Proceedings of BS2015:

14th Conference of International Building Performance Simulation Association, Hyderabad, India, Dec. 7-9, 2015.

CASE STUDIES ON HVAC SYSTEM PERFORMANCE USING A WHOLE

BUILDING SIMULATION BASED REAL-TIME ENERGY EVALUATION
APPROACH WITH BEMS

Hyeun Jun Moon1, Min Seok Choi1, Seung Ho Ryu1, Jung Chul Lee2, Dong Gu Kim2
1
Department of Architectural Engineering, College of Architecture, Dankook University,
Yongin-si, Gyeonggi-do, Republic of Korea
2
R&D Division, Hyundai Engineering & Construction Co. Ltd.,
Yongin-si, Gyeonggi-do, Republic of Korea

performance over the building lifecycle, including the

ABSTRACT design and operation stages. Xu et al. (2004) have
Building systems in the operation stage consume proposed a prototype a hardware-in-the-loop
more energy than building design intent due to simulation environment for controls testing and
various reasons (e.g., faulty construction, training using SPARK (Simulation Problem Analysis
malfunctioning equipment, incorrectly configured and Research Kernel). They have described the
control system and etc.). A whole building energy mechanism that coupled the real-time SPARK
simulation can be used to evaluate building energy simulation with a real building control system using a
performance over the building operation stages. hardware interface. Lee et al. (2007) have conducted
Recently, there is a great deal of research concerning an examination of the use of the ASHRAE simplified
real time performance evaluation using real-time energy analysis procedure in identifying significant
simulation. In this study, real-time simulation is used deviations from expected building energy
for commissioning of HVAC system through a real- consumption and fault detection at the whole building
time comparison between simulation results and level. Clarke et al. (2002) have developed a prototype
monitoring data. The real-time simulation system is control system to integrate simulation using ESP-r
applied to two existing buildings as case studies. A within real-time energy management control system
real-time building system performance is evaluated (EMCS) operation to enhance control capabilities.
by comparing the building energy monitoring data Haves et al. (2001) have explored the application of
with real-time simulation results. Results show that model-based performance assessment at the whole
real-time simulation could help identify the abnormal building level. They have addressed the information
operation of HVAC system. requirements for a simulation to predict the actual
performance of a particular building. They have also
INTRODUCTION discussed the possibilities for using control system
The building sector consumes about 40% of the communication protocols to link real-time simulation
energy used in the world. Knowing this, the theme of and the EMCS. Moreover, recent advancements in a
energy saving in buildings is receiving increasing whole building energy simulation allow for real-time
attention from technical and scientific points of view performance evaluation with monitored data in the
with aim of minimizing the overall energy operational stages (Dong et al., 2012, Moon et al.,
consumption and costs (Mirinejad et al. 2012). More 2012). Pang et al. (2012) proposed a framework for
than half of the energy consumption in a typical simulation based real-time whole building
building results from heating, ventilation, and air performance assessment. Recently, many buildings
conditioning (HVAC) systems (Scherer et al., 2014). have installed and operated building energy
Therefore, improvements in the energy efficiency of management systems (BEMS) which can real-time
building thermal management have the potential for a collect and monitor large amounts of data related to
large economic and environment impact (Chandan building energy consumptions in order to maximize
and Alleyne, 2014). building energy efficiency in the operation stage.
Earlier research shows that building systems consume BEMS would enable more utilization of real-time
more energy than design intent due to various reasons building system and energy data in operation stage.
in existing buildings such as the faculty construction, The objective of this study is to demonstrate a
malfunctioning equipment, incorrectly configured suggested model based real-time energy performance
control systems or inappropriate operating procedures system integrated with BEMS in existing buildings.
(O’Neill et al., 2009). One approach to addressing The paper also tries to identify potential of the
these problems is to compare the predictions of an suggested system as a continuous commissioning tool
energy simulation model of the building to the in a boiler and HVAC systems. The results from a
measured performance and analyse significant real-time simulation model reflecting the design
differences to infer the presence and location of faults. intent is compared to the actual energy data measured
A whole building energy simulation can be from BEMS. In this study, the real-time simulation
continuously used to evaluate building energy system is applied for two case buildings. The first

