Received 7 July 2022, accepted 20 July 2022, date of publication 25 July 2022, date of current version 29 July 2022.

Digital Object Identifier 10.1109/ACCESS.2022.3193774

Intelligent IoT Wind Emulation System Based

on Real-Time Data Fetching Approach
R. RAJA SINGH 1 , SWAPNIL BANERJEE 1 , R. MANIKANDAN 1 , KETAN KOTECHA 2,

V. INDRAGANDHI 1 , AND SUBRAMANIYASWAMY VAIRAVASUNDARAM 3

1 School of Electrical Engineering, Vellore Institute of Technology, Vellore 632014, India
2 Symbiosis Centre for Applied Artificial Intelligence, Symbiosis International (Deemed University), Pune 412115, India
3 School of Computing, SASTRA Deemed University, Thanjavur 613401, India

Corresponding authors: Ketan Kotecha (head@scaai.siu.edu.in) and Subramaniyaswamy Vairavasundaram

(vsubramaniyaswamy@gmail.com)
This work was supported by the Symbiosis International (Deemed University), Pune, India.

ABSTRACT The fundamental challenge for the development of the wind energy conversion system is the
natural instability of this resource, which is now at the forefront of the debate. The Internet of Things (IoT)
has emerged as a smart and practical approach to renewable energy system management. In addition, the
Internet of Things is an important study area with several possibilities for advancement and demanding work.
To improve the dependability of wind energy conversion systems, researchers may simulate all potential wind
turbine operating scenarios and design power electronics topologies and control algorithms appropriately.
The proposed wind emulation platform is wind-turbine-oriented, with significantly increased traceability
and efficiency when compared to existing systems. The wind emulation system is driven with real-time
wind velocity data. The real-time and forecasted wind data is fetched from global nodes using an IoT cloud
application programming interface. The wind emulation system is modeled in the VEE Pro platform and
integrated with IoT cloud API and FPGA controller. Data obtained from IoT is converted into actionable
information to analyze and investigate wind turbine performance. Also, effective security measures have
been taken for ESP8266 communication with the Host Computer. Adapting the proposed IoT algorithm
with an associated hardware prototype, the condition of the wind energy conversion system and the power
generation capability shall be analyzed in a real-time environment.

INDEX TERMS Wind emulator, DC machine, FPGA controller, Internet of Things, Python, application
program interface, web services.

I. INTRODUCTION structures [3]. However, there are some challenges in

Renewable energy sources are an ideal alternative to replace development and innovation, resulting in high motivation
fossil resources in order to fulfill the ever-increasing energy for researchers to develop numerous tools, models, and
demand. Owing to government incentives and technology algorithms in the target domain [4]. Manufacturers of wind
advancements in recent years, wind-based generation is turbines require special tools and appropriate algorithms to
becoming more widespread across the world. Hence, gener- evaluate new models and to investigate their characteristic
ating power from wind energy helps long-term sustainabil- performance with different, wind velocities, pitch angle,
ity [1]. At present, wind power generation supplies about tip-speed ratio, turbine speed, generator load, etc., [5].
2.5 percent of total energy demand [2]. The conversion of A wind turbine emulator is an essential equipment for
wind to electricity comprises various components including evaluating novel wind energy conversion system designs.
turbine, generator, power electronics circuits, and mechanical Without relying on natural wind resources or actual wind
turbines, this emulator simulates steady-state and dynamic
The associate editor coordinating the review of this manuscript and features in a controlled environmen [6], [7]. There are
approving it for publication was Amin Mahmoudi . several methods for emulating a wind turbine in a laboratory,

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
VOLUME 10, 2022 78253
 R. Raja Singh et al.: Intelligent IoT Wind Emulation System Based on Real-Time Data Fetching Approach

however, the most common method is to use an electric be connected, API provides an interface between the internet
motor coupled directly to the wind generator’s shaft. As DC and the things to reveal previously unseen possibilities. IoT
motor’s speed and torque regulation are simple with great cloud API are the points of interaction between an IoT device
dynamic properties, they are the most popular option for a and the internet and other elements within the network [20].
wind turbine emulator [8], [9]. The armature current of a DC To acquire real-time wind speed at a certain place, it is
motor is controlled by the DC chopper to achieve the desired necessary to use a service that offers data continually when
speed and torque. To analyse the power developed by the sought, as well as data that is readily retrieved. This is
wind turbine at various wind velocities and pitch angles, the where the API comes in; once the connection API string
maximum power output is observed using maximum power is known, the API will make an end-to-end connection
point graphs. As the wind velocity data is the key source of with the service and the device seeking the data. The
data for the wind emulation system, the wind velocity data at API connection string is used to authenticate the device,
various sites across the globe is extracted using an IoT cloud preventing access to the API by unidentified devices. The
application programming interface (API). The collected data continuous data from the server is based on the demands
is provided to the controller, which produces control pulses issued by the device after establishing a connection with the
for the DC chopper. The main objective of this work is API [21].
to create an intelligent IoT wind emulation system using a Literature is mainly focused on IoT securities [13]–[21],
data fetching approach. In addition, the output parameters and few of them are domain-specific, including power
of the wind emulator system are also monitored remotely generation sectors, automotive sectors, and other industry
via IoT. applications [18]–[20]. Finding simple IoT security require-
ments has become a challenge for researchers, and this
A. RELATED WORK impending problem impacts the wind emulation system to a
The paper [9] describes the work of a flexible wind emulator great extent. Identifying the appropriate IoT technology and
prototype developed using a DSPACE controller. Similar security protocols is a different issue. The proposed IoT wind
works with significant improvement are addressed in the emulation system based on a real-time data fetching approach
literature [10], [11]. On a global scale, individual embedded addresses the above-mentioned shortcomings while retaining
devices with internet access are capable of communicating their positive attributes. The overall goal of this work is to
with one another, with other services, and with people. provide enhanced security for cyber-attacks and to provide
This entire system is defined as the Internet of Things. security recommendations that are not reliant on protocol
Without a human interface, IoT can transmit information changes and are therefore usable by researchers working in
using unique identities (UIDs). By improving information the field of real-time data transmission.
availability, this degree of connection can improve the In this article, the open-source weather data service
system’s dependability, efficiency, and sustainability [12]. provider ‘‘OpenWeatherMap’’ is used to acquire real-time
Moreover, today’s modern IoT data is analyzed at near- wind data of a specific location. The data obtained from
real-time speeds, and one might argue that successful IoT OpenWeatherMap through API is in JavaScript Object
deployment will be a crucial differentiator for the winners in Notation format and Extensible Markup Language format.
the renewable energy sector’s next phase of expansion [13]. The wind data is saved and then sent to the ThingSpeak
A recent work [14] proposed honeynet-based threat detection Platform using ThingSpeak API which is a public API. The
and situational awareness for secure and resilient artificial graphical user interface (GUI) has been built using Tkinter
IoT. To enhance the secured data transmission through IoT (a GUI building library in python). The wind speed value
numerous advancements are recommended [15], [16]. As a will then be sent to an IoT cloud Platform like ThingSpeak
further step toward lower power consumption, higher energy using the ThingSpeak API. Fig. 1 shows the flow of data
efficiency, and less electromagnetic pollution, the trends in starting from the remote computer. The Remote computer
IoT communications are promoting the use of low-cost and initially sends out a ‘GET’ request to the OpenWeatherMap
easy-to-integrate devices. Ref. [17] explains how real-time API for collecting the wind data. The wind data received is
control and condition monitoring can be achieved through in the form of JSON Object which contains the wind speed
the means of IoT. Similarly, the application programming data. After the needed windspeed data is extracted from the
interfaces are the application enabler, and internet of things JSON Object, the remote computer then sends the wind speed
devices would be useless without them. An application data as a String Object to a ThingSpeak channel using the
programming interface (API) is a set of instructions that ThingSpeak API. The ESP8266 Wi-Fi Module then collects
enables communication between two programs or two com- the wind speed data from the ThingSpeak channel and using
puters [18]. API initially functioned to exchange information UART-Serial communication transfers this information to the
between computer programs. The development of web API Host Computer (inside the Emulator Control).
started in the early 2000s, and the technology has seen The rest of the section explains the main components
consistent growth ever since [19]. Essentially every company involved in collecting real-time wind data for any specified
uses web applications or web services API to develop their location for the wind emulator and also the communication
software. By exposing data that enables multiple devices to interfaces among them.

