- Email: [email protected]
Demand controlled ventilation indoor climate and energy performance in a high performance building with air flow rate controlled chilled beams
Accepted Manuscript Title: Demand Controlled Ventilation Indoor Climate and Energy Performance in a High Performance Building with Air flow Rate Controlled Chilled Beams Author: Kaiser Ahmed Jarek Kurnitski Piia Sormunen PII: DOI: Reference:
S0378-7788(15)30292-9 http://dx.doi.org/doi:10.1016/j.enbuild.2015.09.052 ENB 6172
To appear in:
ENB
Received date: Revised date: Accepted date:
29-6-2015 25-8-2015 20-9-2015
Please cite this article as: K. Ahmed, J. Kurnitski, P. Sormunen, Demand Controlled Ventilation Indoor Climate and Energy Performance in a High Performance Building with Air flow Rate Controlled Chilled Beams, Energy and Buildings (2015), http://dx.doi.org/10.1016/j.enbuild.2015.09.052 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Demand Controlled Ventilation Indoor Climate and Energy Performance in a High Performance Building with Air flow Rate Controlled Chilled Beams
1
Aalto University, Department of Civil and Structural Engineering, Finland
Helsinki Metropolia University of Applied Sciences, Department of Civil Engineering and
3
us
Building Services, Finland
cr
2
ip t
Kaiser Ahmed 1,*, Jarek Kurnitski 1,3, Piia Sormunen 2
Tallinn University of Technology, Faculty of Civil Engineering, Estonia
an
*Corresponding author: Rakentajanaukio 4 A, FI-02150 Espoo, Finland;
te
Abstract
d
M
E-mail: [email protected]
Indoor climate and energy performance of Finnish low energy office building were studied to
Ac ce p
determine optimal control and operation solutions of demand controlled room conditioning and ventilation system with airflow rate controlled active chilled beams. Onsite measured temperature, CO2, occupancy rate were used to calibrate a dynamic simulations model. The results showed an average occupancy rate of 0.55 during office hours (OH) offering a good energy saving potential for Demand Control Ventilation (DCV) system. DCV system used 7 to 8% less total primary energy compared to Constant Air Volume (CAV) system depending on control and operation strategy used. DCV system saving was 33 to 41% if only heating, cooling, fans and pumps primary energy were considered. Supply air temperature controlled according to extract air temperature instead of outdoor air temperature minimized overheating problem which occurred with design solution during both seasons and systems. DCV system with active chilled beam complied at least 94% and
Page 1 of 38
90% of OH with Category II indoor climate with Water Priority Cooling and Air Priority Cooling control respectively. Ceiling cooling panel showed an effect of decoupling cooling from ventilation resulted only in 0.9% additional primary energy saving compared to DCV system with active
ip t
chilled beams.
cr
Keywords
Primary energy use; Thermal comfort; Active chilled beam; Ventilation system; Low energy office
us
building
an
Abbreviation AHU, Air Handling Unit
M
CAV, Constant Air Volume DCV, Demand Control Ventilation
te
IAQ, Indoor Air quality
d
HVAC, Heating, Ventilation and Air Conditioning
Ac ce p
IDA-ICE, IDA Indoor Climate and Energy OH, Office Hours
1. Introduction
European buildings account approximately 37% of the total final energy use whereas residential and commercial buildings are reported for 26% and 11% making buildings the largest energy using sector followed by industry and transport sectors [1]. Energy use in buildings corresponds to 33% of total emission of greenhouse gases [2] released in energy production. Energy use in office buildings is typically more intensive compare to households, a double specific energy use has been reported in Dutch study [3]. To control the energy use of new buildings, EPBD directive requires
Page 2 of 38
that all new buildings in the European Union must be nearly zero energy buildings (nZEB) from 31st December, 2020 and public owned buildings must be nearly zero energy building from 31st December, 2018 [4]. EPBD stresses that energy savings cannot compromise indoor climate,
ip t
therefore in future buildings good indoor climate is achieved with highly energy efficient manner. Energy use of EU office building has increased at a rate of 1.5% per annum due to economic
cr
growth, expansion of building sector, building services [1]. Among building services, heating,
us
ventilation and air conditioning (HVAC) system are the most significant. Lombard et al. [1] identified HVAC as main end use with a weight of nearly 50%. HVAC systems accounted for 48%,
an
55%, 52% of energy usages for United State of America, United Kingdom and Spain office spaces when appliances, lighting and DHW accounted for the rest [5]. Moreover, it is predicted that the
M
energy usages due to HVAC system will increase more than 50% in European Union during next 15
d
years [6]
te
Mechanical supply and exhaust ventilation systems with heat recovery revealed highly suitable for a cold climate [7] [8] supported by high occupant satisfaction rates as for instance 83% reported for
Ac ce p
mechanically ventilated Finnish office building [9]. One possible way to improve energy performance of ventilation system is applying demand controlled ventilation (DCV). Because DCV solutions are expensive, reliable data on achievable energy savings as well as evidence that indoor climate will not be deteriorated is needed for cost benefit analyses. An assessment of DCV energy savings is complicated because it strongly depends on occupancy, i.e. use of building needs to be well specified. This paper is focusing on DCV with air flow controlled active chilled beams which provide an additional complication, because ventilation and cooling is coupled in these room conditioning devices. If cooling need is more dominated than ventilation need, airflow rates cannot be dropped otherwise active chilled beams will lose cooling capacity. For previous literature it is difficult to find an occupancy dependent energy savings of DCV for office buildings however DCV has been studied in many papers.
Page 3 of 38
It is well known that DCV is highly suitable of those places where occupancy rate shows higher fluctuation such as office room, meeting room, copy room, break room, library, sport arena etc. In the same context, Chao and Hu [10] found that DCV-CO2 was highly suitable for densely occupied
ip t
places such as offices, lecture theaters and public buildings. DCV-CO2 system kept the indoor CO2 concentration less than desired level and HVAC unit consumed lower energy [11]. There is also a
cr
good evidence of DCV potential in classrooms. Mysen et al. [12] showed that only 74% of classroom used their design capacity whereas average day time office occupancy was varied from
us
15 to 80% [13] [14]. Mysen reported that DCV with occupancy detection (IR) and DCV with CO2
an
control saved 38% and 51% energy respectively compared to the CAV system [15]. Similar evidence was also found from Finnish sports training arenas [16] where 34% of energy savings
M
were reported with DCV-CO2 system.
From office buildings DCV evidence is less consistent. Energy saving potential was 20 to 30% at
d
low occupancy open plan office and was 3 to 5% at high occupancy open plan office [17]. In
te
addition, a comparison between DCV-CO2 and CAV system were conducted in Montreal office
Ac ce p
building and found energy saving of 12% for DCV-CO2 over CAV system while keeping same quality of thermal comfort and IAQ [18]. Another study [19] found significant reduction of running cost which was nearly 50% for DCV-CO2 system compare to fixed outdoor air flow system in high rise multi-zone office building.
The main objective of this paper was to determine energy saving potential of the cooling coupled DCV system with active chilled beams and its significance in energy balance of a low energy office building. The indoor climate specification was kept exactly the same for DCV and CAV system and necessary control strategies were applied to avoid the deterioration of the thermal comfort in the case of DCV. Dynamic indoor climate and energy simulations were conducted. Actual occupancy, indoor temperatures and CO2 concentrations were measured in heating and cooling seasons which were used to derive the occupancy profile and simulation model calibration. Actual system
Page 4 of 38
operation data were collected and used in simulated model. The simulation model was calibrated with measured data based on derived occupancy profile, lighting and equipment data. Parametric simulations were conducted for full year providing temperature, CO2 concentration, ventilation rate
ip t
and energy data of measured zone and full floor. Based on obtained results the actual control strategy was improved and a comparison of CAV and DCV control and operation options was
cr
conducted allowing to quantify improvements of indoor climate and energy performance.
us
2. Methods
an
2.1 Case study building
Case study building is Skanska Finland’s head office in Helsinki which was completed in
M
February, 2012. The building was designed to achieve LEED (Leadership in Energy and Environmental design) Core & Shell Platinum certification, as well as EU Green Building
d
certification. It has a total leasable area of 9,100 m2, and includes eight above ground floors
te
centered around a glazed atrium and three basement garage levels. The project team used pioneering 4D Building Information Modeling (BIM) to plan the construction of the project with a
Ac ce p
delivery timeline. The building was designed as a low energy building which uses 35% less energy than Finnish energy code requirements. The building envelope has a very high degree of air tightness with a measured air leakage rate of (n50) 0.46 air change per hour. This airtightness was achieved through the construction team’s attention to joint grouting and other junctions in the envelope. The building is a steel frame building whereas the basement floors are made of in-situ concrete. U-value of exterior wall, base floor, and roof are 0.17 W/m2K, 0.24 W/(m2K) and 0.09 W/(m2k) respectively. Furthermore, U-value for windows and glazed exterior wall are 0.8 W/(m2k) and 0.85 W/(m2K). The building indoor climate specifications follows the S2 level in Finnish indoor climate classification [20], which is in between II and I category of prEN 16798-1 [21]. The building has an efficient lighting system with
Page 5 of 38
occupancy sensors for daylight utilization. In figure 1 is shown Skanska head office and measured
ip t
open office space.
cr
Insert Figure 1 here
us
The building is equipped with air conditioning system with demand controlled mechanical supply and exhaust ventilation and active chilled beams including occupancy sensors, air flow rate
an
controlled active chilled beams and low-speed air handling units. The glazed atrium in the middle of the building is used for collecting extract air from office floors around the atrium so that there is
M
almost no extract air ductwork. The heat recovery of the main air handling units has temperature ratio of 85% whereas the SFP is 1.5 kW/(m3s). The schematic diagram of demand controlled
Ac ce p
Insert Figure 2 here
te
d
ventilation system in open offices is shown in Figure 2.
Temperature controller (TC), temperature sensor (TE), moisture sensor (ME), occupancy and photocell sensor (VE) were placed in the zone (Fig. 2). Chilled beams were operated only for cooling, as the radiators in rooms were used as heat emitters. Central air handling unit had another TC controller and TE sensors to control the supply air temperature. District heating and cooling system are connected to central AHU which further conditions the air. District heating was used for all heating in zone radiators, AHU and domestic hot water (DHW). Zone cooling and AHU cooling were provided by district cooling.
Page 6 of 38
2.2 Description of measured zone and floor The open office is located at the 6th floor and orients to North South direction. The total area of the floor is 1097 m2 which has 2 open office spaces, 1 project room, 4 meeting rooms, 2 stair rooms, 2
ip t
WCs, coffee room, 2 corridors, which is shown in Figure 3. Among of them open office spaces, project room, meeting rooms are operated by demand control ventilation (DCV) with air flow
cr
control by active chilled beam. Other spaces i.e. stair room, WC, coffee room, corridor are operated
an
us
by CAV system.
