and miscellaneous items. The equipment in the office rooms mainly includes computers, printers, water dispensers, copiers, etc. General service system refers to lifts, water supply and drainage system, equipment of hot water, etc. The particular area is defined as the regions where special professional equipment is used and high-density terminal energy is consumed. For example, the equipment used in particular areas of a telecommunication building generally includes electronic switching systems, signal launchers, mainframes, servers, and other network and telecommunication equipment. At last, the miscellaneous items are other energy consumption devices not included in the above five categories. The breakdown result is shown in Fig. 1.

Fig. 1 Breakdown of energy consumption in sampled office buildings based on major building services

In the sampled buildings, the energy consumption for the air-conditioning accounted for 43.18%, lower than that of Greece [17], Malaysia [4], and Thailand [13], which were 48%, 57%, and 66.41%, respectively. The energy consumption for professional equipment in the particular area, accounting for 26.9%, was the second largest part. According to the local regulation in Hainan Province, the workers have a rest of 2.5-3 h in the noon. During this period, equipment is generally powered off, including air-conditioning, computers, lights, etc, while equipment in the particular area operates. Therefore, the energy consumption in the particular areas is relatively higher than that of the lighting, equipment in office rooms or general service systems.

3.3 EUI for sampled office buildings

Considering the electricity dominating in the energy use, other types of fuel were converted to electricity. Pearson correlation analysis among energy consumption, GFA, building age (AGE), staff quantity and floor quantity was calculated to determine the primary variable for normalization. As shown in Table 1, except for the weak correlation between energy consumption and AGE, the other three capacity indicators, including GFA, staff quantity and floor quantity, are well correlated with the energy consumption. However, GFA is still the best correlated variable, which is manifested by higher correlation. Therefore, it is chosen as the primary normalization variable for the EUI.

Table 1 Correlation among building energy consumption and main influencing factors in sampled office buildings

In Table 2, the statistical analysis on the EUIs shows significant differences among the sampled office buildings, which reveals the energy-saving potential and necessity in the office buildings.

Table 2 Statistical summary of EUIs for sampled 51 office buildings

3.4 Energy consumption benchmark

3.4.1 Indicators for benchmark

To ensure sustainable use of energy, Chinese government has launched the energy efficiency supervision strategy for the large scale public buildings, which consists of energy consumption statistics, energy audit, energy consumption benchmark, price increase beyond ration and energy efficiency public notice, where energy consumption benchmark was adopted as the classification basis [26]. This work used six indicators involving mean index of total energy consumption (I₁), average of EUIs (I₂), quadratic average of EUIs (I₃), median of EUIs (I₄), percentile of EUIs (I₅), and mode of EUIs (I₆), to analyze the energy consumption benchmark of the sampled office buildings [27-29].

1) Mean index of total energy consumption

I₁ is to illustrate the relationship between the total energy consumption and total GFA, and reveals an overall level of energy consumption. It is determined by

                                  (1)

where C_i is the energy consumption of sampled office building; A_i is the GFA of the sampled office building; n is the sample size. The subscript i denotes the building number.

2) Average of EUIs

I₂ averages the EUIs of the sampled buildings. It is defined as

                                 (2)

where E_i is the energy use index of the sampled building.

3) Quadratic average of EUIs

I₃ is an average value lying between reasonable average level and advanced level. The arithmetic mean represents the reasonable average level of EUIs. The buildings with lower EUI than the arithmetic mean belong to advanced level. I₃ can be obtained by

                        (3)

where E_j is the energy use index for the building with lower EUI than the arithmetic mean; m is the quantity of buildings with lower EUI than the arithmetic mean.

4) Median of EUIs

I₄ is defined as a specific EUI, and is used to divide the EUIs of all the sampled buildings into a higher half and a lower half. It is the median value, when all the EUIs are arranged in ascending order. I₄ reveals the central tendency of energy consumption, and avoids the influence of extremes (maximum or minimum).

5) Percentile of EUIs

I₅ is the value at a certain percentile of EUIs. The certain percentiles in this work are set to be 60% and 75%.