78254 VOLUME 10, 2022

R. Raja Singh et al.: Intelligent IoT Wind Emulation System Based on Real-Time Data Fetching Approach

FIGURE 1. Interpretation of IoT wind emulation system.

B. SIGNIFICANCE OF THE ARTICLE wind emulation system with data fetching, integration, and
The transfer of data from web-based weather API providers emulation. The proposed real-time IoT-based wind emulated
to host PC is playing a significant role in the proposed system is then addressed in detail throughout section IV.
work. Innovative use of IoT in the realm of renewable energy In section V, the experimented results of the proposed IoT
monitoring and wireless data collecting is presented in this cloud API-based wind emulator system are presented. and
research. It is an updated and extended version of the wind finally concluded with future scope in section VI.
turbine emulation system that is given as an alternate option
for collecting wind data from the deployment location and II. MODELLING OF WIND EMULATION SYSTEM
simulating turbine output using a power curve model. The The extensive mathematical modelling of the wind turbine
developed wind turbine has its own IoT device that can show and DC motor with speed control is required to imitate the
its electrical and mechanical output parameters. Moreover, wind turbine system, and it is presented in this section. Using
the created system can show temporal synchronization and the IoT cloud API and ThinkSpeak, dynamic-wind velocity
power analysis of duty-cycled IoT nodes. The following are data is retrieved from OpenWeatherMap and imported into
the work’s most important contributions: the VEE Pro environment. Similarly, using the controller’s
1) Detailed development of data-driven wind emulation
ADC channel, real-time rotor speed data is acquired from a
model in VEE Pro platform.
quadrature encoder pulse sensor. The turbine reference torque
2) Implementation of intelligent data fetching approach
(τr ∗) is calculated using the wind velocity and rotor speed
using IoT cloud API for wind emulation system.
data input into the wind turbine model [8]. The obtained τr∗ is
3) A real-time wind emulation platform that allows
compared with the real torque and the error signals are fed to
researchers to test various downwind topologies
the PI controller. The output signals from the PI controller are
including pitch angle control.
used to control the DC machine’s speed and torque through
4) A user interactive GUI for acquiring wind data
the Class A chopper.
from specific locations to interface with the emulator
for obtaining the different electrical and mechanical
parameters A. MODELLING OF WIND EMULATION SYSTEM
5) Forecasted wind speed is fetched through the Open- The purpose of modelling the wind emulator is to study the
WeatherMap service and API URL, this will help the real-time behaviour of the wind turbine in lab practice and
user understand the power requirement. to analyses the mechanical energy conversion to electrical
6) Effective security measures have been taken for energy at different wind velocities and pitch angles for
ESP8266 communication with the Host Computer. extracting more power from the wind blades by adjusting
Overall, this manuscript is prepared as, section II provides the the pitch angle for obtaining the maximum power point [9].
detailed mathematical modeling of the wind emulation sys- The model relies on the turbine’s steady-state characteristics.
tem with wind turbine mode, DC motor model, and PI control. As the power train’s rigidity is indefinite, the turbine’s friction
Section III depicts the working of the IoT cloud API-based coefficient, as well as momentum, must be mixed with those

VOLUME 10, 2022 78255

 R. Raja Singh et al.: Intelligent IoT Wind Emulation System Based on Real-Time Data Fetching Approach

At the terminals of the armature;

dia
Ua = Ra ia + La +E (6)
dt
Here, back emf, E = K ϕm and K ϕ is the torque. On the
other hand, the armature current is given by,
dia 1
= (Ua − Ra ia − K ϕm ) (7)
dt La
At the excitation inductor, the voltage Uf and Rf are
constants, hence the excitation current If is given by,
Uf
If = (8)
Rf
FIGURE 2. Characteristics curv between Cp and λ. The equation governs the kinematics of the DC motor,
dm
J + f m = τm − τr (9)
dt
of the turbine linked to the generator. The turbine’s generated
Here, f is the friction, τr is the resistant torque. Similarly, the
power is provided as,
mechanical torque τm is given by
ρA 3 τm = K ϕia
Pm = cp (λ, β) v (1) (10)
2 wind
The open-loop transfer function (OLTF) of the DC motor is
Here, Pm_pu is the mechanical output power in p.u of nominal given by,
power, cp_pu is the power coefficient in p.u of nominal power Kϕ
coefficient, vwind_pu is the wind velocity in p.u of the based fRa +Kϕ2
wind speed. The projected wind speed in m/s is used as the H0 (p) =   (11)
JLa (Ra J +fLa
baseline wind speed,kp is the power gain in p.u. The power fRa +Kϕ2
p2 + fRa +Kϕ2
p+1
gain is less than or equal to 1. A generic expression is
used to represent cp (λ, β) and it is depending on the turbine Also, it is possible to write it as a function of the electrical
parameters constant τe and the mechanical constant τm . The function as,
K
Pm_pu = kp cp_pu v3wind_pu (2) H0 (p) = (12)
(1 + τe p)(1 + τm p)
cp (λ, β) = c1 (c2 /λi − c3 β − c4 )e −c5
/λi + c6 λ (3) A PI regulator is used to create the voltage that must be fed
into the DC motor armature to achieve the same rotational
With, speed as the wind turbine. As a result, the PI regulator is
1 1 0.035 chosen since the proportional operates on speed while the
= − 3 (4) integrator reduces static error. R(p) is the regulator given as,
λi λ + 0.08β β +1
1 + τp
 
