2019 IEEE International Conference on Automatic Control and Intelligent Systems (I2CACIS 2019), 29 June 2019, Selangor, Malaysia

IoT-Based Smart Air Conditioning Control for

Thermal Comfort
A.M.Ali1,a, S.A.Abdul Shukor1,b, N.A.Rahim1, K.Kohlhof
Z.M.Razlan1, Z.A.Z.Jamal2 Faculty of Information, Media and Electrical Engineering
1
School of Mechatronic Engineering Cologne University of Applied Science
2
School of Microelectronic Engineering 50679 Cologne, Germany
Universiti Malaysia Perlis
02600, Perlis, Malaysia
a
[email protected]
b
[email protected]

Abstract—Thermal comfort brings a great influence that area tend to ‘over controlled’ the temperature and the
may affect the satisfaction of an occupant about the compressor regulation of the air condition which lead to high
surrounding indoor environment, as this could led to the level power consumption and it may not provide the satisfied
of productivity and social interactions. One of the most suitable thermal comfort [4]. As indicated by World Bank, 85% of
tools that have been utilized in tropical region to create the populace will be situated in the developing countries in
thermally comfortable indoor environment is the air 2030 [5]. This development is prompting an expansion of
conditioning (AC) system. However, the conventional method buildings particularly in the downtown (city centre) area. In
of controlling the AC may not guarantee thermally return, the usage of artificial system will be gradually
comfortable indoor ambience, furthermore excessive cooling
dependent to operate the buildings, as the time in which
chosen by users may also contributed to negative side effect
people spend inside the building is significantly increased.
toward the occupancy’s health. To overcome this problem, an
IoT-based smart system that can control the AC to provide a This has led architects and engineers to consider approaches
suitable thermally comfortable environment is proposed in this to improve the occupancy’s thermal comfortability by
project. The system will interact with the users by recording enhancing the thermal comfort tools in the building, whereas
their feeling toward the environment as the input for the it is been considered that individuals spend at range from
system. Together with sensors’ data, the information will be 80% to 90% of their days’ time inside the buildings [6].
integrated with an enhanced Predicted Mean Vote (PMV)- Figure 1 summarizes the overall scope of this project.
based model to control the AC smartly so that the occupant
This study concentrates on tropical region where it faces high
can feel thermal satisfaction. Based on the results, the IoT-
based smart system can create a thermally comfortable indoor
relative humidity, high air temperature and low air speed. It
environment compared to the maximum cooling setting as focuses on small sized room in which the occupancy is doing
usually being adapted by the room’s occupants. sedentary work. Even though there are many tools that able
to create thermal comfort, in this study, AC system will be
Keywords— Internet-of-Things (IoT), air conditioning system considered as the main tool and it will be controlled by using
control, Predicted Mean Vote (PMV)-based model the enhanced PMV-based algorithm. With the aid of IoT, the
AC can easily and effectively create a suitable thermal
I. INTRODUCTION comfort environment for the occupant. This paper is arranged
in this manner: Section II will cover the background that
In recent years, many researchers around the world have illustrates the basic ideas of thermal comfort and the selected
put more efforts in the field of thermal comfort which is due enhanced PMV-based model. Section III will cover the
to the expanded public argument about the environmental methodology especially on how the integration of the
change. In general, thermal comfort and evaluation of the enhanced PMV-based model with related hardware in
indoor environmental quality is depending on human developing the system. Section IV will cover the results, and
physiological and psychological responses to the surrounding Section V will conclude and recommends future work.
environment integrated with behavioural factors that interact
with space [1]. The developing enhancement of the present
II. RESEARCH BACKGROUND
employees has made social challenges in a workplace
environment. A specific issue has risen which is on how to
A. Thermal Comfort
provide a comfortable environment inside an office
throughout the working time [2]. However, obtaining an The word thermal comfort is utilized to imply data about
optimal office temperature that the worker feel comfortable the thermal condition of human with the indoor environment.
with is challenging because it deals with an integration of the The concept of thermal comfort consists of three main
complex thermodynamic of the human body and a nonlinear disciplines such as physiology, psychology and behavioural
mapping between personal preferences and the factors. Creating a healthy and satisfying condition for the
environmental variables [3]. people is the main objective of thermal comfort researches.
As a whole, indoor thermal comfort researches can be
Air conditioning systems are adopted as a solution for classified into two approaches: adaptive and steady-state
thermal comfort purpose indoors and currently are extremely model. These two approaches have different ideology about
increasing due to the warmer weather. People in the tropical the way individuals react to their surroundings [7].

