Benchmarking energy performance for cooling in large commercial buildings

Benchmarking energy performance for cooling in large commercial buildings

Energy & Buildings 176 (2018) 179–193 Contents lists available at ScienceDirect Energy & Buildings journal homepage: www.elsevier.com/locate/enbuild...

2MB Sizes 0 Downloads 2 Views

Energy & Buildings 176 (2018) 179–193

Contents lists available at ScienceDirect

Energy & Buildings journal homepage: www.elsevier.com/locate/enbuild

Benchmarking energy performance for cooling in large commercial buildingsR Haoru Li a, Xiaofeng Li a,b,∗ a b

Department of Building Science, School of Architecture, Tsinghua University, Beijing 100084, China Key Laboratory of Eco Planning & Green Building, Ministry of Education, Tsinghua University, China

a r t i c l e

i n f o

Article history: Received 31 July 2017 Revised 2 June 2018 Accepted 14 July 2018 Available online 24 July 2018 Keywords: Energy benchmarking Air-conditioning system Multi-level index system Shopping center

a b s t r a c t Urban development rapidly increases the total area and the energy consumption of large commercial buildings. Accordingly, the total energy consumption is the focus of numerous energy saving studies. An optimal benchmarking system provides a recommended energy consumption level for each building as well as identifying and prioritizing challenges to energy saving in each building. Based on detailed submetering system data and building operational data, this paper presents a simplified energy consumption benchmarking method for air-conditioning systems that cool large commercial buildings. Firstly, a multi-level benchmarking index system is established. Next, the energy performance data of eight large shopping centers in China validates the simplified benchmarking method and finally, the energy performance data of the participating shopping centers is analyzed under the present method. Our data focus on the energy saving potential, as well as informing improvement strategies for building energy performance, that can be used for efficient energy benchmarking process for cooling in large-scale commercial buildings. © 2018 Elsevier B.V. All rights reserved.

1. Introduction 1.1. Backgrounds Between 20 0 0 and 2012, the worldwide building energy consumption increased from 102 to 120 EJ, which accounted for more than 30% of total final energy consumption across all economic sectors [1]. During the same period, the building energy consumption in China grew by 37%. Under current consumption trends, this figure is predicted to increase another 70% by 2050 [2]. In 2013, to respond to this issue, the Chinese government announced a “Total Energy Use Control” plan as well as publishing the Energy Development, “Twelfth Five Year Plan”, which mandated a ceiling for energy usage and stronger control of total energy [3]. However, in more recent years, urbanization and overall economic development have significantly improved across China, resulting in a substantial increase in commercial floor space and energy consumption in large-scale commercial buildings. The large-scale commercial buildings consume 20 0–40 0 kWh/(m2 ·a), 4–8 times more than the ordinary public buildings [4], which has raised concern about the energy consumption requirements for these

R ∗

Declarations of interest: None. Corresponding author. E-mail address: xfl[email protected] (X. Li).

https://doi.org/10.1016/j.enbuild.2018.07.039 0378-7788/© 2018 Elsevier B.V. All rights reserved.

types of buildings. Therefore, the establishment of a benchmarking method is necessary to enable scientific evaluation of large-scale commercial building energy performance and inform efficient energy consumption strategies. 1.2. Literature review Building energy benchmarking compares the actual energy performance of similar buildings [5,6]. Considerable researches have investigated building energy benchmarking methods using a variety of classification techniques [7–9]. Wang et al. [7] reviewed and then classified energy performance assessment methods into three categories: calculation-based, measurement-based and hybrid methods. Kinney and Piette [8] classified benchmarking methods into four types: statistical methods, points-based rating systems, simulation methods as well as hierarchical and end-use metrics. Li et al. [9] categorized energy benchmarking methods as black-box, gray-box and white-box. The black box-method uses data fitting techniques rather than physical knowledge. This corresponds to statistical methods per Kinney’s classification which includes regression, artificial neural networking, data envelopment analysis, and other mathematical methods [10–14]. The most commonly used black-box method is the regression method [15–19], which is based on a large sample data size. This method achieves rapid evaluation of building energy consumption through the

180

H. Li, X. Li / Energy & Buildings 176 (2018) 179–193

Nomenclature A e E cp CEsupply CLdemand g G h j n q q0 Q SHGC SR t T U V WWR

Area, m2 Solar radiation absorption factor of the exterior surface Accumulated electricity consumption, kWh Specific heat capacity of water, kJ/(kg·°C) Cooling energy supplied by the air-conditioning system, kWh Cooling demand, kWh Water flow rate, m3 /h Airflow rate, m3 /h Air enthalpy value, kJ/kg Total cooling hours Number of occupants per hour Cooling load from building envelop, W Total heat released by an adult, W Cooling load, kWh Solar heat gain coefficient Solar radiation intensity, W/m2 Chilled water temperature, °C Air temperature, °C Heat transfer coefficient, W/(m2 ·K) Building volume, m3 Window to wall ratio