Ki
Since (3) depends on the wind turbine rotor type, the R(p) = Kp + = Kp (13)
p τp
coefficients c1 − c6 can be different for various turbine
designs. Furthermore, coefficients c2 and c6 also depends The modelling of the PI controller is shown in Fig. 3, and it
on wind velocity and angular velocity of the turbine. is described as,
Z
The coefficient betweenc1 andc6 are chosen according to
c(t) = kp e(t) + ki e(t)dt (14)
references [22], [23], c1 = 0.5176, c2 = 116, c3 = 0.4,
c4 = 5, c5 = 21, c6 = 0.0068. The characteristics curve Here, e(t) = τr∗ (t) − τr (t), c(t) controlled output (reference
between cp and λ for different ranges of pitch angle β is voltage) for the comparator. Laplace transform is applied to
plotted in Fig. 2. The maximum power coefficient value is the above equation for obtaining the transfer function as,
0.48 and it is achieved with β = 0◦ andλ = 8.1. The value of
λ is defined as a nominal value (λ_nom). C(s) Ki
G(s) = = Kp + (15)
E(s) S
B. MATHEMATICAL MODELLING OF WIND TURBINE The proposed wind emulation process with the machine,
The DC motor, in reality, is the most straightforward machine converter, and controller is executed. The laboratory equip-
to operate, requiring either choppers or rectifiers. The field ment and converter setup are modified for use with the
winding flux ϕf is constant since the DC motor model with a proposed IoT-based wind simulation system. Additionally,
distinct excitation is used. the PI controller is tuned using the adaptive trial-and-error
tuning approach, which entails choosing the Kp and Ki
ϕf = kϕif (5) parameters.

78256 VOLUME 10, 2022

R. Raja Singh et al.: Intelligent IoT Wind Emulation System Based on Real-Time Data Fetching Approach

FIGURE 3. Modeling of wind turbine and DC motor control.

III. IOT CLOUD API BASED DATA FETCHING FOR WIND which is mentioned in (1). This relationship is the foundation
EMULATION SYSTEM for creating a wind turbine emulator. There are several
The kinetic energy of all those trillions of air molecules methods for simulating a wind turbine in a laboratory, but
moving together in mass movement termed wind provides the most common method is to use an electric motor coupled
power for a wind turbine. Wind energy appears to be a directly to the wind generator’s shaft [30]. Because, the DC
logically simplistic technology: the wind propels a set of motors are armature current regulated and have a direct link
turbine blades, which spin a mechanical shaft coupled to a with the torque produced by the machine, which makes the
generator, that generates electricity. The rotor blades, gear- DC motor control simple with great dynamic properties.
box, generator, nacelle, and tower are the main components Hence DC motors are the most popular option for a wind
of wind. [24]. Significant technical and material challenges turbine emulator. In this work, a DC motor is controlled
arise during the realization of this conceptually simple by class-A DC to DC chopper. Using a DC motor, the
technology in practice. The general aim of wind energy author built an experimental setup to mimic the dynamic and
engineering is to minimize equipment costs, enhance wind static features of a genuine wind energy conversion system.
energy capture, reduce maintenance, improve dependability, Because of readily available microelectronics, lightweight
and extend system and component lifespan [25]. This communication protocols, open-source servers, visualisation
requires significant efforts in aerodynamics design, materials dashboards, and Web development tools, the Internet of
engineering, construction, endurance, power electronics, Things is a promising approach for remote monitoring. This
controls, and manufacturing techniques. The fundamental section presents the deployment process of an IoT based wind
design of wind turbine aerodynamic effects is extensively emulation system, which involves several stages including
available in the literature [26]–[29]. The mechanical power data fetching, API architecture, web services, and emulation
output of a wind turbine is a function of characteristics such system. Fig. 4 depicts the whole process of IoT cloud API
as wind speed, air density, turbine rotor speed, and swept area, based data fetching for a wind emulation system.

VOLUME 10, 2022 78257

 R. Raja Singh et al.: Intelligent IoT Wind Emulation System Based on Real-Time Data Fetching Approach

FIGURE 4. Detailed structure of IoT cloud API integrated wind emulation system.

A. IOT WIND DATA FETCHING UNIT services across a wide range of application domains. Several
In the IoT data fetching system, the wind velocity data from IoT cloud providers are currently entering the market to
the user’s preferred global point is fetched instantaneously. leverage suitable and specific IoT based services cloud
The wind data fetching unit consists of IoT cloud, gate- platforms can be classified according to their appropriateness
way, API, and web services, which are addressed in this into specific application domains. Such as in the paper [32],
section the authors have used an IoT cloud system to provide
suitable temperature and adequate moisture following the
1) INTERNET OF THINGS growth stages of rice seedlings. In paper [33], the authors
The Internet of Things envisions the total merger of numerous have proposed an IoT based system to detect impurities
‘‘things’’ while employing the internet as the communication and levels of oil for the engine. The IoT fetching system
system’s backbone to create intelligent interactions between which we will be using employs the Wi-Fi based network
people and their surroundings [31]. Cloud, being the crucial layer. In this architecture, the real-time wind data will be
component of IoT, provides essential application-specific collected from the weather API service provider as shown in
78258 VOLUME 10, 2022
R. Raja Singh et al.: Intelligent IoT Wind Emulation System Based on Real-Time Data Fetching Approach

OpenWeatherMap API is of REST architecture. REST APIs

are stateless indicating that the API will not store data or
status between requests. REST APIs can make use of both
XML and JSON data formats. However, in this work, the
response from the REST API endpoint is received in JSON
object format.