- 2735 -
Proceedings of BS2015:
14th Conference of International Building Performance Simulation Association, Hyderabad, India, Dec. 7-9, 2015.

building is a residential building equipped with a consumption deviations from design intent by
night electric boiler. The second building is an office comparing to the measured data. Also, when the
building equipped with AHU. system (e.g., HVAC, lighting etc.) or operation
condition are changed in the existing buildings, the
baseline model that represents these changes can be
used to compare with actual performance.
CASE STUDIES
In this study, real-time simulation reflecting the
design intent is applied to the two buildings. The first
building is a residential building equipped with a
night electric boiler. The second building is an office
building equipped with AHU. In this study, the
results from a real-time simulation model reflecting
the design intent is compared to the actual energy
Figure 1 A framework of commissioning method data measured from BEMS. This real-time energy
using real-time energy performance evaluation enables to identify the abnormal operation
of the HVAC systems in real-time.
COMMISSIONING USING REAL TIME
Case Study 1 : Residential building equipped with
ENERGY SIMULATION MODEL a night electric boiler
Figure 1 shows a framework of the continuous The selected residential building is a two stories
commissioning approach using real-time energy rectangular shape house. The first floor is mainly
performance assessment with a whole building used as a dining space and the second floor is used as
simulation. Real-time building simulation refers to a residential space for children. The second floor has
the use of a reference (virtual) model whose eight conditioned zones (1~7 rooms and a living
simulation time is synchronized in real time. This room). The heating system of this building is
real-time simulation model utilizes measured weather configured to the radiant floor heating system with a
data at each time step. The actual weather data are night storage electric boiler as shown in Figure 2.
used as input values in real-time building energy
simulation through BCVTB (Building Controls
Virtual Test Bed). BCVTB is an open-source
software environment developed for co-simulation
and data exchange among hardware/software. This
software couples different simulation tools for co-
simulations, and enables data exchange with building
automation systems and databases (Wetter, 2011). In
the previous study, Moon (2013) evaluated the effect
of weather data in an integrated real-time building
simulation.
A simulation model reflecting the design intent can Figure 2 Schematic diagram of a radiant floor
be used for commissioning by comparing to the heating system with a night storage electric boiler in
measured data. For newly constructed buildings, a the 1st case building
design intent baseline model can be applied to
identify and quantify the building energy

Figure 3 Real-time simulation system in the 1st case study

- 2736 -
Proceedings of BS2015:
14th Conference of International Building Performance Simulation Association, Hyderabad, India, Dec. 7-9, 2015.

This building is collecting a large amount of building variable flow rate and correctly connect the pipes
energy data through BEMS. Additionally, we with the flow control valves, the simulation results of
installed a BCVTB server connected to BEMS based supply and return water temperature matched with
on a BACnet/IP. The BCVTB was configured to the actual results. Figure 6 shows a comparison of
receive the data associated with a heating system and measured and simulated zone temperature. Actual
environment as shown in Figure 3. The zone air temperatures ranged from 23˚C~27˚C.
BACnetReader actor reads building energy data from However, Simulated result for the zone air
BACnet devices, and these data are sent to a database temperatures were maintained close to the set
program (i.e., PostgreSQL) through the SQL temperature (24˚C). Through this difference, we
Statement actor. could confirm the mismatching between control
valves and pipes. Zone air temperatures were
In addition, a weather station is installed on the site
maintained close to the set temperature (24˚C) after
and collects information including outdoor air dry
commissioning.
bulb and relative humidity, wind speed and direction,
direct normal solar radiation, solar radiation, which
are suitable for energy simulation with EnergyPlus.
The actual data about boiler operation statuses, boiler
set temperature and weather data were used for the
input values in real-time building energy simulation.
The weather data including solar irradiation, outside
air temperature and RH and wind speed and direction
were collected from the on-site weather station. The
real-time monitoring data on boiler water supply and
return temperature are compared with the simulation
results (Table 1). The real-time simulation is
performed during 10th ~12th in Febrary, 2013. This
climate condition is a heating season and The
distribution of the outdoor dry bulb are about - Figure 4 Comparison of measured and simulated
5˚C~10˚C. boiler supply water temperatures

