We are IntechOpen,

the world’s leading publisher of

Open Access books
Built by scientists, for scientists

7,200
Open access books available
191,000
International authors and editors
210M Downloads

Our authors are among the

154
Countries delivered to
TOP 1%
most cited scientists
14%
Contributors from top 500 universities

Selection of our books indexed in the Book Citation Index

in Web of Science™ Core Collection (BKCI)

Interested in publishing with us?

Contact book.department@intechopen.com
Numbers displayed above are based on latest data collected.
For more information visit www.intechopen.com
Chapter

Motor Monitoring and Analysis

System (SMAM) as a Resource for
Predictive Maintenance in
Industrial Environments through
Digital Twin and IIoT
Jhennifer F. dos Santos, Yasmim M.D. Lisboa,
Elielson F.S. Barbosa, Lucas H.B. Santos, Allan R.A. Manito,
Marcelo O. Silva and Wellington S. Fonseca

Abstract

The Motor Monitoring and Analysis System (SMAM) was developed as a tool that
can help with the predictive maintenance of electric motors through Digital Twin
(DT) and the Industrial Internet of Things (IIoT). To do this, SMAM has an embedded
system made up of the ESP32 microcontroller and temperature, voltage and current
sensors, in which the measurements taken are stored in the Firebase web database. It
is worth noting that the sensors are calibrated using the least squares and linear
regression mathematical methods, as well as using the ET-5062 energy analyzer as a
reference to ensure that the measurement is reliable. In addition, strong coupling
simulations are carried out in the Finite Element Method Magnetics (FEMM) soft-
ware, which makes it possible to visualize the magnetic flux and the internal temper-
ature of the motor by entering the measured current data. In this way, SMAM is a
promising resource for the manufacturing environment.

Keywords: induction motors, finite element analysis, industrial internet of things,

digital twin, maintenance

1. Introduction

Electric motors have a significant presence in the industry, acting as essential

components in various production processes [1]. This equipment is considered a
critical asset, as it drives machines and systems that are fundamental to the continu-
ous and efficient operation of factories. Because of its importance, the failure of an
electric motor can result in unplanned downtime, loss of productivity, and high
maintenance costs [2]. For this reason, constant monitoring of electric motors is
crucial to ensure good operating performance.
1
Induction Motors – Latest Research and Applications

In recent years, researchers have been exploring the development of tools for
monitoring induction motors. One example is the article [3] that proposes an auto-
mated induction motor monitoring system with IoT, using a Raspberry Pi 4 and the
Blynk IoT interface to protect the motor in abnormal conditions, such as overvoltage
and high temperature, ensuring continuous operation or automatic shutdown. The
study carried out by Patel et al. [4] monitors the condition of three-phase induction
motors by analyzing the stator current and vibration, using instruments such as the
Digital Storage Oscilloscope and the Vibration Analyzer to diagnose faults with the
support of MATLAB.
In addition, Noyjeen et al. [5] propose the control and monitoring of three-phase
motors with IoT, using the Raspberry Pi to collect data from sensors and send it to a
server, where it is processed to detect faults in components such as the rotor and
stator. The work by Noyjeen et al. [5] presents a system for monitoring three-phase
motor parameters using the ESP32 microcontroller, with cloud storage in ThingSpeak
and notifications via the Line Notify API, allowing remote access and real-time alerts
when performance limits are exceeded. This research reinforces the relevance of
innovative solutions for the efficient monitoring of electric motors in industry.
Within the context of these advances, the Industrial Internet of Things (IIoT) has
emerged as a key technology. The IIoT, which is an extension of the Internet of Things
(IoT), is specifically focused on the industrial environment [6, 7]. While IoT covers
the connectivity and automation of devices in a wide range of applications, IIoT is
aimed at optimizing industrial processes, connecting sensors, machines, and systems
to collect and analyze data in real time. The application of IIoT in the industrial sector
brings significant advantages, such as the possibility of carrying out predictive main-
tenance, avoiding unexpected failures, and extending the useful life of assets [8].
Another emerging technology that is being widely adopted in the industry is the
Digital Twin. A Digital Twin is a virtual representation of a physical asset, which
allows you to simulate and monitor its behavior in real time [9]. There are different
types of Digital Twin, including the Product, Process and Performance Digital Twin,
each focused on different aspects of an asset’s operation [10]. In industry, Digital
Twins have been used to monitor engines, providing a detailed view of their operation
and allowing simulations that predict future behavior based on real data [11]. This
approach allows for more precise control and a rapid response to any anomalies
detected, increasing the reliability and efficiency of electric motors.
In this context, this article will present the optimization of the acquisition module
that makes up the Motor Monitoring and Analysis System (from Portuguese, Sistema
de Monitoramento e Análise de Motores - SMAM) proposed by Santos et al. [12]. This
system integrates IIoT and Digital Twin technologies as a tool that can help with the
predictive maintenance of electric motors. SMAM uses an embedded system with an
ESP32 microcontroller and sensors to measure critical variables such as temperature,
voltage, and current, whose data is stored in a web database. In addition, advanced
simulations are carried out to visualize the internal behavior of the engine and these
results are displayed on a web panel, making SMAM a promising tool for the industry.
The reason for building a new acquisition module is that the work presented by
Santos et al. [12] had a significant error in one of the measurements of the current sensor
that makes up the system and which plays a crucial role in the Digital Twin model. The
current values measured by the module are incremented in the simulation carried out in
the Finite Element Method Magnetics (FEMM) software. Therefore, in addition to the
development of a new module, the contribution of this work is the application of a
calibration methodology using the least squares method and linear regression.
2
Motor Monitoring and Analysis System (SMAM) as a Resource for Predictive Maintenance…
DOI: http://dx.doi.org/10.5772/intechopen.1006914

