Case Studies in Thermal Engineering 28 (2021) 101364

Contents lists available at ScienceDirect

Case Studies in Thermal Engineering

journal homepage: www.elsevier.com/locate/csite

Experimental investigation on the improved cooling seasonal

performance factor by recycling air flow energy from AC
outdoor fans
Yen-Tang Chen, Da-Sheng Lee *
Department of Energy and Refrigerating Air-Conditioning Engineering, National Taipei University of Technology, Taipei, Taiwan

A R T I C L E I N F O A B S T R A C T

Keywords: Due to Taiwan’s constant exposure to high humidity and high temperatures throughout the
AC outdoor fan summer, the demand for air conditioning is immense. As a result, energy consumption is very
Energy recycling high. For improving the energy consumption of air conditioners, the exhaust from the outdoor fan
Urban heat island
of the air conditioner is used to drive the wind-powered generator for energy recovery. The
Improve CSPF
Recycle airflow
recovered energy can offset part of the power consumed by the air conditioner. The exhaust can
Energy efficiency also be guided upward by the generator to suppress the urban heat island effects. In addition, a
numerical simulation was employed to assess various settings of the wind turbine for recovering
kinetic energy. The cooling seasonal performance factor (CSPF) was used to evaluate the recovery
efficiency. The method used discovered three points worth highlighting in the field of energy-
saving research in air conditioning systems: 1. According to the optimized theoretical value,
the CSPF value can be improved by approximately 5.9% using the method. 2. A device can
prevent the backflow of air into the outdoor unit of an air conditioner to avoid reheating the
exhaust that passed through the fins. 3. The device can channel airflow upward, triggering natural
convection and reduce urban heat island effects.

1. Introduction

According to the Global Energy Review 2019 [1] released by the International Energy Agency, the global energy demand increased
by 0.9% (equivalent to 40% of the growth rate observed in 2018) due to the slowdown in global economic growth in 2019 (gross
domestic product growth fell from 3.6% in 2018 to 2.9% in 2019). Another report of the International Energy Agency [2] stated that
from 2018 to 2019, global sales of the entire cooling equipment market have increased by approximately 10% despite a worldwide
economic downturn. Currently, air-conditioning accounts for 20% of total power consumption in buildings and 10% of total global
power consumption. However, the use of air-conditioning equipment is currently only centralized in a few countries, and is expecting
to grow rapidly in emerging economies. Air-conditioning is predicted to become the largest category of energy consumption in the
future, and the global energy need for space cooling will triple by 2050 [3]. Therefore, engineers must develop efficient cooling
equipment that consumes minimal energy, because the world is experiencing massive growth of demand for cooling equipment.
Efficient cooling equipment can slow down the energy consumption growth trend of cooling equipment.
Standards were set to examine the cooling capacity of cooling equipment and to determine the specifications for low energy

* Corresponding author. Tel.: 886227712171x3510; fax: 886227314919.

E-mail addresses: [email protected] (Y.-T. Chen), [email protected] (D.-S. Lee).

https://doi.org/10.1016/j.csite.2021.101364
Received 19 February 2021; Received in revised form 4 August 2021; Accepted 16 August 2021
Available online 21 August 2021
2214-157X/© 2021 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/).
Y.-T. Chen and D.-S. Lee Case Studies in Thermal Engineering 28 (2021) 101364

Fig. 1. Reference outdoor temperature distribution in Korea, Japan, and Taiwan.