978-1-7281-0784-4/19/$31.00 ©2019 IEEE 289

Authorized licensed use limited to: KJ's Educational Institute. Downloaded on October 17,2024 at 09:25:15 UTC from IEEE Xplore. Restrictions apply.
 2019 IEEE International Conference on Automatic Control and Intelligent Systems (I2CACIS 2019), 29 June 2019, Selangor, Malaysia

Figure 1 The scope of this project

The adaptive approach identifies humans as the active E c: evaporation heat exchange at the skin
recipient which can react with a response to maintain their Eres respiratory evaporative heat exchange
thermal comfort whenever any environment changes Cres respiratory convective heat exchange
occurred that can cause discomfort to them [8]. According to
Nicol and Humphreys [9], adaptive approach is based on the
studies of naturally ventilated building. While the steady- PMV deals with a scale from -3 (cool) to +3 (hot), as
state approach is founded by P.O Fanger in 1970s could be appeared in Figure 2. It helps the users in characterizing their
used for indoor environment with AC system [10]. This comfortability towards their surrounding environment.
approach mainly depends on heat balanced model of the According to ISO 7730 [11], 75% of the people are feel
human body. Predicting a mean thermal sensation of the satisfied with a range between PMV = -1 and PMV = +1.
occupancy at indoor environment is the main target of the While the level of people’s satisfaction increased to 90%
Fanger’s model. Hence, most of the standard organization with PMV range between PMV = -0.5 and PMV = + 0.5. An
such as International Standard Organization (ISO) 7730 and additional PPD index which indicates the relationship of
American Society of Heating, Refrigerating and Air- PMV with the percentage of the people who may not satisfy
Conditioning Engineers (ASHRAE) use Fanger’s model to with the characterized scale has also been developed.
set the requirements for indoor thermal conditions. However, since PPD index is usually being used to measure
dissatisfaction of comfortability in a bigger audience, this
1) PMV Thermal Comfort Model PDD index is not usable here as the enhanced PMV-based
P.O Fanger developed the Predicted Mean Vote (PMV) model is designed exclusively for small indoor environment
by integrating equations of heat balance and experimental which occupied one or two people.
studies about human skin temperature and used it to 2) Enhanced PMV-Based Model
determine the thermal comfort. PMV and Predicted PMV has been used by numerous studies in developing
Percentage of Dissatisfied (PPD) are typical indicators that indoor thermal comfort models, but in this project an
use to express the thermal sensation of the people. PMV enhanced PMV developed by [13] is adapted. Since the
model is consider as the main reference for the ISO 7730 enhanced model has been developed for the tropical
[11] and ASHRAE 55 [12]. There are six factors that affect countries and for environment with occupancy who
thermal comfort at indoor environment, four of them are performed sedentary tasks, the model utilizes predefined
environmental-based such as air velocity, relative humidity, parameters as expressed in Table I
mean radiant temperature and air temperature while the other
two factors are personal such as clothing rate and metabolic
rate. PMV can be determined by equation (1). TABLE I. PARAMETERS REQUIRED TO CALCULATE PMV AS
DEFINED BY ISO 7730 [11]
.
0.303 0.028 Parameters Symbol Value
(1) Effective mechanical power W 0 W/m²
Mean radiant temperature tr 25 °C
Where: Person’s activity M 58.15 W/m²
M: metabolic rate Clothing level Id 0.155 m²°C/W
W: effective mechanical power Turbulence Tu 0%
H: dray heat loss

978-1-7281-0784-4/19/$31.00 ©2019 IEEE 290

Authorized licensed use limited to: KJ's Educational Institute. Downloaded on October 17,2024 at 09:25:15 UTC from IEEE Xplore. Restrictions apply.
 2019 IEEE International Conference on Automatic Control and Intelligent Systems (I2CACIS 2019), 29 June 2019, Selangor, Malaysia