2. The motor monitoring and analysis system

The Motor Monitoring and Analysis System (SMAM) consists of an IIoT module
that measures the temperature, voltage, and phase current of induction motors using
non-invasive sensors. The module is equipped with the ESP32 microcontroller, which
transmits measurements in real time to the cloud via a Wi-Fi connection. In addition,
computer simulations of the strong coupling of the monitored motor are carried out.
Figure 1 illustrates the general layout of the SMAM.
As described above, this work proposes the optimization of the embedded system
developed by Santos et al. [12]. Initially, a module was built on a perforated phenolite
plate measuring 12 cm long by 9.3 cm wide, which accommodates the ESP32 micro-
controller and the voltage, current, and temperature sensors, as well as providing
space for the inclusion of new sensors if necessary, as shown in Figure 2.

2.1 Acquisition module

The optimized module is managed by the ESP32-WROOM-32 microcontroller,

which is the backbone of the monitoring since it is responsible for receiving the sensor
measurements and sending them to a cloud database. The ESP32 sends this informa-
tion via the IEEE 802.11 protocol [13].
The ESP32 microcontroller has 32-bit dual-core processors operating at 240 MHz,
analog-to-digital converters, and an internal antenna with a power of 20 dBm,
supporting a data rate of up to 150 Mbps. It also has a sampling rate of 1 kHz, low bit
error rate (BER), and performs high-speed floating point operations. This device has
gained popularity in the industry due to its capabilities, being used, for example, in
the WCD-GT100-ETH and WCD-GT100 gateways from WEG [14].
It is worth noting that the purpose of this work is restricted to optimizing the
existing module, so it was not necessary to replace the sensors and microcontroller
used in the [12] work, and the module’s wiring diagram 3 was maintained.
The acquisition module consists of three current sensors model SCT-013
(Figure 3b), which are current transformers capable of measuring alternating current
(AC) of up to 100 A [15]. These sensors operate on the principles of electromagnetic

Figure 1.
The functioning of the SMAM proposed by Santos et al. [12].

3
Induction Motors – Latest Research and Applications

Figure 2.
New acquisition module built for SMAM.