consumption. The Energy efficiency ratio (EER) was developed to compare the difference between the cooling capacity and the
required power consumption of cooling equipment. However, EER was only designed to evaluate cooling equipment operating under
fixed conditions. Therefore, the Air-Conditioning, Heating, and Refrigeration Institute formulated a seasonal energy efficiency ratio
(SEER) related to climatic conditions in AHRI 210/240 in 2008. In addition, the International Organization for Standardization (ISO)
also published ISO16385-1,2,3 [4–6] that incorporated climatic conditions in 2013, and set a SEER-like related specifications such as
cooling seasonal performance factor (CSPF [7]), heating seasonal performance factor (HSPF), and annual performance factor for
cooling, heating, and annual efficiency.
In 2013, the National Standards of the Republic of China also formulated the CSPF specification, namely CNS15712-1, in accor­
dance with ISO16385-1 to improve the quantitative evaluation of energy efficiency over the years. Conventionally, Taiwan used EER
under fixed conditions to evaluate the energy efficiency and annual energy consumption of fixed-frequency air-conditioner. EER was
applicable because the energy consumption is almost constant at a fixed speed in such air conditioners. However, the newly mature
changeable frequency air-conditioners operate by changing the compressor speed to control the change of refrigerant flow and to
adjust the indoor cooling capacity according to the requirements of different indoor and outdoor environments. This shows that the
efficiency metrics of an inverter air-conditioner are related to climatic conditions.
Cooling capacity ϕt=35◦ C
EER = =
Power input Ṗt=35◦ C
∑n
Total cooling load j=1 ϕ(tj ) · nj
CSPF = = ∑n
Total power consumption j=1 Ṗ(tj ) · nj

where the parameters are defined as:

ϕt=35◦ C denotes cooling capacity at 35 ◦ C; Ṗt=35◦ C is the power input at 35 ◦ C

n = temperature section; j = the number of the temperature section.
ϕ(tj ) is the cooling load of the j temperature section; Ṗ(tj ) is the power consumption of the j temperature section; nj is the time of the
j temperature section.

The CSPF calculated by CNS15712-1 is based on the ratio of the annual cooling seasonal total cooling load to the cooling seasonal
energy consumption. Therefore, CSPF values vary due to different climates and operating conditions in different countries. Moreover,
CSPF is weighted because of the annual occurrence hours in the temperature range (24–37 ◦ C for Taiwan), hence the ratio of its air-
conditioning capacity to power consumption varies with the ambient temperature (as shown in Fig. 1). This study used wind turbine to
offset for the energy consumption of air conditioners, which can reduce the cooling seasonal energy consumption value evenly
regardless of the air-conditioning capacity, so as to improve the CSPF.
At present, energy saving of cooling equipment can be achieved through (1) increasing evaporation temperature and reducing
condensation temperature, (2) improving frequency response of air conditioners, and (3) combining renewable energy for power
supply or energy recycling. The first two methods involve partial improvements of the control system, compressor, and condensing
system of the air conditioning equipment, which have been largely researched; additionally, solar energy and wind energy supply are
topics of various studies.
In the control system, what has been generally commercialized is to change from the past on–off control to the current use of
frequency conversion to reduce the energy consumption of the compressor, which has been a remarkably effective approach.
Furthermore, related applications have been proposed recently with the rapid development of artificial intelligence (AI). Most of the
highlights focus on major commercial heating, ventilation, and air-conditioning (HVAC) systems. For instance, Zheng [8] proposed a
framework to define and learn multiple COP tasks. The COP here measures the energy efficiency of a chiller by its cooling ability under
a certain input power. The chillers usually consume the majority of HVAC energy. A large amount of data was accumulated and
machine learning was used to predict and optimize the chiller control mode to achieve energy saving. In Ref. [9], performing random
neural network training in an empty building can achieve the load estimation of manned building, and the calculation was relatively
easy. In addition, the [10,11] used different AI technologies to manage the system for improved overall operation and greater energy

2
Y.-T. Chen and D.-S. Lee Case Studies in Thermal Engineering 28 (2021) 101364

Fig. 2. The design can create the upward air to bring the floating particles.