Table 1 Real time input data and comparison data in

the residential building case
REAL TIME REAL TIME
INPUT DATA COMPARISON DATA
outdoor air dry bulb boiler supply water
dew point temperature temperature
relative humidity boiler return water
diffuse solar, direct solar, temperature
wind speed, and wind
direction

The real-time simulation was conducted to compare Figure 5 Comparison of measured and simulated
the simulation results and monitoring data of boiler
boiler return water temperatures
supply water temperatures and return water
temperatures in real time.
As a result, simulation results for the boiler supply
water temperature agreed to monitoring data for the
boiler supply water temperature (Figure 4). However,
simulation results for the boiler return water
temperature did not correspond with the actual return
water temperature (Figure 5). The reason for this was
that the actual flow control valves were not set to a
variable flow rate, but to a constant flow rate. These
have been set to a constant flow rate ever since the
control valves were installed.
Thus, we could confirm that the actual flow control
valves were not operated according to the design Figure 6 Comparison of measured and simulated
intent. It can be seen that the energy is wasted as an zone temperature
ineffective system setting. After we change to a

- 2737 -
Proceedings of BS2015:
14th Conference of International Building Performance Simulation Association, Hyderabad, India, Dec. 7-9, 2015.

Case Study 2 : Office building equipped with the The comparison module is constructed to compare
AHU system data associated with the AHU and an absorption
The selected office building is located in Yongin, chiller (such as AHU supply and return fan power,
Korea. This building is equipped with two air AHU supply air temperature, AHU supply water
handling unit (AHU) with an absorption chiller heater temperature, AHU return water temperature, AHU
(Figure 7). This building is collecting a large amount supply air flow rate, AHU water flow rate, CHW
of building energy data through BEMS. We installed pump power, zone temperature, and so on). Table 2
a BCVTB server connected to the BEMS based on shows input data and comparison data used in the
LabVIEW. We installed a BCVTB server connected case study.
to the BEMS (LabVIEW) through XML files.
Figure 8 shows the real-time simulation system using Table 2 Real time input data and comparison data in
BCVTB in the 2nd case study. This real-time the office building case
simulation system is composed of four modules; REAL TIME REAL TIME
BEMS, Simulation, Comparison, and Database. INPUT DATA COMPARISON DATA
BEMS module reads monitoring data from BEMS AHU supply / return fan
system (based LabVIEW) through XML files at each outdoor air dry bulb power
time step (10 min.). The monitoring data include the dew point temperature AHU supply air temperature
real-time input data and comparison data. Simulation relative humidity AHU supply / return water
module performs simulation reflecting the actual diffuse solar, direct solar, temperature
weather data at each time step. The results from the wind speed, and wind Zone temperature
real-time simulation are sent to the comparison direction CHW pump power
module and database module. Comparison module Water flow rate
compares monitoring data and simulation results.
Database module store monitoring data and
simulation results. Database module is based on a Figure 9 shows comparison results from measured
customized XML schema to store building CHW pump power and simulated CHW pump power.
information, building operation data, simulation data The highlighted areas enclosed by the dashed line
and analysis results. represent the differences between measured and
simulated results. As shown in the figure, The CHW
pump operated in the simulation, although it was not
operated in the actual operation. However, at that
same time, AHU 1 supply fan power and simulated
AHU 1 supply fan power operated as shown in
figure 10. It can be seen that the energy is wasted
about 30% due to ineffective operation of the AHU
system. If the actual CHW pump power is monitored
without simulation, it would be difficult to identify
the abnormal operation of AHU system. By
comparing with the simulation results, it could be
easily be found the abnormal operation of the AHU
system.
Figure 7 Schematic diagram of AHU systems with a
chiller heater in the 2nd case building

Figure 8 Real-time simulation system in the 2nd case study

- 2738 -
Proceedings of BS2015:
14th Conference of International Building Performance Simulation Association, Hyderabad, India, Dec. 7-9, 2015.