M
Insert Figure 3 here
d
The measured open office space had floor area of 382.6 m2 and was designed for 30 work stations.
te
The window to wall ratio is 36% in measured office space. DCV with air flow controlled active chilled beam provides cooling and ventilation facilities. Air flow rate from each active chilled beam
Ac ce p
is controlled by occupancy sensor and each active chilled beam provides 30 l/s if the workplace is occupied and 10 l/s if the workplace is not occupied. Each workstation has one desktop computer, one laptop or one extra monitor, double fluorescent ceiling light. 2.3 Onsite measurements
Onsite measurements of indoor temperature and CO2 concentration were taken by two groups in heating and cooling season. Measurements were taken from 11th of June to 20th of June, 2013 during cooling season whereas heating season data were collected from 11th of December to 19th of December, 2014 and 21st of January to 30th of January, 2014. Cooling season data were collected of 30 seconds interval and heating season data were collected of 300 seconds interval. The
Page 7 of 38
measurements of CO2 concentration, indoor temperature were taken by Model 7525 IAQ – CALC Indoor air quality meter. It can measure CO2 concentration up to 5000 ppm with higher accuracy of ± 3 %. In addition, it also measures temperature up to 60 °C with higher accuracy of ± 0.6%. The
ip t
average outdoor temperature during measured period was collected from building automation system which was 13 °C and -7 °C respectively cooling and heating period. Afterward, weekday
cr
office hour data i.e., 08.00 – 18.00 hours was used in analyses.
us
The number of occupants visiting the measured zone was detected by occupancy detection system installed at two entrance points of measured zone. It was capable of detecting the entering and
an
leaving occupant number. The temperature sensor in the system can detect the heat from the moving body and registers the direction of a person if he or she passes the given Red/Green line. In
M
addition, it is able to measure the occupant number at any duration of time. In this study one hour
te
2.4 Simulation model
d
interval period was used to calculate average occupant numbers in the open office.
The onsite measurement was taken for short duration and was not sufficient to draw a conclusion
Ac ce p
about long term IAQ and thermal comfort. IDA Indoor Climate and Energy 4.6.1 (IDA-ICE) simulation tool was used for the indoor climate and energy performance assessment. This simulation tool is capable of modelling multi-zone building with HVAC systems, and is suitable for dynamic simulation, thermal comfort, IAQ and energy assessment which also validated in different cases [22] [23] [24] [25]. Input data of zones is reported in Table 1. IDA-ICE considers the room air is well-mixed and it is not able to evaluate the local thermal discomfort. This study focused on general thermal comfort where indoor thermal condition were evaluated based on prEN 16798-1 standard [21].
Page 8 of 38
Insert Table 1 here
ip t
Helsinki Vantaa test reference year climate data [26] was used in dynamic simulation. In addition, standardized occupant behaviors and internal heat load were considered in simulation work. The
cr
simulation was run during 11th of December to 18th of December, 2014 for model calibration from which period the exact information of HVAC operation, occupancy, lighting and equipment number
us
were also available. In the simulation model CO2 controlled DCV system was used for open office space and ventilation rates were mentioned in Table 1. For model calibration, dynamic ventilation
an
was considered during office hours and maximum allowable CO2 concentration of 700 ppm was given. Active chilled beams operation was controlled by CO2 sensor instead of occupancy sensor
M
(tested in the model calibration phase) because of the limitations of IDA-ICE tools. The maximum
d
design airflow was 525 l/s and respective power of cooling and heating were 7875 W and 4725 W.
te
In addition, the cooling and heating power were considered 395 W and 240 W respectively for zero air flow rate. The supply air temperature was controlled by outdoor air temperature or extract air
Ac ce p
temperature depending on cases as shown in Figure 4.
Insert Figure 4 here
A constant cooling coil supply water temperature of 16 °C was used and the room temperature was controlled with PI controller regulating the cooling coil mass flow. The supply air temperature from air handling unit was controlled according to supply air temperature curve shown in Figure 4b. In IDA-ICE fan model it was used pressure head at nominal flow rate and efficiency of the fan corresponding to SFP 1.5 kW/(m3/s). In DCV case the fan model used polynomial coefficients to
Page 9 of 38
calculate the fan curve according to ASHRAE 90.1 Appendix G Method 2 [27] resulting reduced pressure level in the ductwork in the case of lower flow rates. The specific pump power (SPP) for chiller was 349 W/(l/s). The maximum mass flow was 1.0 kg/s and zone supply temperature set
ip t
point was 16°C.
cr
Parametric simulations
After model calibration a whole year parametric simulations were run. The design occupancy,
us
lighting, equipment were considered according to prEN 16798-1:2015 (Table 1) [21]. Different
an
control and operation strategies were introduced in calibrated model and simulations for DCV and CAV system were run. The comparisons were made based on the thermal comfort, IAQ and energy
M
performance for both systems. Afterwards, simulations were run for the whole floor instead of measured open office spaces and energy results are reported in terms of delivered and primary
d
energy for both systems. Heating and cooling set point specifications and control and operation
Ac ce p
te
strategies used in parametric simulations are shown in Table 2.
Insert Table 2 here
Simulated cases for the measured zone and full floor comprised air priority heating and cooling (Ref.EA24 Case), water priority heating and cooling (Ref.EA.WP Case), ceiling cooling panel (Ref.EA.GC Case). In the primary energy calculation the following Finnish primary energy factors were used: 0.7 for heating, 0.4 for cooling, and 1.7 for electricity [26].
3. Result and Discussion Measured results
Page 10 of 38
Duration curve of measured temperature and CO2 concentration during cooling and heating season are shown in Figure 5. Red lines and blue lines represent the onsite measured day’s temperature (Fig. 5a), CO2 concentration (Fig. 5b) respectively cooling and heating season. Each line indicates the frequency of temperature (Fig. 5a) and CO2 concentration (Fig. 5b) for each single measured
ip t
day.
us
cr
Insert Figure 5 here
an
During heating season the indoor temperature was observed in between 22 and 24 °C and fulfilled the Category II (22.0+2.0 °C) according to prEN 15251 standard [21]. High temperatures indicated
M
high internal load. On the other hand, indoor temperature fulfilled Category I (24.5+1.0 °C) during cooling season if some hours with temperature below the lower limit are not accounted. CO2
d
concentration during office hours fulfilled Category I (< 900 ppm). During this short measurement
te
period no problems were detected and the results were used in the model calibration.
Ac ce p
Occupant number is the indicator of CO2 concentration and relate to internal gain. The variation of occupant number during working hours indicates the suitability of DCV system. Measured office space was designed for 30 people and maximum measured occupant number was 11 which are shown in Figure 6. The blue line, dash line and dotted line are shown respectively maximum, minimum and average occupant number during measured heating season (Fig. 6a). The maximum occupant number was nearly double compared to the minimum occupant number. The difference between minimum and maximum shows energy saving of DCV system over CAV system. The blue filled area and the thick red line in Figure 6b are shown hourly occupancy rate and average occupancy rate during measured period. The average occupancy rate was 55% during office hours.
Page 11 of 38
Insert Figure 6 here
ip t
IDA-ICE simulation tools calculated the air flow rate based on CO2 concentration instead of occupancy sensor. Different control used required to validate the simulation based CO2
cr
concentration with onsite measured data. The measured hourly basis occupancy rate and equivalent number of equipment and lights, actual outdoor temperature, outdoor CO2 concentration (350 ppm)
us
for measurement periods were used in simulation for proper model calibration. Figure 7 shows measured and simulated value. Because of limited number of CO2 measurement points, CO2
an
concentration by occupants was obtained by non-steady state equation of CO2 where previous hour CO2 concentration in space is used to obtain new CO2 concentration for next hour in space (marked
d
Ac ce p
te
Insert Figure 7 here
M
as ‘Calculated’ in Figure 7a).
The model gave reasonably similar result of CO2 concentration to onsite measured and hand calculation result. In addition, simulated indoor temperature was also similar to the onsite measured one. The same calibrated model was run for a similar measured duration of cooling season. Actual occupant, equipment, lighting data were not available and explaining the difference compared to onsite measured results. The cooling set points could be another reason which further improved in ‘Ref. Case’ code (Table 2) and indoor temperature is shown in Figure 8.
Insert Figure 8 here
Page 12 of 38
Performance assessment
ip t
In order to show the thermal and energy performance for DCV and CAV system, the calibrated model was run for the measured office space zone while considering the design occupant number,
cr
i.e., 30 people. The simulation was run for duration of 11th of December to 18th of December, 2013 from which measured and simulated data had good agreement. The designed equipment numbers
us
were equivalent of occupant number and lighting, equipment load per unit area are mentioned in Table 1. The obtained occupant profile from Figure 6b was used. The supply air temperature curve,
an
heating, AHU and active chilled beam cooling set points were considered as in ‘Design Case’ (Table 2) for both system. The air flow rate for DCV and CAV system were 0.6 – 1.6 and 1.6
M
L/s.m2 respectively. In Figure 9 is shown the simulated result of temperature (°C), air flow rate
Ac ce p
Insert Figure 9 here
te
d
(L/s), CO2 (ppm) for DCV and CAV system in the measured zone.
The indoor temperatures for both systems showed higher values because of higher internal gain. The red line, DCV system, obtained higher room temperature compared to the CAV system. The air flow rate in CAV system was constant whereas DCV system gave variable air flow rate based on demand. The variable air flow rate had an effect on indoor temperature (°C) and CO2 concentration (ppm) (Fig. 9). The same calibrated model was used for one year simulation of the measured office zone. It was conducted for cases described in Table 2 in order to find optimal control and operation strategies for
Page 13 of 38
DCV and CAV system and to quantify energy savings of DCV system. Performance was assessed in term of thermal comfort (indoor temperature) footprint, IAQ and energy use for both systems. In
ip t
Figure 10 is shown the simulated result of ‘Design Case’ for DCV system.
us
cr
Insert Figure 10 here
The footprint of indoor temperature was derived according to prEN 15251standard [21] based on
an
lower and higher temperature limits of indoor climate categories (shown in Figure 11). The outdoor and indoor temperatures were simulated with 1 hour interval during one year. The running mean of
M
7 days was taken and if it was less than 10 °C than it was considered as heating season. The white,
d
green, yellow, and red colors indicate Category I, II, III and IV in figures respectively.
te
Calculation logic of the footprint is shown in Figure 11 where all simulated indoor temperature hourly values are plotted against the running mean outdoor temperature (a), and heating and cooling
Ac ce p
season values are shown (b). In the Design Case in Figure 10 and 11 temperatures were out of the range during heating and cooling season and DCV system was not working properly. Worst results were found during heating season where unacceptable office hours were noticed 15% during one year.
Insert Figure 11 here
The minimum temperature limit for Category I, Category II and Category III are presented by ‘+’ sign with proper identifying color (Fig. 11a). On the other hand, the maximum temperature limit for
Page 14 of 38
Category I, Category II and Category III are presented by dotted sign with respective color (Fig. 11a). The indoor temperature presents with gray dots and shows indoor condition during different outdoor temperature. In ‘Design Case’ supply temperature was controlled by outdoor temperature
ip t
which is also shown in Figure 11 b. ‘Design Case’ with CAV system could improve the thermal performance because of constant air
cr
flow rate i.e., 1.6 L/s.m2. Thus, calibrated model was simulated with CAV system while keeping
us
constant of other parameter. Thermal environment during heating and cooling season was better compared to DCV system. However, it is still shown over heating condition during heating season.
an
Higher set point of supplied air curve followed by outdoor temperature could be possible reason. In addition, energy use was higher as can be seen from Table 3. In Figure 12 is shown the simulated
Ac ce p
te
Insert Figure 12 here
d
M
results of ‘Design Case’ for CAV system.