6) Mode of EUIs

The mode in statistics refers to the values with evident central tendency in the statistical distribution, and represents the general level of data. It can also be explained as the value or item occurring most frequently in a series of observations or statistical data. Therefore, I₆ represents the obvious central tendency of energy consumption for the sampled buildings. To get I₆, the sampled data are grouped, and the modal group, i.e., the group with the maximum frequency, is determined. I₆ is obtained by

                (4)

where L is the lower limit of modal group; f is the frequency of modal group; f_-1 is the frequency of previous group of modal group; f₊₁ is the frequency of next group of modal group.

And, d, the interval of groups, is calculated by

                                     (5)

where R is the range, which is the difference between the maximum EUI and the minimum EUI in all the sampled office buildings; n^* is the number of the groups, and is ascertained by [30]:

                               (6)

where n is the sample size.

3.4.2 Developing of benchmark

The six indicators were calculated, and the results are listed in Table 3. It is shown that the minimum indicator is the mode of EUIs, while the maximum     is the mean index of total energy consumption. The  range of indicators lies between 55.26 kW·h/m² and        98.31 kW·h/m², and the percentile range is between 34.10% and 76.42%. Mode of EUIs needs to be elaborated. Firstly, with sample size of 51, the parameter n* was ascertained to be 6.63 through Eq. (6). Secondly, after finding the value of R, d was calculated to be  74.33 kW·h/m² by  Eq. (5). Considering that d is generally the multiples of 5 or 10 [30], d was set to be 75 kW·h/m² in this work. Then, the EUIs of the sampled office buildings were divided to seven groups by the interval of 75 kW·h/m². The frequency of the EUIs in each group was obtained, and the frequency distribution of EUIs of 51 sampled office buildings is shown in Fig. 2, where 28 sampled buildings consumed annually 0-75 kW·h/m², reaching 54.90%, the highest percentage. Therefore, the group of 0-75 kW·h/m² was determined to be the modal group, and I₆ was 55.26 kW·h/m² through Eq. (4).

Table 3 Statistical summary of six indicators

Each indicator has the advantage and disadvantage. Mean index of total energy consumption represents an overall level of energy consumption, and can be used to compare the building energy consumption of different areas, but it threatens to weaken variability of energy consumption. Average of EUIs illustrates an average level of energy consumption, and is also used for the

Fig. 2 Frequency distribution of EUIs of 51 sampled office buildings

comparison, but the extremes endanger the authenticity of average of EUIs. Quadratic average of EUIs is apt to represent the lower energy consumption with sampled buildings, and fails to present the overall energy consumption. Median of EUIs represents the medium level of building energy consumption, but cannot be representative of the overall sampled buildings. Percentile of EUIs reveals energy use at a certain percentile of all the sampled buildings, and the determination of percentile must be in line with the actual situation. Mode of EUIs reflects the central tendency of EUIs in the sampled buildings, and avoids the influence of extremes; however, because the frequency difference between group of 0-75 kW·h/m² and group of 75-150 kW·h/m² is not significant, mode of EUIs becomes less reliable.

Considering the actual situation in Hainan Province and the characteristics of the six indicators, the benchmark for building energy consumption of Hainan Province is more appropriate to be a certain interval than the alternatives, with a lower limit of 55.26 kW·h/m² and an upper limit of 98.31 kW·h/m². The EUI distribution in ascending order, and the scope of upper limit and lower limit are shown in Fig. 3.

Fig. 3 EUI distribution in ascending order, and upper limit and lower limit for benchmark

3.5 Indoor environmental quality

The enhancement of the building energy efficiency should not sacrifice the indoor environmental quality. Field tests were conducted to monitor the level of the indoor environmental quality in sampled office buildings. Portable data loggers were used for a continuous monitoring of three parameters characterizing the indoor environmental quality, namely, the indoor air temperature (from -20 to 70 °C with ±0.4 °C accuracy), the relative humidity (25%-95% range with ±5% accuracy) and the CO₂ content (0–3×10^-3 range with ±3×10^-5 accuracy). The data loggers for indoor air temperature and relative humidity (RH) recorded the data in intervals of 30 min, while the CO₂ content was monitored in intervals of   10 min. With the different sizes and functions among the test office rooms, the measurement was located in 1-5 spots with height of 1-1.2 m, according to the “Energy Audit Guide for Government Office Buildings and Large-scale Public Buildings” [31].