Fig. 3. Flow chart of wind kinetic energy recycling experiment.

efficiency. Many studies have been conducted on HVAC because its scope of use is generally much larger than that of household air
conditioning; knowledge of HVAC is essential in energy saving research.
For the energy saving of a split-type air conditioner used in a home, the use of external support can improve its initial performance
or energy saving. The application of such support to compressors such as in Ref. [12] that used a vacuum solar collector to absorb solar
heat for assisting the compressor to raise the temperature and enter a high-pressure status, which saved some power consumption of
the compressor. Moreover [13], sprayed chilled water on an air-cooled condenser to pre-cool the ambient temperature, which in turn
increased the cooling capacity by approximately 15% and saved 18% of the energy cost. In Ref. [14], an energy storage approach was
applied and microencapsulated phase change material cooling storage system was used to provide cooling through phase change of
part of the time, shorten the air conditioner’s operating time, and achieve 30.5% energy saving efficiency. Furthermore [15], used a
moisture-transferring and quick-drying textile to collect condensate water in the air to cool the compressor, which increased the
dehumidification capacity of the air conditioner by 25.4% and the EER by 7.3%.
Regarding control methods based on machine learning, Lee [16] used cloud-based AI to improve the model-based proportio­
nal–integral–derivative control, and also verified that a cloud-based AI was the most cost-effective method to improve energy effi­
ciency. In Ref. [17], an inverse predicted mean vote was used as a comfort index for AI learning, which is a more effective control
method to save approximately 30% of the energy consumption of the overall system.
The aforementioned studies indicated that current various energy-saving methods begin from improving the four major compo­
nents or system controls of the indoor unit of an air conditioner. In terms of integration with wind turbines [18], used wind energy as
the source of electricity for air-conditioning systems. However, the use of waste heat recovery of outdoor unit to improve energy saving
is currently unavailable. Therefore, this study intended to recycle exhaust gas discharged by air conditioners, and select CSPF as an
evaluation index to improve efficiency.
Typically, an outdoor fan dissipates heat horizontally from the compressor. Because air-conditioning equipment is widely used in
cities, and because buildings in cities are constructed in close proximity to each other, high-temperature air tends to accumulate on the
ground and increase surface temperature. This study applied a vertical-axis wind turbine and a flow augmentation device to the
outdoor fan, which actively channelled the airflow passing through the wind turbine upward, and avoided horizontally discharged
waste heat directly affecting the surrounding buildings. In addition, it was able to increase the upward hot air exhaust rate, promote
the introduction of horizontal airflow around the building, and increase the air circulation efficiency. By channelling upward the waste

3
Y.-T. Chen and D.-S. Lee Case Studies in Thermal Engineering 28 (2021) 101364

Fig. 4. Vertical-axis wind turbine combined with the flow augmentation device.

heat discharged by the substantial use of air conditioners in the city for direct and swift emission, the urban surface temperature and
the impact of urban heat island effect on urban warming can be reduced. Moreover, the designed device can disperse floating particles
in the air upward and outward from the city centre, and can also reduce the excessive concentration of these particles in the city, as
illustrated in Fig. 2.

2. Experimental equipment and research methods

2.1. Experimental equipment and framework

The initial conditions of the experiment set the indoor room temperature to 24 ◦ C and humidity to 60% to maintain constant indoor
temperature and humidity. This study started measuring the entire vertical-axis wind turbine generation system after the split-type air
conditioner was operating at low frequency. With the stable flow rate of the outdoor unit, the pre-adjusted stator and power generation
at different tilt angles were measured. An Arduino Uno board was used as a bridge to save the data in Excel format to a laptop for
subsequent data integration and data analysis. The experimental process of wind kinetic energy recycling is illustrated in Fig. 3.

2.1.1. Air conditioner selection

The air conditioner used in this study was a Hitachi 1–pair–1 split-type air conditioner RAS-22NB/RAC-22NB. The electricity of this
model was single-phase 220 V, rated cooling capacity was 2.2 kW, rated power input was 403 W, dehumidification capacity was 1.4 l/
h, and energy efficiency ratio was 4.33 W/W. The dimensions (W × D × H) of the outdoor case were 730mm × 600mm × 290 mm with
a 355 mm fan inside. The fan was driven by a 36 W and 6HP AC motor. When the air conditioner was switched on the outdoor fan
started to operate and rotate continuously at a constant speed until the air conditioner was turned off. The output air volume measured
at 30 cm was 4.2 m/s.

