0% found this document useful (0 votes)
125 views15 pages

Review of HVAC Systems History and Future Applicat

This concept paper provides a brief history of HVAC systems and discusses their future applications. It notes that HVAC systems have become an integral part of modern buildings and are designed to provide comfortable indoor environments while conserving energy. Advancements in computer control systems and smart technology have made HVAC systems more sophisticated, allowing them to more precisely control indoor environments using various sensors. The paper examines both the historical development of HVAC systems as well as their increasing integration of emerging technologies.

Uploaded by

arie crespo
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
125 views15 pages

Review of HVAC Systems History and Future Applicat

This concept paper provides a brief history of HVAC systems and discusses their future applications. It notes that HVAC systems have become an integral part of modern buildings and are designed to provide comfortable indoor environments while conserving energy. Advancements in computer control systems and smart technology have made HVAC systems more sophisticated, allowing them to more precisely control indoor environments using various sensors. The paper examines both the historical development of HVAC systems as well as their increasing integration of emerging technologies.

Uploaded by

arie crespo
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 15

energies

Concept Paper
Concept Paper

Review of HVAC
HVAC Systems
Systems History
History and
and Future
Future Applications
Applications
DeQuante 1, *, Raymond Charles Tesiero 2,, *, Yaa Takyiwaa Acquaah 1,1, * and Balakrishna Gokaraju11
DeQuante Rashon
Rashon Mckoy
Mckoy ¹,*, Raymond Charles Tesiero ² *, Yaa Takyiwaa Acquaah * and Balakrishna Gokaraju

11 Department
Departmentof ofComputational
ComputationalScience
ScienceandandEngineering,
Engineering,North
North Carolina
CarolinaAgricultural
Agricultural and
and Technical State
Technical State
University, University,
Greensboro, NCGreensboro,
27411, USA;NC 27411, USA; [email protected]
[email protected]
22 Department
Department ofof Civil
Civil Architect
Architect & Environmental Engineering,
Engineering,North
NorthCarolina
CarolinaAgricultural
Agriculturaland
andTechnical State
Technical State
University, University,
Greensboro, NCGreensboro,
27411, USANC 27411, USA
** Correspondence:
Correspondence:[email protected]
[email protected](D.R.M.);
(D.R.M.);[email protected]
[email protected](R.C.T.);
(R.C.T.);
[email protected]
[email protected](Y.T.A.);
(Y.T.A.); Tel.:
Tel.: +1-336-285-3210
+1-336-285-3210(D.R.M.
(D.R.M.&&Y.T.A.)
Y.T.A.)

Abstract: Today,
Abstract: Today,HVAC
HVAC(heating,
(heating,ventilation,
ventilation,andandair
air conditioning)
conditioning)systems
systemshave
have become
become an an integral
integral
part of
part of modern
modern buildings
buildings andand are
are designed
designed to to provide
provide comfortable
comfortable indoor
indoor environments
environments whilewhile con-
con-
serving energy and reducing carbon emissions. With advancement in technology,
serving energy and reducing carbon emissions. With advancement in technology, HVAC systems HVAC systems
have aavariety
have varietyofofsensors
sensorsthatthatare
are used
used to to detect
detect thethe occupants
occupants within
within a controlled
a controlled environment.
environment. Ad-
Advancements
vancements in computer
in computer control
control systemssystems
and the and
usethe
of use
smartof technology
smart technology have
have made madesystems
HVAC HVAC
systems
even even
more more sophisticated,
sophisticated, allowingallowing for approximate
for approximate temperaturetemperature control
control and andmanagement.
energy energy man-
This paper will review the historical development of technology and the current state of state
agement. This paper will review the historical development of technology and the current HVAC of
HVAC systems. With the proper data, development of artificial intelligence models
systems. With the proper data, development of artificial intelligence models can, in theory, improve can, in theory,
improve
the overallthe overall optimization
optimization and reduce and reduce
energy energy consumption
consumption This paperThiswillpaper willa provide
provide review ofa review
HVAC
of HVAC
history andhistory
the keyand the key
concepts concepts
around around theofusefulness
the usefulness using AI fromof using AI from
previous previous
research research
conducted in
conducted in this
this field of study. field of study.

Keywords: HVAC; heating;


Keywords: heating; cooling;
cooling; deep
deep learning;
learning; artificial
artificial intelligence
intelligence

1. Introduction
1.
HVAC systems
systems date
dateback
backtoto ancient
ancient civilizations,
civilizations, where
where systems
systems for heating
for heating and
and cool-
cooling
ing werewere developed
developed forfor comfort
comfort andand health.During
health. Duringthetheancient
ancientRoman
Roman period, central
central
Citation: Mckoy,
Citation: Mckoy,D.R.;
D.R.;Tesiero,
Tesiero,R.C.;
R.C.;
heating
heating systems
systems were
were used
used in
in homes
homes and and public
public baths
baths and
and wind
wind towers
towers were
were used
used to
to cool
cool
Acquaah,Y.T.;
Acquaah, Y.T.;Gokaraju,
Gokaraju,B.B.Review of
homes.
homes. In the 19th century, during
during thethe industrial
industrial revolution,
revolution, the need
need forfor heating
heating and
and
ReviewSystems
HVAC of the History
Historyand
andFuture
Future
Applications. Energies
Applications of HVAC2023, 16, 6109.
Systems.
cooling
coolingsystems
systemsininfactories
factoriesand
andlarge
largebuildings
buildingsincreased,
increased,which
whichledled
to to
thethe
development
development of
https://doi.org/10.3390/en16176109
Energies 2023, 16, x.
the early
of the HVAC
early HVACsystems shown
systems shownin Figure 1. In1.
in Figure the Inearly days,days,
the early coal-fired boilers
coal-fired and large
boilers and
https://doi.org/10.3390/xxxxx
fans
largewere
fansused
weretoused
ventilate these early
to ventilate thesesystems. One of the
early systems. Oneearliest
of the innovations in heating
earliest innovations in
Academic Editors: Wei-Hsin Chen systems was the introduction of the steam radiator, which was invented in the 1820s by a
heating systems was the introduction of the steam radiator, which was invented in the
and Jae-Weon
Academic JeongWei-Hsin Chen
Editors: British engineer named S. A. [1].
1820s by a British engineer named S. A. [1].
and Jae-Weon Jeong
Received: 5 July 2023
Received:165 August
Revised: July 2023
2023
Revised: 16
Accepted: 17August
August2023
2023
Accepted: 17
Published: 22 August
August 2023
2023
Published: date

Copyright: © 2023 by the authors.


Copyright: © 2023 by the authors.
Licensee MDPI, Basel, Switzerland.
Submitted for possible open access
This article is an open access article
publication under the terms and
distributed under the terms and
conditions of the Creative Commons
conditions of the Creative Commons
Attribution (CC BY) license
Attribution (CC BY) license (https://
(https://creativecommons.org/license
creativecommons.org/licenses/by/ Figure 1. Steam HVAC System, Ref. [2].
s/by/4.0/). Figure 1. Steam HVAC System, Ref. [2].
4.0/).

