1-s2.0-S1876610217302035-main

Available online at www.sciencedirect.com
ScienceDirect
Energy Procedia 110 (2017) 486 – 491
1st International Conference on Energy and Power, ICEP2016, 14-16 December 2016, RMIT
University, Melbourne, Australia
Performance assessment of desiccant air conditioning system in an
institutional building in subtropical climate
Mohammad Golam Rasul* and Christopher Doring
Central Queensland University, School of Engineering and Technology, Rockhampton, Queensland 4702, Australia
Abstract
In hot and humid climate, the heating, ventilation and air conditioning (HVAC) system consumes more than 40% of the total
energy consumption of the building. Therefore, it is necessary to find out the ways of reducing overall energy consumption of
such power-hungry systems. The goal of this paper was to model and analyze the performance of desiccant air conditioning
system in an institutional building in subtropical climate similar to that in Central Queensland and investigate whether the
incorporation of a desiccant cooling system is able to reduce the energy consumption of the existing HVAC system. The analysis
involved modelling the reference building using OpenStudio to determine its cooling/heating requirements. This is then followed
by HVAC performance assessment of the existing VRF system, installed in the reference building, through developing a model
in EnergyPlus (EP). The model is validated with on-site measured data recorded in building management system (BMS). Then,
include desiccant air conditioning (DAC) system into the model to provide the dehumidification for the reference building to
reach “thermal comfort” in terms of both relative humidity and temperature.
©
Published
by Elsevier
Ltd. This
©2017
2017The
TheAuthors.
Authors.
Published
by Elsevier
Ltd. is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of the organizing committee of the 1st International Conference on Energy and Power.
Peer-review under responsibility of the organizing committee of the 1st International Conference on Energy and Power.
Keywords: Air conditioning; desiccant cooling systems; building systems; performance measurement; subtropical climate
1. Introduction
Since the beginning of the 20th century, the cooling and dehumidification of homes and commercial buildings
has transitioned from being a luxury to a necessity [1, 2]. Because of this, a significant amount of electricity being
* Corresponding author. Tel.: +61 7 4930 9676; fax: +61 7 4930 9382.
E-mail address: m.rasul@cqu.edu.au
1876-6102 © 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/4.0/).
Peer-review under responsibility of the organizing committee of the 1st International Conference on Energy and Power.
doi:10.1016/j.egypro.2017.03.173
Mohammad Golam Rasul and Christopher Doring / Energy Procedia 110 (2017) 486 – 491
consumed through the use of air conditioning systems especially during peak hour i.e. in middle of the day when the
maximum ambient temperature occurs also peaks. Consumption during these times increases the usage cost of the
electricity which is typically generated using non-renewable energy source. In fact, approximately 86% of electricity
supplied in Australia is generated from these fuel types, being 73% coal and 13% natural gas [3]. Hence, due to the
increase in energy demands, this directly contributes to more fossil fuels being burnt. As an example, HVAC and
refrigeration systems consume 46% of the South Australia’s total produced electricity. With such a large percentage
of power usage, HVAC and refrigeration systems can be linked to the contribution to Australia’s carbon emissions.
Considering that Australia’s energy is generated centrally with heavy reliance on traditional energy sources such
as “cheap and abundant” coal, this manner of supplying power encourages the emission of greenhouse gases which
is detrimental particularly towards the surrounding environment and the community. Attempts were made to
increase the efficiency of existing AC designs by using economizers to help alleviate power consumption hence less
fossil fuels. However, it operates by recycling air within the building with encourages virus, dust, germ, and mould
to be transmitted through the system’s ventilation [4]. With the integration of desiccants into the air conditioning
system this problem can be mitigated as desiccants have properties that enable the removal of moisture and latent
heat from the process air, consequently, improving indoor air quality. Additionally, this process can be done with
much less energy consumption due to the solar-assisted heating [5-7] Therefore, the carbon footprint and the power
usage by traditional air conditioning systems can be significantly reduced. The goal of the project is to analyse the
performance of desiccant air conditioning system in a subtropical climate similar to that of Central Queensland.