Greek symbols ∂ out Convective heat transfer coefficient of the exterior surface, W/(m2 ·K) ε Building shape factor η Correction factor of the motor installation position ηm Electrical efficiency of fans and pumps ρ Water density, kg/m3 ϕ Percentage of adult male, adult woman and children. τ Time interval, hour Subscript lighting and equip Lighting and equipment exf Exfiltration fan Terminal system i Orientation, which represents east, west, south, north and the horizontal plane (i.e., roof) in Indoor air inf Infiltration inlet Inlet chilled water l The number of chillers ma Mechanical fresh airflow me Mechanical exhausted airflow out Outdoor air outlet Outlet chilled water pump,c Condensate water pumps pump,ch Chilled water pumps r Refrigeration system total Total fresh outdoor air tower Cooling tower wall Exterior wall win Exterior window

input of finite parameters. However, the reliability of this model is highly dependent on sample data accuracy which makes it fundamentally unusable in real-world applications. The white-box method is based on physical principles whereby constraints are embedded into the modelling of building components. The most

common white-box method is the detailed simulation method [20–22] in which the built-in simulation software is very detailed; as such, it is able to provide accurate benchmarking results. Moreover, the input parameters can be used to analyze different working conditions as well as comparing the influence of different factors on energy consumption. Here, a disadvantage is that each building must be separately modeled, which is a complex and a nonreplicable process. The gray-box method combines both physical knowledge and data-fitting techniques to derive a model that typically utilizes Bayesian and RC networks for air-conditioning loads. The Bayesian networks can only be applied in the buildings with a similar building usage pattern and energy consumption profile, while RC networks are only applicable in heating and cooling load calculations [9]. Given that the type and proportion of both business and operational conditions in each building significantly vary, each building must set individual energy consumption targets. While all three methods can be used to benchmark building energy consumption, only the results obtained by the white-box method reflect the actual management level for each target building as well as guiding further energy saving strategies. In general, calculations are simplified to promote efficiency. Yan et al. [23] proposed a simplified assessment method to calculate cooling loads in existing buildings. They used an optimization algorithm to disaggregate total building energy consumption into per-user consumption, i.e., airconditioning consumers, internal consumers and other consumers. However, the assessment only occurred at the main system level and it did not provide an evaluation criterion.

1.3. Purpose of the paper The energy consumed by an air-conditioning system usually accounts for 50% of total energy consumption of a building [24]; as such, air-conditioning has the greatest potential for energy reduction. Therefore, this paper presents a simplified method for benchmarking the energy consumption of air-conditioning systems that cool large commercial buildings. A multi-level benchmarking index system of energy consumption is established using simplified calculation formulas based on fundamental principles. The benchmark provides recommended energy consumption levels for each building, which represents the average building energy performance within the same type of buildings in China. Then, the building operation and management level can be evaluated from the deviation between the actual value and recommended value as well as the energy-saving potential. A field test validates the reliability of the calculation method by providing actual operational data from sample buildings. The energy performance of the cooling systems in eight large shopping centers is evaluated using the index system, and then the energy saving potential is analyzed.

2. Methodology The energy performance of an air-conditioning system in a building is determined by the cooling demand of the building and the energy efficiency of the air-conditioning system. The cooling demand indicates the cooling load of a building under controlled conditions and the energy efficiency describes the operational level of the air-conditioning system and the related devices. As such, benchmarking the cooling energy performance of a commercial building follows these steps: benchmark cooling energy; rate energy efficiency; and benchmark energy consumption. The following subsections describe the calculation methods.

H. Li, X. Li / Energy & Buildings 176 (2018) 179–193

2.1. Cooling energy calculation The cooling load is established per the requirements for a given building to maintain environmental parameters within a certain range. Distribution equipment consumes electricity and generates heat in the process of cooling transmission, which results in additional cooling demands. Therefore, the cooling supply of an airconditioning system in a commercial building is used to meet the cooling load demands and offset the heat generated from the cooling transmission process. The balance between the cooling demand and supply can be expressed as:

C Esupply = C Ldemand + Qtds

(1)

Typically, the cooling load of a building comes from the envelope, ventilation-based outdoor air, occupants as well as internal lighting and equipment, which can be represented as:

C Ldemand = Qenvelope + Qventilation + Qoccupant + Qelectricity

(2)

2.1.1. Envelope Setting aside the thermal mass and time response of the envelope, the cooling load from a building envelope corresponding to a selected orientation, W, is usually calculated by:

qi =

  Uwal l ,i Awal l ,i + Uwin,i Awin,i (Tout − Tin ) i

+

  e Uwal l ,i Awal l ,i + SHGC · Awin,i SR ∂out

i

Eq. (3) can also be expressed in the form:

qi = εV {[Uwall · (1 − W W Ri ) + Uwin · W W Ri ] · (Tout − Tin )   e + Uwall · (1 − W W Ri ) + SHGC · W W Ri · SR

∂out

(4)

Therefore, the accumulated heat gain from the building envelope during an entire cooling season (Qenvelope kWh) can be calculated by:

Qenvelope =

 j

i

qi · τ 10 0 0

 1.2Gtotal (hout − hin ) · τ 3600

Qtds = E f an η + E pump,ch ηm

(6)

2.2. Energy efficiency calculation Usually, an air-conditioning system comprises the refrigeration system, chilled water transport system and terminal system. Hence, the energy consumption of an air-conditioning system, EHVAC , is:

EHVAC = Er + E pump,ch + E f an

Qoccupant

(7)

j

2.1.4. Lighting and equipment The cooling load caused by heat from lighting and equipment (Qelectricity , kWh) can be directly calculated by electricity consumption data recorded by the sub-metering system:

Qelectricity = Elighting

and equip

(8)

(11)

The energy efficiency of an air-conditioning system is an indicator of the building operation and management level. Evaluation indicator indexes include the energy efficiency ratio of the air-conditioning system, EERs; the energy efficiency ratio of the refrigeration system, EERr; the energy efficiency ratio of the terminal system, EERt; and the water transport factor of chilled water, WTFchw. The calculation formulae are shown below.