CONCLUSION
This study successfully demonstrated the integrated
whole building simulation based real-time energy
evaluation with BEMS in two case studies. Using the
comparison between simulation results and
monitoring data in real time, we could identify faulty
operation, control problems, sensor errors. The key
findings from two case studies are summarized as
follows:
Figure 9 Comparison of measured and simulated
CHW pump power 1. Two case studies showed that real-time
simulation system can help identify the
abnormal operation of HVAC systems and
sensor errors.
2. Through the suggested real-time simulation-
based commissioning approach, anomalies in
the heating system and AHU system were
corrected and potential energy savings were
identified.

In the future study, we will use the integrated real-

Figure 10 Comparison of measured and simulated time simulation approach for FDD in existing
AHU 1 fan power buildings and analyze causes of problems. It is
required the evaluation criteria from the difference
In addition, figure 11 and 12 show the comparison of between measured and simulated data that can
measured and simulated zone temperature. The determin the abnormal operation or errors of building
highlighted areas as shown in figure 11 and 12 systems.
represent the differences between measured and
simulated results. Through the differences, we could
identify and confirm sensor errors of the zone
ACKNOWLEDGEMENT
temperature sensors. This work was supported by the Energy Efficiency
& Resources Core Technology Program of the
Korea Institute of Energy Technology Evaluation
and Planning (KETEP) grant funded by the Korea
government Ministry of Knowledge Economy (No.
20142020103190)

This work was supported by the Human Resources

Program in Energy Technology of the Korea
Institute of Energy Technology Evaluation and
Planning (KETEP) grant funded by the Korea
Figure 11 Comparison of measured and simulated 2F government Ministry of Knowledge Economy (No.
zone temperature 20154030200830)

REFERENCES
Chandan, V. and Alleyne, A.G. 2014. Decentralized
predictive thermal control for buildings. Journal
of Process Control (24), 820-835.
Clarke, J.A., Cockroft, J., Conner. S., Hand, J.W.,
Kelly. N.J., Moore. R. 2002. Simulation-Assisted
control in building energy Management systems.
Energy and Buildings (33), 933-940.
Figure 12 Comparison of measured and simulated 3F Dong, B., O’Neill, Z., Li, Z., Luo, D., Madhusudana,
zone temperature S., Ahuja, S. and Bailey, T. 2012. An integrated
infrastructure for real-time building energy

- 2739 -
Proceedings of BS2015:
14th Conference of International Building Performance Simulation Association, Hyderabad, India, Dec. 7-9, 2015.

modeling and fault detection and diagnostics, with a building energy monitoring system,
SimBuild 2012, Madison, USA. Sensor Letters (12), 1-7.
Haves. P., Salsbury. T., Claridge. D., Liu. M. 2001. O’Neill, Z., Haves, P. 2009. Automated continuous
Use of whole building simulation in online commissioning of commercial buildings, final
performance assessment: modeling and report.
implementation issues. Proceedings of building Pang, X., Wetter, M., Bhattacharya P. and Haves, P.
simulation 2001, Rio de Janeiro, Brazil. 2012. A framework for simulation-based real-
Lee, S., Painter, F.L., Claridge, D.E. 2007. Whole- time whole building performance assessment,
Building commercial HVAC system simulation Building and Environment (54). 100-108.
for use in energy consumption fault detection. Scherer, H.F., Pasamontes, M., Guzman, J.L.,
ASHRAE Transaction (113) n.2. 52-61. Alvarez, J.D., Camponogara, E., Normey-Rico,
Mirinejad, H., Welch, K., Spricer, L. 2012. A review J.E. 2014. Efficient building energy management
of intelligent control techniques in HVAC using distributed model predictive control.
systems, Proceedings of the IEEE Energytech, Journal of Process Control (24). 740-749.
Cleveland, USA. Xu, P., Haves, P., Deringer, J.A. 2004. Simulation-
Moon, H.J. and Choi, M.S. 2013. The effect of Based testing and training environment for
weather data in an integrated real-time building building controls. Proceedings of SimBuild 2004,
simulation connected to a monitoring system, Boulder, USA.
Proceedings of ISOEN, Daegu, Korea. Wetter, M. 2011. Co-simulation of building energy
Moon, H.J. and Choi, M.S. 2014. Continuous and control systems with the Building Controls
commissioning through real-time simulation Virtual Test Bed, Journal of Building
Performance Simulation (4) n.3. 185-203.

- 2740 -