During the period of testing, the average    outdoor temperature reached 30.2 °C, while the indoor temperature averaged 27.28 °C. Indoor relative humidity averaged 65.68%, and the corresponding outdoor value was 69.93%. The ranges between the maximum and minimum of the indoor temperature and relative humidity were 8.2 °C and 28.59%, respectively. The large ranges revealed the potential to improve the indoor thermal condition of office buildings in Hainan Province. A graphical method is provided in the ASHRAE standard 55—2004 [32] for determining comfortable thermal conditions in occupied spaces. An indoor thermal comfort zone, i.e. indoor temperature of 22-26 °C and indoor relative humidity of 25%-65%, was ascertained. The upper limit of CO₂ content is 1×10^-3, according to the ASHRAE standard. The overall distributions for average indoor temperature, average relative humidity, and average CO₂ content of each sampled office building are illustrated in Figs. 4-6, and the responding limits are also marked. As is shown, only approximate 20% of sampled buildings were in the temperature range of ASHRAE standard. About 59% of sampled buildings exceeded the upper limit of indoor relative humidity of ASHRAE standard, but no building exceeded the lower limit. And all the average CO₂ content met ASHRAE standard. It should be noted that the CO₂ content remained low at periods of non-occupancy.

Fig. 4 Average indoor temperature distribution of sampled office buildings and comparison between average indoor temperature (T_up and T_low are upper limit and lower limit, respectively, according to ASHRAE comfort zone)

Fig. 5 Average relative humidity distribution of sampled office buildings and comparison between average relative humidity (RH_up and RH_low are upper limit and lower limit, respectively, according to ASHRAE comfort zone)

Fig. 6 Average CO₂ content distribution of sampled office buildings and comparison between average CO₂ content and limit according to ASHRAE standard

3.6 Benchmark of building energy consumption combined with indoor environmental quality

Air conditioning and ventilation systems are applied in the buildings to create a comfort indoor environment [33]. The indoor environmental quality can explicitly or implicitly influence the thermal comfort, health, and productivity of humans in the buildings [34-35]. However, blind pursuit of high indoor environmental quality poses difficulties for building energy efficiency. The indoor environmental quality and building energy consumption need to be kept at a reasonable balance. The best performance office buildings are those with low energy consumption and advanced indoor environmental quality.

China issued “Chinese Indoor Air Quality Standard GB/T 18883—2002” [36], which regulated acceptable thermal comfort condition, i.e. indoor temperature of 22-28 °C and relative humidity of 40%-80%, based on Chinese national conditions. Compared with ASHRAE standard for comfortable condition, these limits were considered as an acceptable condition. Buildings with the indoor environmental quality conforming to ASHRAE standard were classified as a good rank, while the buildings incompatible with the ASHRAE standard but in accordance with GB/T 18883—2002 were classified as normal rank, and the remaining buildings were classified as the bad rank. Correspondingly, the building energy consumption was also divided into three ranks: good (BEC<55.26 kW·h/m²), normal (55.26 kW·h/m²≤ BEC≤98.31 kW·h/m²), and bad (BEC>98.31 kW·h/m²). Table 4 lists the results of benchmark of energy consumption combined with indoor environmental quality, which have the following characteristics:

1) Only two buildings, demonstrating excellent indoor thermal comfort while consuming small amounts of energy, were defined as the best performance buildings. It implied the large potential and the urgency of indoor environmental quality enhancement.

2) Fifteen buildings, being the max percentage of 29.41%, were classified into the both normal ranks, which represented a majority situation.

3) Only 7.84% of sampled buildings met the ASHRAE standard and 70.69% of sampled buildings conformed to GB/T 18883—2002, displaying a large gap of indoor environmental quality of office buildings between the Hainan Province and advanced area in the world.

Table 4 Summary for benchmark of energy consumption combined with indoor environmental quality

4 Conclusions

1) Three types of fuels, i.e. electricity, natural gas and diesel were used in office buildings of Hainan Province, and the electricity dominated in the energy consumption with a share of 99.79%.