E E Rs =

C Esupply EHVAC

(12)

E E Rr =

C Esupply Er

(13)

E E Rt =

C Esupply E f an

(14)

W T F chw =

C Esupply E pump,ch

(15)

In the air-conditioning system, water-cooled chillers are considered as the cold sources, and the refrigeration system consists of chillers, condensate water pumps and cooling tower. Given that the energy consumption of a cooling tower comprises only a small proportion of total energy consumption of the refrigeration system, further analysis indicators for the refrigeration system contain a Coefficient of Performance (COP) for the chillers and Water Transport Factor (WTF) for the condensate water, WTFcw:

COP =

 q 0 nϕ = · τ 10 0 0

(10)

The refrigeration system includes chillers, condensate water pumps and a cooling tower. The energy consumption of the refrigeration system, Er , is:

j

2.1.3. Occupants Heat gain released by occupants (Qoccupant , kWh) is calculated by:

(9)

When the motor is installed inside the transferred fluid, then the value of η is 1. When the motor is installed outside the transferred fluid, then the value of η is equal to ηm .

(5)

2.1.2. Ventilation Outdoor air enters a building via mechanical ventilation and infiltration. Usually, outdoor air is filtered and cooled by large Air Handling Units (AHUs), Makeup Air Units (MAUs) and Pre-cooling Air handling Units (PAUs), which then mechanically transfer air into the building. Although commercial buildings are designed as enclosed structures, outdoor air can enter a building through doorways and unintentional openings. Therefore, the ventilation load from outdoor air (Qventilation , kWh) in a cooling season is:

Qventilation =

2.1.5. Transmission system As mentioned before, in addition to the building cooling demand, extra heat is generated by the air-conditioning system in the transmission and distribution cooling processes. This extra cooling consumption (Qtds , kWh) is determined by the energy consumption of the related equipment:

Er = Echil l er + E pump,c + Etower (3)

181

C Esupply Echiller

W T F cw =

C Esupply + Echiller E pump,c

(16) (17)

2.3. Energy performance evaluation Fig. 1 shows the systematic benchmarking index system of energy performance for cooling purposes. Fig. 2 illustrates a benchmarking flowchart process using the methodology presented in this section. All required input data, including energy consumption data, building design data, weather data and building operation data, are acquired from the building landloards. The actual

182

H. Li, X. Li / Energy & Buildings 176 (2018) 179–193

Fig. 1. Benchmarking index system of energy performance for cooling purposes.

Fig. 2. Detailed energy performance benchmarking process.

and recommended benchmarking index values are obtained by inputting the actual and recommended values of the required input parameters into the above-mentioned calculation formulae. The energy consumption index is used to benchmark the energy performance of the target building, with detailed cooling energy and energy efficiency indexes providing additional information to interpret the energy performance. Thus, the simplified benchmarking energy consumption method for an air-conditioning system (with a cooling purpose) is established. The recommended value of cooling energy can be calculated according to the cooling energy calculation of envelope, ventilation, occupants, lightning and equipment as well as transmission system. The values for the envelope criteria from the China national standard, GB 50189-2015Design standard of energy efficiency of public buildings [25], are used to calculate the recommended envelope cooling energy. For the recommended ventilation cooling energy, the total fresh air rates are calculated by actual

occupant number and the fresh air rate per person (derived from national standard). For a shopping center, the shoppers, lighting and equipment are directly related to business conditions. Therefore, the recommended cooling energy from occupants, lighting, equipment and transit system is equal to the actual value. For energy efficiency, the China national standard, GB 501892015, provides a recommended value for energy efficiency under typical working conditions; however, it lacks an annual limit value. Another China national standard, GB/T 17981-2007 Economic operation of air conditioning systems [26] (issued in 2007), which is based on GB 50189-2005 Design standard of energy efficiency of public buildings (issued in 2005) [27], provides annual evaluation criteria for energy efficiencies. Due to improvements in chiller units over the past decade, the value established in 2007 is no longer suitable. Therefore, to conduct further benchmarking, this paper uses the relationship between values in standards GB/T 17981-2007

H. Li, X. Li / Energy & Buildings 176 (2018) 179–193

183

Fig. 3. Method verification process.

and GB 50189-2005 to obtain an annual limit value for energy efficiency per standard GB 50189-2015. These standards are developed based on a large number of actual project measurement and investigation and obtain the annual average energy efficiency of actual projects in China at present. Indicators have taken into account factors such as partial load, outdoor air temperature changes, changes in the number of cold machines and other factors. Therefore, these indicators are used as recommended values in this benchmark. 3. Measurement and verification The energy performance and system operational data from eight large shopping centers, loacted in China, are used to verify the cooling energy calculation method presented in Section 2. Fig. 3 presents the method verification process. Firstly, the centers were selected based on different climate zones. Next, necessary data was investigated in each of the selected shopping centers and a series of measurements were used to acquire key parameters that were not provided in the standard operational records, including outdoor air volume from mechanical ventilation systems and infiltration process. Then, the calculated cooling energy (from the present method) is compared with the measured cooling energy from the operational records and parameters, with a relative error within the acceptable range and finally, the simplified cooling energy calculation method is verified. 3.1. Sample building selection The China national standard GB 50176 identifies five major climate zones: severe cold region, cold region, hot summer and cold winter region, hot summer and warm winter region, and mild region [28]. The zoning criteria are based on the average temperature in the coldest and hottest months of the year, and on the number of days that the daily average temperature is below 5 °C or above 25 °C [29]. The building envelope criteria and some airconditioning design guidelines (recommended by national standards) are based on climate zones [30]. Fig. 4 shows the locations of all sampled buildings, which are located across the different climate zones. 3.2. Data acquisition All of the required data can be classified into three types, including building design, building operation and energy consumption data. Building design data includes floor area, thermal parameters of envelope, building shape factors and so on, which

Fig. 4. Positions of sampled shopping centers.