3) WEATHER WEB SERVICES

OpenWeatherMap is a service that provides real-time weather
data of a place using API. Weather Stations distributed
across the world have instruments and sensors such as
thermometer (temperature data), hygrometer (humidity),
anemometer (wind speed), that transmit this data to servers
of OpenWeatherMap [36]. The data collected is stored in
the OpenWeatherMap database and can be accessed using
their API. The data obtained using OpenWeatherMap API
is in JSON and XML format. Each weather data of the city
or state called is a parameter of the JSON data [37]. The
JSON format data is used due to its simplicity. The wind field
data contains wind.deg (wind direction in degrees), wind.gust
(wind gust data), and wind.speed (wind speed of an area).
For this work, the ‘speed’ parameter inside the ‘wind’ field
is called from API to obtain the wind speed. Fig. 5 shows the
data communication flow between cloud and controller. The
wind speed in meters per second obtained after selecting the
FIGURE 5. Data communication flow between cloud and controller. place will be considered as an API call using the place (which
might be a city, state, or village). To request information using
the API, the HTTP method must be used, named ‘GET’. The
Fig. 5. The end node of the proposed system is called wind city name and API key are concatenated to the URL and
data fetching units (WDFUs). In this system, the location ‘GET’ method is used to retrieve the data. The graphical user
data is processed and transmitted to a Wi-Fi gateway using interface has been built using Tkinter (a GUI building library
2.4 GHz (802.11 n.). The wireless communication between a in python). The wind speed value will then be sent to an IoT
WDFU and a Wi-Fi gateway is a single-hop link. The Wi-Fi cloud Platform like ThingSpeak using the ThingSpeak API.
gateways connect to the Internet via a network switch and
from there to a central IoT-WDFU server. The IoT enabled B. INTEGRATING UNIT
gateway’s high-level application collects wind physical data In the proposed work, ThingSpeak is chosen as the cloud
and automatically executes the duty of transferring the data to platform for monitoring wind speed data. ThingSpeak
the cloud with fewer development overheads than traditional (https://thingspeak.com) is a public cloud-based, IoT data
embedded systems. platform. Using an Open API, ThingSpeak enables us to
collect real-time and analyse real-time data. Various third-
2) ARCHITECTURE OF APPLICATION PROGRAMMING party platforms, including Arduino, ioBridge and Matlab,
INTERFACE (API) can also be integrated with ThingSpeak for visualizing,
On the basics of access privileges, there are four types of web monitoring, and storing data [38]. The word ’Channel,’ which
APIs; internal APIs, partner APIs, open APIs, and composite has fields for data, location, and status for various sensed data,
APIs [34]. Among these open API is used to develop this is the most important component of ThingSpeak capability.
work because we need access to information that is openly Once the channels have been created in ’ThingSpeak,’ the
and freely accessible to everyone. These open types of APIs data can be implemented, or the data may be processed and
also known as public API are accessible to everyone inside visualized using Matlab, and the user can be notified about the
and outside an organization. Similarly, there are three types data with tweets and other forms of alerts [39]. Fig. 5 shows
of API architectures and protocols; representational state the real-time wind speed data fetched from OpenWeatherMap
transfer (REST) API is an architecture in which the server website API to the ThingSpeak cloud channel field. The
responds solely to the client’s requests; Remote procedure transfer of data securing from web-based weather API
call (RPC) APIs are used when the API calls a known providers to host PC is playing a significant role in the
procedure on the web service; Simple object access protocol proposed work. In the proposed work, the system comprises
(SOAP) APIs allows computers built on different operating of 5 crucial components. The data flow starts from the remote
systems to communicate with each other [35]. Among these computer which will request the OpenWeatherMap Service

VOLUME 10, 2022 78259

 R. Raja Singh et al.: Intelligent IoT Wind Emulation System Based on Real-Time Data Fetching Approach

FIGURE 6. Communication interface between I/O and Application.

for windspeed for a particular location using their open API.

The service will then provide a JSON object containing
the windspeed. This information is further communicated
to ThingSpeak.com which is a cloud platform for analytics.
ESP8266 Wi-Fi module receives the windspeed data from
ThingSpeak and using UART-Serial communication sends
the data to the host computer. Using development of Python
interface to acquire data through the cloud and send it to FIGURE 7. Control process of wind emulation unit.
personal computer in remote location. In the remote computer
python code environment to be setup to collect wind data.
The developed python code collected the rea-time wind speed the system; (i) fixed wind mode, analysis of wind energy
from weather API service provider. The app developed using conversion system by varying the system parameters for a
python environment transfers the wind data to the cloud. particular wind velocity, (ii) variable wind mode, analysis
The host system interfaced with Wi-Fi module will receive of wind energy conversion system by varying the wind
the wind speed data from the cloud platform. It is further velocity for the particular design parameters. Fig. 7 shows
transferred to VEE Pro IDE to control the FPGA development the detailed control process of the wind emulation unit. Once
board. The wind velocity data collected by Wi-Fi module has the process is initialized with design parameters including,
to be transferred to wind turbine model which is developed pitch angle (β), swept area (A), power coefficient (Cp), and
using VEE pro software. That can be achieved by connecting air density (ρ), the controller provides enable signal to the
the Wi-Fi module in the COM port of the host PC. Further field excitation system. Likewise, the switching pulse of
from the COM (USB) port with the help of VISA driver the ready input of the step-down chopper is enabled by the
technology the data will be sent to the VEE pro application. controller for driving the armature current of the DC motor.
VISA (Virtual Instrument Software Architecture) is a driver The user has requested the selecting operating mode (Fixed
technology that is more abstract than instrument drivers. wind velocity mode or variable wind velocity mode). If the
An input, output (I/O) library is installed in the computer user opts for fixed mode, the wind velocity should be fed
to connect with and operate the vector adapter. I/O libraries manually, between8 m/sec and 14 m/sec. If the user opts
are a set of libraries, APIs, and utility applications offered for variable mode, the wind velocity data will be fetched
by Keysight. The Keysight I/O libraries package allows directly from the cloud. Under this model, the output profile
for instrument connection in a variety of development of various parameters including torque, speed and power shall
contexts, including USB, LAN, PXI, AXIe, and VXI test be obtained. Finally, the system will be stopped by reducing
instruments from various manufacturers. The communication the load and de-energizing the chopper input power with the
interface flow between I/O and the application is shown in external interrupt signal.
Fig. 6
IV. EXPERIMENTAL SETUP AND GUI INTEGRATION
C. EMULATING UNIT WITH FORECASTING AND SECURITY SERVICES
The purpose of this emulation unit is to analyse the wind In this section, the detailed process of implementing the
energy generation system’s performance through the char- experimental setup as well as the development of the
acteristic curves. There are two possibilities to investigate graphical user interface is discussed.