2) The energy consumption of the air-conditioning dominated in the energy use, accounting for 43.18%, followed by the professional equipment in the particular areas, accounting for 26.9%.

3) Considering the condition of Hainan Province, the benchmark for office buildings was defined as an interval with an upper limit of 98.31 kW·h/m² and a lower limit of 55.26 kW·h/m². The building energy consumption was classified by three ranks: good, normal and bad.

4) The good environmental quality is conducive for physical health and work performance, which is considered as a factor in the development of an energy benchmark. Therefore, indoor environmental quality of the sampled buildings was analyzed under the ASHRAE standard (comfortable standard) and GB/T 18883—2002 (acceptable standard), which also divided the performance of indoor environmental quality into three ranks of good, normal and bad.

5) By combining building energy consumption with indoor environmental quality, the analysis revealed that only two buildings could be defined as best performance buildings with both good ranks, and with a share of 29.41%, the majority of sampled buildings were in the both normal ranks.

Acknowledgements

The authors would like to acknowledge the Department of Housing and Urban-Rural Development of Hainan Province in cooperating in building energy audit and providing building energy consumption data and valuable comments for this work.

References

[1] LOMBARD L P, ORTIZ J, POUT C. A review on buildings energy consumption information [J]. Energy and Building, 2008, 40: 394-398.

[2] GEOFFREY K F T, KELVIN K W Y. A study of domestic energy usage patterns in Hong Kong [J]. Energy, 2003, 28: 1671-1682.

[3] RADHI H. A systematic methodology for optimizing the energy performance of buildings in Bahrain [J]. Energy and Building, 2008, 40: 1297-1303.

[4] SAIDUR R. Energy consumption, energy savings, and emission analysis in Malaysian office buildings [J]. Energy Policy, 2009, 37: 4104-4113.

[5] National Bureau of Statistics of China. China statistical yearbooks (1999~2010) [M]. Beijing: China Statistics Press, 1999~2010. (in Chinese)

[6] DINCER I, DOST S. Energy and GDP [J]. Int J Energy Res, 1997, 21: 153-67.

[7] KADOSHIN S, NISHIYAMA T, ITO T. The trend in current and near future energy consumption from a statistical perspective [J]. Applied Energy, 2000, 67: 407-17.

[8] EGERLIOGLU F, MOHAMAD A A, GUVEN H. Economic variables and electricity consumption in Northern Cyprus [J]. Energy, 2001, 26: 355-362.

[9] National Bureau of Statistics of China (NBS). China statistical yearbook 2006 [M]. Beijing: NBS, 2006. (in Chinese)

[10] LI Jun. Towards a low-carbon future in China’s building sector—A review of energy and climate models forecast [J]. Energy Policy, 2008, 36: 1736-1747.

[11] National Bureau of Statistics of China, National Energy Administration. China energy statistical yearbooks (2004~2010) [M]. Beijing: China Statistics Press, 2004~2010. (in Chinese)

[12] KAYA A, YALCINTAS M. Energy consumption trends in Hawaii [J]. Energy, 2010, 35: 1363-1367.

[13] CHIRARATTANANON S, TAWEEKUN J. A technical review of energy conservation programs for commercial and government buildings in Thailand [J]. Energy Conversion and Management, 2003, 44: 743-762.

[14] YANG L, JOSEPH C L, TSANG C L. Energy performance of building envelopes in different climate zones in China [J]. Applied Energy, 2008, 85: 800-817.

[15] BALARAS C A, DROUTSA K, ARGIRIOU A A, WITTCHEN K. Assessment of energy and natural resources conservation in office buildings using TOBUS [J]. Energy and Buildings, 2002, 34: 135-153.

[16] ROULET C A, BLYUSSEN P M, DUCARME D, FERNANDES E D O, RIBERON J, WOUTERS P. Ventilation performance and energy consumption in European office buildings [J]. Proceedings of Healthy Buildings, 1995, 95: 1299-1304.

[17] SPYROPOULOS G N, BALARAS C A. Energy consumption and the potential of energy savings in Hellenic office buildings used as bank branches—A case study [J]. Energy and Buildings, 2011, 43: 770-778.