Energies2023,
Energies 2023,16,
16,6109.
x. https://doi.org/10.3390/xxxxx
https://doi.org/10.3390/en16176109 www.mdpi.com/journal/energies
https://www.mdpi.com/journal/energies
Energies 2023, 16, x FOR PEER REVIEW 2 of 1

Heating steam radiators used a central boiler to heat water and distribute steam
Energies 2023, 16, 6109 throughout a building via a network of pipes and radiators. This system allowed 2 of 15 for
more even distribution of heat compared with the fireplaces and stoves that were previ
ously used. Ventilation systems in the 19th century were typically designed to provid
fresh air to steam
Heating the building
radiators and to aremove
used central stale
boilerair [3]. Openable
to heat water and windows distribute and
steamchimney
throughout
were the most a building
common via a network
methodsofof pipes and radiators.
ventilation at theThis
time,system allowed
although for abuildings
some more als
even distribution of heat compared with the fireplaces and stoves
used mechanical ventilation systems that used fans to circulate air. The forced-air furnac that were previously
used. Ventilation
was another systems in
important the 19th of
invention century
the 19thwerecentury.
typicallyThisdesigned
system to provide fresh by Al
was invented
air to the building and to remove stale air [3]. Openable windows and chimneys were
phonse Beau de Rochas in France and circulates hot air throughout a building with th
the most common methods of ventilation at the time, although some buildings also used
help of a blower.
mechanical ventilation Compared
systems that with gravity
used fans toheating
circulatesystems, which relied
air. The forced-air on was
furnace natural con
vectionimportant
another to circulate air, this
invention of was a significant
the 19th century. This improvement. In the mid-19th
system was invented by Alphonse century, th
use of air conditioning systems began to emerge. The
Beau de Rochas in France and circulates hot air throughout a building with the help offirst air conditioning system
a wa
patented
blower. in 1851 (shown
Compared above
with gravity in Figure
heating 1) bywhich
systems, John Gorrie,
relied on a natural
physician and inventor
convection to from
circulate
Florida.air, Histhis was aused
system significant improvement.
ice to cool air and was In primarily
the mid-19th usedcentury, the useto
in hospitals ofhelp
air reduc
conditioning
the spread of systems begandiseases
infectious to emerge.[3].The first airthe
Overall, conditioning
19th century systemwaswasa timepatented
of greatin innova
1851 (shown above in Figure 1) by John Gorrie, a physician and inventor
tion in the field of HVAC systems. The invention of the steam radiator, forced-air furnace from Florida. His
system used ice to cool air and was primarily used in hospitals to help reduce the spread
and air conditioning system laid the foundations for the modern HVAC systems used to
of infectious diseases [3]. Overall, the 19th century was a time of great innovation in the
day.of HVAC systems. The invention of the steam radiator, forced-air furnace, and air
field
In the 20th
conditioning system century,
laid the the use of electricity
foundations for the modernand the HVACdevelopment
systems used oftoday.
modern technolo
giesInled theto20th
significant
century, theadvancements
use of electricityin HVAC
and the systems.
development Theofinvention of the first modern
modern technologies
air to
led conditioner
significant by Willis Carrier
advancements in 1902
in HVAC revolutionized
systems. The invention the HVAC industry
of the first modern (Figure 2)
air conditioner by Willis Carrier in 1902 revolutionized the HVAC
making it possible to cool large buildings. His invention removed the moisture from th industry (Figure 2),
making
air, he it possible
called to cool large for
it “Apparatus buildings.
Treating HisAir”,
invention removed this
and through the moisture
invention from
he the
was able t
air, he called it “Apparatus for Treating Air”, and through this invention
notice that it also cooled the air at the same time. From the beginning, Carrier realized he was able to tha
notice that it also cooled the air at the same time. From the beginning, Carrier realized
his invention had a wider range of applications than just dehumidifying industrial build
that his invention had a wider range of applications than just dehumidifying industrial
ings. At first, Carrier did not refer to his invention as an air conditioner. The term “ai
buildings. At first, Carrier did not refer to his invention as an air conditioner. The term
conditioner”
“air conditioner” camecame from
fromStuart
StuartCramer,
Cramer, he he invented
invented aamachine
machinethat thatadded
added humidity to th
humidity
airthe
to toairmake yarnyarn
to make easier to work
easier to workin in
the themanufacturing
manufacturing plants plants [3].[3]. In
In1904,
1904,atatthetheSt.St. Loui
World
Louis Fair,Fair,
World thethe public
publicwas wasfirst
firstintroduced
introduced to tothe
thewonders
wonders of conditioning.
of air air conditioning. The The ai
air conditioningunit
conditioning unitwaswas able
able to circulate
circulateair airtotoover
over 35,000 cubic
35,000 feetfeet
cubic of air
ofper
air minute
per minutein in th
the Missouri
Missouri StateBuilding
State Buildingthatthat hosted
hosted thetheSt.St.Louis
Louis World
World FairFair
[4]. [4].

Figure2.2.Electric
Figure Electric HVAC
HVAC System,
System, Ref. Ref.
[4]. [4].
Energies 2023, 16, 6109 3 of 15

The air conditioning machine cooled all rooms on two floors except for the libraries and
bathrooms. At that time, nobody could have imagined anything such as this was possible.
This system was installed by the United Iron Works Co. of Springfield (Springfield, MO,
USA), Missouri. Unfortunately, the boiler for the system exploded in November of that year
and set the building on fire. In 1906, Carrier patented his invention, the “air conditioner”,
and then patented the control system for the air conditioner in 1914. After conducting
more research and critiques of his first version of the air conditioning system, Carrier was
able to downsize it to make it more manageable. In 1920, the first version was able to be
downsized and by 1925 it was able to be installed at the Rivoli Theater in New York City. As
technology improved, air conditioning units became smaller. A window air conditioning
unit was invented in 1931 by H.H. [5]. Today’s air conditioners are based on this technology.
In the years following World War II, air conditioning development boomed and small
companies became household names. In 1956, Carrier won a contract to provide central
air conditioning in 700 upscale Pennsylvania homes. By the 1970s, central air conditioning
and ventilation had become mainstream. In response to the demand for cool indoor air,
Paramount installed air conditioners in another 300 movie theaters. Additionally, in the
late 1920s and early 1930s, Frigidaire, General Electric, and General Motors joined the A/C
game. The development of refrigerants and better insulation materials also improved the
efficiency of air conditioning systems. HVAC systems play a crucial role in maintaining a
comfortable indoor environment in residential, commercial, and industrial buildings. These
systems control the temperature, humidity, and air quality of indoor spaces. However, the
proper functioning of HVAC systems depends on several factors, such as proper installation,
maintenance, and repair. The early detection of malfunctions or inefficiencies in HVAC
systems can help prevent significant energy waste and ensure occupant comfort. The
design of HVAC systems in the 19th century was limited by the technology available at
the time [5]. Nevertheless, these systems represented a significant improvement over the
earlier heating and ventilation systems, which relied on fireplaces and stoves for heating
and natural ventilation methods for ventilation.

2. Related Topics
There have been many studies using machine learning techniques to enhance the
efficiency and performance of HVAC systems. One common area of research focuses
on predictive maintenance for HVAC equipment. By leveraging historical sensor data
and machine learning algorithms, researchers have developed models that can predict
equipment failures and malfunctions before they occur. These predictive maintenance
models allow HVAC technicians to perform timely maintenance, reducing downtime and
preventing costly breakdowns and leading to improved system reliability and energy
savings. From some of the recently published research conducted on this topic, different
techniques have been used to help build these deep learning models. In a study conducted
by Yang, Shiyu, et al., machine learning techniques were used by taking the collected online
building data to help improve the system performance. The systems implemented in their
research uses multi-objective functions to help improve indoor thermal comfort levels [6].
In a similar study, Du, Yan, et al. used reinforced learning techniques to build optimal
controls for multi-zoned control residential HVAC systems for user comfort levels. The
simulation results showed that energy consumption was reduced by 15% and comfort level
violation reduced by 79%, and when compared with a rule-based HVAC control strategy,
the comfort violation can be reduced by 98% [6].
Some more related studies focus on using similar techniques. Later, I will discuss
building deep learning models implemented with Pi cameras and thermal infrared cameras
for the detection of human occupancy. In both publications, the papers reference using dif-
ferent techniques to improve AI-based techniques to improve heating and air-conditioning
comfort. Beltran, Alex, Varick L. Erickson, and Alberto E. Cerpa’s study used software
called ThermoSensor to detect the occupancy of users [7]. The system regulates ventilation
using conditioning regimens trained from occupant usage patterns and near real-time
Energies 2023, 16, 6109 4 of 15