Figure 2 Seven Point PMV Thermal Sensation Scales

Figure 3 The enhanced PMV used in this project, with the simulation analysis at relative humidity RH = 79.4%
Figure 3 shows the graph representing this enhanced While if the user choose PMV = -0.5, then T° = 19°C, v =
PMV-based model that has been utilized here. Users can 0.212m/s (low). Meanwhile, for PMV = 0, then T° = 21°C, v
easily choose a particular PMV scale ranging from -1 to +1 = 0.255m/s (medium). And for PMV = +0.5, then T° = 23°C,
with 0.5 increment or decrement as desired and the model v = 0.310m/s (high); PMV = +1, then T° = 24°C, v =
will set a fitting estimation of the air temperature and air 0.360m/s (high).
velocity (or fan speed) of the air conditioning system to
develop a thermally comfortable environment. In this study, TABLE II. VALUE OF THE AC AIR TEMPERATURE AND FAN SPEED
the analysis are done at an office room in the School of FOR PMV VALUES WHEN RH= 79% BY [13]
Mechatronic Engineering, Universiti Malaysia Perlis
(UniMAP) which is located in Pauh, Perlis, Malaysia where PMV value AC air temperature AC fan speed
its annual average relative humidity (RH) = 79.4% according -1 17°C Low
to Malaysia Meteorological Department [14]. As shown in -0.5 19°C Low
Table II, if the user selects PMV = -1, then the value of AC 0 21°C Meduim
air temperature (T°) and fan speed (v) of the AC should +0.5 23°C High
operate is around 17°C and 0.180m/s (low) respectively.
+1 24°C High

978-1-7281-0784-4/19/$31.00 ©2019 IEEE 291

Authorized licensed use limited to: KJ's Educational Institute. Downloaded on October 17,2024 at 09:25:15 UTC from IEEE Xplore. Restrictions apply.
 2019 IEEE International Conference on Automatic Control and Intelligent Systems (I2CACIS 2019), 29 June 2019, Selangor, Malaysia