To improve the situation cooling and heating set points were changed to ‘Ref. Case’ values which are shown in Table 2. The ceiling active chilled beam cooling set point was considered 25 °C instead of 26 °C (Design Case). The thermal environment during cooling season was improved relative to ‘Design Case’ for DCV system so that 96% of time was fallen to Category I and 4% to Category II. During heating season the performance was only slightly improved because of heating set point was changed from 22.5 °C to 22 °C, but major overheating still occurred. The same specifications of ‘Ref. Case’ were used for CAV system which resulted in drop of Category III excess hours from 30% to 23% and from 55% to 52% in Category II. To fix the overheating problem during heating season the supply air temperature which was controlled by outdoor
Page 15 of 38
temperature in previous cases was changed in ‘Ref.EA Case’. In this case supply air temperature curve controlled by extract air for DCV and CAV system and results are shown in Figure 13 for
ip t
DCV and CAV system.
us
cr
Insert Figure 13 here
The performance of thermal comfort during heating season was improved relative to ‘Ref Case’
an
for both systems. For DCV system 66% of time was fallen to Category I, II which was 50% in ‘Ref Case’ and Category IV occupied hours were reduced from 11 to 2%. Similar results also found for
M
CAV system so that Category I occupied hours was increased from 25 to 60% and significant reduction of Category III, i.e., 23 to 6% relative to ‘Ref Case’. The changes of supply air
te
DCV system.
d
temperature curve solved the problem for CAV system but still required additional improvement for
Ac ce p
All previous cases represented good thermal comfort during cooling season but indoor temperature was reached into higher values during heating season indicating a cooling need during heating season. In previous cases the air priority cooling was considered which means that AHU first increases the air flow for cooling and water cooling in active chilled beams will follow. AHU cooling set point was considered 24.0 °C for all previous cases which was same as maximum limit of Category II during heating season. Afterward, temperature reached cooling set point of active chilled beams, i.e., 25 °C for all previous cases which was same as maximum limit of Category III. These could be the reason of poor indoor thermal performance during heating season. To avoid overheating during heating season more strict set points of AHU and active chilled beam cooling were introduced in ‘Ref.EA24 Case’ which results are shown in Figure 14.
Page 16 of 38
ip t
Insert Figure 14 here
cr
As the initial cooling demand was fulfilled by air priority AHU unit, air flow rate showed cooling caused variation during heating season. During cooling season the indoor temperature crossed the
us
AHU cooling set point (23.5 °C) and reached active chilled beam cooling set point (24.0 °C). Thus, air flow rate during cooling season from AHU unit was nearly constant and additional cooling
an
demand was covered by active chilled beams. CO2 concentration was going low during cooling season because total air flow was used for both services of ventilation and cooling. However,
M
footprint result gave 11% of occupied office hours were in Category III which required further improvement. Furthermore, AHU cooling priority increased the fan power electricity use which has
te
d
higher conversion factor from delivered to primary energy as can be seen from Table 3. To minimize Category III hours the water priority cooling of active chilled beams were applied in
Ac ce p
‘Ref.EA.WP Case’ instead of air priority cooling. In Figure 15 air flow rate by AHU was found smaller than ‘Ref.EA24 Case’ because it was only used for ventilation purposes and the upper limit of CO2 concentration was 700 ppm kept of a year round. Water priority cooling of active chilled beam was started when indoor temperature reached to 23.5 °C. This solution saved fan energy (Table 3) and almost 96% of office hours satisfied Category I, II during heating season and 100% in cooling season.
Insert Figure 15 here
17 Page 17 of 38
To analyze the effect of individual ventilation and cooling, active chilled beams were replaced with ceiling cooling panel in ‘Ref.EA.GC Case’ and results are shown in Figure 16. The cooling capacity of a ceiling cooling panel did not depend on supply airflow rate that was a feature of active
ip t
chilled beam cases. Because of more strict cooling capacity of a ceiling cooling panel, it was necessary to change cooling season temperature set points as shown in Table 2. The selection of
cr
season’s duration was based on the outdoor temperature so that the heating season were considered
us
from 1st of January to 30th of April and 10th of September to 31st of December and cooling season was from 1st of May to 9th of September. The results of ‘Ref.EA.GC Case’ with these variables had
M
‘Ref.EA.WP Case’ (only 4% of occupied hours).
an
shown 10% of occupied hours were classified as category III which was higher compared to
te
d
Insert Figure 16 here
Ac ce p
Ref.EA.GC’ case introduced variable set point reducing cooling set point during heating season. In all other cases it was possible to avoid overheating problem during heating season with proper supply air temperature curve, i.e. by utilization of free cooling via outdoor air instead of active cooling via cooling coils of chilled beams. In such a way it was possible to provide Category II thermal comfort within minimum energy use. The results in figures 14 (d), 15 (d) and 16 (b) show a good agreement, i.e. equal conditions in different cases. Delivered and primary energy results for all simulated cases of measured office zone and full office floor are reported in Table 3. The breakdown of the delivered energy is given for measured office space and total delivered and primary energy values are given for both the measured office space and full office floor. 18 Page 18 of 38
ip t
Insert Table 3 here
For all cases CAV system used at least 12 to 14% more primary energy compared to DCV system
cr
for the measured open office zone. For the full 6th floor this difference was from 7 to 8%. Results
us
show that applied cases dropped primary energy from 95.9 of ‘Design Case’ to 86.4 kWh/m2/a, i.e., nearly 11% of energy savings. For both systems thermal comfort was not adequate in ‘Design Case’
an
and ‘Ref Case’ because of the overheating problem especially during heating season. The most significant improvement of energy use and thermal comfort were found due to extract air instead of
M
outdoor air controlled supply air temperature curve. ‘Ref.EA Case’ solved the overheating problem for CAV system but still required improvement for DCV system. Ventilation system coupled with
d
active chilled beam ‘Ref.EA.24 Case’ and ‘Ref.EA.WP Case’ saved primary energy use compared
te
to CAV system and provided Category II indoor climate for most of the time. Individual ventilation
Ac ce p
system and ceiling cooling panel (Ref.EA.GC Case) showed lowest usages of primary energy among all cases. It saved only 0.9% of primary energy and slightly poor thermal performance compared to ‘Ref.EA.WP Case’. 4. Conclusion
Thermal condition, IAQ and energy performance were investigated in high performance building which had demand controlled ventilation system coupled with active chilled beam. Onsite measured data and dynamic simulation were used to assess the building thermal comfort, IAQ and energy performance. Based on the result, the following conclusion can be drawn: • Occupancy rate showed high variation and average occupancy rate was 0.55 during office hours. Such occupancy profile provided a good energy saving potential for DCV system. 19 Page 19 of 38
• Indoor CO2 concentration was fallen to Category I (< 900 ppm) indoor climate for all control and operation strategies of CAV and DCV system.
ip t
• Overheating problem was noticed during both seasons. DCV system with ‘Design Case’ specification resulted only 32% and 39% of office hours falling into Category I, II during
cr
cooling and heating seasons respectively.
us
• Supply air temperature was controlled by extract air (Ref.EA Case) instead of outdoor air, showed significant improvement of indoor thermal performance. For DCV and CAV system
an
100% of office hours complied with Category I, II during cooling season. In heating season with same control strategy DCV and CAV system were fulfilled 66% and 94% of office hours
M
respectively as Category I, II indoor climate.
• Some cases (Design Case and Ref.EA Case) DCV failed to keep IAQ and thermal comfort
d
equal with CAV. But in the main cases it was achieved equal IAQ and thermal comfort with
te
DCV system by using proper set points for CO2 and temperatures.
Ac ce p
• Water priority cooling system (Ref.EA.WP Case) gave better thermal performance compared to air priority cooling system (Ref.EA.24 Case). Water priority cooling and Air priority cooling fulfilled at least 94% and 89% of office hours respectively as Category II indoor climate during both seasons. In addition, Air priority cooling system showed only 0.6% of primary energy savings compared to Water priority cooling system for the measured office zone.
• Cooling not coupled with ventilation system (Ref.EA.GC Case) provided only small additional energy saving potential of 1.0% (Ref.EA.WP Case) compared to DCV coupled with active chilled beam system (Ref.EA.WP Case) for the measured office zone. However,
20 Page 20 of 38
10% of occupied hours during heating season were classified as Category III indoor climate whereas only 4% was found for Ref.EA.WP Case.
ip t
• For the whole office floor CAV system consumed 15 to 19% more delivered energy compared to DCV system in the reference cases. Furthermore, DCV system saved 7 to 8% of
cr
primary energy calculated with Finnish primary energy factors. The total primary energy savings were 7.7%, 7.1%, 8.2% for Ref.EA.24, Ref.EA.WP, Ref.EA.GC DCV cases
us
respectively compared to ‘Ref.EA Case’ with CAV system. These savings were 35, 33 and
an
41% if only heating, cooling and HVAC primary energy was considered.
M
Acknowledgement
d
The research work was funded by RYM sisäilmasto. In addition, K.V. Lindholms Stiftelse
Ac ce p
for their nice cooperation.
te
foundation was also supported with grants. The authors also wish to thank the authority of Skanska
References
[1] L. Pérez-Lombard, J. Ortiz and C. Pout, "A review on buildings energy consumption information," Energy and Buildings, vol. 40, no. 3, p. 394–398, 2008. [2] "Energy efficiency requirements in building codes and energy efficiency policies for new buildings," International Energy Agency, 2008. [3] R. Clocquet, "Kwantitatief data onderzoek Energie-innovaties," Agentschap NL, The Netherlands, 2012. [4] M. Almeida, "nZEB training in the Southern EU countries maintaining building traditions," SouthZEB, 2014.
21 Page 21 of 38
[5] "Best practice programme energy consumption guide 19, energy use in offices," Carbon Trust, 2000.
ip t
[6] "Project for the directorate general transportation-energy of the commission of the European Union, final report," Energy Efficiency and Certification of Central Air Conditioners (EECCAC), 2003.
cr
[7] M. A. Alalawi and M. Krarti, "Experimental evaluation of CO2-based demand controlled ventilation strategies," ASHRAE Transactions, vol. 108, 2002.
us
[8] M. Krarti and M. Al-Alaw, "Analysis of the Impact of CO2 -Based Demand-Controlled Ventilation Strategies on Energy Consumption," ASHRAE Transactions, vol. 110, no. 1, p. 274, 2004.
an
[9] J. Kurnitski, "Measurement and assessment of ventilation systems’ performance and indoor climate," HVAC-Technology, Helsinki University of Technology, 2008.
M
[10] C. Chao and J. Hu, "Development of a dual-mode demand control ventilation strategy for indoor air quality control and energy saving," Building and Environment, vol. Volume 39, no. Issue 4, p. 385–397, 2004.
te
d
[11] V. Congradac and F. Kulic, "HVAC system optimization with CO2 concentration control using genetic algorithms," Energy and Buildings, vol. 41, no. 5, p. 571–577, 2009.