78260 VOLUME 10, 2022

R. Raja Singh et al.: Intelligent IoT Wind Emulation System Based on Real-Time Data Fetching Approach

to the controller. Depending upon the sensor’s response

and the IoT cloud input, the Xilinx FPGA Spartan 6LX25
controller generates the switching pulses. The VEE (Visual
Engineering Environment) Pro software is used to deploy
the developed maximum power point control algorithm in
the controller. Agilent Technologies’ VEE Pro is a graphical
dataflow programming software development environment
for automated testing, measurement, data analysis, and
reporting. It works well with test and measurement equipment
and the virtual instrument software architecture (VISA) is
a commonly used application programming interface (API)
for interfacing with instruments from a computer. The
communication architecture of VISA with instrument and
software application. The VISA standard includes specifica-
tions for communication with resources over T&M specific
I/O interfaces such as GPIB and VXI. The IoT system
is integrated with an FPGA controller for providing the
FIGURE 8. Experimental arrangement of IoT based wind emulation.
instantaneous wind velocity data of the selected place and to
visualize the output of the wind emulation system. The power
TABLE 1. Machine, converter and controller specifications. meter (WT 333E) measures the output electrical quantities
of the AC generator and is used to calibrate the output of
the emulation system. The ESP8266 provides a complete
and self-contained Wi-Fi networking solution, allowing it
to either host or delegate all Wi-Fi networking functions to
a supplemental application processor. It is frequently used
in many IoT applications and refers to software rather than
the creation of kits that enable access to these GPIOs of
the ESP8266. Because the ESP8266 is the only ’application
processor’ in the device, it boots up right away from an
external source. One serial port on the ESP8266 is allocated
to communication with the computer to which the ESP is
attached. A serial port is USB, which stands for Universal
Serial Bus. This USB connection on the ESP8266 is separated
into two digital pins, GPIO 0 and GPIO 1, which may be
utilized in projects that entail serial communication with
devices other than the computer, thanks to onboard hardware.

B. DEVELOPMENT OF GRAPHICAL USER INTERFACE

A. EXPERIMENTAL ARRANGEMENT APPLICATION
The experimental arrangement of the proposed IoT based Fig. 9 depicts the Tkinter-based GUI for fetching wind
wind emulation system is shown in Fig. 8. The system speed data. Tkinter, a Python library, is used to develop
consists of a 3.7 kW DC drive setup (prime mover) coupled an interactive application. In this application, various node
with a 3 kVA AC generator (test machine). Also, the machine points have been established in global locations. The real-
setup is integrated with a quadrature encoder pulse (QEP) time wind speed data, as well as time coordinates, are
sensor and load cell for measuring the speed and torque collected from the respective global nodes when navigating
respectively. The rating of the machine, converter, and and selecting.
controller are given in Table 1. In the proposed system, the In Fig. 9, a popup window displays real-time data retrieved
DC motor is driven by a chopper converter and excited by an from Esbjerg, Denmark (55.4765◦ N, 8.4594◦ E). This data
external excitation system. The power circuit of the converter is supplied to the controller using the procedure outlined
consists of a three-phase diode bridge rectifier with DC link in section 3.1. Whenever users click the wind emulator
capacitors connected to IGBT switches and inductors. Also, icon, a new popup window appears with the wind emulator
the converter is integrated with drivers, sensors, and signal output parameters, such as output power, speed, and torque.
conditioning circuits. The input electrical quantities of the By default, all of these data are refreshed every 15 seconds;
converter are measured using the multifunction meter (MFM) however, if the data is to be renewed when the user requests
and the output electrical quantities are sensed and fed back it, the user must click the refresh button.

VOLUME 10, 2022 78261

 R. Raja Singh et al.: Intelligent IoT Wind Emulation System Based on Real-Time Data Fetching Approach

FIGURE 9. Front panel of graphical user interface for wind data fetching and monitoring.

Usage of API key will prevent the Distributed Denial

of Service attacks (DDoS). Once a channel is created in
ThingSpeak, a Read and Write API key is assigned to it.
The communication to the API server has been achieved
securely using HTTPS to prevent Man-in-the-Middle attacks.
ThingSpeak has incorporated an API rate limiter which can
be a strategy to reduce the frequency of requests sent to the
server hence it can limit access of bots especially when a huge
number of requests is sent. API caching is also implemented
in the ThingSpeak API to enhance the performance of the
FIGURE 10. Forecasted wind data fetched from OpenWeatherMap.
application. Hence these mechanisms can help in reducing
the most prevalent attacks on APIs and secure the data from
the wind emulator. The communication between the ESP8266
C. FORECASTED WIND DATA
module and the host computer has also been secured. The
There is extreme uncertainty in weather, and having an ESP8266 is a Wi-Fi system on chip (SoC) with a fully
accurate forecast of the weather is of utmost importance. integrated TCP/IP protocol stack. Due to its connectivity,
Forecasting the windspeed will help the user understand the it is susceptible to intrusions such as eavesdropping and
power requirement of the wind turbine in the upcoming days unauthorized access to the device or the data. To ensure
and adjust accordingly to save power. This section deals the data passed by the ESP8266 is secured, the data is
with Forecasted wind data fetched from OpenWeatherMap. AES (Advanced Encryption Standard) encrypted so that the
The OpenWeatherMap service also provides a forecast of message cannot be decoded very easily. The process is as
the weather of any given location only by specifying the follows, the string data is converted into binary data with the
longitude and latitude of the location in the API URL. help of UTF-8, and the binary data is AES encrypted and
To obtain the wind speed value of each day in the forecast then encoded using base64 encoding. The host computer has
we make use of a for loop to iterate over the ‘daily’ key. a decrypt function that will decode the base64 payload into
Fig. 10 shows the 12 windspeed data points forecast that is binary ciphertext and then decrypt the binary ciphertext to
obtained. Day 1 denotes the previous day; Day 2 is the current get the original message.
day while from Day 3 to Day 5 is the windspeed forecast.

D. SECURITY SERVICES V. WIND TURBINE EMULATION AND IOT RESULTS

ThingSpeak like other web APIs uses a unique API key to In this section, the experimented results of the proposed wind
authorize the users to secure the content of the application. emulator system are presented. To analyze the wind turbine

78262 VOLUME 10, 2022

R. Raja Singh et al.: Intelligent IoT Wind Emulation System Based on Real-Time Data Fetching Approach

FIGURE 11. Maximum power point of wind emulation system for different velocities and different pitch angle.

with the new designs and control this realization is made with TABLE 2. Maximum power at different pitch angle and wind velocity.
real-time wind data.

A. MAXIMUM POWER POINT OF WIND EMULATION

SYSTEM
The performance of wind turbines can be evaluated using
power curves. For realizing the maximum power point of the
proposed system, the control process of the wind emulation
system is followed, as discussed earlier (Fig. 7). Selecting the
fixed wind velocity mode, (i.e.), feeding the wind velocity
(Vw ) constant, the load torque (τ _L ) is varied from zero
to maximum. During this process, the pitch angle (β) is
also maintained constant. The same procedure is repeated
for different wind velocities from 8 m/sec to 14 m/sec with
the interval of 2 m/sec, and different pitch angles from 0 to for different velocities and different pitch angles. Based on
6 degrees with the interval of degree. All these results are the collected data, the maximum power point for various
obtained with the maximum power coefficient (C_(p_maz) = wind velocities and pitch angles is calculated and reported
0.48) and with the nominal tip-speed ratio (λ_nom = 8.1). in Table 2. The results reveal that the wind turbine’s output
Pitch angle directly affects the power coefficient of the wind power is influenced directly by the pitching angle and wind
turbine. The power coefficient is the percentage of energy velocity. The maximum power point must also be attained by
converted by the wind turbine. Fig. 11 shows the power altering the reference speed in the controller, as can be seen
curves plotted between power generated and turbine speed from the data. This aids in the development of various control
with a maximum power point of the wind emulation system algorithms aimed at maximizing power output.