Figure 3.
Electrical schematic of the module [12].

induction, where the magnetic field generated by the alternating current in the pri-
mary conductor induces a proportional current in the sensor’s secondary circuit. Each
SCT-013 is connected to one of the motor’s phases, allowing precise measurement of
the current in each phase.
For voltage measurement, the ZMPT101B module is used (Figure 4a). This sensor
is capable of measuring alternating voltages (AC) of up to 250 V, with a margin of
error of 1% [16]. The ZMPT101B includes a voltage transformer and an RC circuit

4
Motor Monitoring and Analysis System (SMAM) as a Resource for Predictive Maintenance…
DOI: http://dx.doi.org/10.5772/intechopen.1006914

with amplifiers, which condition the voltage signal into a form suitable for reading by
the ESP32 microcontroller.
For temperature measurement, the system relies on a type K thermocouple, which
has a measuring range from 0 to 600°C. The K-type thermocouple operates on the
basis of the Seebeck Effect, where the junction of two different metals, in this case,
Cromel and Alumel, generates an electrical voltage proportional to the temperature
difference between the hot junction and the reference junction. The signal generated
by the thermocouple is in millivolts, which makes it necessary to use the MAX6675, a
specialized analog-to-digital converter.
The MAX6675 reads the voltage difference generated by the thermocouple and
compares it with the voltage of an internal temperature sensor diode, converting these
values into an accurate temperature reading. The integration of the type K thermo-
couple with the MAX6675 allows the system to obtain reliable and accurate tempera-
ture measurements. The type K thermocouple and the MAX6675 are shown in
Figure 4c and d, respectively.

2.2 Sensor calibration

Linear regression and least squares statistical methods were used to adjust the
measurements taken by the on-board system. Linear regression is a specific applica-
tion of the least squares method, where the relationship between variables is modeled
by a linear equation and the parameters of this equation are estimated using the least
squares method.
Linear regression aims to model the relationship between two variables using a
linear equation, where X is the independent variable and y is the dependent variable.
Eq. 1 illustrates this model, where i represents each of the n observations in the
database, a is the intercept of the regression line on the y-axis, b is the slope of the line
and ε is the residual representing the values of y not explained by the model [17].

yi ¼ axi þ b þ εi (1)

Figure 4.
Voltage (a) and current (b) sensors, as well as the k-type thermocouple (c) and MAX6675 integrated circuit (d)
used in the SMAM.

5
Induction Motors – Latest Research and Applications

The least squares method is used to estimate the parameters a and b of the linear
regression equation. This method seeks to minimize the sum of squares of the residuals
(SSE), which are the differences between the observed values of y and the values
predicted by the regression equation [18]. With this, it is possible to determine the values
of aaa and b that minimize the sum of the squares of the residuals, as shown in Eq. 2.
n 
X 2
SSE ¼ yi ðmxi þ bÞ (2)
i¼1

To find the SSE, calculate the partial derivatives in relation to a and b and equal
them to zero to find the values that minimize the function. Once the values of a and b
have been estimated using the least squares method, the linear regression equation can
be used to make predictions of y based on the values of x.This results in predictions
that represent the best possible estimate of the relationship between the variables,
given the distribution of the observed data.
As mentioned earlier, the engine was monitored by the ET-5062 and the optimized
SMAM system. Based on this, a Python script was developed to automate the process of
calibrating the SMAM’s current sensors, using linear regression and least squares, in order
to make it easier to identify deviations and improve the accuracy of measurements.
The script used the Pandas library for manipulating and analyzing structured data,
allowing data to be loaded from different formats such as CSV, xlsx, and SQL [19]. In
addition, the Matplotlib library was used to visualize the data. According to Ref. [20],
this library generates high-quality customizable graphics, allowing for intuitive and
harmonious data visualization.
The calibration methods were applied using the sklearn:linearm odel library, which
has the LinearRegression function of adjusting the linear regression model to the data
using the least squares method [21]. The fit method of the LinearRegression class
calculates the regression coefficients that minimize the sum of the squares of the
differences between the actual observations and the model’s predictions. In this way,
the calculated coefficients represent the parameters of the regression line, solving the
least squares problem.
Figure 5 shows the use of LinearRegression and the fit method for the values of the
current sensor that monitors Phase A. It is important to note that these functions were
also applied to the motor’s Phase B and C sensors.