Ac ce p
[12] M. Mysen, J. Rydock and P. Tjelflaat, "Demand controlled ventilation for office cubicles—can it be profitable?," Energy and Buildings, vol. 35, no. 7, p. 657–662, 2003. [13] J. Halvarsson, "Occupancy pattern in office buildings consequences for HVAC system design and operation," Department of Energy and Process Engineering, Norwegian University of Science and Technology, Trondheim, 2012. [14] A. Bernard, J. Villenave and M. Lemaire, "Potential of savings for demand controlled ventilation (DCV) in office buildings," in 24th AIVC and BETEC Conference, Washington D.C., 2003. [15] M. Mysen, S. Berntsen, P. Nafstad and P. G. Schild, "Occupancy density and benefits of demand-controlled ventilation in Norwegian primary schools," Energy and Buildings, vol. 37, no. 12, p. 1234–1240, 2005. [16] T. Lu, X. Lü and M. Viljanen, "A novel and dynamic demand-controlled ventilation strategy for CO2 control and energy saving in buildings," Energy and Buildings, vol. 43, no. 9, p. 2499–2508, 2011.
22 Page 22 of 38
[17] J. Lavergea, N. V. D. Bossche, N. Heijmans and A. Janssens, "Energy saving potential and repercussions on indoor air quality of demand controlled residential ventilation strategies," Building and Environment, vol. 46, no. 7, p. 1497–1503, 2011.
ip t
[18] Y. Huo, F. Haghighat, J. Zhang and C. Shaw, "A systematic approach to describe the air terminal device in CFD simulation for room air distribution analysis," Building and Environment, vol. 35, no. 6, p. 563–576, 2000.
us
cr
[19] Z. Sun, S. Wang and Z. Ma, "In-situ implementation and validation of a CO2-based adaptive demand-controlled ventilation strategy in a multi-zone office building," Building and Environment, vol. 46, no. 1, p. 124–133, 2011. [20] "Classification of Indoor Climate 2000," Finnish society of indoor air quality and climate (FiSIAQ), 2001.
an
[21] "prEN:16798-1, Indoor environmental input parameters for the design and assessment of energy performance of buildings," European Committee for standardization, 2015.
M
[22] S. Moinard and G. Guyon, "Empirical validation of EDF ETNA and GENEC test-cell, report of task 22, building energy analysis tools," International energy Egency - solar heating and cooling programme, 2000.
Ac ce p
te
d
[23] J. Travesi, G. Maxwell, C. Klaassen and M. Holtz, "Empirical validation of Iowa energy resource station building energy analysis simulation models, report of task 22, subtask building energy analysis tools," Interbational energy agency - solar heating and cooling programme, 2001. [24] G. Zweifel and M. Achermann, "Radtest - the extension of program validation towards radiant heating and cooling," in Eighth International IBPSA Conference, Eindhoven, Netherlands, 2003. [25] IDA ICE. [Online]. [Accessed 28 02 2015]. [26] "D3 Energy management in buildings regulations and guidelines 2012," National building code of Finland, Finnish Ministry of the Environment, 2011. [27] "ASHRAE 90.1 Appendix G," ASHRAE Standard, 2004.
23 Page 23 of 38
us
cr
ip t
Graphs and Figures
(b)
an
(a)
Ac ce p
te
d
M
Figure 1: a) Skanska head quarter in Finland b) Measured open office in 6th floor
**TC-Temperature controller, TE-Temperature sensor, ME-Moisture sensor, VE-Occupancy and photocell sensor
24 Page 24 of 38
Figure 2: Schematic diagram of demand control ventilation system coupled with active chilled
M
an
us
cr
ip t
beam.
d
**MOO-Measured open office space, OO-Open office space, C-Corridor, M-Meeting room, R-Rest room, W-WC, SStair case
Ac ce p
te
Figure 3: Floor plan of the studied office floor.
(a)
(b)
25 Page 25 of 38
Figure 4: a) 6th floor model in IDA-ICE simulation tool b) Supply air temperature curve in simulations and was controlled according to outdoor or extracts air temperature depending on cases
te
(a)
d
M
an
us
cr
ip t
(Table 2).
(b)
Ac ce p
Figure 5: Duration curve of measured parameters during cooling and heating season a) Temperature (°C), b) CO2 concentration (ppm)
26 Page 26 of 38
(a)
(b)
te
(a)
d
M
an
us
cr
ip t
Figure 6: Measured occupancy during heating season a) Occupant number, b) Occupancy rate
(b)
Ac ce p
Figure 7: Model calibration during office hours at heating season a) CO2 concentration (ppm) b) Temperature (°C)
27 Page 27 of 38
(b)
Ac ce p
(a)
te
d
M
an
us
cr
ip t
Figure 8: Temperature calibration of model during office hours during cooling season
(c)
28 Page 28 of 38
Figure 9: Simulated result for DCV and CAV system in the measured open plan office zone a)
Ac ce p
te
d
M
an
us
cr
ip t
Temperature (°C), b) Air flow rate (l/s), and c) CO2 concentration (ppm)
(a)
(b)
Figure 10: Simulated results of ‘Design Case’ for DCV system a) Temperature (°C), b) Footprint
29 Page 29 of 38
ip t cr us
(a)
(b)
an
Figure 11: Simulated indoor temperature for ‘Design Case’ a) Category I, II and III limits and office
Ac ce p
te
d
M
hour indoor temperature b) Running mean, class I and III limits and office hour indoor temperature
(a)
(b)
Figure 12: Simulated results of ‘Design Case’ for CAV system a) Temperature (°C), b) Footprint
30 Page 30 of 38
ip t cr (b)
us
(a)
Ac ce p
te
d
M
an
Figure 13: Simulated footprint of ‘Ref.EA Case’ for a) DCV, b) CAV system
(a)
(b)
31 Page 31 of 38
ip t cr us
(c)
(d)
an
Figure 14: Simulated results of ‘Ref.EA24 Case’ for a) Air flow rate (l/s), b) CO2 concentration
Ac ce p
te
d
M
(ppm), c) Temperature (°C) and d) Footprint
(a)
(b) 32 Page 32 of 38
ip t cr (d)
us
(c)
an
Figure 15: Simulated value of ‘Ref.EA.WP Case’ for a) Air flow rate (l/s), b) CO2 concentration
Ac ce p
te
d
M
(ppm), c) Temperature (°C) and d) Footprint
33 Page 33 of 38
(a)
(b)
Ac ce p
te
d
M
an
us
cr
ip t
Figure 16: Simulated results of ‘Ref.EA.GC Case’ for a) Temperature (°C) and b) Footprint
34 Page 34 of 38
Tables Table 1: Description of main parameters in office zones Floor Area m2
AHU system
Supply Occupancy Lighting Equipment air m2/person W/m2 W/m2 L/s.m2
Open office (measured)
382.6
DCV, temp+ CO2 control
0.6-1.6
12.5
10
Open office
360.4
DCV, temp+ CO2 control
0.6-1.6
12.5
10
Project room
20.7
DCV, temp+ CO2 control
0.6-1.6
3.3
Meeting room - 1
19.8
DCV, temp+ CO2 control
0.6-4.0
Stair room - 1
16.6
CAV
1.5
WC - 1
31.9
CAV
Coffee room
34.2
CAV
Stair room - 2
12.5
CAV
Meeting room - 2
22.6
WC - 2
Ext. window area, m2 71.7
12
60.0
10
12
2.2
10
12
2.9
0
8
0
1.9
1.5
0
10
0
0.0
1.5
5
10
8
8.2
1.5
0
8
0
4.0
DCV, temp+ CO2 control
0.6-4.0
3.3
10
12
6.2
19.4
CAV
1.5
0
10
0
0.0
Corridor - 1
74.8
CAV
1.5
0
8
0
0.0
Corridor - 2
61.8
CAV
1.5
0
8
0
0.0
Meeting room - 3
19.8
DCV, temp+ CO2 control
0.6-4.0
3.3
10
12
2.9
Meeting room - 4
19.8
DCV, temp+ CO2 control
0.6-4.0
3.3
10
12
2.9
Ac ce p
te
d
3.3
M
us
cr
12
an
ip t
Zone name
35 Page 35 of 38
Table 2: Heating and cooling set point specifications used in parametric simulations Supply air
Heating
temperature set point
beam Cooling
set point
set point
Ref.
Calibrated reference case
Outdoor air
Ref.EA
Ref. + EA controlled SA
Extract air
Ref.EA24
Ref. + EA controlled SA
Extract air
Ref.EA.WP
Ref. + EA controlled SA +
Extract air
Ref. + EA controlled SA +
Extract air
d
Ref.EA.GC
M
water priority
22.5
26.0
22.0
24.0
25.0
22.0
24.0
25.0
22.0
23.5
24.0
22.0
25.0
23.5
*
*
*
22.0
**
22.0
25.0
**
26.0
23.5
**
24.5
te
water priority + ceiling cooling
24.0
us
Outdoor air
an
Design specification
Active chilled
Cooling
curve Design
AHU
ip t
Explanation
cr
Case Code
Ac ce p
panel
EA- Extract air, SA - Supply air, *Heating season set point, **Cooling season set point
36 Page 36 of 38
Table 3: Primary and delivered energy use (kWh/m2/a) in simulated cases.
DHW
Zone
AHU
Equipmen
7.4
6.4
7.7
4.1
3.8
18.8
21.7
6.1
0.4
76.4
98.1
69.4
86.4
Design
CAV
10.9
12.1
7.7
13.9
3.6
18.8
21.6
7.8
0.8
97.2
111.8
85.4
95.9
Ref.
DCV
5.6
6
7.7
7.9
3.7
18.7
21.6
5.7
77.4
97.2
70.3
85.9
Ref.
CAV
8.6
12.7
7.7
13
3.6
18.7
21.6
7.8
0.8
94.5
110.1
83
94.3
Ref.EA
DCV
7.2
1.7
7.7
3.6
4.2
an
0.5
18.7
21.6
4.8
0.4
69.9
92.1
64.7
81.8
Ref.EA
CAV
15.1
5.8
7.7
6.9
4.1
18.7
21.7
7.8
0.6
88.4
107.4
77.1
91.5
Ref.EA24
DCV
8.6
2
7.7
7.2
4.1
18.7
21.6
5.2
0.6
75.7
95.7
69.4
84.9
Ref.EA.WP
DCV
9.0
1.8
7.7
20.5
1.8
18.7
21.7
1.8
1.6
84.6
96.3
76.7
85.4
Ref.EA.GC
DCV
9.5
1.8
1.6
18.4
21.3
1.7
0.2
84.7
95.4
77.5
84.6
M
d
te 19.4
ip t
cr
Pump
us
Fan
t Light
Ac ce p
7.7
Delivered energy full floor Primary energy full floor
AHU
DCV
Electricity cooling
Delivered energy measured zone Primary energy measured zone
Zone
Design
Case Code
Ventilation system
District District heating
37 Page 37 of 38
Highlights • Higher fluctuation of occupancy rate and average was 0.55 during office hours.
• DCV system saved 7 to 8% of primary energy compared to CAV system.
ip t
• IAQ is fallen to Category I but overheating problem was found during both seasons.
cr
• Water priority cooling system, ‘Ref.EA.WP Case’, gave better thermal performance.