2.1.2. The design of the wind turbine

The output air volume measured on the outdoor unit was approximately 4.2 m/s, which was a fixed-speed wind source. The wind
speed was lower than that of an average wind energy generator, and the experiment focused on exploring the airflow and energy
generated by outdoor fan to reduce the effect of CSPF. Therefore, a vertical-axis wind turbine was selected. In addition, the vertical-axis
wind turbine is suitable for this experiment because it has the advantages of easy production, easy maintenance, low operating noise,
low initial wind speed, and can generate a small amount of electricity at approximately 1 m/s. In terms of configuration, we used three-
dimensional printing to produce eight C-shaped blades with a thickness of 1 mm and a height of 240 mm.

4
Y.-T. Chen and D.-S. Lee Case Studies in Thermal Engineering 28 (2021) 101364

Fig. 5. Geometric model and mesh in 3D simulation.

2.1.3. The design of flow augmentation device

To allow the airflow entering from the outdoor fan to be channelled to drive the wind turbine, and to ensure the airflow flowing
through the wind turbine to generate electricity can discharge in the intended direction, we designed a flow augmentation device on
the basis of performance enhancements on vertical axis wind percentage to stabilize the flow direction of the airflow discharged from
the outdoor fan and achieve the most effective application. The outer diameter of this device is approximately 460 mm in width and
328 mm in height, and the inner rotation area is approximately 280 mm in diameter, and the assembled model is as shown in Fig. 4.

2.1.4. The selection of wind energy conversion system

In the review of [19], wind energy conversion system can be divided into fixed speed wind turbines, limited variables speed wind
turbines, variable speed with partial scale frequency converter, and variable speed with full-scale frequency converter according to the
power delivering capabilities and type of machine involved wind turbines. Because a constant rotating speed of the air conditioner
(from the moment it was switched on until it was switched off) was maintained in this study, wind energy and electrical energy
conversion was achieved using a fixed speed wind turbine. This type of turbine can be easily acquired and controlled, and has simple
motor structure.

2.2. Numerical calculation method

2.2.1. Three-dimensional flow model and grid independence test

To reduce the simulation time, the present researchers appropriately simplified the wind turbine model, and the simulation results
should not be excessively different from the actual situation. The three-dimensional model designed by this study fixed the inflow
direction at X axis with a low blade rotating speed, and the controlled experimental equipment was placed in a static indoor envi­
ronment. The simulation model can be simplified as shown in Fig. 5 to observe the airflow direction channelled by the guide vanes after
the airflow drove the turbine fan blades, and the wind energy generated by the turbine.
Finally, based on the recommendations generated by the software we can adjust the quality of the skewness and orthogonal size of
the mesh accordingly. If the mesh quality is too poor, make corrections and reinforcements. The model with skewness greater than 0.95
should not be used in calculating, so keep it below 0.95. The orthogonal should be close to 1. The skewness quality of this study was
distributed between 1.7912 × 10− 6 and 0.873, and the orthogonal quality is between 0.1403 and 0.9974. For this study, the grid
quality was within an acceptable range.

2.2.2. Initial and boundary conditions

For the simulations, the CFD software ANSYS Fluent had been used. All cases used 4.2 m/s steady, unidirectional incoming airflow
as the wind from the outdoor fan. The air is considered as incompressible Newtonian Fluid and with gravity in the environment. The
settings of boundary conditions included: (1) the air inlet was set as wind blowing in from infinity, and the air inlet speed was set in the
setting section; (2) the air outlet was set as free outflow to observe the effect of turbine blades on the airflow; and (3) the interface
boundary was set to interfaces of the regions, whereby the control body regions could exchange values.