# Upload sensor and commercial sensor and SMAM data

medicao_SMAM = pd.read_excel(medicao_SMAM)
medicao_anali = pd.read_excel(medicao_anali)

# Separate data into sensor readings and corresponding

# measurements for PhaseA
FaseA_SMAM = medicao_SMAM[['Fase A']]

# Initialize and adjust the linear regression model for Phase A

modelo_FaseA = LinearRegression()
modelo_FaseA.fit(FaseA_SMAM, FaseA_anali)

# Regression coefficients for Phase A

coeficiente_FaseA = modelo_FaseA.coef_[0][0]
intercepto_FaseA = modelo_FaseA.intercept_[0]

6
Motor Monitoring and Analysis System (SMAM) as a Resource for Predictive Maintenance…
DOI: http://dx.doi.org/10.5772/intechopen.1006914

Figure 5.
Arrangement of equipment for testing the embedded system (a) and thermographic camera used in the
experiment (b).

2.3 Functional testing methodology

The motor selected for this study is WEG’s high-performance Plus model, with a
rated current of 4.42 A. This motor operates in conjunction with an electromagnetic
break, which consists of coils and a metal disk controlled by an electrical control
bench. The function of the brake is to act as a load, simulating real motor operation in
industry. The connection between the motor and the brake can be seen in Figure 5a.
To test operation, the motor was monitored by the optimized SMAM module and
the ET-5062 energy analyzer, used as a reference meter. According to the manufac-
turer [22], this analyzer complies with the IEC/EN61010-1 standard, Category IV
600 V Overvoltage, and allows the user to choose between 1-second and 60-minute
measurements. It also offers real-time visualization of general network parameters
such as voltage, current, and power. It is worth noting that the motor is monitored
simultaneously with both devices so that measurements can be compared.
The test followed the methodology adopted by Santos et al. [12], sequentially
applying four load levels, identified as load 1, 2, 3, and 4. At the initial load (load I),
the motor operated at no load, without the Eddy Brake coils energized. At load levels
2, 3, and 4, the motor’s phase currents reached values of 3 A, 3.5 A, and 4 A, respec-
tively. Figure 5 shows the test layout.
To monitor the external temperature of the engine, a FLIR T650sc thermographic
50 camera was used (Figure 5b), capable of measuring temperatures from 40 to
650°C. In addition, a Minipa multimeter model ET-2042E was used, with a thermal
sensitivity of 40 to 1000°C [23]. The data was stored manually, with the measure-
ments from the reference equipment being compared with the measurements from
the optimized module’s temperature sensor at each load transition.

2.4 Web application structure

Communication is carried out via a web application developed in the Python,

HTML, and JavaScript programming languages and stored on the Replit online plat-
form. Replit is a platform that stores web applications and offers resources for testing
and deploying applications in various programming languages [24].
7
Induction Motors – Latest Research and Applications

Figure 6.
Structure of the application developed by [12].

The web application receives the data sent by ESP32 in JSON format, using the
HTTP POST method. The measurements received are stored in Firebase, a non-
relational database from Google that offers various functionalities, including a real-
time database [25]. The measurements are stored in Firebase in JSON format.
Socket.io middleware was also implemented in the application. This technology
enables real-time communication between the registered client and the server [26].
Thus, as soon as the measurement data is received by the application, the socket
connection sends this data to a Virtual Machine (VM).
The VM uses the Windows operating system, as the FEMM software used for
computer simulation is native to this system. This VM was configured to detect
variations in the motor current and perform a thermomagnetic simulation based on
these variations. The result of the simulation is sent to Firebase. The general flow
diagram of the communication described is shown in Figure 6.
The control panel, developed in HTML and JavaScript, is integrated into the appli-
cation to display the results of the motor current simulations in real time. Once the
simulations are complete, images are sent from the VM to Firebase and the application
via a socket connection. The JavaScript panel then automatically checks and displays the
results on the web page, eliminating the need for the user to manually update the page.