[18] NIKOLAOU T, SKIAS I, KOLOKOTSA D, STAVRAKAKIS G. Virtual building dataset for energy and indoor thermal comfort benchmarking of office buildings in Greece [J]. Energy and Buildings, 2009, 41: 1409-1416.

[19] FEDERSPIEL C, ZHANG Qiang, ARENS E. Model-based benchmarking with application to laboratory buildings [J]. Energy and Buildings, 2002, 34: 203-214.

[20] CHUNG W, HUI Y V, LAM M Y. Benchmarking the energy efficiency of commercial buildings [J]. Applied Energy, 2006, 83: 1-14.

[21] ZHANG Xiao-ling. Establishing energy consumption quota and reducing energy consumption of public buildings to implement the effective supervision to the government office buildings and large-scale public buildings [J]. Construction Science and Technology, 2009, 8: 22-23. (in Chinese)

[22] HAN Guang, MA Liang-dong, CHEN Yong-pan. Statistics and analysis of the energy consumption of typical government office buildings in Dalian [J]. Refrigeration Air Conditioning & Electric Power Machinery, 2009, 4: 77-80. (in Chinese)

[23] ZHAO Shu-jie. Energy consumption status and energy-saving of typical office building in Suzhou [J]. Building Efficiency, 2010, 38: 73-76. (in Chinese)

[24] SUN Li-xin, YAN Zeng-feng , YANG Li-ping. Investigation and analysis on energy consumption of public buildings in Xi'an [J]. Building science, 2008, 24: 25-28. (in Chinese)

[25] YANG Yong, ZHU Qing-song. The energy consumption survey analysis of municipal government office buildings and large public buildings [J]. Chinese and Overseas Architecture, 2010, 7: 193-195. (in Chinese)

[26] JIN Zhen-xing, WU Yong, LI Bai-zhan, GAO Ya-feng. Energy efficiency supervision strategy selection of Chinese large-scale public buildings [J]. Energy Policy, 2009, 37: 2066-2072.

[27] CHUNG W. Review of building energy-use performance benchmarking methodologies [J]. Applied Energy, 2011, 88: 1470-1479

[28] ZHOU Zhi-yong, FU Xiang-zhao, LIU Jun-yue, HE Hua. Energy consumption quota of public buildings based on statistics data [J]. Gas & Heat, 2009, 29: 14-17. (in Chinese)

[29] TANG Fang. Reasonable application of arithmetic mean, mode and median [J]. China Statistics, 2008, 3: 49-50. (in Chinese)

[30] XIONG Jun-shun, WANG Juan, LIU Gan. Statistics tutorial [M]. Hangzhou: Zhejiang University Press. 2007: 33-37. (in Chinese)

[31] Ministry of Housing and Urban-Rural Development of People’s Republic of China. Energy audit guide for government office buildings and large-scale public buildings [EB/OL]. http://www.gov.cn/zwgk/ 2007-11/02/con-tent_793903.htm. (in Chinese)

[32] ANSI/ASHRAE 55-2004. Thermal environmental conditions for human occupancy [S]. Atlanta, USA, 2004.

[33] ZHANG L Z, NIU J L. Indoor humidity behaviors associated with decoupled cooling in hot and humid climates [J]. Building and Environment, 2003, 38: 99-107.

[34] HOLM D, ENGELBRECHT F A. Practical choice of thermal comfort scale and range in naturally ventilated buildings in South Africa [J]. Journal of the South African Institution of Civil Engineering, 2005, 47: 9-14.

[35] HWANG R L, LIN T P, CHENG M J, CHIEN J H. Patient thermal comfort requirement for hospital environments in Taiwan [J]. Building and Environment, 2007, 42: 2980-2987.

[36] Ministry of Environmental Protection of the People’s Republic of China, Ministry of Health of the People’s Republic of China. Chinese indoor air quality standard. GB/T 18883—2002 [S]. (in Chinese)

(Edited by YANG Bing)

Foundation item: Project(2011BAJ01B05) supported by the National Science and Technology Pillar Program during the Twelfth Five-year Plan Period of China

Received date: 2011-07-26; Accepted date: 2011-11-14

Corresponding author: L? Shi-lei, Associate Professor, PhD; Tel: +86-22-27402177; E-mail: lvshilei@tju.edu.cn