Therefore it is ideal for the simulated experiment to be conducted based from the location of the CQU main campus
at North Rockhampton, Queensland. The reference building in the experiment is Building 29, one of the engineering
buildings at CQU.
The analysis involves modelling the reference building using OpenStudio to determine its cooling/heating
requirements. This is followed by HVAC performance assessment of the existing VRF system installed in the
reference building through EnergyPlus (EP) [8]. Then, inclusion of a desiccant air-conditioning (DAC) system into
the model is done to provide the dehumidification for the reference building to reach “thermal comfort” in terms of
relative humidity. Additionally, the current energy consumption of the reference building is collected to serve as a
basis of comparison and validation against the simulated results in OpenStudio and EnergyPlus. Appropriate
conclusions and recommendations are then provided in the completion of this investigation.
With the completion of this investigation, valuable information, data, key measures and decision-making tools
can be provided to designers, developers, and engineers about the viability of the desiccant air conditioning system
to operate in a harsh subtropical geographical location. With further studies succeeding this investigation in
improving efficiency and performance competitiveness of the desiccant system can be used as a potential energysaving for the energy-hungry traditional vapour compression air conditioning units with the assumption that
financial expenditures can be brought down to a more competitive level.
2. Reference Buildings
Building 29 of Rockhampton campus of Central Queensland University, Australia, was used as a reference
building for simulation. The location details of the building are; latitude -23.4, longitude 150.48, elevation 15 m and
time zone 10. The building consists of 3 levels and each floor is completely air-conditioned. Access to the upper
floors of the building can used at the south end where a stairwell is located. There is also a walkway that connects
the upper floors to the adjacent building 28, another 3-story building. The ground floor consists of staff 3 staff
offices, a lecture theatre, toilets for male, female, and the disabled, and a shower room. It is important to note that
the toilets and the shower room are not being air conditioned. The first and second floors have a similar floor plan
consisting of two studio rooms per floor equipped with several monitors (CPU and accessories not included) for
student use and large interactive projectors per studio room. There are also 2 breakout rooms per floor adjacent to
one kitchenette per floor. Typical usage of the building is for student use e.g. lectures/tutorials/workshops. The
ground floor consists of Energy consumption of the building can come from university computers, screen projectors,
monitors, power outlets, lighting, kitchen appliances in the kitchenette, and exhaust fans for ventilation of non-air
conditioned rooms such as the shower room at the ground floor. The HVAC system is the main energy consumer of
the building. The HVAC system being used in the building is a Variable Refrigerant Flow (VRF) system.
487
488
Mohammad Golam Rasul and Christopher Doring / Energy Procedia 110 (2017) 486 – 491
3. Methodology
Initially, the cooling and heating loading requirement of the reference building was investigated using
OpenStudio. This was done to create the boundary condition of the building. This includes its geometry to calculate
the volume of the space that is required to be conditioned and the addition of shading present for the building which
is a significant contributor of the reduction of the solar heat that is absorbed by the building. Construction materials
of the building was also defined based on ASHRAE standards. Occupancy rates, lighting and equipment loading,
and air infiltration based on ASHRAE can also be added through this software. Secondly, the existing DX VRF
HVAC system was included into the OpenStudio model. It is important to note that the OpenStudio model is an
OSM file type. This is not recognisable by EnergyPlus therefore it must be exported as an IDF file type first to
overcome compatibility issues. Parameters measured includes the power consumption of the system. This is then
compared with the power consumption recorded by the BMS system for validation of results. Thirdly, the desiccant
system into the DX VRF HVAC system was incorporated. The power consumption of this system is observed and
compared to the consumption of other cases. Relative humidity levels are also monitored to verify if the EnergyPlus
model is able to provide thermal comfort conditions based in humidity levels.