estimates of occupancy and temperature. ThermoSense uses a cutting-edge multisensor


node that combines a passive infrared sensor with a low-cost, low-power thermal sensor
array. We demonstrate that our system is capable of measuring occupancy with an RMSE
of about 0.35 persons by utilizing a unique processing pipeline and sensor fusion. We
demonstrate that we can conserve 25% of annual energy while retaining the efficiency
of room temperature by conditioning areas based on occupancy. One case study used
techniques to analyze the response of comfort levels in office spaces [8]. Similar to data
collection, other methods were used to help develop deep learning models to improve the
comfort levels of the occupants [9,10]. For example, in a study conducted by Tien, Paige
Wenbin, et al., CO2 sensors were used to help accurately detect occupants for indoor air
quality. Because the model for counting individuals was simpler, it was able to detect
intersections over unions with an average accuracy of about 98.9%, which was higher than
the accuracy of the activity detection model, which was approximately 88.5%. According
to the real-time data regarding the population and their activities, count-based occupancy
profiles were created during the detection.[7] Through the use of building energy simula-
tion, scenario-based modeling of the case study building under four different ventilation
scenarios was performed to determine the impact of this strategy on indoor air quality and
energy consumption. The results showed that the suggested method may give data on
dynamic changes in occupancy for demand-driven ventilation controls to enhance indoor
air quality and address the issue of under- or over-estimation of the ventilation demand
when employing the static or fixed profiles [11]. In all the relevant research conducted
around this growing area, there has yet to be a systematic system built from the ground
up showing the proposed research. In this paper, I will discuss the development of my
prototype using related techniques used in previous research.
The proposed prototype HVAC system aims to use Raspberry Pi cameras and IR
cameras to control butterfly valves, enabling an intelligent and responsive HVAC control
system. The system leverages the capabilities of the Pi cameras to capture visual data, such
as occupancy patterns and thermal profiles, whereas the IR cameras provide information
about temperature variations within the space [12]. Using this data, the system employs
machine learning algorithms to analyze and make real-time decisions on valve control. The
butterfly valves function as actuators, regulating the flow of air or water within the HVAC
system based on the data received from the cameras. The valves can be automatically ad-
justed to open or close, optimizing the airflow or water flow distribution, thus maintaining
an ideal indoor environment.

3. Energy Usage in Homes


Figure 3 shows that, according to the Department of Energy (DOE), approximately
48% of a home’s total energy consumption comes from HVAC systems. In addition to the
size of the home, efficiency of the HVAC system, and outdoor temperature, the energy
consumption of HVAC systems can vary widely. Heating and air conditioning systems
can consume as much as 60% of a home’s energy in warm climates. In colder climates,
where heating is the primary use, HVAC systems can account for up to 30% of home energy
consumption [13].
Energies 2023, 16, x FOR PEER REVIEW 5 of 15
Energies 2023, 16, 6109 5 of 15

Figure 3. Production of Energy and Consumption [14].


Figure 3. Production of Energy and Consumption [14].
According to a survey conducted by the National Association of Home Builders, in
According
the United States,to90%a survey
of newconducted
homes areby the National
equipped Association
with central of Home Builders,
air conditioners. According in
thethe
to United
survey,States,
new 90%homes of are
newalso
homes
larger,areincreasing
equippedthe withneedcentral air conditioners.
for HVAC Accord-
systems. Moreover,
ing increasing
the to the survey,
demand newforhomes
larger are
homesalsoandlarger, increasing
the use the need HVAC
of more advanced for HVAC systemssystems.
with
Moreover, the increasing demand for larger homes and the use
advanced features such as zoning and smart thermostats have resulted in higher energy of more advanced HVAC
systems
usage with
[15]. advancedtofeatures
According such as zoning
the International EnergyandAgency,
smart thermostats
the energy have resulted of
consumption in
higher energy usage [15]. According to the International Energy
HVAC systems in buildings is expected to increase by 30% by 2040, highlighting the need Agency, the energy con-
sumption
for of HVACHVAC
energy-efficient systems in buildings
systems is expected
and practices to increase
to mitigate by 30%
the energy by 2040, high-
demand.
lighting the need
A house’s for energy-efficient
HVAC system ensuresHVAC systems
the comfort and
and practices
quality to inside
of air mitigate thethe energy
home by
demand. heat and air. However, HVAC systems consume a significant amount of energy,
distributing
A house’s
accounting HVAC half
for nearly system ensures
of the energy theusage
comfortin aand quality
typical of air inside
American home. theThere
homeare by
distributing heat and air. However, HVAC systems consume a significant
several factors that affect the energy usage of HVAC systems, such as the size of the home, amount of en-
ergy,
the accounting
efficiency of theforsystem,
nearly the
halfoutdoor
of the energy usage and
temperature, in athetypical
indoor American
temperaturehome. There
settings.
are several factors
Inefficiency of HVAC that affect the
systems canenergy
result in usage
higherof energy
HVAC billssystems, such asemissions.
and carbon the size ofThe the
home, theof
reduction efficiency of the system,and
energy consumption theimprovement
outdoor temperature,
of indoorand the indoor
air quality can temperature
be achieved
settings.
by Inefficiency
installing of HVAC
high-efficiency HVACsystems
systemscanandresult in higher
ensuring energy
proper bills and carbon
maintenance [15]. The emis-
use
of programmable
sions. The reduction thermostats
of energyinconsumption
residential homes can reduce theofenergy
and improvement indoorusage of HVAC
air quality can
systems. When the occupants are sleeping or the building is
be achieved by installing high-efficiency HVAC systems and ensuring proper mainte- unoccupied, homeowners can
use
nanceprogrammable
[15]. The usethermostats
of programmableto adjust temperature
thermostats settings according
in residential homes can to their
reduceoccupancy
the en-
patterns,
ergy usage thereby
of HVAC reducing
systems.energy
When usage. Additionally,
the occupants regular HVAC
are sleeping maintenance,
or the building such
is unoccu-
as cleaning
pied, or replacing
homeowners can useair programmable
filters and checking ductwork,
thermostats can improve
to adjust temperaturethe efficiency
settings ac-of
the HVAC
cording system,
to their reducing
occupancy energy thereby
patterns, consumption reducingandenergy
improvingusage.indoor air quality
Additionally, [16].
regular
Proper
HVAC insulation
maintenance, and such
weatherization
as cleaningcan or also prevent
replacing airair leakage
filters andand reduceductwork,
checking the workload can
on the HVAC system, further reducing energy consumption.
improve the efficiency of the HVAC system, reducing energy consumption and improving Overall, reducing the energy
usage
indoorofairHVAC
quality systems in residential
[16]. Proper insulation homes requires a combination
and weatherization of energy-efficient
can also prevent air leakage
HVAC
and reducesystems,
the regular
workload maintenance,
on the HVAC andsystem,
energy-saving
furtherpractices
reducingby homeowners.
energy consumption.
Overall, reducing the energy usage of HVAC systems in residential homes requires a com-
4. IR Camera
bination of energy-efficient HVAC systems, regular maintenance, and energy-saving prac-
tices The history of infrared cameras dates to the mid-20th century. The first infrared
by homeowners.
cameras were developed in the late 1920s and early 1950s, following the development of
Energies 2023, 16, x FOR PEER REVIEW 6 of 15

Energies 2023, 16, 6109 4. IR Camera 6 of 15

The history of infrared cameras dates to the mid-20th century. The first infrared cam-
eras were developed in the late 1920s and early 1950s, following the development of mod-
modern technologies
ern technologies that that
mademade it possible
it possible to detect
to detect andand measure
measure infrared
infrared radiation.
radiation. The The
ear-
earliest infrared
liest infrared cameras
cameras useda aspecial
used specialtype
typeofoffilm
filmthat
thatwas
wassensitive
sensitive to
to infrared
infrared radiation.
Shown
Shownin inthe
theimage
imageabove
aboveFigure
Figure4,4,this
thisspecial
specialtype
typeof
offilm
filmwas
wasplaced
placedin infront
frontof
ofthe
thelens
lens
of
of the
the IR
IRcamera,
camera,which
whichfocused
focusedthetheinfrared
infraredradiation
radiationonto
ontothe
thefilm.
film. The
The film
film was
was then
then
developed
developed in in aa darkroom,
darkroom, producing
producing an an image
image that
that showed
showed the
the distribution
distribution of of infrared
infrared
radiation in the scene.
radiation in the scene.