Ac ce p
te
d
M
an
us
• Ref.EA.WP used 0.6 - 1.0% higher primary energy compare to Ref.EA.24 and Ref.EA.GC.
38 Page 38 of 38
S0378-7788(15)30292-9 http://dx.doi.org/doi:10.1016/j.enbuild.2015.09.052 ENB 6172
To appear in:
ENB
Received date: Revised date: Accepted date:
29-6-2015 25-8-2015 20-9-2015
Please cite this article as: K. Ahmed, J. Kurnitski, P. Sormunen, Demand Controlled Ventilation Indoor Climate and Energy Performance in a High Performance Building with Air flow Rate Controlled Chilled Beams, Energy and Buildings (2015), http://dx.doi.org/10.1016/j.enbuild.2015.09.052 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Demand Controlled Ventilation Indoor Climate and Energy Performance in a High Performance Building with Air flow Rate Controlled Chilled Beams
1
Aalto University, Department of Civil and Structural Engineering, Finland
Helsinki Metropolia University of Applied Sciences, Department of Civil Engineering and
3
us
Building Services, Finland
cr
2
ip t
Kaiser Ahmed 1,*, Jarek Kurnitski 1,3, Piia Sormunen 2
Tallinn University of Technology, Faculty of Civil Engineering, Estonia
an
*Corresponding author: Rakentajanaukio 4 A, FI-02150 Espoo, Finland;
te
Abstract
d
M
E-mail: [email protected]
Indoor climate and energy performance of Finnish low energy office building were studied to
Ac ce p
determine optimal control and operation solutions of demand controlled room conditioning and ventilation system with airflow rate controlled active chilled beams. Onsite measured temperature, CO2, occupancy rate were used to calibrate a dynamic simulations model. The results showed an average occupancy rate of 0.55 during office hours (OH) offering a good energy saving potential for Demand Control Ventilation (DCV) system. DCV system used 7 to 8% less total primary energy compared to Constant Air Volume (CAV) system depending on control and operation strategy used. DCV system saving was 33 to 41% if only heating, cooling, fans and pumps primary energy were considered. Supply air temperature controlled according to extract air temperature instead of outdoor air temperature minimized overheating problem which occurred with design solution during both seasons and systems. DCV system with active chilled beam complied at least 94% and
Page 1 of 38
90% of OH with Category II indoor climate with Water Priority Cooling and Air Priority Cooling control respectively. Ceiling cooling panel showed an effect of decoupling cooling from ventilation resulted only in 0.9% additional primary energy saving compared to DCV system with active
ip t
chilled beams.
cr
Keywords
Primary energy use; Thermal comfort; Active chilled beam; Ventilation system; Low energy office
us
building
an
Abbreviation AHU, Air Handling Unit
M
CAV, Constant Air Volume DCV, Demand Control Ventilation
te
IAQ, Indoor Air quality
d
HVAC, Heating, Ventilation and Air Conditioning
Ac ce p
IDA-ICE, IDA Indoor Climate and Energy OH, Office Hours
1. Introduction
European buildings account approximately 37% of the total final energy use whereas residential and commercial buildings are reported for 26% and 11% making buildings the largest energy using sector followed by industry and transport sectors [1]. Energy use in buildings corresponds to 33% of total emission of greenhouse gases [2] released in energy production. Energy use in office buildings is typically more intensive compare to households, a double specific energy use has been reported in Dutch study [3]. To control the energy use of new buildings, EPBD directive requires
Page 2 of 38
that all new buildings in the European Union must be nearly zero energy buildings (nZEB) from 31st December, 2020 and public owned buildings must be nearly zero energy building from 31st December, 2018 [4]. EPBD stresses that energy savings cannot compromise indoor climate,
ip t
therefore in future buildings good indoor climate is achieved with highly energy efficient manner. Energy use of EU office building has increased at a rate of 1.5% per annum due to economic
cr
growth, expansion of building sector, building services [1]. Among building services, heating,
us
ventilation and air conditioning (HVAC) system are the most significant. Lombard et al. [1] identified HVAC as main end use with a weight of nearly 50%. HVAC systems accounted for 48%,
an
55%, 52% of energy usages for United State of America, United Kingdom and Spain office spaces when appliances, lighting and DHW accounted for the rest [5]. Moreover, it is predicted that the
M
energy usages due to HVAC system will increase more than 50% in European Union during next 15
d
years [6]
te
Mechanical supply and exhaust ventilation systems with heat recovery revealed highly suitable for a cold climate [7] [8] supported by high occupant satisfaction rates as for instance 83% reported for
Ac ce p
mechanically ventilated Finnish office building [9]. One possible way to improve energy performance of ventilation system is applying demand controlled ventilation (DCV). Because DCV solutions are expensive, reliable data on achievable energy savings as well as evidence that indoor climate will not be deteriorated is needed for cost benefit analyses. An assessment of DCV energy savings is complicated because it strongly depends on occupancy, i.e. use of building needs to be well specified. This paper is focusing on DCV with air flow controlled active chilled beams which provide an additional complication, because ventilation and cooling is coupled in these room conditioning devices. If cooling need is more dominated than ventilation need, airflow rates cannot be dropped otherwise active chilled beams will lose cooling capacity. For previous literature it is difficult to find an occupancy dependent energy savings of DCV for office buildings however DCV has been studied in many papers.
Page 3 of 38
It is well known that DCV is highly suitable of those places where occupancy rate shows higher fluctuation such as office room, meeting room, copy room, break room, library, sport arena etc. In the same context, Chao and Hu [10] found that DCV-CO2 was highly suitable for densely occupied
ip t
places such as offices, lecture theaters and public buildings. DCV-CO2 system kept the indoor CO2 concentration less than desired level and HVAC unit consumed lower energy [11]. There is also a
cr
good evidence of DCV potential in classrooms. Mysen et al. [12] showed that only 74% of classroom used their design capacity whereas average day time office occupancy was varied from
us
15 to 80% [13] [14]. Mysen reported that DCV with occupancy detection (IR) and DCV with CO2
an
control saved 38% and 51% energy respectively compared to the CAV system [15]. Similar evidence was also found from Finnish sports training arenas [16] where 34% of energy savings
M
were reported with DCV-CO2 system.
From office buildings DCV evidence is less consistent. Energy saving potential was 20 to 30% at
d
low occupancy open plan office and was 3 to 5% at high occupancy open plan office [17]. In
te
addition, a comparison between DCV-CO2 and CAV system were conducted in Montreal office
Ac ce p
building and found energy saving of 12% for DCV-CO2 over CAV system while keeping same quality of thermal comfort and IAQ [18]. Another study [19] found significant reduction of running cost which was nearly 50% for DCV-CO2 system compare to fixed outdoor air flow system in high rise multi-zone office building.
The main objective of this paper was to determine energy saving potential of the cooling coupled DCV system with active chilled beams and its significance in energy balance of a low energy office building. The indoor climate specification was kept exactly the same for DCV and CAV system and necessary control strategies were applied to avoid the deterioration of the thermal comfort in the case of DCV. Dynamic indoor climate and energy simulations were conducted. Actual occupancy, indoor temperatures and CO2 concentrations were measured in heating and cooling seasons which were used to derive the occupancy profile and simulation model calibration. Actual system
Page 4 of 38
operation data were collected and used in simulated model. The simulation model was calibrated with measured data based on derived occupancy profile, lighting and equipment data. Parametric simulations were conducted for full year providing temperature, CO2 concentration, ventilation rate
ip t
and energy data of measured zone and full floor. Based on obtained results the actual control strategy was improved and a comparison of CAV and DCV control and operation options was
cr
conducted allowing to quantify improvements of indoor climate and energy performance.
us
2. Methods
an
2.1 Case study building
Case study building is Skanska Finland’s head office in Helsinki which was completed in
M
February, 2012. The building was designed to achieve LEED (Leadership in Energy and Environmental design) Core & Shell Platinum certification, as well as EU Green Building
d
certification. It has a total leasable area of 9,100 m2, and includes eight above ground floors
te
centered around a glazed atrium and three basement garage levels. The project team used pioneering 4D Building Information Modeling (BIM) to plan the construction of the project with a
Ac ce p
delivery timeline. The building was designed as a low energy building which uses 35% less energy than Finnish energy code requirements. The building envelope has a very high degree of air tightness with a measured air leakage rate of (n50) 0.46 air change per hour. This airtightness was achieved through the construction team’s attention to joint grouting and other junctions in the envelope. The building is a steel frame building whereas the basement floors are made of in-situ concrete. U-value of exterior wall, base floor, and roof are 0.17 W/m2K, 0.24 W/(m2K) and 0.09 W/(m2k) respectively. Furthermore, U-value for windows and glazed exterior wall are 0.8 W/(m2k) and 0.85 W/(m2K). The building indoor climate specifications follows the S2 level in Finnish indoor climate classification [20], which is in between II and I category of prEN 16798-1 [21]. The building has an efficient lighting system with
Page 5 of 38
occupancy sensors for daylight utilization. In figure 1 is shown Skanska head office and measured
ip t
open office space.
cr
Insert Figure 1 here
us
The building is equipped with air conditioning system with demand controlled mechanical supply and exhaust ventilation and active chilled beams including occupancy sensors, air flow rate
an
controlled active chilled beams and low-speed air handling units. The glazed atrium in the middle of the building is used for collecting extract air from office floors around the atrium so that there is
M
almost no extract air ductwork. The heat recovery of the main air handling units has temperature ratio of 85% whereas the SFP is 1.5 kW/(m3s). The schematic diagram of demand controlled
Ac ce p
Insert Figure 2 here
te
d
ventilation system in open offices is shown in Figure 2.
Temperature controller (TC), temperature sensor (TE), moisture sensor (ME), occupancy and photocell sensor (VE) were placed in the zone (Fig. 2). Chilled beams were operated only for cooling, as the radiators in rooms were used as heat emitters. Central air handling unit had another TC controller and TE sensors to control the supply air temperature. District heating and cooling system are connected to central AHU which further conditions the air. District heating was used for all heating in zone radiators, AHU and domestic hot water (DHW). Zone cooling and AHU cooling were provided by district cooling.
Page 6 of 38
2.2 Description of measured zone and floor The open office is located at the 6th floor and orients to North South direction. The total area of the floor is 1097 m2 which has 2 open office spaces, 1 project room, 4 meeting rooms, 2 stair rooms, 2
ip t
WCs, coffee room, 2 corridors, which is shown in Figure 3. Among of them open office spaces, project room, meeting rooms are operated by demand control ventilation (DCV) with air flow
cr
control by active chilled beam. Other spaces i.e. stair room, WC, coffee room, corridor are operated
an
us
by CAV system.