3.1. OpenStudio (through SketchUp)
The cooling load of the reference building is calculated using SketchUp with OpenStudio plug in. All relevant
data were collected from Division of Facilities Management (DFM) of CQU. The geometry of the building was
prepared based on the floor plans provided by DFM. Specific details such as building materials have not been
acquired hence default settings according to ASHARE Standard 189.1 2009 - Standard for the Design of HighPerformance Green Buildings that is built in within the SketchUp software has been used. Additionally, the climate
zone was separated based on the ASHRAE Standard 169 2006. Factors that affect the total heat gain of the building
such as solar irradiance during the summer season, heat transfer from the outside environment to the inside
conditioned space e.g. infiltration rates, and other heat sources was also included into the model based from the
default settings. The calculated data is exported into EnergyPlus to simulate the response of the HVAC system. The
boundary conditions of the building was drawn within the SketchUp software whilst using the OpenStudio plug-in
tools. This creates the physical model including the controlled zones of the building and incorporate the cooling
loading settings such as the heat transfer surfaces generated in the boundaries of the analysis. Figs. 1 and 2 shows
the images of the shading that was added into the building model.
Fig. 1. Shading of Building 29 (south west view)
Fig. 2. Shading of Building 29 (north east view)
3.2. Existing VRF Systems
Energy Plus was used to simulate the HVAC behaviour of the system based on the cooling and heating load of
the conditioned space that was determined using OpenStudio. After the reference building has been modelled in
OpenStudio, the file was exported as an .idf file and imported in EnergyPlus. An .idf file is simply a file type that is
Mohammad Golam Rasul and Christopher Doring / Energy Procedia 110 (2017) 486 – 491
489
recognisable by EnergyPlus and modifications relevant to the project can be made accordingly. To simulate the
current VRF system, a VRF template that is in an .idf format is imported into EnergyPlus. Simultaneously, the .idf
file of the building was also imported. Firstly the location and climate setting were set up. Some of these
information have been acquired from the Rockhampton weather file that is available in the EnergyPlus website. The
internal gains have been determined based from the default settings in OpenStudio. This is in accordance with
ASHRAE Standard 55a 1995 Edition which defines an “office building”. The ASHRAE Standard 55a 1995 Edition
has also included a zone airflow design into the OpenStudio model as default so this has been added into the
simulation. The HVAC templates have been modified to suit the 10 thermal zones that were construction in the
OpenStudio. These zone have their respective evaporator and condenser linked. This section is the main setting that
defines the existing model of the reference building and its VRF system. Other HVAC settings include the design
objects, and the controls and thermostats. From here, the model is prepared and ready to be simulated. The results
are provided in the Results and Discussion section of the report.
3.3. Desiccant Systems
The desiccant system that was used is the desiccant dehumidifier package that provides a model that packages
components for a desiccant-based subsystem that dehumidifies an air stream, normally called the process air stream.
A second heated air stream, called the regeneration air stream, was used to remove the collected moisture from the
desiccant heat exchanger and this moisture-laden air is then usually exhausted from the building [8]. To model the
desiccant system, the functional VRF model was merged with the desiccant relative humidity control template
(DesiccantCVRhZoneRHCtrl) that is available in the ExampleFiles that were included in the installation of the
program. The structure of this model is based from the desiccant template. The envelope of the building is then
added into the model. The building has been split into 3 zones (the ground floor, first floor, and the second floor).
The VRF systems are then assigned to each of the zones. Dehumidification was controlled using default settings.
4. Results and Discussion
The data was collected from DFM of CQU. The data that was considered is between mid-June 2015 and midJune 2016 (Fig. 3). This range has been selected as this presents the most recent available data from Building 29.
The Fig. 3 represents the daily measurements at 12pm of the energy consumed. The difference between the first data
collected at 12th June 2016 (403,831 kWh) against the data in 11 June 2016 (550050 kWh) is 146219 kWh.
Therefore within a span of about a year the energy consumed by the building is 146 MWh. The active energy uses
for cooling and heating is presented in this section. Cooling load is the demand needed to keep the building at the set
point required. Therefore, this analysis is important to check if the VRF model is working correctly. The net site
energy used was 149072.36 kWh and EUI (based on net site energy and total building area) was 179.81 kWh/m 2.