can be obtained directly from the building plans. Building operation data includes ventilation, shopping traffic data, obtained using real-time measurements. Energy consumption data was obtained directly from sub-metered systems, reflecting different operation needs in different buildings. Detailed analysis of the data acquisition is shown below. 3.2.1. Building design data Building design data were obtained directly from the building plans, as shown in Table 1. For instance, the Building C structure type is cast-in-place frame-shear wall structure. Its exterior wall is made of 30 mm stone, 90 mm insulation material and 220 mm concrete. The roof is composed of 4 mm waterproof material, 40 mm insulation material and 210 mm concrete. All windows are double glazed window with the glass thickness of 6 mm. Three centrifugal chillers and one screw chiller are selected as the cold source. Although, the similar structures and envelope materials are used for the other buildings, the material thickness varies due to the different climatic zones. Water-cooled chillers are also selected as the cold source in these buildings, when the different buildings with different cooling demand have variable refrigerating capacity. In all of the sampled shopping centers, the designed indoor temperature is 24 °C and designed indoor relative humidity is within 40%−60%. The shopping centers are usually operated from 9:0 0 to 22:0 0. All the individual premises inside the shopping

184

H. Li, X. Li / Energy & Buildings 176 (2018) 179–193 Table 1 General description of sampled shopping centers. Building ID Total floor area (m2 ) Floors Building volume (m3 ) Building shape factor Thermal parameter [W/(m2 ·K)]

WWR

Exterior wall Window Roof Skylight East West South North Horizontal

Climate zone Total cooling days Number of chillers Total refrigerating capacity rate (kW) CDD 24 (°C·d)

A 57,996 3 173,988 0.03

B 69,278 3 207,834 0.09

C 93,259 5 279,777 0.08

D 79,0 0 0 3 237,0 0 0 0.09

E 67,327 4 201,981 0.08

F 55,410 3 166,230 0.13

G 52,271 3 156,813 0.10

H 51,950 3 155,850 0.10

0.35 2.9 0.31 2.6 0.27 0.17 0.39 0.22 0.16 SC 113 4 11,005 82

0.75 2.27 0.41 1.49 0.19 0.24 0.19 0.22 0.15 SC 96 4 11,156 30

0.5 2.28 0.44 2.28 0.29 0.34 0.4 0.35 0.28 C 151 4 11,887 219

0.82 3.9 0.66 3 0.02 0.01 0.25 0.05 0.07 HSCW 184 4 13,360 279

0.72 2.7 0.58 2.6 0.33 0.10 0.17 0.17 0.15 HSCW 153 3 14,097 334

0.89 2.6 0.66 2.6 0.04 0.33 0.31 0.28 0.17 HSCW 191 4 11,630 334

1.3 4.37 1.02 3.3 0.17 0.06 0.04 0.21 0.14 HSWW 250 4 13,362 496

1.22 3.5 0.65 3.5 0.42 0.51 0.48 0.47 0.35 HSWW 259 4 10,738 594

Climate zone abbreviations: SC – Severe Cold region; C – Cold region; HSCW – Hot Summer and Cold Winter region; HSWW – Hot Summer and Warm Winter region.

centers, including restaurants, retail shops, and entertainments, are operated as the same timetable. There are supermarkets underneath the ground floor, which do not belong to the shopping centers and are operated by another air-conditioning system independently. The indoor design parameters of the supermarkets are the same as the shopping centers. Therefore, the heat transfer from this space is not considerable. This is the common case in the shopping centers in China. If not, then the cooling load from the basement should be considered.

3.2.2. Building operation data Airflow rate measurement. As shown in Eq. (6), the total fresh outdoor airflow rate is the necessary parameter to obtain the ventilation load. Generally, outdoor air comes into a building by mechanical ventilation systems or infiltration. In this paper, the thermal balloon anemometry is used to measure the mechanical ventilation wind speed directly at a long and straight section of the duct (Beijing Detector Instrument Co., Ltd. Model D30JS, measurement error is 0.1 m/s or 5% of the measurement value) of which the airflow is relatively uniform. The measured duct’s section is divided into several rectangles. Wind speed is measured at the center of each rectangular. The average wind speed is calculated and then, the mechanical ventilation rate is ascertained. This rate is constant, while the ventilation unit operation mode does not change. Therefore, the measurement is only conducted once, and measurement error of the mechanical ventilation airflow rate is about 10%. The traditional infiltration rate measurement methods (such as the fan pressurization method and the tracer gas dilution method) are not suitable for the large-scale shopping centers. According to the investigation, the outer doors of the shopping centers are usually kept opened during business hours to meet the center operational requirements. These doors are the main access points for infiltration or exfiltration airflows. Therefore, the infiltration rate is measured at entrances by thermal balloon anemometry (Beijing Detector Instrument Co., Ltd. Model D30JS). The entrances are selected as the measurement section as wind speed at the other sections is generally less than 0.2 m/s, leading to a larger measurement error. Measurement is conducted every 0.25 m × 0.25 m. Many anemometers are used to measure the wind speed at all entrances at the same time. The measurement takes 1 h, data is recorded every 30 s, and, the average wind speed per hour is calculated. Then, the infiltration or exfiltration airflow rate is ascertained. The measurement error is about 25%.

Infiltration or exfiltration airflow rates are affected by indoor and outdoor temperature difference, wind velocity and operational strategy of the mechanical ventilation system. To determine the actual ventilation environment, ventilation rates from all mechanical ventilation systems and outer entrances are averagely measured, including 15 AHUs and PAUs, 33 exhausters, and 13 entrances in each shopping center. The total inlet and outlet airflow mass balance relationships are used to validate the reliability of the measurement results. Taking the combined building and systems as a control (for volume), the airflow mass balance can be expresses as [31]:

Gma + Ginf = Gme +Gex f ,

(18)

Table 2 lists the measurement results. The relative error between the total inlet and outlet airflow rates are all less than 20%, which is within the acceptable error range. Hence, the measurement results are reliable. In each of the eight shopping centers, the exfiltration rate comprises a small proportion of the total outlet airflow, while the mechanical exhaust airflow rate accounts for a much larger proportion. Fig. 5 compares the measured infiltration rate and the difference between the mechanical exhaust airflow and fresh airflow rates. The average relative error is 18%. The results confirm that infiltration in these buildings is mainly dependent upon the ventilation difference between the mechanical exhaust and the supply systems, such that:

Gtotal = Gma + Ginf ≈ Gme

(19)

Given that the operation model of the mechanical ventilation system remains unchanged during the cooling season (in most shopping centers in China), the infiltration rate value can be treated as fixed. Therefore, when using Eq. (6) to calculate the actual ventilation cooling energy for an entire cooling season, the total outdoor air volume value can be directly derived from the mechanical exhaust airflow rate, Gme . Counting the number of occupants. The number of occupants in the shopping centers are counted by infrared counters, which are installed at every entrance. Hourly occupants getting in and out of the buildings are distinguished and counted separately. Then the number of occupants staying in the buildings every hour can be calculated. Fig. 6 presents the hourly number of people in a shopping center during a typical day.

H. Li, X. Li / Energy & Buildings 176 (2018) 179–193

185

Table 2 Airflow mass balance in the shopping centers. Building ID

A

B

C

D

E

F

G

H

Gma (× 104 m3 /h) Ginf (× 104 m3 /h) Gme (× 104 m3 /h) Gexf (× 104 m3 /h) Relative deviation between total inlet and outlet airflow rate(%)

2.0 12.5 −16.0 −1.3 −16.0

4.0 13.5 −12.8 −1.4 19.0

1.0 33.7 −35.0 −5.5 −17.0

19.0 34.0 −41.5 −5.0 12.0

4.0 25.5 −30.0 −4.3 −17.0

3.0 25.1 −29.5 −4.2 −18.0

11.0 16.8 −20.0 −3.3 17.0

1.0 21.3 −24.5 −1.4 −17.0

Fig. 5. Comparison of measured infiltration rate and difference in mechanical ventilation airflow rates.

Table 3 Average measurement start-up temperature in every month in eight buildings in four climate zones (°C) .

9000 8000 7000

person

6000 5000 4000 3000 2000

Climate zone Month\Building

SC A

B

C C

HSCW D E

F

HSWW G H

May June July August September October

21.9 22.8 23.5 21.8

21.8 22.9 23.2 22.0

24.0 24.6 25.5 25.7 24.1

23.2 24.3 25.5 24.7 24.4 23.1

23.4 24.3 24.4 25.9 24.9 22.6

24.7 24.8 24.4 24.8 24.8 25.0

22.8 23.8 25.4 24.9 22.8 22.0

25.6 24.9 24.8 25.3 25.9 25.3

1000 01:00 02:00 03:00 04:00 05:00 06:00 07:00 08:00 09:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00 19:00 20:00 21:00 22:00 23:00 00:00

0

Fig. 6. Hourly number of people in a shopping center during a typical day.

3.2.3. Energy consumption data Sub-metered system is used to record the detailed energy consumption data. National standard in China regulates the principle of sub-metered system installation [32]. Fig. 7 shows the classification method of sub-metered system in a typical shopping center. Energy consumption of each type of air-conditioning equipment can be measured and recorded separately, which is convenient for the data analysis. In individual tenants, only the total energy consumption is measured, including lighting and appliances. Considering that electricity used by the lighting and appliances in the tenants is almost transferred into air-conditioning load, the total energy consumption data is adequate for analysis. The time-step of sub-metered system for recording energy consumption data is 5 min. Fig. 8 presents the hourly energy consumption of total tenants in a shopping center during a typical day.

3.3. Method verification 3.3.1. Comparison with dynamic simulation Thermal mass and time response of the envelope may influence the indoor air temperature as the air-conditioning system is not operated 24 h in the shopping centers. However, ventilation through openable skylights during the nights is commonly adopted energy saving strategy which helps to reduce the night and startup temperature. Table 3 shows the average start-up temperature in every month in buildings in four climate zones. The most deviation between the start-up temperature and indoor design temperature is 2.2 °C in severe cold region. In order to evaluate the influence of the start-up temperature deviation on the total cooling energy, a series of calculation is conducted. The energy consumption simulation software DeST is used to dynamically simulate the cooling load of the eight buildings during the operation time [33]. In the simulation process, the ventilation rates during the night are adjusted to make the simulation start-up temperature, consistent with the measured results. During the business time, the indoor air temperature is set to be the designed temperature. The calculation results are shown in Table 4, with a relative difference of −8% to 6% between the cooling energy calculated by the simplified method proposed in this paper,

186

H. Li, X. Li / Energy & Buildings 176 (2018) 179–193

Fig. 7. Sub-metered system in a typical shopping center. Table 4 Relative deviation between cooling energy calculated by simplified method and cooling energy Simulated by DeST. Climate zone Building

SC A

B

C C

HSCW D

E

F

HSWW G H

Relative deviation(%)

7.8

7.6

−3.2

−3.5

−2.2

−2.9

−4.2

1800

Energy consumption (kWh)

1600 1400 1200 1000 800 600 400 200 01:00 02:00 03:00 04:00 05:00 06:00 07:00 08:00 09:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00 19:00 20:00 21:00 22:00 23:00 00:00

0

Fig. 8. Hourly energy consumption of total tenants in a shopping center during a typical day.

and the cooling energy simulated by DeST. Therefore, it is acceptable to neglect the influence of starting load in the cooling load calculation process for engineering usage, referring to the relative researches [34–36]. 3.3.2. Verification by measurement To verify the cooling energy calculation method, the actual cooling energy is calculated based on the following equation:

C Esupply =

 c p ρ g(tinlet − toutlet ) 3600

(20)

j

The flow rate and the water temperature sensors are commonly installed in the cooling system of the commercial buildings.