2.2.3. Controlling equations

This study used the SST k-ω turbulence model that contains the advantages of conventional k-ω and k-ε, which strengthens the
calculation of near-wall boundary and free shear flow based on the k-ω turbulence model near-wall simulation effect. The system has
no additional damping functions in the low-Re turbulence model. Therefore, this model is suitable for calculating the airflow close to

5
Y.-T. Chen and D.-S. Lee Case Studies in Thermal Engineering 28 (2021) 101364

Fig. 6. The difference in air velocity and static pressure between the experiment and simulation.

the near-wall blades and near-field airflow. Compared with k-ε that is excellent at calculating far-field parameters, SST k-ω is more
accurate in measuring the speed of changes with pressure gradient, has more satisfactory results in processing eddy viscosity, and can
effectively infer the position of separate flow. Therefore, the SST k-ω turbulence model was used to calculate the aerodynamics of the
turbine blades, which can maintain the pressure around the fan blades, ensure stability of the airflow calculation, and accurately
analyse the airflow direction after the air left the fan blades.

2.2.4. Convergence
This study considered (1) The number of grids and the uniformity and non-uniformity of the spacing when performing simulation
calculations. (2) The error before and after the value of each numerical calculation should be less than 10^-3. (3) During the iterative
calculation, the under-relaxation parameter was to be set lower than the original Fluent setting value to avoid the divergence of the
value change or increase its convergence. According to the difference of the grid models, the time step of each model should be 0.01 s
or less to ensure the stability of the simulation process. In this study, when the error of the inlet and outlet airflow was less than 1%, it
was regarded as meeting the requirement of convergence.

2.2.5. Calculation
This study selected the finite volume method to determine numerical simulation calculations because of its easy configuration and
suitability for unstructured grids, which can provide favourable simulation results for the adopted complex fan blades and diversion
structures. When performing calculations, this method can effectively distinguish the computing grid according to the control body
without overlapping to other parts, and integrate the differential equations of different control bodies to obtain a discrete equation
group. In addition, this study also used sliding mesh to solve the control bodies between relative motion, which can directly solve the
multi-body issue.

2.2.6. First case for validating simulation results

The first experiment of the research had been compared to the simulation results, which represented the guide vane angle of 40◦
and fan blade angles of 0◦ –40◦ . The velocity and static pressure derived by this experiment were mutually verified with the data
obtained by the simulation software, which is displayed in Fig. 6. After the applicability of the simulation parameter settings had been
determined, the simulation could be expanded to more different angles.
The actual measured velocity of the rotation area gradually decreased as the angle of the guide vanes increased. Regarding the
angles of fan blades, 10◦ presented the highest velocity, whereas 30◦ showed the lowest velocity. The overall static pressure can be
verified by the Bernoulli’s principle, which was consistent with the actual results of wind energy recycling.

3. Calculated results

3.1. Calculation result and analysis for the generated power

This study sorted the calculation results of the recycled and collected energy from the designed equipment by adjusting angles of
guide vanes of the flow augmentation device and wind turbine blade angles. Under a fixed angle of the guide vanes of the flow
augmentation device, the maximum wind energy recycling can be achieved at the blade inclination angle of 30◦ . However, the
maximum wind energy recycling decreased after the inclination angle exceeded 30◦ .
According to the various angles of the guide vanes of the flow augmentation device and the angles of the wind turbine fan blades,
the recycled energy was applied to the air conditioner used in this study to calculate CSPF increases, as summarized in Table 1.

6
Y.-T. Chen and D.-S. Lee Case Studies in Thermal Engineering 28 (2021) 101364

Table 1
Power generation (Unit: W) and CSPF increased percentages (Unit: %) under of various flow augmentation device guide vane angles and wind turbine
fan blade angles.
Fan blades 0◦ 10◦ 20◦ 30◦ 40◦
Guide vanes
W CSPF W CSPF W CSPF W CSPF W CSPF

0◦ 2.0 1.4 2.5 1.8 2.7 1.9 3.1 2.2 2.3 1.6
10◦ 2.6 1.9 3.1 2.2 5.0 3.6 5.1 3.7 4.8 3.5
20◦ 4.0 2.9 5.1 3.7 6.5 4.8 7.0 5.1 5.5 4.0
30◦ 5.4 3.9 7.4 5.5 7.8 5.8 8.0 5.9 7.5 5.5
40◦ 4.2 3.0 6.5 4.8 6.8 5.0 7.3 5.4 6.6 4.8