Figure4.4. IR
Figure IRCamera
CameraVision,
Vision,Ref.
Ref.[17].
[17].

The first
The firstthermographic
thermographic camera,
camera, aa groundbreaking
groundbreaking inventioninvention in in the
the field
field ofofinfrared
infrared
technology,emerged
technology, emergedin inthe
theearly
early1920s.
1920s.Developed
Developed by bythetheHungarian
HungarianphysicistphysicistKálmán
Kálmán
Tihanyi,the
Tihanyi, thedevice
device revolutionized
revolutionized howhow we perceive
we perceive and measure
and measure heat. Theheat. The thermo-
thermographic
graphicharnessed
camera camera harnessed the principles
the principles of infrared of infrared
radiation, radiation,
capturing capturing and converting
and converting the in-
the invisible
visible thermal thermal
energyenergy
emitted emitted
by objectsby objects into visible
into visible images.images. This pioneering
This pioneering technologytech-
nology enabled
enabled scientistsscientists and engineers
and engineers to peerthe
to peer beyond beyond the limitations
limitations of humanofvision, human vision,
opening
opening
new new for
avenues avenues for research
research in various inindustries
various industries
[17]. The[17]. The thermographic
thermographic camera’scamera’s
signifi-
significance
cance quicklyquickly
extended extended
beyondbeyond the scientific
the scientific community.
community. It found It practical
found practical appli-
applications
in military
cations reconnaissance
in military during during
reconnaissance World World
War II,Warenabling soldiers
II, enabling to detect
soldiers hidden
to detect ene-
hidden
mies, vehicles,
enemies, and and
vehicles, eveneven
identify potential
identify hotspots
potential on battlefields.
hotspots on battlefields. In subsequent
In subsequent decades,
dec-
the technology evolved further, becoming more compact and
ades, the technology evolved further, becoming more compact and affordable, which led affordable, which led to its
integration into various sectors, including building inspections, firefighting,
to its integration into various sectors, including building inspections, firefighting, indus- industrial
maintenance,
trial maintenance, and medical diagnosis.
and medical diagnosis.Since its inception,
Since its inception, the the
firstfirst
thermographic
thermographic camera
cam-
has
era paved
has paved the the
wayway for for
a new
a neweraeraof of
innovation,
innovation,continuously
continuouslypushing pushingthe theboundaries
boundaries
of
of what
what we we can
cansee seeand
andunderstand
understandabout aboutthetheworld
worldaround
aroundus us[17].
[17]. Its
Itsimpact
impact remains
remains
evident
evident in modern-day thermal imaging devices, which continue to play a vital role
in modern-day thermal imaging devices, which continue to play a vital role in
in
countless industries, contributing to safety, efficiency, and problem
countless industries, contributing to safety, efficiency, and problem solving and now solving and now using
us-
machine
ing machine learning techniques
learning in our
techniques in ever-evolving
our ever-evolving technological
technological landscape.
landscape.
In
In the 1960s, the first solid-state infrared cameras were developed, using
the 1960s, the first solid-state infrared cameras were developed, using recent
recent tech-
tech-
nologies
nologies suchsuch as ascharge-coupled
charge-coupled devices
devices (CCDs)
(CCDs) and anduncooled
uncooledthermalthermaldetectors.
detectors. In In the
the
1980s and 1990s, the development of digital imaging technologies
1980s and 1990s, the development of digital imaging technologies led to the widespread led to the widespread
use
useofofdigital
digital infrared
infraredcameras.
cameras. These
These cameras
cameras captured
captured infrared
infrared radiation
radiation as as digital
digital data,
data,
which
which could be easily processed and displayed on a computer screen. Since thethe
could be easily processed and displayed on a computer screen. Since turn turn of
of the
the 21st century, advances in digital imaging technologies have continued to drive the
development of infrared cameras. Today, there are a wide variety of infrared cameras
available, ranging from handheld devices used for industrial and military applications to
specialized cameras used in scientific research.
Energies 2023, 16, x FOR PEER REVIEW 7 of 15

21st century, advances in digital imaging technologies have continued to drive the devel-
Energies 2023, 16, 6109
opment of infrared cameras. Today, there are a wide variety of infrared cameras available, 7 of 15
ranging from handheld devices used for industrial and military applications to special-
ized cameras used in scientific research.
In 2012, the Raspberry Pi foundation launched the Raspberry tool to help educate
In 2012, the Raspberry Pi foundation launched the Raspberry tool to help educate
students and programming professionals. The Raspberry Pi was created with the goal of
students and programming professionals. The Raspberry Pi was created with the goal of
education in mind. This ultra-tiny computer was designed to be small and cheap so that
education in mind. This ultra-tiny computer was designed to be small and cheap so that
schools could easily afford to teach students about microprocessors in the classroom [18].
schools could easily afford to teach students about microprocessors in the classroom [18].
In the image above Figure 5, you can see each of the ports for a Raspberry Pi. A powerful
In the image above Figure 5, you can see each of the ports for a Raspberry Pi. A powerful
feature of the Raspberry Pi is the row of GPIO (general-purpose input/output) pins along
feature of the Raspberry Pi is the row of GPIO (general-purpose input/output) pins along
the top edge of the board. The Raspberry Pi is great for two reasons, the first is that it
the top edge of the board. The Raspberry Pi is great for two reasons, the first is that it
provides extremely cheap access to a computer, the second is that it is a great tool for
provides extremely cheap access to a computer, the second is that it is a great tool for
learning more about computers. In today’s modern environment, detecting occupancy in
learning more about computers. In today’s modern environment, detecting occupancy in
automation systems for several reasons can be a critical task. Artificial intelligence has a
automation systems for several reasons can be a critical task. Artificial intelligence has a
wide range of uses; machine learning is one of them. It empowers any task to learn from
wide range of uses; machine learning is one of them. It empowers any task to learn from
and
andimprove
improveon onprevious
previousexperiences
experiencesand andoutcomes
outcomeswithout
withoutthetherequirement
requirementfor forexplicit
explicit
programming.
programming. It focuses on the creation of a variety of other programs that canacquire
It focuses on the creation of a variety of other programs that can acquire
data
dataand
anduse
useitittototrain
trainand
andlearn
learnon ontheir
theirown
ownthrough
throughiterative
iterativeprocedures
proceduresbefore
beforetesting
testing
themselves
themselves[18].
[18].AtAtthis
thistime,
time,occupancy
occupancydetection
detectionbased
basedonona amachine
machinelearning
learningalgorithm
algorithm
isisquite
quitebeneficial
beneficialfor foraavariety
varietyofofapplications.
applications.The Thehistory
historyofofinfrared
infraredcameras
camerashas hasbeen
been
marked
marked by the steady progression of technological advancements that have madethese
by the steady progression of technological advancements that have made these
cameras
camerasmore
moreversatile,
versatile,easier
easiertotouse,
use,and
andmore
morewidely
widelyavailable.
available.Today,
Today,infrared
infraredcameras
cameras
are
are a useful tool used in a wide range of fields, including industrial inspection,scientific
a useful tool used in a wide range of fields, including industrial inspection, scientific
research,
research,and
andmilitary
militaryand andsecurity
securityapplications.
applications.