The snippet taken from the EnergyPlus formatted result from the Open studio model shows the similar result with
regards to the net site energy. The net site energy of the model is 149075 kWh and the EUI is 181.58 kWh/m 2. This
means that the model results are comparable with on-site (BMS) data. This validated model was used for inclusion
of desiccant system. The annual active energy consumption of the case study building (Building 29 at CQU
Rockhampton campus) is shown in Fig. 3. The heating and the cooling load is 66% of the total load. Therefore the
HVAC total site energy and EUI is 98389 kWh and the EUI is 119.84 kWh/m 2.
4.1. Existing HVAC Model – DX VRFF Systems
The simulation results from EnergyPlus regarding the inclusion of the DX VRF system in the OpenStudio Model
is presented in this section. The total site energy per unit area of the conditioned space is 288.8 GJ and the total
source energy per unit area of the conditioned space is 914.68 GJ. Therefore, the total site energy used is 80227.78
kWh. The EUI is then 97.72 kWh/m2. The temperature of 23°C in the thermal zones have been meet during
operational hours as shown in the graph below for Studio 1 located in the first floor.
490
Mohammad Golam Rasul and Christopher Doring / Energy Procedia 110 (2017) 486 – 491
Notably, the temperature goes up in the summer
and down in the winter in non-operational such as
night time and the weekends. Operating hours are
between 7am and 5pm. However, there is a system
warm up stage that starts at 6am. A sample summer
week temperature graph is shown in Fig. 4. Fig. 5
shows the relative humidity of Studio 1 in the First
Floor. Since the DX VRF system is not designed to
control the relative humidity of the zones therefore
the measured relative humidity is heavily influenced
by the outside relative humidity. This is a noticeable
change in the relative humidity during operational
hours however this is only caused by coincidental
dehumidification when the passing moist air contacts
the cold cooling coils creating moisture in its surface.
Fig. 4. Temperature line of Studio 1 in January (sample summer
time – VRF system)
Fig. 3. Active Energy Use of Building 29
Fig. 5. Relative Humidity Line of Studio 1 in January
(sample summer time – VRF system)
4.2. Desiccant HVAC Model
The total site energy per unit area of the conditioned space is 615.46 GJ and the total source energy per unit area
of the conditioned space is 1635.09GJ. Therefore, the total site energy used is 170961 kWh. The EUI is then 206.48
kWh/m2. Similar to the VRF system, the temperature of 23°C in the thermal zones have been meet during
operational hours as shown in the graph below for Studio 1 located in the first floor. The relative humidity in Zone 1
during the summer has remained in the optimum point of 50% R.H. These results indicate that thermal comfort has
been achieved in terms of temperature and relative humidity. Summary of the energy consumption of building 29 is
shown in Fig. 6. It has been determined that the installation of the desiccant HVAC system based on the simulation
dramatically increases the energy consumption of the building. The DX VRF system, being one of the most energy
efficient systems on the market is still the preferred option. There is also a relatively significant difference between
the measured building energy consumption data compared to the OpenStudio model. This is due to the configuration
of the logging system of Building 29 which only record the overall energy consumption of the building rather than
just the existing VRF system. With regards to the difference between the OpenStudio model and the DX VRF
system that was simulated in EnergyPlus, there are lot of factors that played a role. Firstly, there were some settings
that were lost between the conversion of the building model from an OSM file (recognised by OpenStudio) into an
IDF file (recognised by EnergyPlus). Secondly, There were some default settings from the VRF template that were
kept as is due to the several errors that were encountered during the multiple attempts to have a functional
EnergyPlus model.
Mohammad Golam Rasul and Christopher Doring / Energy Procedia 110 (2017) 486 – 491
Finally, the efficiency of the VRF system
compared to traditional system have slightly
reduced the energy required to cool the
building. The desiccant system has the highest
energy consumption amongst all the collected
and calculated energy consumption data. The
desiccant and the VRF system EnergyPlus
templates were heavily used which resulted in
the inaccuracies of the model. A large portion of
the HVAC components were auto-sized and
several of the default settings were used.