VOLUME 10, 2022 78263

 R. Raja Singh et al.: Intelligent IoT Wind Emulation System Based on Real-Time Data Fetching Approach

FIGURE 12. Experiment results of wind emulation parameters.

FIGURE 13. Wind emulator results in ThingSpeak platform; (a) wind velocity, (b) turbine speed, (c) output power, (d) torque.

B. PARAMETER VARIATION WITH CHANGE IN WIND The PI controller makes the DC machines’ torque follow
VELOCITY the reference torque that can be observed in Fig. 12b. Also,
The performance of wind turbines also can be evaluated from the model, the tip speed ratio and power coefficient are
using output parameters for changes in wind velocity. For extracted through a digital to analog converter and acquired
realizing the wind turbine’s performance with change in in Fig. 12b. This aids in the development of various control
parameters including wind velocity, and pitch angle, given algorithms aimed at maximizing power output.
in Table 2. The control process of the wind emulation
system is adapted, (Fig. 7). The variable wind mode is C. IOT OUTPUT RESULTS
selected to analyse the developed wind energy conversion ThingSpeak is chosen as the cloud platform for monitoring
system while varying the wind velocity for the particular wind speed data. ThingSpeak (https://thingspeak.com) is
design parameters. Fig. 12 shows the 10 minutes results of a public cloud-based, open IoT data platform. With an
electrical and mechanical parameters concerning the dynamic Open API, ThingSpeak allows for real-time data collecting,
wind velocity fetched from Esbjerg, Denmark (55.4765◦ N, analysis, and actuation. The word ’Channel,’ which has fields
8.4594◦ E). As the system operated in islanded mode, the for data, location, and status for various sensed data, is the
generator speed varied with the input wind velocity, which most important component of ThingSpeak capability. Once
can be observed in Fig. 12a. Similarly, as the load is main- the channel has been formed in ’ThingSpeak,’ data can be
tained constant, the power output profile of the developed implemented, or the data may be processed and visualized
system is varied with reference to the speed and wind velocity. using Matlab, and the data can be responded to with tweets
Depending upon the developed model in section 2, the and other types of alerts. Fig. 5 shows the real-time wind
reference torque is generated with different wind velocities. speed data fetched from OpenWeatherMap website API to the

78264 VOLUME 10, 2022

R. Raja Singh et al.: Intelligent IoT Wind Emulation System Based on Real-Time Data Fetching Approach

ThingSpeak cloud channel field. The transfer of data securing the client. A globally unique identifier (GUID) generator
from web-based weather API providers to host PC is playing will generate a 128-bit or 256-bit secret key using base64
a significant role in the proposed work. The remote computer encoding. This is based on the possible unique hash values
transfers the data from the emulator to the ThingSpeak server that can be generated using an n-set of rounds. On the other
with 15 second sampling rate for 10 minutes period. The side, by integrating the output of wind turbines into a common
channel fields have been written by making API Requests to bus, energy waste will be eliminated. This IoT-enabled real-
the ThingSpeak API using a unique Write key. Wind emulator time wind turbine simulation system’s regenerative control
results including wind velocity, turbine speed, output power, will improve security whilst also reducing electricity waste.
and actual torque are shown in Fig. 13. Similar to the pitch angle control, the yaw angle variation also
can be studied.
VI. CONCLUSION WITH FUTURE SCOPE
The wind power industry’s fast growth depends on the NOMENCLATURE
Symbol Explanation
dependable functioning and steady control of wind turbines.
τ∗ Turbine Reference Torque.
The proposed method ensures the IoT based wind emulator
Pm Mechanical Output Power.
works with a real-time wind velocity of different global
cp Power Coefficient.
locations. Also, a hardware prototype has been developed to
vwind Wind velocity.
test the controller and embedded algorithms. The obtained
kp Power gain.
result demonstrates the high performance of the proposed
β Pitch angle.
IoT based wind emulation system that fetches the wind
ϕf Field winding flux.
data from the cloud, as well as feeds the wind emulator
E Back emf.
output parameters to the cloud. The system also provides
Uf , Rf Constants.
a forecast of the wind speed based on which the power
If Excitation current.
consumption of the wind emulator can be estimated. Also,
f Friction.
for the transmission of data securely inside the system,
τr Resistant torque.
the data from ESP8266 is AES encrypted. Experimental
τm Mechanical torque.
and simulation results are found to be in good agreement.
H0 (p) Open-loop transfer function (OLTF).
In this way, authenticated remote users can monitor the
τe Electrical constant.
conditions and take a proper and timely asset management
R(p) Regulator.
decision to avoid any potential consequences to the wind
A Swept area.
energy conversion system. Also, the proposed system is based
ρ Air density.
on an intelligent decentralized application with identical
kW Kilowatt.
mechanical characteristics to a real wind turbine, and it also
kVA Kilovolt-Ampere.
produces adjustable wind energy equivalent. As a result,
Nm Newton-meter.
it can be applied to the design, estimating, and testing of the
hp Horsepower.
wind power generation system, considerably increasing the
p.u Per-unit.
efficiency and validity of research development.
τ _L Load torque.
The future scope of this research will focus on improving
rpm Revolution per minute.
IoT security since more advanced and proactive measures are
GHz Gigahertz.
required. The signature based intrusion detection system can
be used to detect specific attack patterns in the IoT network.
This will help in gathering data on new malware attacks.
ACKNOWLEDGMENT
Once, the data is gathered about the new malware, it will be
The authors thank the Vellore Institute of Technology (VIT),
easier to predict whether or not there might be an attack. Also,
Vellore, India, for providing ‘VIT SEED GRANT’ for
complex event processing technology can be deployed along
carrying out this work in the Advanced Drives Laboratory.
with ML techniques for detecting malicious IoT devices
connected to the system. Moreover, adopting a private API
as compared to a public one will ensure greater safety. The REFERENCES
use of AI approaches will be made in the context of wind [1] Z. Xu, J. Wei, S. Zhang, Z. Liu, X. Chen, Q. Yan, and J. Guo, ‘‘A state-
of-the-art review of the vibration and noise of wind turbine drivetrains,’’
velocity forecasting. The user will be able to comprehend the Sustain. Energy Technol. Assessments, vol. 48, Dec. 2021, Art. no. 101629,
power required with the aid of a wind speed forecast. The doi: 10.1016/j.seta.2021.101629.
wind speed at a certain place can be retrieved via API. Private [2] S. Dawn, P. K. Tiwari, A. K. Goswami, A. K. Singh, and R. Panda,
‘‘Wind power: Existing status, achievements and government’s initiative
API is used within an organization or to a limited group of towards renewable power dominating India,’’ Energy Strategy Rev., vol. 23,
people such as private APIs usage in banking operations, pp. 178–199, Jan. 2019, doi: 10.1016/j.esr.2019.01.002.
private home IoT devices, etc. The API key is randomly [3] W. Cao, Y. Xie, and Z. Tan, ‘‘Wind turbine generator technolo-
gies,’’ in Advances in Wind Power. London, U.K.: Intechopen, 2012.
generated and SHA 256 results in a secure message and [Online]. Available: https://www.intechopen.com/chapters/38933, doi:
secret key. This secret key will be shared by the service and 10.5772/51780.