2.5 Thermomagnetic simulation of an induction motor

This research involved performing a multiphysics thermomagnetic simulation uti-

lizing FEMM software. By coupling these simulations, the outcomes from one serve as
inputs for another, thereby enhancing the fidelity to actual physical phenomena [27].
The induction motor model features a rotor with 28 slots and a stator with 36 slots,
composed of aluminum and copper, respectively.
The simulation process is automated through a Python script leveraging the
pyFEMM package, facilitating easy execution of simulations in FEMM. The script
defines the motor materials, creates the finite element mesh, and initiates the mag-
netic simulation. In this stage, Eq. 1 is numerically solved to find the potential of the
three magnetic vectors within all generated finite elements.

1
 
∂A
∇ ∇A ¼ σ σ∇V þ J src (3)
μðBÞ ∂t

8
Motor Monitoring and Analysis System (SMAM) as a Resource for Predictive Maintenance…
DOI: http://dx.doi.org/10.5772/intechopen.1006914

In this context, μðBÞ denotes the magnetic permeability of the medium, and J src
represents the current density applied by an external source. The resistive losses in the
copper conductors are calculated using Eq. 4.

1
q ¼ J2 (4)
σ

Subsequently, the thermal simulation phase commences, solving Eq. 5 numerically

across all finite elements. This equation incorporates the resistive losses from the
magnetic simulation as heat sources.

dT
ρcp ∇
ðk∇T Þ ¼ q (5)
dt

Here, ρ0 is the mass density, cp signifies the specific heat capacity, k represents the
thermal conductivity, and T stands for temperature. The electrical conductivities are
updated in accordance with Eq. 6. Where ρ0 is the electrical resistivity at 0°C, and β0
is the temperature coefficient of resistivity.

1
σ ðT Þ ¼ , (6)
ρ0 ð1 þ β0 ÞT

3. Results and discussions

The test carried out to verify the performance of the SMAM’s optimized embedded
system was satisfactory, since both the ESP32 microcontroller received all the sensor
measurements and sent them to Firebase, and the ET-5062 stored all the measure-
ments in its internal memory. In addition, during the test, as described in Section 2.3,
the temperature reached by the engine was monitored and it was found that there was
no need to apply statistical techniques for calibration, since this sensor showed an
average error of around 1%. Figure 7 shows the thermogram of the temperature
measurement for load 3.
With the testing stage successfully completed, we proceeded to calibration, in
which the script described in the previous section came into play, transforming the
files generated into a data frame by the Pandas library for studying the data. The first

Figure 7.
Thermogram of the motor operating with load 3 in the performance test of the optimized SMAM module.

9
Induction Motors – Latest Research and Applications

Without calibration With calibration

Monitored phases MAE MAE % MAE MAE %

Phase A 0.5992 14.87 0.0681 1.88

Phase B 0.6518 15.48 0.0622 1.79

Phase C 0.5944 14.26 0.0367 1.00

Table 1.
Absolute error of the SMAM current sensors.

analysis made of this was the voltage analysis, in which the mean absolute error of the
voltage read by the SMAM and ET-5062 sensor was calculated, the result of which was
a mean absolute error (MAE) of 1.5, around 1%. The errors presented by the voltage
and temperature sensor follow those described in the work by Santos et al. [12].
The second analysis took place with the current data, in which the mean absolute
error was first calculated without calibration to assess the need to apply static tech-
niques. The result of this is shown in Table 1 and as can be seen the errors presented
by each sensor are high compared to the voltage and temperature measurements.
Once the difference between the sensors was noted, the LinearRegression functions
and the fit method were applied to determine the equations of the regression line for
each sensor, which resulted in Eqs. 7, 8, and 9.

yi ¼ 0:83x þ 0:08 (7)

yi ¼ 0:82x þ 0:10 (8)
yi ¼ 0:85x þ 0:04 (9)

After this, the calibrated measurements were calculated using the equations found,
in which the measurement made by the SMAM’s current sensor is the input to the
regression model represented by the x variable. After carrying out this process, the
average absolute error of the calibrated measurements was measured to indicate the
effectiveness of the model and the result obtained is that after using the static tech-
niques the error decreased significantly, since it is below 2% as seen in Table 1.
In addition, Figure 8 shows the graphs for each phase relating the SMAM mea-
surements to the ET-5062 data. As you can see, both the SMAM and the ET-5062 did

Figure 8.
Graph showing the measurements of the motor’s three phases made by the ET-5062 and the SMAM after
calibration.