Table 2
Direction and velocity of airflow under various guide vane angles and wind turbine fan blade angles of the flow augmentation device (X-Z section).
Fan blades 0◦ 10◦ 20◦ 30◦ 40◦
Guide vanes

0◦

10◦

20◦

30◦

40◦

3.2. Calculation result and analysis for the air flow and direction

The flow-field velocity graphs of the X-Z section showed that the velocity ranged from 0 m/s to 10 m/s; blue, green, yellow, and red
indicated a greater velocity change in sequence. The initial boundary condition of the simulation was 4.2 m/s at the inlet velocity. After
the airflow passed through the vertical-axis wind turbine, the changes in the airflow rising rates derived from the flow augmentation
device was observed. The graphs are as shown in Table 2.
The rising rate of the derived airflow was extremely small when the angles of the guide vanes and turbine blades were both 0◦ .
Subsequently, the rising rate of the derived airflow increased as the angle of the guide vanes increased, but the velocity obviously
began to decrease at 40◦ . The maximum rising rate obtained was approximately 3.1 m/s in overall when the guide vane angle was 20◦ .
However, when the guide vane angle was 30◦ with the turbine blade angle between 0◦ and 40◦ produced a relatively stable rising rate
greater than 2.5 m/s, and the airflow can be effectively discharged out of the area of the outdoor unit.

4. Discussion

In this study, a wind turbine was installed at outdoor fan with stable wind outlet to recycle the wind kinetic energy. Outside the
wind turbine, a flow augmentation device was used for airflow channelling, and various angles of the guide vanes on the device were
simulated using simulation software. With the adjustment of the inclination angles of the internal wind turbine blades, the wind energy

7
Y.-T. Chen and D.-S. Lee Case Studies in Thermal Engineering 28 (2021) 101364

recycling status was discussed.

According to the airflow simulation results, the X-Z section at the angles of 0◦ and 20◦ for different guide blade angles yielded the
highest velocity and the smallest static pressure in the rotating region, indicating the lowest wind energy recycling. At the inclination
angle of 30◦ , the rotating region witnessed the lowest velocity and the maximum static pressure, representing that the wind energy
recoverable was the largest, and was consistent with the actual wind energy recycling curve.

5. Conclusions

This study used a wind turbine to recycle the wind kinetic energy of outdoor fan, simulated various angles of the guide vanes on the
flow augmentation device, and adjusted the blade angles of the internal wind turbine. The obtained wind energy recycling conclusions
were as follows:

1. The increase of CSPF value verified the feasibility of using wind energy to recycle the kinetic energy of exhaust gases from air
conditioners.
2. This study provided a design concept that integrated a diversion structure with a wind turbine. Observed from the X-Z section,
airflow can effectively rise after passing through the flow augmentation device; air can be uniformly mixed with the outdoor at­
mosphere at a high place, effectively reducing the external heat effect on the environment, so as to avoid urban heat island effects.
3. In subsequent studies, we plan to conduct small-scale experiments in primary and secondary schools to verify the proposed
method’s effectiveness in avoiding the increase in surface temperature.

Authorship statement

Conception and design of study: Yen-Tang Chen, Da-Sheng Lee

Acquisition of data: Yen-Tang Chen
Analysis and/or interpretation of data: Yen-Tang Chen, Da-Sheng Lee.
Drafting the manuscript: Yen-Tang Chen
Revising the manuscript critically for important intellectual content: Yen-Tang Chen, Da-Sheng Lee.
Approval of the version of the manuscript to be published (the names of all authors must be listed): Yen-Tang Chen, Da-Sheng Lee.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to
influence the work reported in this paper.