VOLUME 10, 2022 78265

 R. Raja Singh et al.: Intelligent IoT Wind Emulation System Based on Real-Time Data Fetching Approach

[4] J. D. M. De Kooning, A. E. Samani, S. De Zutter, J. De Maeyer, [24] M. G. Molina and P. E. Mercado, ‘‘Modelling and control design of pitch-
and L. Vandevelde, ‘‘Techno-economic optimisation of small wind controlled variable speed wind turbines,’’ in Wind Turbines. London, U.K.:
turbines using co-design on a parametrised model,’’ Sustain. Energy IntechOpen, 2011. [Online]. Available: https://www.intechopen.com/
Technol. Assessments, vol. 45, Jun. 2021, Art. no. 101165, doi: chapters/14810, doi: 10.5772/15880.
10.1016/j.seta.2021.101165. [25] A. Elia, M. Taylor, B. Ó Gallachóir, and F. Rogan, ‘‘Wind tur-
[5] V. Sohoni, S. C. Gupta, and R. K. Nema, ‘‘A critical review on wind bine cost reduction: A detailed bottom-up analysis of innovation
turbine power curve modelling techniques and their applications in wind drivers,’’ Energy Policy, vol. 147, Dec. 2020, Art. no. 111912, doi:
based energy systems,’’ J. Energy, vol. 2016, pp. 1–18, Jun. 2016, doi: 10.1016/j.enpol.2020.111912.
10.1155/2016/8519785. [26] P. J. Schubel and R. J. Crossley, ‘‘Wind turbine blade design,’’ Energies,
[6] K. S. Ajirlo, P. H. Tari, K. Gharali, and M. Zandi, ‘‘Development of vol. 5, no. 9, pp. 3425–3449, Sep. 2012, doi: 10.3390/en5093425.
a wind turbine simulator to design and test micro HAWTs,’’ Sustain. [27] J. J. Chattot, ‘‘Wind turbine aerodynamics: Analysis and design,’’ Int. J.
Energy Technol. Assessments, vol. 43, Feb. 2021, Art. no. 100900, doi: Aerodyn., vol. 1, nos. 3–4, p. 404, 2011, doi: 10.1504/ijad.2011.038853.
10.1016/j.seta.2020.100900. [28] T. Mojtaba, S. Tahmine, R. Kiana, and A. Pouria, ‘‘Aerodynamic optimal
[7] D. Zouheyr, B. Lotfi, and B. Abdelmadjid, ‘‘Improved hardware imple- design of wind turbine blades using genetic algorithm,’’ Energy Equip.
mentation of a TSR based MPPT algorithm for a low cost connected Syst., vol. 2, no. 2, pp. 185–193, 2014.
wind turbine emulator under unbalanced wind speeds,’’ Energy, vol. 232, [29] H. Snel, ‘‘Review of aerodynamics for wind turbines,’’ Wind Energy, vol. 6,
Oct. 2021, Art. no. 121039, doi: 10.1016/j.energy.2021.121039. no. 3, pp. 203–211, Jul. 2003, doi: 10.1002/we.97.
[8] M. Monfared, H. M. Kojabadi, and H. Rastegar, ‘‘Static and dynamic [30] K. K. M. S. Kariyawasam, K. K. N. P. Karunarathna,
wind turbine simulator using a converter controlled DC motor,’’ Renew. R. M. A. Karunarathne, M. P. D. S. C. Kularathne, and
Energy, vol. 33, no. 5, pp. 906–913, 2008, doi: 10.1016/j.renene.2007. K. T. M. U. Hemapala, ‘‘Design and development of a wind turbine
06.007. simulator using a separately excited DC motor,’’ Smart Grid Renew.
[9] R. Teodorescu, F. Iov, and F. Blaabjerg, ‘‘Flexible development and Energy, vol. 4, no. 3, pp. 259–265, 2013, doi: 10.4236/sgre.2013.43031.
test system for 11 kW wind turbine,’’ in Proc. IEEE 34th Annu. Conf. [31] H. Xu, W. Yu, D. Griffith, and N. Golmie, ‘‘A survey on industrial Internet
Power Electron. Spec., Jun. 2003, pp. 67–72, doi: 10.1109/PESC.2003. of Things: A cyber-physical systems perspective,’’ IEEE Access, vol. 6,
1218275. pp. 78238–78259, 2018, doi: 10.1109/ACCESS.2018.2884906.
[10] H. M. Kojabadi, L. Chang, and T. Boutot, ‘‘Development of a novel wind [32] P. P. Ray, ‘‘A survey of IoT cloud platforms,’’ Futur. Comput. Informat. J.,
turbine simulator for wind energy conversion systems using an inverter- vol. 1, nos. 1–2, pp. 35–46, 2016, doi: 10.1016/j.fcij.2017.02.001.
[33] A. Pantelopoulos and N. G. Bourbakis, ‘‘A survey on wearable sensor-
controlled induction motor,’’ IEEE Trans. Energy Convers., vol. 19, no. 3,
based systems for health monitoring and prognosis,’’ IEEE Trans. Syst.,
pp. 547–552, Sep. 2004, doi: 10.1109/TEC.2004.832070.
[11] B. Rabelo, W. Hofmann, and M. Gluck, ‘‘Emulation of the static and Man, Cybern. C, Appl. Rev., vol. 40, no. 1, pp. 1–12, Jan. 2010, doi:
dynamic behaviour of a wind-turbine with a DC-machine drive,’’ in Proc. 10.1109/TSMCC.2009.2032660.
[34] H. Elhoone, T. Zhang, M. Anwar, and S. Desai, ‘‘Cyber-based design
IEEE 35th Annu. Power Electron. Spec. Conf., Jun. 2004, pp. 2107–2112,
for additive manufacturing using artificial neural networks for industry
doi: 10.1109/PESC.2004.1355443.
[12] P. P. Ray, ‘‘A survey on Internet of Things architectures,’’ J. King Saud 4.0,’’ Int. J. Prod. Res., vol. 58, no. 9, pp. 2841–2861, May 2020, doi:
Univ., Comput. Inf. Sci., vol. 30, no. 3, pp. 291–319, Jul. 2018, doi: 10.1080/00207543.2019.1671627.
[35] K. Wagh and R. Thool, ‘‘A comparative study of SOAP vs REST web
10.1016/j.jksuci.2016.10.003.
[13] S. Kumar, P. Tiwari, and M. Zymbler, ‘‘Internet of Things is a revolutionary services provisioning techniques for mobile host,’’ J. Inf. Eng. Appl., vol. 2,
approach for future technology enhancement: A review,’’ J. Big Data, no. 2, pp. 12–17, 2012.