B. AC system as the thermal comfort tool usage, electricity demand increase when more people enter
1) Usage of conventional AC system to create thermal the room. Furthermore, since their system set the PMV at
specific value, this may not provide a thermally comfortable
comfort
ambiance as desired.
AC system is an electrical machine that is used to
improve the comfort level of an indoor environment by
controlling the air temperature of that closed area. AC III. SMART AC CONTROL
consists of inner and outer units and using condensation and The structure of the proposed smart system is as shown in
evaporation properties of the refrigerant gas, it does cooling Figure 4, involves of hardware platform and programming
and heating function. According to Ozkan and Aybar [15], algorithm in suitable software. DHT11 sensor, Esp8266
by operating the compressor as on/off, AC attempts to NodeMcu Wi-Fi microcontroller, relay circuit and IR
maintain room temperature within a range of temperature remote, smart phone and 1HP portable AC are part of the
value that has been set by the user. For instance, during hardware used in this smart system. The software includes
summer season when the room temperature is higher, user the programming of the air temperature and relative humidity
will set a cooling temperature (Tc), and the compressor will DHT11 using Arduino. When the user insert his / her thermal
operate to blow cool air until ambient temperature reaches feeling (PVM preference) from range (-1 to +1), the mobile
Tc. The compressor will stop when room temperature application will communicate with Esp8266 NodeMcu Wi-Fi
exceeds Tc. The inner fan will keep on working in order to for sending data to the cloud server to control the AC as
maintain the air flow within the room. If the cooling desired.
parameters are measured properly, the AC system could be
used to control the room's air draft to generate a thermally
comfortable condition in the room [13].
Noticeably, user tends to set the AC system into
maximum cooling (lowest temperature at highest fan speed)
especially when he / she comes from a warmer place.
However, this will not guarantee a thermally comfortable
environment for the occupant, and furthermore, it can
increase the energy utilization of air conditioning system
[16]. In few nations, ‘reduce thermostat temperature setting’
becomes one of the inspiring campaigns. For example, in
Malaysia, minimum thermostat temperature of 24 °C for
cooling governmental offices has been promoted by the
government [17]. Instructions on the control of indoor
regulator (thermostat) setting are particularly important for
developing nations located in a tropical area as reviewed by
Damiati et al. [18]. However, people are not following the
instruction of thermostat, especially when occupant arrived
from outside where high solar radiation and high relative Figure 4 Structure of the proposed smart AC system
humidity at daytime tend to utilize excessive cooling which
as a result it may led them to discomfort and illness for The collected data of the sensors must pass through a
prolong usage. processing stage in order to effectively control the AC and
achieve satisfied thermal comfort. The most essential part of
2) AC control using IoT-Based the data processing is designing a suitable programming of
Thermal comfort has been exposed to enhance the Android Application by utilizing enhanced PMV-based
pleasure of workers as well as their efficiency at work and model. Arduino software is used to program the system
their social collaborations. Many researchers suggested an which have three input variables which are users’ feeling or
Internet of Thing (IoT)-based system to control AC system user PMV preference at range of -1 to +1, relative humidity
so as to create a thermally comfortable environment with low at range from 40% to 90%, and room temperature at range
power consumption. Project by Song, Feng, Tian, & Fong, from 19°C to 34°C. Two output variables which are
[19] proposed an IoT-based controlling system for the air temperature of the AC (from 17°C to 26°C) and fan speed
conditioner to reduce the power utilization, the system at range from 0.1m/s to 0.5m/s are used.
concentrate on energy reducing rather than thermal
comfortability of the occupancy and their project have lack IV. RESULTS AND DISCUSSION
of standard algorithm such as PMV. Creating a comfortable environment is very significant in
In an experiment that was done by Godo, Haase, & Nishi order to enhance personal satisfaction as well as efficiency of
[20], effective air conditioner control was achieved by means work. The proposed smart system will monitor the air
of self-controlled sensor (pressure sensor) which are fitted in temperature and relative humidity of the room using DHT11
seats and can sense whenever individual takes a seat or stand sensor. A mobile application which acts as a controller
up. The data from wireless sensor network (temperature and allows the user to control the AC system in creating a
humidity sensor) and pressure sensor are administered using thermally comfortable ambiance. The graphical user
a cloud server which allow the smart system to regulate the interface (GUI) allows the user to insert his/her feeling. The
air conditioner function accordingly. The system can be integration of the user feeling together with the enhanced
fitted into current residential house. All through PMV index relation is as shown in Table III. If the user feels
experimentation, utilization of electricity was decreased. cool in the room then the PMV index should be assigned to
Although the results emphasis decreasing on the total energy warm, and if the user feel slightly cool then the PMV will

978-1-7281-0784-4/19/$31.00 ©2019 IEEE 292

Authorized licensed use limited to: KJ's Educational Institute. Downloaded on October 17,2024 at 09:25:15 UTC from IEEE Xplore. Restrictions apply.
 2019 IEEE International Conference on Automatic Control and Intelligent Systems (I2CACIS 2019), 29 June 2019, Selangor, Malaysia