Additionally, a simple 3-zone model in
OpenStudio was created to minimise the
chances of errors and make the troubleshooting
process easier. In contrast, the DX VRF model
that was done earlier had a 10-zone model.
Fig. 6. Summary of the Energy Performance of Building 29
5. Conclusions
Upon the completion of this investigation, it has been determined that the desiccant VRF system does not
improve the energy consumption of building 29. In fact there was a significant increase in the overall energy
consumption. This was proven by the total site energy usage and the energy per conditioned building area values.
However, it was able to achieve the thermal comfort requirements of the building in terms of relative humidity
levels of approximately 50%. It is also important to take into consideration that the values that were determined still
require further fine turning to significantly improve the accuracy of the findings. Incorporate solar heating for the
regeneration of the moist desiccant since solar power is a free source heating energy [9, 10]. Further study is
required into accurately model a liquid desiccant model. From the literature review, this is a more effective system
with regards to dehumidification and lower required regeneration temperatures.
References
[1] Origin Energy Limited 2015, Energy in Australia, viewed 15 September 2015, https://www.originenergy.com.au/blog/about-energy/energyin-australia.html.
[2] Baniyounes, A., Ghadi, Y.Y., Rasul, M.G. & Khan, M.M.K. (2013), An Overview of Solar Assisted Air Conditioning in Queensland’s
Subtropical Regions, Australia, Renewable and Sustainable Energy Reviews, 26, 781-804, Elsevier.
[3] Lowenstein, A, Slayzak, & S, Kozubal, E 2006, A Zero Carryover Liquid-Desiccant Air Conditioner for Solar Applications, To be presented
at ASME International Solar Energy Conference (ISEC2006), viewed 10 September 2015, http://www.nrel.gov/buildings/pdfs/39798.pdf
[4] Abdel-Salam, A.H., G. Ge, and C.J. Simonson, Thermo-economic performance of a solar membrane liquid desiccant air conditioning system.
Solar Energy, 2014. 102(0): p. 56-73
[5] Dieckmann, J, Roth, K.W, Brodick, J, 2004, ‘Liquid Desiccant Air Conditioners’, ASHRAE (American Society of Heating, Refrigerating,
and Air Conditioning Engineers) Journal, p. 1, viewed 1 October 2015, bookstore.ashrae.biz/journal/download.php?file=Emerging102008.pdf
[6] Dean, J, Kozubal, E, Herrmann, L, Miller, J, Lowenstein, A, Barker, G & Slayzak, S 2012, Solar-Powered, Liquid Desiccant Air Conditioner
for Low-Electricity Humidity Control, Energy and Water Projects Demonstration Plan SI-0822, Environmental Security Technology
Certification Program (ESTCP)
[7] Baniyounes, A., Liu, G., Rasul, M.G., and Khan, M.M.K. (2013), Comparison Study of Solar Cooling Technologies for an Institutional
Building in Subtropical Queensland (Australia) Absorption vs. desiccant cooling systems, Renewable and Sustainable Energy Reviews, 23,
421-430, Elsevier..
[8] EnergyPlus 2015, Engineering Reference The Reference to EnergyPlus Calculations, viewed 10 June 2016,
https://energyplus.net/sites/default/files/pdfs_v8.3.0/EngineeringReference.pdf
[9] Baniyounes, A., Rasul, M.G., and Khan, M.M.K. (2013), Experimental Assessment of a Solar Desiccant Cooling Systems for an Institutional
Building in Subtropical Queensland, Australia, Energy and Buildings, 62, 78-86, Elsevier.
[10]Baniyounes, A., Rasul, M.G. and Khan, M.M.K. (2013), Assessment of Solar Assisted Air Conditioning in Central Queensland’s Subtropical
Climate, Renewable Energy, 50, 334-242. Elsevier.
491