RaspberryPiPi4 4Model
Figure5.5.Raspberry
Figure ModelB,B,Ref.
Ref.[18].
[18].

A single person did not discover Arduino, rather it was created by a team of researchers
A single person did not discover Arduino, rather it was created by a team of research-
and students at the Interaction Design Institute Ivrea (IDII) in Ivrea, Italy in 2003. The team
ers and students at the Interaction Design Institute Ivrea (IDII) in Ivrea, Italy in 2003. The
was led by Massimo Banzi, a professor at the institute, and included David Cuartielles,
team was led by Massimo Banzi, a professor at the institute, and included David
Tom Igoe, Gianluca Martino, and David Mellis. Their goal was to create a low-cost and
Cuartielles, Tom Igoe, Gianluca Martino, and David Mellis. Their goal was to create a low-
easy-to-use microcontroller board that could be used by artists, designers, hobbyists, and
cost and easy-to-use microcontroller board that could be used by artists, designers, hob-
anyone interested in creating interactive projects [19]. The first prototype of the Arduino
byists, and anyone interested in creating interactive projects [19]. The first prototype of
board was based on the wiring platform that was developed by Hernando Barragán. Since
the Arduino board was based on the wiring platform that was developed by Hernando
then, the Arduino project has grown into a global community of makers, artists, designers,
Barragán. Since then, the Arduino project has grown into a global community of makers,
educators, and hobbyists who use the platform to create a wide range of projects, from
artists, designers, educators, and hobbyists who use the platform to create a wide range
interactive art installations to smart home devices.
of projects, fromUno
Arduino interactive art installations
is a popular to smart
microcontroller home
board thatdevices.
is widely used in various elec-
tronics projects. The board is based on the ATmega328P microcontroller chip and has a
variety of input/output pins, analog inputs, and other useful features. With the help of
Arduino IDE 2.1.1 software, users can write code to control various electronic components
connected to the board, such as motors, sensors, and displays [19]. This makes the Arduino
Uno ideal for prototyping and experimenting with in various electronics projects. One of
Arduino Uno is a popular microcontroller board that is widely used in various elec-
tronics projects. The board is based on the ATmega328P microcontroller chip and has a
variety of input/output pins, analog inputs, and other useful features. With the help of
Arduino IDE 2.1.1 software, users can write code to control various electronic components
connected to the board, such as motors, sensors, and displays [19]. This makes the Ar-
Energies 2023, 16, 6109 8 of 15
duino Uno ideal for prototyping and experimenting with in various electronics projects.
One of the key features of the Arduino Uno is its versatility. The Arduino Uno board can
be used to create a wide range of projects, from simple circuits to complex robotics and
the key features
automation systems. ofFigure
the Arduino
6 showsUno is its
that versatility.
Arduino The fourteen
uno has Arduino digital
Uno board can be used
input/output
to create a wide range of projects, from simple circuits to complex robotics
pins, six analog inputs, and a 16 MHz quartz crystal; the board can easily communicate and automation
systems. Figure 6 shows that Arduino uno has fourteen digital input/output
with and control other electronic components [20]. Additionally, the board’s USB connec- pins, six
tion and power jack make it easy to power and program. Another advantage of the Ar-and
analog inputs, and a 16 MHz quartz crystal; the board can easily communicate with
control
duino Uno other electronic
is its ease of use.components
The board is [20]. Additionally,
designed the board’s
to be user-friendly, USB
and evenconnection
beginnersand
power jack make it easy to power and program. Another advantage
with no experience in electronics or programming can start using it. The Arduino IDE of the Arduino Uno
is its ease
software comes of use.
with The boardofisbuilt-in
a variety designed to be user-friendly,
functions and librariesandthateven beginners
simplify with no
the process
experience in electronics or programming can start using it. The Arduino
of programming the board. This makes the Arduino Uno a great learning tool for students IDE software
andcomes withwho
hobbyists a variety
want ofto built-in
explore functions
the worldand libraries that
of electronics andsimplify the process
programming. of pro-
Overall,
the Arduino Uno is a versatile and user-friendly microcontroller board that offers a wideand
gramming the board. This makes the Arduino Uno a great learning tool for students
hobbyists who want to explore the world of electronics and programming. Overall, the
range of possibilities for electronics enthusiasts and professionals alike.
Arduino Uno is a versatile and user-friendly microcontroller board that offers a wide range
of possibilities for electronics enthusiasts and professionals alike.

Figure 6. Arduino Uno, Ref. [20].


Figure 6. Arduino Uno, Ref. [20].
5. HVAC Capabilities
5. HVAC Capabilities
Infrared cameras can be used to diagnose various HVAC problems and occupancy and
improve wasted
Infrared camerasenergy costused
can be and to
climate control
diagnose for inactive
various HVACspaces. By identifying
problems temper-
and occupancy
andature variations
improve wastedof energy
occupied or and
cost unoccupied spaces, infrared
climate control cameras
for inactive canBy
spaces. help technicians
identifying
improve the
temperature overall efficiency
variations of occupiedof or
HVAC systems,
unoccupied as shown
spaces, in Figure
infrared 7 [21].
cameras can help tech-
By detecting
nicians improve temperature
the overall variations
efficiency on equipment
of HVAC systems, assurfaces,
shown ininfrared
Figurecameras
7 [21]. can help
technicians identify problems that may be causing the equipment to operate inefficiently.
There are many use cases for human occupancy recognition. In energy consumption
reduction, knowing if a space is occupied or not can be used to control ventilation and
temperature systems. As the demand increases for comfort levels in all spaces, being able
to provide a new development in technology around comfort levels for hospitals, office
spaces, and museums with priceless artifacts is important. These types of commercial spaces
require a more complex HVAC system to maintain the most approximate temperature
to the required specification needed [21]. Human occupancy recognition data can be
used to optimize room-booking systems. It can also be used for monitoring purposes, for
example in medical facilities and in security systems; a system can be established to detect
the number of occupants who entered and left the facility, as shown in Figure 7. HVAC
units, which are a key target of building automation, account for about half of the energy
consumed in both residential and commercial buildings. Building automation systems,
Energies 2023, 16, 6109 9 of 15

Energies 2023, 16, x FOR PEER REVIEW 9 of 15


in general, are designed to intelligently operate building facilities in response to dynamic
external conditions while maintaining acceptable energy usage and comfort levels [22].

Figure 7. Diagram of user-friendly HVAC system [21].