−5.7

The sensors record data every five-minute. In the meantime, the sub-metered system also records the energy consumption of each equipment every five-minute. The measured data of the installed sensors are calibrated using the standard measurement instrument. Then, the actual annual accumulated cooling energy is calculated by the calibrated flow rate and water temperature data using Eq. (20), used to calculate the energy efficiency. The calculated cooling energy is obtained by substituting the relative parameters from Eq. (1) into Eq. (9). Next, the calculated cooling energy and the measured cooling energy are compared, as shown in Fig. 9. From Fig. 10, the biggest difference between the calculated and the actual accumulated cooling energy in the cooling season exists in building A and relative error is 14.7%. The smallest difference exists in building B and relative error is −1.4%. Given the purpose of the paper and the relative deviation in other similar researches and standards [37–39], the relative error of 14.7% in this study is considered to be acceptable. Therefore, the simplified cooling energy calculation method is reliable and feasible. 4. Case studies This section describes the index system that benchmarks the energy performance for cooling in the shopping centers, as shown in Fig. 2. 4.1. Cooling energy evaluation Fig. 11 shows the cooling energy evaluation results. For shopping centers A, B, C and E, the total actual cooling energy is close to the recommended cooling energy. For the shopping centers D,

H. Li, X. Li / Energy & Buildings 176 (2018) 179–193

187

Fig. 9. Verification of cooling energy calculation method.

Fig. 10. Relative error of calculation results.

F and G, the total actual cooling energy is higher than the recommended value. In the shopping center H, the actual cooling energy is lower than the recommended value. However, based on the total energy consumption, the benchmarking results fail to determine a rationale for the actual cooling energy and its consumption. Therefore, a detailed analysis of the cooling energy components is required. Fig. 12 compares the actual and recommended envelope cooling energy for the selected shopping centers. For most of the shopping centers, the actual and recommended envelope cooling energies are similar to the extent that there is no significant difference. Human health and indoor thermal comfort require the sufficient amounts of fresh outdoor air. For shopping centers with high personnel density ratings, the required amount of fresh air per person is 15–19 m3 /h [40]. Therefore, the recommended ventilation cooling load is primarily dependent on shopper traffic in the shopping center. Fig. 13 compares the actual and recommended ventilation cooling energy for the selected shopping centers. In different buildings, the ventilation cooling load varies due to the

different ventilation rates. According to the investigations, the shopping centers use the constant air volume system to guarantee the air uniformity and fresh air volume. Therefore, the mechanical ventilation rate only relates to the number of running units, and then, the annual ventilation rates can be obtained by the equipment performance measurement and the annual operation records. Based on the investigation, the fresh air units for shopping centers B and H are rarely open during the cooling season. For both shopping centers, the fresh outdoor air is almost entirely supplied from natural ventilation and infiltration caused by the indoor and outdoor pressure difference, while the total fresh air volume is significantly beneath the national requirements. However, in the other buildings, a large amount of smoke is exhausted from the kitchens into the restaurants with closed make-up air fans. The entire building is under significant negative pressure which brings a large amount of outdoor air from openings and doors, leading to the total mechanical ventilation and the infiltration rates exceeding requirements. These two conditions can be directly learned from the equipment operation records.

188

H. Li, X. Li / Energy & Buildings 176 (2018) 179–193

Fig. 11. Comparison of actual and recommended cooling energy evaluation.

Fig. 12. Comparison of actual and recommended envelope cooling energy.

4.2. Energy efficiency evaluation The energy efficiency in the test shopping centers is evaluated using the present index system; the results are discussed below. Figs. 13–17 compares the energy efficiencies of the actual and recommended values. For most shopping centers, the actual energy efficiency ratio of the air-conditioning systems exceed the recommended value given by the national standard. However, the energy efficiency ratio of the refrigeration system in shopping centers E and G are below the recommended value. For shopping center E, the energy consumption of the condensate water pumps is too high, such that WTFcw is much lower than the recommended value. For shopping center G, the energy consumption of the chillers is high and the COP is slightly lower than the recommended value. For shopping centers A, B, C and D, WTFchw is lower than the recommended value and the energy consumption of the chilled water pumps is higher.

Erpelding B and Hartman T provided a detailed benchmark for the building cooling efficiency, as shown in Fig. 19 (corresponding to index EERs) and the chiller plant efficiency, as shown in Fig. 20 (corresponding to index EERr). The recommended value used above is in the “fair” level of the two benchmarks. This means that the energy efficiency in these commercial buildings can be further improved. Fig. 21 presents a comparison of actual and “good” values for EERr. Fig. 22 shows a comparison of actual and “good” values for EERs. Although the evaluation results are similar, as shown in Fig. 22 and Fig. 18, the majority of current EERs are below the “good” value. 4.3. Energy performance evaluation Fig. 23 illustrates the evaluation results of the energy performance for cooling purposes. For shopping centers D, E and G, the actual energy consumption is larger than the recommended value,

H. Li, X. Li / Energy & Buildings 176 (2018) 179–193

Fig. 13. Comparison of actual and recommended ventilation cooling energy.

Fig. 14. Coefficient of Performance of chiller evaluation.

Fig. 15. Water transport factor of condensate water evaluation.

Fig. 16. Energy efficiency ratio of refrigeration system evaluation.

189

190

H. Li, X. Li / Energy & Buildings 176 (2018) 179–193

Fig. 17. Water transport factor of chilled water evaluation.

Fig. 18. Energy efficiency ratio of air-conditioning system evaluation.

Fig. 19. Average annual building cooling efficiency [41].