References

[1] International Energy Agency IEA, Global Energy Review, 2020, 2019.
[2] International energy agency cooling, Available online, https://www.iea.org/reports/cooling. (Accessed 22 June 2020).
[3] the Organization for Economic Co-Operation and Development/International Energy Agency, The Future of Cooling Opportunities for Energy-Efficient Air
Conditioning, 2018, p. 92.
[4] International Organization for Standardization ISO 16358-1:2013, Air-cooled Air Conditioners and Air-To-Air Heat Pumps — Testing and Calculating Methods
for Seasonal Performance Factors — Part 1: Cooling Seasonal Performance Factor, 2013, p. 13.
[5] International Organization for Standardization ISO 16358-3:2013 Air-cooled Air Conditioners and Air-To-Air Heat Pumps — Testing and Calculating Methods
for Seasonal Performance Factors — Part 3: Annual Performance Factor.
[6] International Organization for Standardization ISO 16358-2:2013 Air-cooled Air Conditioners and Air-To-Air Heat Pumps — Testing and Calculating Methods
for Seasonal Performance Factors — Part 2: Heating Seasonal Performance Factor.
[7] Bureau of Standards, Metrology and Inspection, M.O.E.A., R.O.C, CNS15712-1 Air Conditioners and Heat Pumps – Testing and Calculating Methods for Seasonal
Performance Factors -Part 1: Cooling Seasonal Performance Factor, 2013, pp. 1–24.
[8] Zheng, Z.; Xie, D.; Pu, J.; Wang, F. In Proceedings of the Proceedings of the Eleventh ACM International Conference on Future Energy Systems. ACM: New York,
NY, USA (2020), pp. 47–56.
[9] A. Javed, H. Larijani, A. Ahmadinia, D. Gibson, Smart random neural network controller for HVAC using cloud computing technology, IEEE Transact. Industr.
Inform. 13 (2017) 351–360.
[10] A. Afram, F. Janabi-Sharifi, A.S. Fung, K. Raahemifar, Artificial neural network (ANN) based model predictive control (MPC) and optimization of HVAC systems:
a state of the art review and case study of a residential HVAC system, Energy Build. 141 (2017) 96–113.
[11] P.W. Tien, J.K. Calautit, J. Darkwa, C. Wood, S. Wei, C.A.J. Pantua, W. Xu, A deep learning framework for energy management and optimisation of HVAC
systems, in: Proceedings of the IOP Conference Series: Earth and Environmental Science, vol. 463, Institute of Physics Publishing, 2020.
[12] Y.V. Vankov, A.K. Al-Okbi, M.H. Hasanen, Solar hybrid air conditioning system to use in Iraq to save energy, E3S Web Conf. 124 (2019) 1024, https://doi.org/
10.1051/e3sconf/201912401024.
[13] S.A. Khan, F. Wang, S. Al-Zubaidy, Energy recovery through the condensate of air conditioning machines, Int. J. Eng. Res. Technol. 3 (2014) 548–555.
[14] L. Zheng, W. Zhang, L. Xie, W. Wang, H. Tian, M. Chen, Experimental study on the thermal performance of solar air conditioning system with MEPCM cooling
storage, Int. J. Low Carbon Technol. 14 (2019) 83–88.
[15] W.H. Chen, H.E. Mo, T.P. Teng, Performance improvement of a split air conditioner by using an energy saving device, Energy Build. 174 (2018) 380–387.
[16] D. Lee, F.-P. Tsai, Air conditioning energy saving from cloud-based artificial intelligence: case study of a split-type Air conditioner, Energies 13 (2020) 2001.
[17] K.L. Ku, J.S. Liaw, M.Y. Tsai, T.S. Liu, Automatic control system for thermal comfort based on predicted mean vote and energy saving, IEEE Trans. Autom. Sci.
Eng. 12 (2015) 378–383.
[18] G. Catalano, A wind electric-driven combined heating, cooling, and electricity system for an office building in two Italian cities, Energies (2012) 241–244.
[19] S. Bukhari, D. Bhutto, J. Ansari Wind Energy Conversion Systems (Wecs) Generators: A Review. 2019 International Conference on Computing, Mathematics and
Engineering Technologies, 2019.