Figure 7. Diagram of user-friendly HVAC system [21].
Deep Learning image recognition is a subfield of computer vision and artificial intelli-
gence Bythat
detecting
deals with temperature variations
training artificial neuralon networks
equipment surfaces, objects
to recognize infraredand cameras
patterns canin
help technicians identify problems that may be causing the equipment
images. This is achieved by using large datasets of labeled images to train the network to to operate ineffi-
ciently.
recognize There are many
features, useand
objects, cases for human
patterns in new occupancy
images. The recognition. In energy
training process consump-
involves using
tion reduction, knowing
an optimization algorithmifto a adjust
space is theoccupied
weights of or the
notneurons
can be usedin thetonetwork
controlto ventilation
minimize
and temperature
the difference systems.
between the As the demand
predicted increases
and actual outputforfor
comfort
a given levels
image. in all
Oncespaces, being
the images
able to provide a new development in technology around
are trained, deep learning image recognition models can be used for a variety of taskscomfort levels for hospitals,
office
such as spaces,
objectand museums
detection, imagewithclassification,
priceless artifacts is important. [22].
and segmentation TheseFor types of commer-
example, in the
cial
imagespaces
shown require a more
in Figure 8 ancomplex
infrared HVAC
imagesystem to maintain
was captured using theobject
most detection
approximate tem-
methods
perature
to capture to heat
the required specification
signatures. neededmodels
Deep learning [21]. Human
have the occupancy
potential recognition
to improve data
HVAC can
be used to
systems byoptimize
enablingroom-booking systems.operation,
more energy-efficient It can alsopredictive
be used for monitoring purposes,
maintenance, improved
for example
comfort in medical
control, facilitiesfault
more accurate anddetection
in security andsystems;
diagnosis, a system
and bettercan indoor
be established
air qualityto
detect the number
management. Someofstudies
occupants
showwho that entered and for
energy costs left heat
the facility,
and cooling as shown in Figure
residential or com-7.
HVAC
mercialunits, which
buildings are asignificantly
could key target ofchange building automation,
if tools account
such as deep for about
learning half of
are used in the
the
advancement
energy consumed of HVAC
in bothsystems.
residentialFor and
example, a studybuildings.
commercial publishedBuilding
in the IEEE Transactions
automation sys-
on Industrial
tems, in general, Electronics
are designedin 2021 to proposed
intelligently a method
operatefor detecting
building and tracking
facilities human
in response to
activity in
dynamic video sequences
external conditions using an IR cameraacceptable
while maintaining and a deep learning
energy usagemodel. The study
and comfort lev-
showed
els [22]. that the proposed method was able to accurately detect and track human activity
in a variety of scenarios,
Deep Learning image including
recognition low-light conditions
is a subfield [22].
of computer vision and artificial intel-
In Figure 9, you can see a deep learning model
ligence that deals with training artificial neural networks to recognize process a batch of images
objects andand make
patterns
an output detecting multiple objects, two people, and a person
in images. This is achieved by using large datasets of labeled images to train the network and object. These models
can
to be applied
recognize to real-world
features, objects, problems
and patterns suchinasnewself-driving
images. The cars,training
medicalprocess
image involves
analysis,
facial an
using recognition,
optimization and algorithm
human recognition.
to adjust the Deep learning
weights of is
thea subset
neurons ofin machine learning
the network to
that involves training artificial neural networks on substantial
minimize the difference between the predicted and actual output for a given image. Once amounts of data to make
predictions
the images are or decisions.
trained, deep It has the potential
learning to significantly
image recognition improve
models can be heating,
used for ventilation,
a variety
and air conditioning (HVAC) systems in several ways:
of tasks such as object detection, image classification, and segmentation [22]. For example,
• theEnergy
in imageEfficiency:
shown in Deep
Figurelearning algorithms
8 an infrared imagecanwas
analyze building
captured usingenergy
objectusage data
detection
and to
methods predict when
capture heatHVAC systems
signatures. should
Deep be activated
learning basedthe
models have onpotential
factors such as occu-
to improve
HVAC pancy, weather,
systems and energy
by enabling moreprices. This can result
energy-efficient in more
operation, energy-efficient
predictive operation,
maintenance, im-
reducing energy costs and reducing the environmental impact of HVAC systems.
proved comfort control, more accurate fault detection and diagnosis, and better indoor air
quality management. Some studies show that energy costs for heat and cooling residential
or commercial buildings could significantly change if tools such as deep learning are used
in the advancement of HVAC systems. For example, a study published in the IEEE Trans-
actions on Industrial Electronics in 2021 proposed a method for detecting and tracking
Energies 2023, 16, 6109 10 of 15

• Predictive Maintenance: Deep learning algorithms can be trained on data from HVAC
systems to predict when components are likely to fail. This can allow for proactive
maintenance, reducing the risk of unexpected breakdowns and improving the overall
reliability of HVAC systems.
• Comfort Control: Deep learning algorithms can analyze building occupancy data and
make predictions about temperature and air flow requirements. This can result in
more accurate and responsive temperature control, improving occupant comfort.
• Fault Detection and Diagnosis: Deep learning algorithms can be trained on data from
Energies 2023, 16, x FOR PEER REVIEW HVAC systems to identify and diagnose faults. This can reduce the time and cost 10 of of
15
maintenance and improve the overall performance of HVAC systems.
• Indoor Air Quality: Deep learning algorithms can analyze indoor air quality data
and make predictions about the levels of pollutants in the air. This can help building
human activity in video sequences using an IR camera and a deep learning model. The
managers to make informed decisions about when to activate air purifiers, ventilation
study showed that the proposed method was able to accurately detect and track human
systems, and other air quality control measures.
activity in a variety of scenarios, including low-light conditions [22].

Energies 2023, 16, x FOR PEER REVIEW

Figure8.8.IR
Figure IRImage
Image[23].
[23].

In Figure 9, you can see a deep learning model process a batch of images and make
an output detecting multiple objects, two people, and a person and object. These models
can be applied to real-world problems such as self-driving cars, medical image analysis,
facial recognition, and human recognition. Deep learning is a subset of machine learning
that involves training artificial neural networks on substantial amounts of data to make
predictions or decisions. It has the potential to significantly improve heating, ventilation,
and air conditioning (HVAC) systems in several ways:
• Energy Efficiency: Deep learning algorithms can analyze building energy usage data
and predict when HVAC systems should be activated based on factors such as occu-
pancy, weather, and energy prices. This can result in more energy-efficient operation,
reducing energy costs and reducing the environmental impact of HVAC systems.
• Predictive Maintenance: Deep learning algorithms can be trained on data from
HVAC systems to predict when components are likely to fail. This can allow for pro-
active maintenance, reducing the risk of unexpected breakdowns and improving the
overall reliability of HVAC systems.
Figure
•Figure Sample
9. 9.
Comfort of Thermal
Sample Images
of Thermal
Control: Deep [23]. algorithms
Images
learning [23]. can analyze building occupancy data and
make predictions about temperature and air flow requirements. This can result in
more accurate
6. Prototype and responsive temperature control, improving occupant comfort.
Concept
• Fault Detection and Diagnosis: Deep learning algorithms can be trained on data from
To use
HVAC this DL
systems model,and
to identify wediagnose
must use a system
faults. This canof wiring
reduce the to connect
time and costaofRaspbe
multiple Arduino
maintenance and Uno forthe
improve communication.
overall performanceThis process
of HVAC is known as serial com
systems.
•tion.Indoor
SerialAir Quality: Deep learning
communication algorithms can analyze
is a communication methodindoor airuses
that quality
onedata
orand
two trans
Energies 2023, 16, 6109 11 of 15
Figure 9. Sample of Thermal Images [23].

6. Prototype Concept
6. Prototype Concept
To use this DL model, we must use a system of wiring to connect a Raspberry Pi to
To use
multiple this DLUno
Arduino model, we must use a system
for communication. of wiring
This process to connect
is known a Raspberry
as serial Pi to
communica-
multiple
tion. Arduino
Serial Uno for communication.
communication is a communication This method
process is known
that as serial
uses one or twocommunication.
transmission
Serialtocommunication
lines send and receiveis adata;
communication method that uses
this data is continuously one or
sent and two transmission
received one bit at a lines
time
to send and receive data; this data is continuously sent and received
by a device shown in Figure 10 [23]. Communication between a Raspberry Pi and an Ar- one bit at a time
by a device
duino Uno isshown in Figure
a popular 10 [23]. Communication
and straightforward method to between
exchangeadata Raspberry
between Pi the
andtwoan
Arduino Uno is a popular and straightforward method to exchange
devices. The Raspberry Pi, running a Linux-based operating system, communicates withdata between the
twoArduino
the devices. Uno,
The Raspberry
which uses Pi,an
running
Atmel amicrocontroller
Linux-based operating
over a UARTsystem, communicates
(Universal Asyn-
with the Arduino
chronous Uno, which uses
Receiver/Transmitter) serialaninterface.
Atmel microcontroller over a UART (Universal
Asynchronous Receiver/Transmitter) serial interface.

Figure 10.
Figure Raspberry Pi
10. Raspberry Pi via
via USB
USB Arduino
Arduino Uno
Uno [24].
[24].