Fig. 20. Average annual chiller plant efficiency [42].

which means the energy performance should be improved. For the remaining shopping centers, the actual energy consumption is lower than the recommended value. However, per the analyses in Sections 4.1 and 4.2, there are still some sub-items that require further improvement in these shopping centers, the results are shown in Table 5. Per the above analysis, during the benchmarking energy performance process, the total amount of energy in the target build-

ing reaches the standards whereas some of the sub-items do not reach the standards. In these cases, benchmarking based on the total quantity ignores the potential problems that exist in the building, which is not conducive to further improving the energy savings of the building. Therefore, the benchmarking method presented in this paper has instructive significance for practical applications.

H. Li, X. Li / Energy & Buildings 176 (2018) 179–193

191

Fig. 21. Energy efficiency ratio of refrigeration system evaluation 2.

Fig. 22. Energy efficiency ratio of air-conditioning system evaluation 2.

Fig. 23. Energy performance evaluation.

Table 5 Relative deviation between actual energy performance and recommended value. Building ID Cooling energy

Energy efficiency

Energy consumption

Envelope Ventilation Total COP WTFcw WTFchw EERr EERs

A

B

C

D

E

F

G

H

0% 38% 1% −5% 91% −18% −1% 20% −15%

4% −50% −1% 8% 11% −36% 4% 15% −13%

−1% 76% 1% 15% 92% −21% 20% 11% −9%

18% 567% 27% 7% 17% −17% 3% 12% 12%

8% 26% 3% −29% −43% 4% −38% −18% 24%

−4% 311% 12% −4% 84% 62% −1% 14% −2%

11% 146% 17% −16% 1% 4% −18% −1% 18%

−3% −67% −7% 7% 50% 31% 11% 18% −21%

+: higher than recommended value; −: lower than recommended value.

192

H. Li, X. Li / Energy & Buildings 176 (2018) 179–193

5. Discussion The outdoor temperatures influence the energy performance of chillers. Therefore, in the Chinese National Standard, there are different energy efficiency limited value in different climate zones. For example, the deviation between energy efficiency of centrifugal chillers with the capacity lower than 1163 kW in the severe cold region, and hot summer and warm winter region is 8%. According to our investigation, most large commercial buildings in China are equipped with several chillers; usually one screw chiller with small cooling capacity, and three centrifugal chillers with large cooling capacities. Different chiller operation strategies are conducted to meet different load condition and to provide a wide range of the cooling energy supply. When the cooling load of the building is at a low level, a screw chiller is only operated. When the cooling load increases, the centrifugal chillers are put into operation to gradually supply the cooling energy. Therefore, in most load conditions, the chillers are operated at a relatively high load rate, and the building load conditions have little impact on the energy efficiency of the specific equipment. Considering that different kinds of chillers have different energy efficiency standards, actual operation duration is counted to calculate the comprehensive energy efficiency of the system. Based on the previous calculation results, this section provides an analysis of the cooling energy characteristics and the main points for energy savings in the shopping centers. 1. In most shopping centers, the envelope cooling energy accounts for 20% or less of the total cooling energy. Therefore, the energy saving potential via envelope improvements is limited (Fig. 11 shows the statement). 2. Ventilation cooling energy significantly varies in different shopping centers. Therefore, the management protocols of mechanical ventilation systems should be improved and standardized for reasonable ventilation cooling energy and better indoor environments (Fig. 13 illustrates the statement). 3. The combined cooling load of shopper traffic, lighting and equipment comprises the largest proportion of total cooling energy, which is directly associated with business operations. Considering commercial operation and energy management requirements, further investigation is needed to determine an optimal design (Fig. 11 illustrates the statement). 4. The water transport factors in some buildings are lower than the recommended value because water pumps usually run at a fixed frequency. The actual water flow rate generally exceeds what is required to ensure a terminal load without a detailed adjustment. Lower efficiency causes higher air-conditioning system energy consumption and generates additional heat during the transmission and distribution cooling processes, which decreases the energy performance (Fig. 17 shows the statement). 6. Conclusion Currently, most benchmarking systems solely focus on whole building performance while lacking multi-level evaluations. For the management group, the reasonable energy consumption indicators should be set for each building per the different levels (i.e., sizes and types) of business. For the operations group, the actual performance of the building systems and equipment should be measured and analyzed to develop better operational strategies. Therefore, a multi-level benchmarking system is necessary for optimal energy performance management and a detailed evaluation method should be simplified to improve benchmarking efficiency. Based on detailed sub-metering system data and building operational data, this paper presents a simplified benchmarking energy consumption method for air-conditioning systems that cool