M
Insert Figure 3 here
d
The measured open office space had floor area of 382.6 m2 and was designed for 30 work stations.
te
The window to wall ratio is 36% in measured office space. DCV with air flow controlled active chilled beam provides cooling and ventilation facilities. Air flow rate from each active chilled beam
Ac ce p
is controlled by occupancy sensor and each active chilled beam provides 30 l/s if the workplace is occupied and 10 l/s if the workplace is not occupied. Each workstation has one desktop computer, one laptop or one extra monitor, double fluorescent ceiling light. 2.3 Onsite measurements
Onsite measurements of indoor temperature and CO2 concentration were taken by two groups in heating and cooling season. Measurements were taken from 11th of June to 20th of June, 2013 during cooling season whereas heating season data were collected from 11th of December to 19th of December, 2014 and 21st of January to 30th of January, 2014. Cooling season data were collected of 30 seconds interval and heating season data were collected of 300 seconds interval. The
Page 7 of 38
measurements of CO2 concentration, indoor temperature were taken by Model 7525 IAQ – CALC Indoor air quality meter. It can measure CO2 concentration up to 5000 ppm with higher accuracy of ± 3 %. In addition, it also measures temperature up to 60 °C with higher accuracy of ± 0.6%. The
ip t
average outdoor temperature during measured period was collected from building automation system which was 13 °C and -7 °C respectively cooling and heating period. Afterward, weekday
cr
office hour data i.e., 08.00 – 18.00 hours was used in analyses.
us
The number of occupants visiting the measured zone was detected by occupancy detection system installed at two entrance points of measured zone. It was capable of detecting the entering and
an
leaving occupant number. The temperature sensor in the system can detect the heat from the moving body and registers the direction of a person if he or she passes the given Red/Green line. In
M
addition, it is able to measure the occupant number at any duration of time. In this study one hour
te
2.4 Simulation model
d
interval period was used to calculate average occupant numbers in the open office.
The onsite measurement was taken for short duration and was not sufficient to draw a conclusion
Ac ce p
about long term IAQ and thermal comfort. IDA Indoor Climate and Energy 4.6.1 (IDA-ICE) simulation tool was used for the indoor climate and energy performance assessment. This simulation tool is capable of modelling multi-zone building with HVAC systems, and is suitable for dynamic simulation, thermal comfort, IAQ and energy assessment which also validated in different cases [22] [23] [24] [25]. Input data of zones is reported in Table 1. IDA-ICE considers the room air is well-mixed and it is not able to evaluate the local thermal discomfort. This study focused on general thermal comfort where indoor thermal condition were evaluated based on prEN 16798-1 standard [21].
Page 8 of 38
Insert Table 1 here
ip t
Helsinki Vantaa test reference year climate data [26] was used in dynamic simulation. In addition, standardized occupant behaviors and internal heat load were considered in simulation work. The
cr
simulation was run during 11th of December to 18th of December, 2014 for model calibration from which period the exact information of HVAC operation, occupancy, lighting and equipment number
us
were also available. In the simulation model CO2 controlled DCV system was used for open office space and ventilation rates were mentioned in Table 1. For model calibration, dynamic ventilation
an
was considered during office hours and maximum allowable CO2 concentration of 700 ppm was given. Active chilled beams operation was controlled by CO2 sensor instead of occupancy sensor
M
(tested in the model calibration phase) because of the limitations of IDA-ICE tools. The maximum
d
design airflow was 525 l/s and respective power of cooling and heating were 7875 W and 4725 W.
te
In addition, the cooling and heating power were considered 395 W and 240 W respectively for zero air flow rate. The supply air temperature was controlled by outdoor air temperature or extract air
Ac ce p
temperature depending on cases as shown in Figure 4.
Insert Figure 4 here
A constant cooling coil supply water temperature of 16 °C was used and the room temperature was controlled with PI controller regulating the cooling coil mass flow. The supply air temperature from air handling unit was controlled according to supply air temperature curve shown in Figure 4b. In IDA-ICE fan model it was used pressure head at nominal flow rate and efficiency of the fan corresponding to SFP 1.5 kW/(m3/s). In DCV case the fan model used polynomial coefficients to
Page 9 of 38
calculate the fan curve according to ASHRAE 90.1 Appendix G Method 2 [27] resulting reduced pressure level in the ductwork in the case of lower flow rates. The specific pump power (SPP) for chiller was 349 W/(l/s). The maximum mass flow was 1.0 kg/s and zone supply temperature set
ip t
point was 16°C.
cr
Parametric simulations
After model calibration a whole year parametric simulations were run. The design occupancy,
us
lighting, equipment were considered according to prEN 16798-1:2015 (Table 1) [21]. Different
an
control and operation strategies were introduced in calibrated model and simulations for DCV and CAV system were run. The comparisons were made based on the thermal comfort, IAQ and energy
M
performance for both systems. Afterwards, simulations were run for the whole floor instead of measured open office spaces and energy results are reported in terms of delivered and primary
d
energy for both systems. Heating and cooling set point specifications and control and operation
Ac ce p
te
strategies used in parametric simulations are shown in Table 2.
Insert Table 2 here
Simulated cases for the measured zone and full floor comprised air priority heating and cooling (Ref.EA24 Case), water priority heating and cooling (Ref.EA.WP Case), ceiling cooling panel (Ref.EA.GC Case). In the primary energy calculation the following Finnish primary energy factors were used: 0.7 for heating, 0.4 for cooling, and 1.7 for electricity [26].
3. Result and Discussion Measured results
Page 10 of 38
Duration curve of measured temperature and CO2 concentration during cooling and heating season are shown in Figure 5. Red lines and blue lines represent the onsite measured day’s temperature (Fig. 5a), CO2 concentration (Fig. 5b) respectively cooling and heating season. Each line indicates the frequency of temperature (Fig. 5a) and CO2 concentration (Fig. 5b) for each single measured
ip t
day.
us
cr
Insert Figure 5 here
an
During heating season the indoor temperature was observed in between 22 and 24 °C and fulfilled the Category II (22.0+2.0 °C) according to prEN 15251 standard [21]. High temperatures indicated
M
high internal load. On the other hand, indoor temperature fulfilled Category I (24.5+1.0 °C) during cooling season if some hours with temperature below the lower limit are not accounted. CO2
d
concentration during office hours fulfilled Category I (< 900 ppm). During this short measurement
te
period no problems were detected and the results were used in the model calibration.
Ac ce p
Occupant number is the indicator of CO2 concentration and relate to internal gain. The variation of occupant number during working hours indicates the suitability of DCV system. Measured office space was designed for 30 people and maximum measured occupant number was 11 which are shown in Figure 6. The blue line, dash line and dotted line are shown respectively maximum, minimum and average occupant number during measured heating season (Fig. 6a). The maximum occupant number was nearly double compared to the minimum occupant number. The difference between minimum and maximum shows energy saving of DCV system over CAV system. The blue filled area and the thick red line in Figure 6b are shown hourly occupancy rate and average occupancy rate during measured period. The average occupancy rate was 55% during office hours.
Page 11 of 38
Insert Figure 6 here
ip t
IDA-ICE simulation tools calculated the air flow rate based on CO2 concentration instead of occupancy sensor. Different control used required to validate the simulation based CO2
cr
concentration with onsite measured data. The measured hourly basis occupancy rate and equivalent number of equipment and lights, actual outdoor temperature, outdoor CO2 concentration (350 ppm)
us
for measurement periods were used in simulation for proper model calibration. Figure 7 shows measured and simulated value. Because of limited number of CO2 measurement points, CO2
an
concentration by occupants was obtained by non-steady state equation of CO2 where previous hour CO2 concentration in space is used to obtain new CO2 concentration for next hour in space (marked
d
Ac ce p
te
Insert Figure 7 here
M
as ‘Calculated’ in Figure 7a).
The model gave reasonably similar result of CO2 concentration to onsite measured and hand calculation result. In addition, simulated indoor temperature was also similar to the onsite measured one. The same calibrated model was run for a similar measured duration of cooling season. Actual occupant, equipment, lighting data were not available and explaining the difference compared to onsite measured results. The cooling set points could be another reason which further improved in ‘Ref. Case’ code (Table 2) and indoor temperature is shown in Figure 8.
Insert Figure 8 here
Page 12 of 38
Performance assessment
ip t
In order to show the thermal and energy performance for DCV and CAV system, the calibrated model was run for the measured office space zone while considering the design occupant number,
cr
i.e., 30 people. The simulation was run for duration of 11th of December to 18th of December, 2013 from which measured and simulated data had good agreement. The designed equipment numbers
us
were equivalent of occupant number and lighting, equipment load per unit area are mentioned in Table 1. The obtained occupant profile from Figure 6b was used. The supply air temperature curve,
an
heating, AHU and active chilled beam cooling set points were considered as in ‘Design Case’ (Table 2) for both system. The air flow rate for DCV and CAV system were 0.6 – 1.6 and 1.6
M
L/s.m2 respectively. In Figure 9 is shown the simulated result of temperature (°C), air flow rate
Ac ce p
Insert Figure 9 here
te
d
(L/s), CO2 (ppm) for DCV and CAV system in the measured zone.
The indoor temperatures for both systems showed higher values because of higher internal gain. The red line, DCV system, obtained higher room temperature compared to the CAV system. The air flow rate in CAV system was constant whereas DCV system gave variable air flow rate based on demand. The variable air flow rate had an effect on indoor temperature (°C) and CO2 concentration (ppm) (Fig. 9). The same calibrated model was used for one year simulation of the measured office zone. It was conducted for cases described in Table 2 in order to find optimal control and operation strategies for
Page 13 of 38
DCV and CAV system and to quantify energy savings of DCV system. Performance was assessed in term of thermal comfort (indoor temperature) footprint, IAQ and energy use for both systems. In
ip t
Figure 10 is shown the simulated result of ‘Design Case’ for DCV system.
us
cr
Insert Figure 10 here
The footprint of indoor temperature was derived according to prEN 15251standard [21] based on
an
lower and higher temperature limits of indoor climate categories (shown in Figure 11). The outdoor and indoor temperatures were simulated with 1 hour interval during one year. The running mean of
M
7 days was taken and if it was less than 10 °C than it was considered as heating season. The white,
d
green, yellow, and red colors indicate Category I, II, III and IV in figures respectively.
te
Calculation logic of the footprint is shown in Figure 11 where all simulated indoor temperature hourly values are plotted against the running mean outdoor temperature (a), and heating and cooling
Ac ce p
season values are shown (b). In the Design Case in Figure 10 and 11 temperatures were out of the range during heating and cooling season and DCV system was not working properly. Worst results were found during heating season where unacceptable office hours were noticed 15% during one year.
Insert Figure 11 here
The minimum temperature limit for Category I, Category II and Category III are presented by ‘+’ sign with proper identifying color (Fig. 11a). On the other hand, the maximum temperature limit for
Page 14 of 38
Category I, Category II and Category III are presented by dotted sign with respective color (Fig. 11a). The indoor temperature presents with gray dots and shows indoor condition during different outdoor temperature. In ‘Design Case’ supply temperature was controlled by outdoor temperature
ip t
which is also shown in Figure 11 b. ‘Design Case’ with CAV system could improve the thermal performance because of constant air
cr
flow rate i.e., 1.6 L/s.m2. Thus, calibrated model was simulated with CAV system while keeping
us
constant of other parameter. Thermal environment during heating and cooling season was better compared to DCV system. However, it is still shown over heating condition during heating season.
an
Higher set point of supplied air curve followed by outdoor temperature could be possible reason. In addition, energy use was higher as can be seen from Table 3. In Figure 12 is shown the simulated
Ac ce p
te
Insert Figure 12 here
d
M
results of ‘Design Case’ for CAV system.