10
Motor Monitoring and Analysis System (SMAM) as a Resource for Predictive Maintenance…
DOI: http://dx.doi.org/10.5772/intechopen.1006914

Figure 9.
The magnetic flux of the electric motor when operating at load I (a), load II (b), load III (c), and load IV (d).

not capture the motor’s peak current, only the permanent regime in which each step
represents the four load levels described in Section 2.3.
As for the magnetic simulation results, it was possible to visualize the magnetic
flux present in the stator and rotor, as well as the formation of the four poles inside the
motor, using a triangular mesh with 46,426 nodes and 92,294 elements. It was also
found that the maximum magnetic induction was 1.5 T, 1.8 T, 1.9 T, and 1.95 T for
loads I, II, III, and IV, respectively. As can be seen in Figure 9.
In the thermal simulation, the values of the resistive losses in the rotor and stator
windings calculated in the magnetic simulation are incorporated into the thermal

Figure 10.
Resistive losses calculated in the simulation of the motor operating under load IV.

11
Induction Motors – Latest Research and Applications

Figure 11.
Temperature in Kelvin of the engine when operating with load I (a), load II (b), load III (c), and load IV (d).

simulation as described above. The graph in Figure 10 shows the resistive losses (in
watts) per slot in the stator and rotor for all loads.
Figure 11 shows the thermal behav of the motor for all loads, and shows the
concentration of heat in the center of the poles due to resistive losses and the high
concentration of magnetic flux.
In addition, by carrying out the thermal simulation, it is possible to calculate the
conductivity of the materials that make up the motor and which are influenced by the
temperature which, when it increases, decreases the conductivity of the materials and
causes the resistance to increase. Figure 12 shows the reduction in conductivity in the
motor’s stator and rotor as the load increases.

Figure 12.
Electrical conductivity of the stator and rotor.

12
Motor Monitoring and Analysis System (SMAM) as a Resource for Predictive Maintenance…
DOI: http://dx.doi.org/10.5772/intechopen.1006914

Finally, as in Ref. [12], the result of each simulation was visualized in the dash-
board developed. In this way, it was possible to follow the status of the monitored
motor in real time, since the panel is web responsive, defined as a page that adapts to
the type of device used according to França [28], it was possible to monitor the
thermal and magnetic performance of the motor in operation on both smartphones
and computers.

4. Conclusions

This work proposed improvements in the measurements of the sensors present in

the Motor Monitoring and Analysis System (SMAM), a Digital Twin (DT) system that
can assist in the predictive analysis of faults in three-phase electric motors. The focus
was on optimizing the measurements of the current sensors responsible for monitor-
ing the motor’s three phases, using mathematical methods to adjust the sensors and
validate the results with reference equipment.
The results obtained showed that the proposed improvements were effective, since
after calibration, the measurements showed errors of less than 2%. Thus, the perfor-
mance test of the new SMAM module was satisfactory and provided accurate and
reliable measurements. As a result, this work contributes to the advancement of IIoT
in industry by showing effective methods for adjusting and calibrating sensors, which
is essential for the successful implementation of IIoT and for improving the efficiency
and reliability of industrial processes.

Acknowledgements

The authors of this article would like to thank the Financier of Studies and Projects
(FINEP) of the Ministry of Science, Technology and Innovation (MCTIC) for its
support in carrying out this work, within the scope of the Project entitled “Techno-
logical Development and Innovative Extension of Advanced Materials in Energy and
Mobility applied to the Amazonian context” (EMOB-AMAZON), carried out through
the National Science and Technology Plan of Public Call MCTI/FINEP/AT—Advanced
Materials and Strategic Minerals 2020 n° 066/21. The authors would also like to thank
the Coordination for the Improvement of Higher Education Personnel (CAPES) and
the Center of Excellence in Energy Efficiency in the Amazon (CEAMAZON).