large-scale commercial buildings. A multi-level benchmarking index system for energy consumption is established using simplified calculation formulas based on fundamental principles. The calculation method is validated using data from eight large shopping centers in China. Given that business conditions vary across commercial buildings, the recommended index value is first customized and then calculated based on actual business conditions. The energy performance in eight shopping centers is analyzed and evaluated to determine further performance improvements. The results of this study realize a scientific and efficient energy benchmarking process for cooling in large-scale commercial buildings with a focus on energy-saving potential as well as informing improvement strategies for building energy performance. Acknowledgement This study is funded by the Innovative Research Groups of the National Natural Science Foundation of China (grant no. 51521005). References [1] IEA. Building energy performance metrics. 2015. https://www.iea.org/ publications/. [2] IEA. Building energy use in China. 2015. https://www.iea.org/publications. [3] C. Peng, D. Yan, S. Guo, et al., Building energy use in China: ceiling and scenario, Energy Build. 102 (2015) 307–316. [4] THUBERC, Annual Report on China Building Energy Efficiency, China Building Industry Press, Beijing, China, 2014. [5] T. Nikolaou, D. Kolokotsa, G. Stavrakakis, Review on methodologies for energy benchmarking, rating and classification of buildings, Adv. Build. Energy Res. 5 (1) (2011) 53–70. [6] L. Pérez-Lombard, J. Ortiz, R. González, et al., A review of benchmarking, rating and labelling concepts within the framework of building energy certification schemes, Energy Build. 41 (3) (2009) 272–278. [7] S. Wang, C. Yan, F. Xiao, Quantitative energy performance assessment methods for existing buildings, EnergyBuild. 55 (2012) 873–888. [8] S. Kinney, M.A. Piette, Development of a California Commercial Building Benchmarking Database, Lawrence Berkeley National Laboratory, 2002. [9] Z. Li, Y. Han, P. Xu, Methods for benchmarking building energy consumption against its past or intended performance: an overview, Appl. Energy 124 (2014) 325–334. [10] W. Chung, Review of building energy-use performance benchmarking methodologies, Appl. Energy 88 (5) (2011) 1470–1479. [11] P.A. Mathew, L.N. Dunn, M.D. Sohn, et al., Big-data for building energy performance: lessons from assembling a very large national database of building energy use, Appl. Energy 140 (2015) 85–93. [12] C. Chang, J. Zhao, N. Zhu, Energy saving effect prediction and post evaluation of air-conditioning system in public buildings, Energy Build. 43 (11) (2011) 3243–3249. [13] W.S. Lee, Benchmarking the energy performance for cooling purposes in buildings using a novel index-total performance of energy for cooling purposes, Energy 35 (1) (2010) 50–54. [14] W.S. Lee, C.K. Kung, Using climate classification to evaluate building energy performance, Energy 36 (3) (2011) 1797–1801. [15] W. Chung, Y.V. Hui, Y.M. Lam, Benchmarking the energy efficiency of commercial buildings, Appl. Energy 83 (1) (2006) 1–14. [16] W Chung, Using the fuzzy linear regression method to benchmark the energy efficiency of commercial buildings, Appl. Energy 95 (2012) 45–49. [17] P. Bohdanowicz, I. Martinac, Determinants and benchmarking of resource consumption in hotels - Case study of Hilton International and Scandic in Europe, Energy Build. 39 (1) (2007) 82–95. [18] W. Chung, Y.V. Hui, A study of energy efficiency of private office buildings in Hong Kong, Energy Build. 41 (6) (2009) 696–701. [19] J.C. Lam, K.K.W. Wan, D. Liu, et al., Multiple regression models for energy use in air-conditioned office buildings in different climates, Energy Convers. Manag. 51 (12) (2010) 2692–2697. [20] C.H. Philip, W.K. Chow, Energy use in commercial buildings in Hong Kong, Appl. Energy 69 (4) (2001) 243–255. [21] N. Fumo, P. Mago, R. Luck, Methodology to estimate building energy consumption using EnergyPlus Benchmark Models, Energy Build. 42 (12) (2010) 2331–2337. [22] E.H. Borgstein, R. Lamberts, Developing energy consumption benchmarks for buildings: bank branches in Brazil, Energy Build. 82 (2014) 82–91. [23] C. Yan, S. Wang, F. Xiao, A simplified energy performance assessment method for existing buildings based on energy bill disaggregation, Energy Build. 55 (2012) 563–574. [24] K.J. Chua, S.K. Chou, W.M. Yang, et al., Achieving better energy-efficient air conditioning-a review of technologies and strategies, Appl. Energy 104 (2013) 87–104. [25] Design standard of energy efficiency of public buildings. GB 50189-2015. (in Chinese)

H. Li, X. Li / Energy & Buildings 176 (2018) 179–193 [26] Economic operation of air conditioning systems. GB/T 17981-2007. (in Chinese) [27] Design standard of energy efficiency of public buildings. GB 50189-2005. (in Chinese) [28] Thermal design code for civil building. GB 50176-2016. (in Chinese) [29] J.C. Lam, K.K.W. Wan, C.L. Tsang, et al., Building energy efficiency in different climates, Energy Convers. Manag. 49 (8) (2008) 2354–2366. [30] T. Hong, A close look at the China design standard for energy efficiency of public buildings, Energy Build. 41 (4) (2009) 426–435. [31] H. Li, X. Li, A new method to assess the infiltration rate in large commercial complex in Beijing, China, Procedia Eng. 121 (2015) 628–634. [32] Ministry of Housing and Urban-Rural Development. Technical guidance of submetering design and installation for energy consumption monitoring system in government office buildings and large-scale public buildings. 2008 [33] D. Yan, J. Xia, W. Tang, et al., DeST—An Integrated Building Simulation Toolkit Part I: Fundamentals//Building Simulation, 1, Tsinghua University Press, co-published with Springer-Verlag GmbH, 2008, pp. 95–110. (2). [34] Y. Lu, Practical Heating and Air Conditioning Design Manual, China Architecture & Building Press, 1993.

193

[35] D. Lin, Z. Wang, Z.J. Zhai, X Li, A simplified method to predict hourly building cooling load for urban energy planning, Energy Build. 58 (3) (2013) 281–291. [36] S. Sholahudin, H Han, Simplified dynamic neural network model to predict heating load of a building using taguchi method, Energy 115 (2016) 1672–1678. [37] M. Shin, S.L. Do, Prediction of cooling energy use in buildings using an enthalpy-based cooling degree days method in a hot and humid climate, Energy Build. (2016) 110. [38] J.C. Wang, A study on the energy performance of hotel buildings in taiwan, Energy Build. 49 (2) (2012) 268–275. [39] The methods of performance test for water-chilling units and heat pump package using the vaper compression cycle. GB/T 10870-2014. (in Chinese) [40] Design code for heating ventilation and air conditioning of civil buildings. GB 50736-2012. (in Chinese) [41] B. Erpelding, Ultra-efficient all-variable speed buildings: introducing a benchmark for entire-building HVAC system efficiency, Heating/Piping/Air Cond. Eng.: HPAC 80 (11) (2008). [42] T. Hartman, All-variable speed centrifugal chiller plants, ASHRAE J. 43 (9) (2001) 43.