To improve the situation cooling and heating set points were changed to ‘Ref. Case’ values which are shown in Table 2. The ceiling active chilled beam cooling set point was considered 25 °C instead of 26 °C (Design Case). The thermal environment during cooling season was improved relative to ‘Design Case’ for DCV system so that 96% of time was fallen to Category I and 4% to Category II. During heating season the performance was only slightly improved because of heating set point was changed from 22.5 °C to 22 °C, but major overheating still occurred. The same specifications of ‘Ref. Case’ were used for CAV system which resulted in drop of Category III excess hours from 30% to 23% and from 55% to 52% in Category II. To fix the overheating problem during heating season the supply air temperature which was controlled by outdoor
Page 15 of 38
temperature in previous cases was changed in ‘Ref.EA Case’. In this case supply air temperature curve controlled by extract air for DCV and CAV system and results are shown in Figure 13 for
ip t
DCV and CAV system.
us
cr
Insert Figure 13 here
The performance of thermal comfort during heating season was improved relative to ‘Ref Case’
an
for both systems. For DCV system 66% of time was fallen to Category I, II which was 50% in ‘Ref Case’ and Category IV occupied hours were reduced from 11 to 2%. Similar results also found for
M
CAV system so that Category I occupied hours was increased from 25 to 60% and significant reduction of Category III, i.e., 23 to 6% relative to ‘Ref Case’. The changes of supply air
te
DCV system.
d
temperature curve solved the problem for CAV system but still required additional improvement for
Ac ce p
All previous cases represented good thermal comfort during cooling season but indoor temperature was reached into higher values during heating season indicating a cooling need during heating season. In previous cases the air priority cooling was considered which means that AHU first increases the air flow for cooling and water cooling in active chilled beams will follow. AHU cooling set point was considered 24.0 °C for all previous cases which was same as maximum limit of Category II during heating season. Afterward, temperature reached cooling set point of active chilled beams, i.e., 25 °C for all previous cases which was same as maximum limit of Category III. These could be the reason of poor indoor thermal performance during heating season. To avoid overheating during heating season more strict set points of AHU and active chilled beam cooling were introduced in ‘Ref.EA24 Case’ which results are shown in Figure 14.
Page 16 of 38
ip t
Insert Figure 14 here
cr
As the initial cooling demand was fulfilled by air priority AHU unit, air flow rate showed cooling caused variation during heating season. During cooling season the indoor temperature crossed the
us
AHU cooling set point (23.5 °C) and reached active chilled beam cooling set point (24.0 °C). Thus, air flow rate during cooling season from AHU unit was nearly constant and additional cooling
an
demand was covered by active chilled beams. CO2 concentration was going low during cooling season because total air flow was used for both services of ventilation and cooling. However,
M
footprint result gave 11% of occupied office hours were in Category III which required further improvement. Furthermore, AHU cooling priority increased the fan power electricity use which has
te
d
higher conversion factor from delivered to primary energy as can be seen from Table 3. To minimize Category III hours the water priority cooling of active chilled beams were applied in
Ac ce p
‘Ref.EA.WP Case’ instead of air priority cooling. In Figure 15 air flow rate by AHU was found smaller than ‘Ref.EA24 Case’ because it was only used for ventilation purposes and the upper limit of CO2 concentration was 700 ppm kept of a year round. Water priority cooling of active chilled beam was started when indoor temperature reached to 23.5 °C. This solution saved fan energy (Table 3) and almost 96% of office hours satisfied Category I, II during heating season and 100% in cooling season.
Insert Figure 15 here
17 Page 17 of 38
To analyze the effect of individual ventilation and cooling, active chilled beams were replaced with ceiling cooling panel in ‘Ref.EA.GC Case’ and results are shown in Figure 16. The cooling capacity of a ceiling cooling panel did not depend on supply airflow rate that was a feature of active
ip t
chilled beam cases. Because of more strict cooling capacity of a ceiling cooling panel, it was necessary to change cooling season temperature set points as shown in Table 2. The selection of
cr
season’s duration was based on the outdoor temperature so that the heating season were considered
us
from 1st of January to 30th of April and 10th of September to 31st of December and cooling season was from 1st of May to 9th of September. The results of ‘Ref.EA.GC Case’ with these variables had
M
‘Ref.EA.WP Case’ (only 4% of occupied hours).
an
shown 10% of occupied hours were classified as category III which was higher compared to
te
d
Insert Figure 16 here
Ac ce p
Ref.EA.GC’ case introduced variable set point reducing cooling set point during heating season. In all other cases it was possible to avoid overheating problem during heating season with proper supply air temperature curve, i.e. by utilization of free cooling via outdoor air instead of active cooling via cooling coils of chilled beams. In such a way it was possible to provide Category II thermal comfort within minimum energy use. The results in figures 14 (d), 15 (d) and 16 (b) show a good agreement, i.e. equal conditions in different cases. Delivered and primary energy results for all simulated cases of measured office zone and full office floor are reported in Table 3. The breakdown of the delivered energy is given for measured office space and total delivered and primary energy values are given for both the measured office space and full office floor. 18 Page 18 of 38
ip t
Insert Table 3 here
For all cases CAV system used at least 12 to 14% more primary energy compared to DCV system
cr
for the measured open office zone. For the full 6th floor this difference was from 7 to 8%. Results
us
show that applied cases dropped primary energy from 95.9 of ‘Design Case’ to 86.4 kWh/m2/a, i.e., nearly 11% of energy savings. For both systems thermal comfort was not adequate in ‘Design Case’
an
and ‘Ref Case’ because of the overheating problem especially during heating season. The most significant improvement of energy use and thermal comfort were found due to extract air instead of
M
outdoor air controlled supply air temperature curve. ‘Ref.EA Case’ solved the overheating problem for CAV system but still required improvement for DCV system. Ventilation system coupled with
d
active chilled beam ‘Ref.EA.24 Case’ and ‘Ref.EA.WP Case’ saved primary energy use compared
te
to CAV system and provided Category II indoor climate for most of the time. Individual ventilation
Ac ce p
system and ceiling cooling panel (Ref.EA.GC Case) showed lowest usages of primary energy among all cases. It saved only 0.9% of primary energy and slightly poor thermal performance compared to ‘Ref.EA.WP Case’. 4. Conclusion
Thermal condition, IAQ and energy performance were investigated in high performance building which had demand controlled ventilation system coupled with active chilled beam. Onsite measured data and dynamic simulation were used to assess the building thermal comfort, IAQ and energy performance. Based on the result, the following conclusion can be drawn: • Occupancy rate showed high variation and average occupancy rate was 0.55 during office hours. Such occupancy profile provided a good energy saving potential for DCV system. 19 Page 19 of 38
• Indoor CO2 concentration was fallen to Category I (< 900 ppm) indoor climate for all control and operation strategies of CAV and DCV system.
ip t
• Overheating problem was noticed during both seasons. DCV system with ‘Design Case’ specification resulted only 32% and 39% of office hours falling into Category I, II during
cr
cooling and heating seasons respectively.
us
• Supply air temperature was controlled by extract air (Ref.EA Case) instead of outdoor air, showed significant improvement of indoor thermal performance. For DCV and CAV system
an
100% of office hours complied with Category I, II during cooling season. In heating season with same control strategy DCV and CAV system were fulfilled 66% and 94% of office hours
M
respectively as Category I, II indoor climate.
• Some cases (Design Case and Ref.EA Case) DCV failed to keep IAQ and thermal comfort
d
equal with CAV. But in the main cases it was achieved equal IAQ and thermal comfort with
te
DCV system by using proper set points for CO2 and temperatures.
Ac ce p
• Water priority cooling system (Ref.EA.WP Case) gave better thermal performance compared to air priority cooling system (Ref.EA.24 Case). Water priority cooling and Air priority cooling fulfilled at least 94% and 89% of office hours respectively as Category II indoor climate during both seasons. In addition, Air priority cooling system showed only 0.6% of primary energy savings compared to Water priority cooling system for the measured office zone.
• Cooling not coupled with ventilation system (Ref.EA.GC Case) provided only small additional energy saving potential of 1.0% (Ref.EA.WP Case) compared to DCV coupled with active chilled beam system (Ref.EA.WP Case) for the measured office zone. However,
20 Page 20 of 38
10% of occupied hours during heating season were classified as Category III indoor climate whereas only 4% was found for Ref.EA.WP Case.
ip t
• For the whole office floor CAV system consumed 15 to 19% more delivered energy compared to DCV system in the reference cases. Furthermore, DCV system saved 7 to 8% of
cr
primary energy calculated with Finnish primary energy factors. The total primary energy savings were 7.7%, 7.1%, 8.2% for Ref.EA.24, Ref.EA.WP, Ref.EA.GC DCV cases
us
respectively compared to ‘Ref.EA Case’ with CAV system. These savings were 35, 33 and
an
41% if only heating, cooling and HVAC primary energy was considered.
M
Acknowledgement
d
The research work was funded by RYM sisäilmasto. In addition, K.V. Lindholms Stiftelse
Ac ce p
for their nice cooperation.
te
foundation was also supported with grants. The authors also wish to thank the authority of Skanska
References
[1] L. Pérez-Lombard, J. Ortiz and C. Pout, "A review on buildings energy consumption information," Energy and Buildings, vol. 40, no. 3, p. 394–398, 2008. [2] "Energy efficiency requirements in building codes and energy efficiency policies for new buildings," International Energy Agency, 2008. [3] R. Clocquet, "Kwantitatief data onderzoek Energie-innovaties," Agentschap NL, The Netherlands, 2012. [4] M. Almeida, "nZEB training in the Southern EU countries maintaining building traditions," SouthZEB, 2014.
21 Page 21 of 38
[5] "Best practice programme energy consumption guide 19, energy use in offices," Carbon Trust, 2000.
ip t
[6] "Project for the directorate general transportation-energy of the commission of the European Union, final report," Energy Efficiency and Certification of Central Air Conditioners (EECCAC), 2003.
cr
[7] M. A. Alalawi and M. Krarti, "Experimental evaluation of CO2-based demand controlled ventilation strategies," ASHRAE Transactions, vol. 108, 2002.
us
[8] M. Krarti and M. Al-Alaw, "Analysis of the Impact of CO2 -Based Demand-Controlled Ventilation Strategies on Energy Consumption," ASHRAE Transactions, vol. 110, no. 1, p. 274, 2004.
an
[9] J. Kurnitski, "Measurement and assessment of ventilation systems’ performance and indoor climate," HVAC-Technology, Helsinki University of Technology, 2008.
M
[10] C. Chao and J. Hu, "Development of a dual-mode demand control ventilation strategy for indoor air quality control and energy saving," Building and Environment, vol. Volume 39, no. Issue 4, p. 385–397, 2004.
te
d
[11] V. Congradac and F. Kulic, "HVAC system optimization with CO2 concentration control using genetic algorithms," Energy and Buildings, vol. 41, no. 5, p. 571–577, 2009.