By using this technique, it will allow for communication to be conducted between the
By using this technique, it will allow for communication to be conducted between the
Raspberry Pi and Arduino Uno. To effectively use this method, you must first pip install
Raspberry Pi and Arduino Uno. To effectively use this method, you must first pip install
Raspberry Pi configuration. This command will allow you to connect the Raspberry Pi to
Raspberry Pi configuration. This command will allow you to connect the Raspberry Pi to
the Arduino to receive the transmitted data [24]. Once the connection is established, the
the
nextArduino
step is totocall
receive
for thetheconnection
transmittedports
datato[24]. Once
check forthe connection is
an established established,
connection. the
Once
next step is to call for the connection ports to check for an established connection.
these steps are conducted properly, the installation of Arduino IDE (integrated development Once
these steps arecan
environment) conducted
occur. properly, the installation of Arduino IDE (integrated develop-
mentBy environment)
using intricate canwiring
occur.systems, we will be able to map the wiring of the Raspberry
By using intricate wiring
Pi to the Arduino using Pi camera systems, we will
images. be systems
The able to map the wiring
of wires ofaligned
will be the Raspberry
across
Pi
the prototype connecting to the blower and each of the four butterfly valves shownthe
to the Arduino using Pi camera images. The systems of wires will be aligned across in
prototype connecting
Figure 11. The butterfly tovalve
the blower andbeeach
will also of the four
connected withbutterfly valves
a Pi cameras shownoccupants.
to detect in Figure
11.
ThisThe
Energies 2023, 16, x FOR PEER REVIEW butterfly
requires many valve will also
different pip be connected
installs such aswith a Pi cameras
OpenCV and otherto detect
programsoccupants.12 This
required offor
15
requires many different pip installs such as OpenCV and other
the Arduino to be able to seamlessly communicate with the Raspberry Pi [25]. programs required for the
Arduino to be able to seamlessly communicate with the Raspberry Pi [25].

Figure 11. Model of Prototype System, developed in Autodesk.


Figure 11. Model of Prototype System, developed in Autodesk.

OpenCV (Open-Source Computer Vision Library) is an open-source computer vision


and machine learning software library. It was originally developed by Intel and later sup-
ported by Willow Garage and Itseez (now part of Intel). OpenCV is written in C++ and has
interfaces for C++, Python, Java, and several other programming languages, making it
Energies 2023, 16, 6109 12 of 15

OpenCV (Open-Source Computer Vision Library) is an open-source computer vision


and machine learning software library. It was originally developed by Intel and later
supported by Willow Garage and Itseez (now part of Intel). OpenCV is written in C++ and
has interfaces for C++, Python, Java, and several other programming languages, making
it highly versatile and accessible to developers [26]. The primary goal of OpenCV is to
provide a comprehensive set of tools and functions for various computer vision tasks,
enabling developers to build applications that can understand, analyze, and manipulate
visual data. My proposal concept will use these techniques to help develop a prototype to
simulate a HVAC system. By using the OpenCV for communication, code can be developed
to send serial messages between the devices to control the flow of air through the system.
A snippet of the example code developed shows how communication between the devices
will be processed when receiving a message.

def getObjects(img, thres, nms, draw=True, objects=[]):


classIds, confs, bbox = net.detect(img,confThreshold=thres,nmsThreshold=nms)
#print(classIds,bbox)
if len(objects) == 0: objects = classNames
objectInfo =[]
if len(classIds) != 0:
for classId, confidence,box in zip(classIds.flatten(),confs.flatten(),bbox):
className = classNames[classId - 1]
if className in objects:
objectInfo.append([box,className])
if (draw):
cv2.rectangle(img,box,color=(0,255,0),thickness=2)
cv2.putText(img,classNames[classId-1].upper(),(box[0]+10,box[1]+30),
cv2.FONT_HERSHEY_COMPLEX,1,(0,255,0),2)
cv2.putText(img,str(round(confidence*100,2)),(box[0]+200,box[1]+30),
cv2.FONT_HERSHEY_COMPLEX,1,(0,255,0),2)
# Temporary Attempt to write-in code for Valve Controls Opening @ 90 degrees
if camera_1, objects = True:
V_1.angle = 90
print("Human Detected")
human_check = 1
else:
V_1.angle = -90

return img, objectInfo, human_check

From the above code, this portion of the program will take the loaded image from
the Pi camera to detect any occupants that may trigger a serial message to be sent to the
butterfly valve. When an occupant is detected, this will activate the butterfly values to
go from the rest angle (180◦ ) to 90◦ angles. When the serial message is sent, one (person
detected) or 0 (no motion), a message will be sent to activate the sensor. This process
will cycle every 5 min to allow the store data to be dumped to reserve memory space.
Serial communication between the Raspberry Pi and Arduino Uno enables real-time data
exchange, allowing the Arduino Uno to manage precise control of the HVAC equipment
based on inputs from the Pi camera [26]. The Arduino Uno’s ability to interface with
various actuators and sensors makes it well-suited for fine-tuning the HVAC system’s
parameters according to the data provided.
By using the methods suggested in this paper, the system could be developed to adapt
to changing occupancy patterns, ensuring efficient heating or cooling only in areas with
occupants. Secondly, the IR cameras can detect temperature variations, allowing the system
to respond to localized thermal conditions and adjust the valves accordingly [27]. Lastly,
Energies 2023, 16, 6109 13 of 15

the use of machine learning enables predictive control, anticipating environmental changes
and adjusting the valves in advance to improve energy efficiency and occupant comfort.

7. Conclusions
In conclusion, the use of infrared cameras in diagnosing HVAC problems can improve
the efficiency and effectiveness of HVAC maintenance and repair. HVAC systems play
a vital role in maintaining comfortable and healthy indoor environments, especially in
buildings where people spend a lot of time. Over the years, HVAC technology has made
significant advancements, revolutionizing the way we control and optimize indoor envi-
ronments. These advancements have resulted in increased energy efficiency, improved
comfort, enhanced air quality, and smarter system management. However, in some cases,
these systems can develop faults or inefficiencies that can lead to higher energy costs, poor
indoor air quality, and reduced system performance. The introduction of smart technology
has also had a profound impact on HVAC systems. With the rise of the Internet of Things
(IoT), HVAC systems can now be connected to a network, enabling remote control and
monitoring [28]. Smart thermostats, for example, have the capability to learn occupants’
preferences and adjust the temperature settings accordingly, leading to energy savings and
personalized comfort [29]. Additionally, data analytics and machine learning algorithms
are employed to optimize HVAC system performance, identify potential issues, and pro-
vide predictive maintenance, reducing downtime and improving overall efficiency. This
proposed research, with the integration of a Pi camera and serial communication between
a Raspberry Pi and an Arduino Uno, holds great promise for enhancing HVAC systems.
The Pi camera, with its imaging capabilities, can provide valuable visual information about
occupancy patterns, thermal profiles, and environmental conditions within a space. This
data, when combined with machine learning algorithms, can optimize HVAC control strate-
gies, leading to improved energy efficiency and occupant comfort IR (infrared) cameras,
on the other hand, can be a useful tool for diagnosing and assessing the performance of
HVAC systems. By detecting temperature variations and thermal patterns, IR cameras can
identify areas of the system that may be malfunctioning or where energy may be wasted.
For example, during certain periods of the year, university and grade schools may shut
down; when these buildings are unoccupied, the HVAC system still performs as if the
building has occupants unless it is regulated by a maintenance worker to adjust the settings.
Implementing a system that could detect whether a building is occupied in specific area
would play vital role in saving energy. This can help prevent significant energy waste,
improve indoor air quality, and ensure occupant comfort. Therefore, the use of smart
cameras such as IR cameras and Pi cameras in HVAC systems can play a crucial role in
maintaining a comfortable indoor environment and reducing overall energy costs.

8. Important Terms
• HVAC (heating, ventilation, and air conditioning),
• Energy consumption
• Energy output
• Occupancy detection
• Human occupancy detection
• Infrared radiation
• Deep learning
• Air Quality
• Arduino Uno
• Raspberry Pi
• GPIO.