assigned to slightly warm. However, if the user feels

comfortable with the environment then he/she can choose
neutral and the PMV sensation will be neutral.
This mobile application system would be more effective
to be used than utilizing conventional remote to control the
AC system. This is because conventionally, users will try to
use the AC system in maximum cooling by setting the AC
temperature to the lowest value and at high fan speed [4]
[16]. Moreover, the value of the temperature may not
represent the desired thermally comfortable environment.
Furthermore, the mobile application implemented in this
project is designed according to the enhanced PMV-based
model which is developed based on standards such as ISO
7730 and ASHRAE 55. A corresponding value of the AC
temperature and fan speed will be displayed at the GUI for
users’ convenience. In terms of the security and privacy, the
mobile application is fully secured because it is designed
using MIT App inventor and there is no users’ personal data Figure 6 Relative humidity measurement taken by DHT11 sensor
needs to be uploaded on the internet, i.e. ThingSpeak website Figure 7 shows the GUI of the Android-based application
[21]. The only data needs to be uploaded are users’ feeling for AC control. The GUI is classified into three main parts:
toward their environment, air temperature and relative the first section is for displaying the current relative
humidity inside the room.
humidity and air temperature inside the office room. While
the second section is for the users’ feeling about the
TABLE III. RELATIONS BETWEEN USERS FEEELING AND PMV environment and consists of five button – cool, slightly cool,
THERMAL SENSATION
natural, slightly warm, and warm, so that the occupant can
Users feeling PMV sensation PMV Value choose any button according to his/her feelings. However,
Cool Warm -1 when the temperature of the room is below 20°C, and if the
Slightly cool Slightly warm -0.5 user selects ‘feel warm’, the system will be at neutral mood
Neutral Neutral 0 to avoid excessive cooling. The third section in the GUI is
Slightly warm Slighly cool +0.5 assigned to display the corresponding value of the
Warm Cool +1 temperature (from 17°C to 26°C) and fan speed (low,
Figure 5 and 6 show the reading of air temperature and medium, high) that the AC should work according the user’s
relative humidity inside the room, the data collected using feeling.
DHT11 and sent to the internet cloud using ThingSpeak
website that act as communication platform between the
Esp8266 NodeMcu Wi-Fi microcontroller and the Android-
based application. It has been cleared from Figure 5 the air
temperature of the room when the smart system applied
became at the range of 23°C to 25°C (PMV= -0.5 and
PMV= 0) which consider the comfortable zone according to
[22]. Meanwhile, Figure 6 shows that value of relative
humidity varies from 83% to 93% as expected due to the
area of our study is located nearby Chuping, Perlis, Malaysia
which is known to be high in relative humidity according to
Malaysia Meteorological Department [14].

Figure 7 GUI of the Android-based Application for AC control

Figure 5 Air temperature reading taken by DHT11 sensor V. CONCLUSION

The main contribution of this project is to propose an

978-1-7281-0784-4/19/$31.00 ©2019 IEEE 293

Authorized licensed use limited to: KJ's Educational Institute. Downloaded on October 17,2024 at 09:25:15 UTC from IEEE Xplore. Restrictions apply.
 2019 IEEE International Conference on Automatic Control and Intelligent Systems (I2CACIS 2019), 29 June 2019, Selangor, Malaysia