13
Induction Motors – Latest Research and Applications

Author details

Jhennifer F. dos Santos1*†, Yasmim M.D. Lisboa1†, Elielson F.S. Barbosa2,

Lucas H.B. Santos1, Allan R.A. Manito1, Marcelo O. Silva1 and Wellington S. Fonseca1

1 Institute of Technology, Federal University of Pará, Belém, Pará, Brazil

2 University Center of Pará, Belém, Pará, Brazil

*Address all correspondence to: jhennifer.santos@itec.ufpa.br

† These authors contributed equally.

© 2024 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of
the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0),
which permits unrestricted use, distribution, and reproduction in any medium, provided
the original work is properly cited.
14
Motor Monitoring and Analysis System (SMAM) as a Resource for Predictive Maintenance…
DOI: http://dx.doi.org/10.5772/intechopen.1006914

References

[1] Tiwari D, Miscandlon J, Tiwari A, DOI: 10.1109/ICACCS57279.2023.

Jewell GW. A Review of Circular 10112825
Economy Research for Electric Motors
and the Role of Industry 4.0 [7] Singh I, Hodgins S, Gadhrri A,
Technologies. Sustainability. 2021;13: Schmidt R. Iiot machine health
9668. DOI: 10.3390/su13179668 monitoring models for education and
training. In: Singh I, Gao Z, Massarelli C,
[2] Antonino-Daviu J. Electrical editors. IoT Applications Computing.
monitoring under transient conditions: A London, UK: IntechOpen; 2021.
new paradigm in electric motors p. ch. 12. DOI: 10.5772/intechopen.
predictive maintenance. Applied 99032
Sciences. 2020;10(17):6137
[8] Javaid M, Haleem A, Singh RP, Rab S,
[3] Subhashini N, Mouli M, Mugunthan J,
Suman R. Upgrading the manufacturing
Kumar RP, Revanth M, Tejaa M.
sector via applications of Industrial
Automated Induction Motor Monitoring
Internet of Things (IIoT). Sensors
System Using IoT. In: 2023 9th
International. 2021;2:100129. DOI:
International Conference on Advanced
10.1016/j.sintl.2021.100129
Computing and Communication
Systems (ICACCS), Coimbatore, India;
[9] Sharma S, Monga C. Introduction to
2023. pp. 166-169. DOI: 10.1109/
Different Simulation Techniques of
ICACCS57279.2023.10112942
Digital Twin Development. In: Anand A,
[4] Patel RA, Bhalja B, Alam MA.
Sardana A, Kumar A, Mohapatra SK,
Gupta S, editors. Simulation Techniques
Condition Monitoring of Three-Phase
of Digital Twin in Real-Time
Induction Motor. In: 2020 IEEE 1st
Applications: Design Modeling and
International Conference for
Implementation. 2024. pp. 1-24
Convergence in Engineering (ICCE),
Kolkata, India. 2020. pp. 16-20. DOI:
10.1109/ICCE50343.2020.9290540 [10] Grieves MW. Digital twins: Past,
present, and future. In: The digital twin.
[5] Noyjeen E, Tanita C, Panthasarn N, Cham: Springer International
Chansri P, Pukkham J. Monitoring Publishing; 2023. p. 97-121
Parameters of Three-Phase Induction
Motor Using IoT. In: 2021 9th [11] Singh RR, et al. Building a digital
International Electrical Engineering twin powered intelligent predictive
Congress (iEECON), Pattaya, Thailand; maintenance system for industrial AC
2021. pp. 483-486. DOI: 10.1109/ machines. Machines. 2023;11(8):796
iEECON51072.2021.9440368
[12] Santos JF, Lisboa YMD, Santos LH,
[6] Gokilakrishnan G, Dhanamurugan Barbosa E, Manito AR, Fonseca WS,
VMA, Bhasha A, Subbiah R. A Review of et al. Three-Phase Induction Motor
Applications, Enabling Technologies, Monitoring System for Predictive Failure
Growth Challenges and Solutions for Analysis Based on Digital Twin, IIoT and
IoT/IIoT. In: 2023 9th International Finite Element Method. In: 2023 15th
Conference on Advanced Computing IEEE International Conference on
and Communication Systems (ICACCS), Industry Applications (INDUSCON),
Coimbatore, India. 2023. pp. 2241-2250. São Bernardo do Campo, Brazil; 2023.