Author Contributions: Review and editing performed by D.R.M. and Y.T.A.; Funding provided by
S4G: clean energy workforce development grant; review performed by B.G., R.C.T. and Y.T.A. All
authors have read and agreed to the published version of the manuscript.
Energies 2023, 16, 6109 14 of 15

Funding: The funding was provided S4G and the APC was funded by the department of Computa-
tional Data Science and Engineering department at North Carolina A&T State University.
Data Availability Statement: Data sharing is not applicable to this article as no new data were
created or analyzed in this study.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Taheri, S.; Hosseini, P.; Razban, A. Model predictive control of heating, ventilation, and air conditioning (HVAC) systems: A
state-of-the-art review. J. Build. Eng. 2022, 60, 105067. [CrossRef]
2. Modern Building Services. Available online: https://modbs.co.uk/news/archivestory.php/aid/2799/Perkins_high-pressure_
apparatus_.html (accessed on 17 December 2022).
3. Pérez-Lombard, L.; Ortiz, J.; Coronel, J.F.; Maestre, I.R. A review of HVAC systems requirements in building energy regulations.
Energy Build. 2011, 43, 255–268. [CrossRef]
4. Jeon, E.S.; Choi, J.S.; Lee, J.H.; Shin, K.Y.; Kim, Y.G.; Le, T.T.; Park, K.R. Human detection based on the generation of a background
image by using a far-infrared light camera. Sensors 2015, 15, 6763–6788. [CrossRef] [PubMed]
5. Melikov, A.K. Advanced air distribution: Improving health and comfort while reducing energy use. Indoor Air 2016, 26, 112–124.
[CrossRef] [PubMed]
6. Prokoski, F. History, current status, and future of infrared identification. In Proceedings of the IEEE Workshop on Computer
Vision Beyond the Visible Spectrum: Methods and Applications, Hilton Head, SC, USA, 16 June 2000; (Cat. No. PR00640). IEEE:
New York, NY, USA, 2000. [CrossRef]
7. Beltran, A.; Erickson, V.L.; Cerpa, A.E. Thermosense: Occupancy thermal based sensing for hvac control. In Proceedings of the
5th ACM Workshop on Embedded Systems for Energy-Efficient Buildings, Roma, Italy, 11–15 November 2013. [CrossRef]
8. Axley, J. Multizone airflow modeling in buildings: History and theory. HVACR Res. 2007, 13, 907–928. [CrossRef]
9. Kalman Tihanyi’s 1926 Patent Application Radioskop. UNESCO Memory of the World. 2001. Available online: https://handwiki.
org/wiki/Biography:K%C3%A1lm%C3%A1n_Tihanyi (accessed on 15 March 2023).
10. Tien, P.W.; Wei, S.; Darkwa, J.; Wood, C.; Calautit, J.K. Machine learning and deep learning methods for enhancing building
energy efficiency and indoor environmental quality—A review. Energy AI 2022, 10, 100198. Available online: https://www.
sciencedirect.com/science/article/pii/S2666546822000441 (accessed on 15 March 2023). [CrossRef]
11. Yildirim, E.; Saraç, Ş.; Aslan, A. Energy consumption and economic growth in the USA: Evidence from renewable energy. Renew.
Sustain. Energy Rev. 2012, 16, 6770–6774. [CrossRef]
12. Du, Y.; Zandi, H.; Kotevska, O.; Kurte, K.; Munk, J.; Amasyali, K.; Mckee, E.; Li, F. Intelligent multi-zone residential HVAC control
strategy based on deep reinforcement learning. Appl. Energy 2021, 281, 116117. [CrossRef]
13. Yang, S.; Wan, M.P.; Chen, W.; Ng, B.F.; Dubey, S. Model predictive control with adaptive machine-learning-based model for
building energy efficiency and comfort optimization. Appl. Energy 2020, 271, 115147. [CrossRef]
14. Statista. World Energy Outlook. IEA. 2022. Available online: https://www.statista.com/statistics/192579/us-energy-
consumption-and-production/ (accessed on 18 March 2023).
15. Borda, D.; Bergagio, M.; Amerio, M.; Masoero, M.C.; Borchiellini, R.; Papurello, D. Development of Anomaly Detectors for HVAC
Systems Using Machine Learning. Processes 2023, 11, 535. [CrossRef]
16. Gi, K.; Sano, F.; Hayashi, A.; Tomoda, T.; Akimoto, K. A global analysis of residential heating and cooling service demand and
cost-effective energy consumption under different climate change scenarios up to 2050. Mitig. Adapt. Strateg. Glob. Chang. 2018,
23, 51–79. [CrossRef]
17. Speedir Thermal Vision. The History of Thermal Imaging Cameras. Speedir. Available online: https://speedir.com/blog/the-
history-of-thermal-imaging-cameras/ (accessed on 1 March 2023).
18. Raspberry Pi. Available online: https://www.raspberrypi.com/documentation/computers/os.html (accessed on 7 May 2023).
19. Raspberry Pi Tutorials. Available online: https://roboticsbackend.com/raspberry-pi-arduino-serial-communication/ (accessed
on 13 April 2023).
20. Arduino Uno. Available online: https://www.arduino.cc/en/hardware (accessed on 12 March 2023).
21. Wen, J.T.; Mishra, S. Intelligent Building Control Systems; Springer: Berlin/Heidelberg, Germany, 2018. [CrossRef]
22. Gupta, S.K.; Atkinson, S.; O’boyle, I.; Drogo, J.; Kar, K.; Mishra, S.; Wen, J.T. BEES: Real-time occupant feedback and environmental
learning framework for collaborative thermal management in multi-zone, multi-occupant buildings. Energy Build. 2016, 125,
142–152. [CrossRef]
23. Acquaah, Y.; Steele, J.B.; Gokaraju, B.; Tesiero, R.; Monty, G.H. Occupancy detection for smart HVAC efficiency in building energy:
A deep learning neural network framework using thermal imagery. In Proceedings of the 2020 IEEE Applied Imagery Pattern
Recognition Workshop (AIPR), Washington, DC, USA, 13–15 October 2020; IEEE: New York, NY, USA, 2020. [CrossRef]
24. Chandan, G.; Jain, A.; Jain, H. Real Time Object Detection and Tracking Using Deep Learning and OpenCV. In Proceedings of the
2018 International Conference on Inventive Research in Computing Applications (ICIRCA), Coimbatore, India, 11–12 July 2018;
pp. 1305–1308. [CrossRef]
Energies 2023, 16, 6109 15 of 15

25. Xukyo. Serial Communication between Raspberry Pi and Arduino. Available online: https://www.aranacorp.com/en/serial-
communication-between-raspberry-pi-and-arduino/ (accessed on 23 April 2023).
26. Sharma, A.; Pathak, J.; Prakash, M.; Singh, J.N. Object detection using opencv and python. In Proceedings of the 2021 3rd
International Conference on Advances in Computing, Communication Control and Networking (ICAC3N), Greater Noida, India,
17–18 December 2021; IEEE: New York, NY, USA, 2021. [CrossRef]
27. Choi, H.; Um, C.Y.; Kang, K.; Kim, H.; Kim, T. Application of vision-based occupancy counting method using deep learning and
performance analysis. Energy Build. 2021, 252, 111389. [CrossRef]
28. Farooq, M.U.; Waseem, M.; Mazhar, S.; Khairi, A.; Kamal, T. A review on internet of things (IoT). Int. J. Comput. Appl. 2015, 113, 1–7.
29. Du, Y.; Li, F.; Munk, J.; Kurte, K.; Kotevska, O.; Amasyali, K.; Zandi, H. Multi-task deep reinforcement learning for intelligent
multi-zone residential HVAC control. Electr. Power Syst. Res. 2021, 192, 106959. [CrossRef]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.

You might also like