IoT-based smart system that able to control the AC in order [17] L. C. Lau, K. T. Tan, K. T. Lee, and A. R. Mohamed, “A
comparative study on the energy policies in Japan and Malaysia
to produce an appropriate, thermally comfortable indoor in fulfilling their nations’ obligations towards the Kyoto
environment by monitoring air temperature and relative Protocol,” Energy Policy, vol. 37, no. 11, pp. 4771–4778, Nov.
humidity inside the office room. It has been realized that 2009.
most of the occupancy discomfort is resulting from the [18] S. A. Damiati, S. A. Zaki, H. B. Rijal, and S. Wonorahardjo,
“Field study on adaptive thermal comfort in office buildings in
method of controlling the AC. Working the AC at maximum Malaysia, Indonesia, Singapore, and Japan during hot and humid
cooling for an extended time may affect the occupancy’s season,” Build. Environ., vol. 109, pp. 208–223, Nov. 2016.
health and welfare. Therefore, by applying the developed [19] W. Song, N. Feng, Y. Tian, and S. Fong, “An IoT-Based Smart
IoT-based smart AC controller with the selected enhanced Controlling System of Air Conditioner for High Energy
Efficiency,” 2017 IEEE Int. Conf. Internet Things IEEE Green
PMV-based model, the AC can create a comfortable Comput. Commun. IEEE Cyber, Phys. Soc. Comput. IEEE Smart
environment that can increase user’s satisfaction about the Data, pp. 442–449, 2017.
environment and this will have a great impact of the [20] S. Godo, J. Haase, and H. Nishi, “Air conditioning control using
workers’ productivity. The system could be improved by self-powered sensor considering comfort level and occupant
location,” IECON 2015 - 41st Annu. Conf. IEEE Ind. Electron.
integrating the artificial intelligent element and be tested in Soc., pp. 2497–2502, 2015.
a more variety of rooms and conditions, with respect to [21] “IoT Analytics - ThingSpeak Internet of Things.” [Online].
different type of AC. Available: https://thingspeak.com/. [Accessed: 13-May-2019].
[22] N. Jamaludin, N. I. Mohammed, M. F. Khamidi, and S. N. A.
Wahab, “Thermal Comfort of Residential Building in Malaysia at
REFERENCES Different Micro-climates,” Procedia - Soc. Behav. Sci., vol. 170,
pp. 613–623, Jan. 2015.
[1] K. C. Parsons, “Environmental ergonomics: A review of
principles, methods and models,” Appl. Ergon., vol. 31, no. 6, pp.
581–594, 2000.
[2] R. F. Rupp, N. G. Vásquez, and R. Lamberts, “A review of
human thermal comfort in the built environment,” Energy Build.,
vol. 105, pp. 178–205, Oct. 2015.
[3] E. Laftchiev and D. Nikovski, “An IoT system to estimate
personal thermal comfort,” in 2016 IEEE 3rd World Forum on
Internet of Things, WF-IoT 2016, 2017, pp. 672–677.
[4] A. S. Ahmad, S. A. A. Shukor, K. Kohlhof, Z. A. Z. Jamal, and
A. Mohamad, “Analyzing and Evaluating a PMV-Based Thermal
Comfort Model in Controlling Air Conditioning System,” Jurnal,
Teknol. (Sciences Eng., pp. 99–106, 2015.
[5] “World Bank Open Data | Data.” [Online]. Available:
https://data.worldbank.org/. [Accessed: 26-Feb-2019].
[6] “United States Environmental Protection Agency | US EPA.”
[Online]. Available: https://www.epa.gov/. [Accessed: 26-Feb-
2019].
[7] S. Merabti, B. Draoui, and F. Bounaama, “A review of control
systems for energy and comfort management in buildings,” in 8th
International Conference on Modelling, Identification and
Control, 2016, pp. 478–486.
[8] F. Nicol, M. Humphreys, and S. Roaf, Adaptive thermal comfort:
Principles and practice, 1st Editio. London: Routledge,
Abingdon, 2012.
[9] J. F. Nicol and M. A. Humphreys, “Adaptive thermal comfort and
sustainable thermal standards for buildings,” Energy Build., vol.
34, no. 6, pp. 563–572, 2002.
[10] R. F. Rupp, N. G. Vásquez, and R. Lamberts, “A review of
human thermal comfort in the built environment,” Energy Build.,
vol. 105, pp. 178–205, 2015.
[11] (ISO) International Organization for Standardization,
“Ergonomics of the thermal environment-analytical
determination and interpretation of thermal comfort using
calcualtion of the PMV and PPD indices and local thermal
comfort criteria (Standard No. ISO 7730:2205).,” Geneva, Switz.
ISO, 2005.
[12] ANSI/ASHRAE, “ANSI/ASHRAE 55:2004 Thermal
Environmental Conditions for Human Occupancy,” Ashrae, vol.
2004, p. 30, 2004.
[13] S. A. A. Shukor, K. Kohlhof, and Z. A. Z. Jamal, “Development
of a PMV-based thermal comfort modelling,” Proc. IASTED Int.
Conf. Model. Simul., pp. 670–675, 2007.
[14] MetMalaysia, “Daily Mean TT, Mean RH, MSPD.” 2016.
[15] H. A. Ozkan and A. Aybar, “A smart air conditioner in smart
home,” EEEIC 2016 - Int. Conf. Environ. Electr. Eng., 2016.
[16] R. Saidur, “Energy consumption, energy savings, and emission
analysis in Malaysian office buildings,” Energy Policy, vol. 37,
no. 10, pp. 4104–4113, Oct. 2009.

978-1-7281-0784-4/19/$31.00 ©2019 IEEE 294

Authorized licensed use limited to: KJ's Educational Institute. Downloaded on October 17,2024 at 09:25:15 UTC from IEEE Xplore. Restrictions apply.