15
Induction Motors – Latest Research and Applications

pp. 766-773. DOI: 10.1109/ [19] Chen DY. Análise de Dados Com
INDUSCON58041.2023.10374867 Python e Pandas, 1st ed. São Paulo,
Brazil: NOVATEC; 2018; ISBN 978-85-
[13] Espressif Systems. ESP32-WROOM- 7522-699-5
32D and ESP32-WROOM-32U
Datasheet. 2023. Available from: https:// [20] Nelli F. Data Visualization with
encr.pw/22pzH [Accessed: March 25, matplotlib and Seaborn. In: Python Data
2023] Analytics: With Pandas, NumPy, and
Matplotlib. Berkeley, CA: Apress; 2023.
[14] WEG. WCD-GT100 and WCD- pp. 183-257
GT100-ETH. 2022. Available from:
https://static.weg.net/medias/ [21] Haposan D, Hidayat MN, Santosa A,
downloadcenter/ha9/ha2/WEG- Ibrahim MA, Majiid RN. Estimated
GATEWAY-GT100-manual-do-usuario- Twitch Streamer Revenue Using Linear
10009992380-pt.pdf [Accessed: Regression Algorithm. In: 2023 5th
March 25, 2023] International Conference on Cybernetics
and Intelligent System (ICORIS).
[15] YHDC. Split Core Current Pangkalpinang, Indonesia; 2023. pp. 1-5.
Transformer Model SCT013–100 DOI: 10.1109/ICORIS60118.
Datasheet. 2015. Available from: https:// 2023.10352194
en.yhdc.com/comp/file/download.do?
id=941 [Accessed: March 25, 2023] [22] Minipa Electric. Analisador de
Energia Modelo ET-5062. 2020.
[16] Qingxian Zeming Langxi Electronic. Available from: https://www.
ZMPT101B Current-Type Voltage minipaelectric.com.br/images/
Transformer. 2020. Available from: propostatecnica/ET50621313BR.pdf
https://5nrorwxhmqqijik.leadongcdn. [Accessed: April 10, 2024]
com/ZMPT101B+specification-
aidijBqoKomRilSqqokpjkp.pdf [23] Minipa Electric. Multimetro Digital
[Accessed: April 23, 2023] Modelo: ET-2042E. 2016. Available
from: https://www.minipa.com.br/
images/Manual/ET-2042E-1101-BR.pdf
[17] Liang J. Multivariate linear
[Accessed: April 10, 2024]
regression method based on SPSS
analysis of influencing factors of CPI
[24] Replit. Replit. 2023. Available from:
during epidemic situation. In: 2020 2nd
https://replit.com/ [Accessed: April 29,
International Conference on Economic
2023]
Management and Model Engineering
(ICEMME), Chongqing, China; 2020.
[25] Firebase. Firebase Realtime
pp. 294-297. DOI: 10.1109/
Database. 2023. Available from: https://
ICEMME51517.2020.00062
acesse.one/jckNW [Accessed: April 29,
2023]
[18] Pengfei L, Shun T, Liting Y,
Yeniu Q , Jian S. An assessment method [26] Socket.IO. What Socket.IO Is. 2023.
of power system harmonic impedance Available from: https://socket.io/docs/
based on mixed total least squares. In: v4/ [Accessed: April 29, 2023]
2016 IEEE PES Asia-Pacific Power and
Energy Engineering Conference [27] Wang Y, Li S, Meng W. Strong
(APPEEC), Xi’an; 2016. pp. 1694-1698. coupling analysis of fluid–solid for
DOI: 10.1109/APPEEC.2016.7779839 magnetorheological fluid braking
16
Motor Monitoring and Analysis System (SMAM) as a Resource for Predictive Maintenance…
DOI: http://dx.doi.org/10.5772/intechopen.1006914

system. Journal of Intelligent Material

Systems and Structures. 2018;29(8):
1586-1599

[28] França SDS. Web design responsivo:

Caminhos para um site adaptÁvel.
Interfaces Científicas—Exatas e
Tecnológicas. 2015;1(2):75-84. Available
from: https://periodicos.set.edu.br/
exatas/article/view/2220

17