Ac ce p
[12] M. Mysen, J. Rydock and P. Tjelflaat, "Demand controlled ventilation for office cubicles—can it be profitable?," Energy and Buildings, vol. 35, no. 7, p. 657–662, 2003. [13] J. Halvarsson, "Occupancy pattern in office buildings consequences for HVAC system design and operation," Department of Energy and Process Engineering, Norwegian University of Science and Technology, Trondheim, 2012. [14] A. Bernard, J. Villenave and M. Lemaire, "Potential of savings for demand controlled ventilation (DCV) in office buildings," in 24th AIVC and BETEC Conference, Washington D.C., 2003. [15] M. Mysen, S. Berntsen, P. Nafstad and P. G. Schild, "Occupancy density and benefits of demand-controlled ventilation in Norwegian primary schools," Energy and Buildings, vol. 37, no. 12, p. 1234–1240, 2005. [16] T. Lu, X. Lü and M. Viljanen, "A novel and dynamic demand-controlled ventilation strategy for CO2 control and energy saving in buildings," Energy and Buildings, vol. 43, no. 9, p. 2499–2508, 2011.
22 Page 22 of 38
[17] J. Lavergea, N. V. D. Bossche, N. Heijmans and A. Janssens, "Energy saving potential and repercussions on indoor air quality of demand controlled residential ventilation strategies," Building and Environment, vol. 46, no. 7, p. 1497–1503, 2011.
ip t
[18] Y. Huo, F. Haghighat, J. Zhang and C. Shaw, "A systematic approach to describe the air terminal device in CFD simulation for room air distribution analysis," Building and Environment, vol. 35, no. 6, p. 563–576, 2000.
us
cr
[19] Z. Sun, S. Wang and Z. Ma, "In-situ implementation and validation of a CO2-based adaptive demand-controlled ventilation strategy in a multi-zone office building," Building and Environment, vol. 46, no. 1, p. 124–133, 2011. [20] "Classification of Indoor Climate 2000," Finnish society of indoor air quality and climate (FiSIAQ), 2001.
an
[21] "prEN:16798-1, Indoor environmental input parameters for the design and assessment of energy performance of buildings," European Committee for standardization, 2015.
M
[22] S. Moinard and G. Guyon, "Empirical validation of EDF ETNA and GENEC test-cell, report of task 22, building energy analysis tools," International energy Egency - solar heating and cooling programme, 2000.
Ac ce p
te
d
[23] J. Travesi, G. Maxwell, C. Klaassen and M. Holtz, "Empirical validation of Iowa energy resource station building energy analysis simulation models, report of task 22, subtask building energy analysis tools," Interbational energy agency - solar heating and cooling programme, 2001. [24] G. Zweifel and M. Achermann, "Radtest - the extension of program validation towards radiant heating and cooling," in Eighth International IBPSA Conference, Eindhoven, Netherlands, 2003. [25] IDA ICE. [Online]. [Accessed 28 02 2015]. [26] "D3 Energy management in buildings regulations and guidelines 2012," National building code of Finland, Finnish Ministry of the Environment, 2011. [27] "ASHRAE 90.1 Appendix G," ASHRAE Standard, 2004.
23 Page 23 of 38
us
cr
ip t
Graphs and Figures
(b)
an
(a)
Ac ce p
te
d
M
Figure 1: a) Skanska head quarter in Finland b) Measured open office in 6th floor
**TC-Temperature controller, TE-Temperature sensor, ME-Moisture sensor, VE-Occupancy and photocell sensor
24 Page 24 of 38
Figure 2: Schematic diagram of demand control ventilation system coupled with active chilled
M
an
us
cr
ip t
beam.
d
**MOO-Measured open office space, OO-Open office space, C-Corridor, M-Meeting room, R-Rest room, W-WC, SStair case
Ac ce p
te
Figure 3: Floor plan of the studied office floor.
(a)
(b)
25 Page 25 of 38
Figure 4: a) 6th floor model in IDA-ICE simulation tool b) Supply air temperature curve in simulations and was controlled according to outdoor or extracts air temperature depending on cases
te
(a)
d
M
an
us
cr
ip t
(Table 2).
(b)
Ac ce p
Figure 5: Duration curve of measured parameters during cooling and heating season a) Temperature (°C), b) CO2 concentration (ppm)
26 Page 26 of 38
(a)
(b)
te
(a)
d
M
an
us
cr
ip t
Figure 6: Measured occupancy during heating season a) Occupant number, b) Occupancy rate
(b)
Ac ce p
Figure 7: Model calibration during office hours at heating season a) CO2 concentration (ppm) b) Temperature (°C)
27 Page 27 of 38
(b)
Ac ce p
(a)
te
d
M
an
us
cr
ip t
Figure 8: Temperature calibration of model during office hours during cooling season
(c)
28 Page 28 of 38
Figure 9: Simulated result for DCV and CAV system in the measured open plan office zone a)
Ac ce p
te
d
M
an
us
cr
ip t
Temperature (°C), b) Air flow rate (l/s), and c) CO2 concentration (ppm)
(a)
(b)
Figure 10: Simulated results of ‘Design Case’ for DCV system a) Temperature (°C), b) Footprint
29 Page 29 of 38
ip t cr us
(a)
(b)
an
Figure 11: Simulated indoor temperature for ‘Design Case’ a) Category I, II and III limits and office
Ac ce p
te
d
M
hour indoor temperature b) Running mean, class I and III limits and office hour indoor temperature
(a)
(b)
Figure 12: Simulated results of ‘Design Case’ for CAV system a) Temperature (°C), b) Footprint
30 Page 30 of 38
ip t cr (b)
us
(a)
Ac ce p
te
d
M
an
Figure 13: Simulated footprint of ‘Ref.EA Case’ for a) DCV, b) CAV system
(a)
(b)
31 Page 31 of 38
ip t cr us
(c)
(d)
an
Figure 14: Simulated results of ‘Ref.EA24 Case’ for a) Air flow rate (l/s), b) CO2 concentration
Ac ce p
te
d
M
(ppm), c) Temperature (°C) and d) Footprint
(a)
(b) 32 Page 32 of 38
ip t cr (d)
us
(c)
an
Figure 15: Simulated value of ‘Ref.EA.WP Case’ for a) Air flow rate (l/s), b) CO2 concentration
Ac ce p
te
d
M
(ppm), c) Temperature (°C) and d) Footprint
33 Page 33 of 38
(a)
(b)
Ac ce p
te
d
M
an
us
cr
ip t
Figure 16: Simulated results of ‘Ref.EA.GC Case’ for a) Temperature (°C) and b) Footprint
34 Page 34 of 38
Tables Table 1: Description of main parameters in office zones Floor Area m2
AHU system
Supply Occupancy Lighting Equipment air m2/person W/m2 W/m2 L/s.m2
Open office (measured)
382.6
DCV, temp+ CO2 control
0.6-1.6
12.5
10
Open office
360.4
DCV, temp+ CO2 control
0.6-1.6
12.5
10
Project room
20.7
DCV, temp+ CO2 control
0.6-1.6
3.3
Meeting room - 1
19.8
DCV, temp+ CO2 control
0.6-4.0
Stair room - 1
16.6
CAV
1.5
WC - 1
31.9
CAV
Coffee room
34.2
CAV
Stair room - 2
12.5
CAV
Meeting room - 2
22.6
WC - 2
Ext. window area, m2 71.7
12
60.0
10
12
2.2
10
12
2.9
0
8
0
1.9
1.5
0
10
0
0.0
1.5
5
10
8
8.2
1.5
0
8
0
4.0
DCV, temp+ CO2 control
0.6-4.0
3.3
10
12
6.2
19.4
CAV
1.5
0
10
0
0.0
Corridor - 1
74.8
CAV
1.5
0
8
0
0.0
Corridor - 2
61.8
CAV
1.5
0
8
0
0.0
Meeting room - 3
19.8
DCV, temp+ CO2 control
0.6-4.0
3.3
10
12
2.9
Meeting room - 4
19.8
DCV, temp+ CO2 control
0.6-4.0
3.3
10
12
2.9
Ac ce p
te
d
3.3
M
us
cr
12
an
ip t
Zone name
35 Page 35 of 38
Table 2: Heating and cooling set point specifications used in parametric simulations Supply air
Heating
temperature set point
beam Cooling
set point
set point
Ref.
Calibrated reference case
Outdoor air
Ref.EA
Ref. + EA controlled SA
Extract air
Ref.EA24
Ref. + EA controlled SA
Extract air
Ref.EA.WP
Ref. + EA controlled SA +
Extract air
Ref. + EA controlled SA +
Extract air
d
Ref.EA.GC
M
water priority
22.5
26.0
22.0
24.0
25.0
22.0
24.0
25.0
22.0
23.5
24.0
22.0
25.0
23.5
*
*
*
22.0
**
22.0
25.0
**
26.0
23.5
**
24.5
te
water priority + ceiling cooling
24.0
us
Outdoor air
an
Design specification
Active chilled
Cooling
curve Design
AHU
ip t
Explanation
cr
Case Code
Ac ce p
panel
EA- Extract air, SA - Supply air, *Heating season set point, **Cooling season set point
36 Page 36 of 38
Table 3: Primary and delivered energy use (kWh/m2/a) in simulated cases.
DHW
Zone
AHU
Equipmen
7.4
6.4
7.7
4.1
3.8
18.8
21.7
6.1
0.4
76.4
98.1
69.4
86.4
Design
CAV
10.9
12.1
7.7
13.9
3.6
18.8
21.6
7.8
0.8
97.2
111.8
85.4
95.9
Ref.
DCV
5.6
6
7.7
7.9
3.7
18.7
21.6
5.7
77.4
97.2
70.3
85.9
Ref.
CAV
8.6
12.7
7.7
13
3.6
18.7
21.6
7.8
0.8
94.5
110.1
83
94.3
Ref.EA
DCV
7.2
1.7
7.7
3.6
4.2
an
0.5
18.7
21.6
4.8
0.4
69.9
92.1
64.7
81.8
Ref.EA
CAV
15.1
5.8
7.7
6.9
4.1
18.7
21.7
7.8
0.6
88.4
107.4
77.1
91.5
Ref.EA24
DCV
8.6
2
7.7
7.2
4.1
18.7
21.6
5.2
0.6
75.7
95.7
69.4
84.9
Ref.EA.WP
DCV
9.0
1.8
7.7
20.5
1.8
18.7
21.7
1.8
1.6
84.6
96.3
76.7
85.4
Ref.EA.GC
DCV
9.5
1.8
1.6
18.4
21.3
1.7
0.2
84.7
95.4
77.5
84.6
M
d
te 19.4
ip t
cr
Pump
us
Fan
t Light
Ac ce p
7.7
Delivered energy full floor Primary energy full floor
AHU
DCV
Electricity cooling
Delivered energy measured zone Primary energy measured zone
Zone
Design
Case Code
Ventilation system
District District heating
37 Page 37 of 38
Highlights • Higher fluctuation of occupancy rate and average was 0.55 during office hours.
• DCV system saved 7 to 8% of primary energy compared to CAV system.
ip t
• IAQ is fallen to Category I but overheating problem was found during both seasons.
cr
• Water priority cooling system, ‘Ref.EA.WP Case’, gave better thermal performance.
Ac ce p
te
d
M
an
us
• Ref.EA.WP used 0.6 - 1.0% higher primary energy compare to Ref.EA.24 and Ref.EA.GC.
38 Page 38 of 38