[36] M. G. de Vos, W. Hazeleger, D. Bari, J. Behrens, S. Bendoukha,
vol. 6, no. 1, pp. 1–21, Dec. 2019, doi: 10.1186/s40537-019-0268-2.
[14] L. Tan, K. Yu, F. Ming, X. Cheng, and G. Srivastava, ‘‘Secure and resilient I. Garcia-Marti, R. van Haren, S. E. Haupt, R. Hut, F. Jansson, A. Mueller,
artificial intelligence of things: A HoneyNet approach for threat detection P. Neilley, G. van den Oord, I. Pelupessy, P. Ruti, M. G. Schultz, and
and situational awareness,’’ IEEE Consum. Electron. Mag., vol. 11, no. 3, J. Walton, ‘‘Open weather and climate science in the digital era,’’ Geosci.
pp. 69–78, May 2022, doi: 10.1109/MCE.2021.3081874. Commun., vol. 3, no. 2, pp. 191–201, Aug. 2020, doi: 10.5194/gc-3-191-
[15] Y. Djenouri, G. Srivastava, A. Belhadi, and J. C. Lin, ‘‘Intelligent 2020.
blockchain management for distributed knowledge graphs in IoT 5G envi- [37] P. Bourhis, J. L. Reutter, and D. Vrgoč, ‘‘JSON: Data model and
ronments,’’ Trans. Emerg. Telecommun. Technol., Jul. 2021, Art. no. e4332, query languages,’’ Inf. Syst., vol. 89, Mar. 2020, Art. no. 101478, doi:
doi: 10.1002/ett.4332. 10.1016/j.is.2019.101478.
[16] L. Malina, P. Dzurenda, S. Ricci, J. Hajny, G. Srivastava, R. Matulevicius, [38] A. Giessmann and C. Legner, ‘‘Designing business models for cloud
A.-A.-O. Affia, M. Laurent, N. H. Sultan, and Q. Tang, ‘‘Post-quantum platforms,’’ Inf. Syst. J., vol. 26, no. 5, pp. 551–579, Sep. 2016, doi:
era privacy protection for intelligent infrastructures,’’ IEEE Access, vol. 9, 10.1111/isj.12107.
[39] A. A. H. Mohamad, N. K. Jumaa, and S. H. Majeed, ‘‘ThingSpeak cloud
pp. 36038–36077, 2021, doi: 10.1109/ACCESS.2021.3062201.
[17] R. R. Singh, S. M. Yash, S. C. Shubham, V. Indragandhi, V. Vijayakumar, computing platform based ECG diagnose system,’’ Int. J. Comput. Digit.
P. Saravanan, and V. Subramaniyaswamy, ‘‘IoT embedded cloud-based Syst., vol. 8, no. 1, pp. 11–18, Jan. 2019, doi: 10.12785/ijcds/080102.
intelligent power quality monitoring system for industrial drive applica-
tion,’’ Future Gener. Comput. Syst., vol. 112, pp. 884–898, Nov. 2020, doi:
10.1016/j.future.2020.06.032.
[18] A. B. Letaifa, ‘‘SSIM and ML based QoE enhancement approach in SDN R. RAJA SINGH received the B.Tech. degree in
context,’’ in Advances in Computers, vol. 114, The Netherlands, 2019, electrical engineering from Pondicherry Univer-
ch. 4, pp. 151–196. sity, India, in 2004, the M.E. degree in power
[19] J. Kopecký, P. Fremantle, and R. Boakes, ‘‘A history and future of web
electronics and drives from Anna University, India,
Apis,’’ Inf. Technol., vol. 56, no. 3, pp. 90–97, Jun. 2014, doi: 10.1515/itit-
in 2008, and the Ph.D. degree in electrical drives
2013-1035.
[20] W. Alam, D. Sarma, R. J. Chakma, M. J. Alam, and S. Hossain, ‘‘Internet from the Indian Institute of Technology (IIT)
of Things based smart vending machine using digital payment system,’’ Roorkee, India, in 2017.
Indonesian J. Electr. Eng. Informat., vol. 9, no. 3, pp. 719–731, Aug. 2021, He is currently working as an Associate Pro-
doi: 10.52549/.V9I3.3133. fessor with the Department of Energy and Power
[21] L. C. C. De Biase, P. C. Calcina-Ccori, G. Fedrecheski, G. M. Duarte, Electronics, Vellore Institute of Technology (VIT),
P. S. S. Rangel, and M. K. Zuffo, ‘‘Swarm economy: A model for transac- Vellore, India. He has more than 12 years of teaching experience in India
tions in a distributed and organic IoT platform,’’ IEEE Internet Things J., and Tanzania, as well as, five years of research experience at the IIT
vol. 6, no. 3, pp. 4561–4572, Jun. 2019, doi: 10.1109/JIOT.2018.2886069. Roorkee. He has developed various energy efficient control strategies for
[22] P. Anderson and A. Bose, ‘‘Stability simulation of wind turbine systems,’’
IEEE Trans. Power App. Syst., vol. PAS-102, no. 12, pp. 3791–3795,
industrial drives. He has published many research papers in various journals
Dec. 1983, doi: 10.1109/TPAS.1983.317873. and conferences of international repute. His research interests include the
[23] B. Amlang, D. Arsurdis, W. Leonhard, W. Vollstedt, and K. Wefelmeier, IoT, wind energy conversion systems, multilevel power converter drives for
‘‘Elektrische energieversorgung mit windkraftanlagen,’’ BMFT- renewable energy applications, power quality control, starting transient and
Forschungsvorhaben, Brunswick, Germany, Final Rep. 032-8265-B, fault analysis on industrial drives, electrical vehicles systems, and smart
1992. metering systems.

78266 VOLUME 10, 2022

R. Raja Singh et al.: Intelligent IoT Wind Emulation System Based on Real-Time Data Fetching Approach

SWAPNIL BANERJEE is currently pursuing the V. INDRAGANDHI received the B.E. degree
B.Tech. degree in electrical and electronics engi- in electrical and electronics engineering from
neering with the Vellore Institute of Technology Bharadhidasan University, in 2004, the M.E.
(VIT), Vellore, India. His current research inter- degree in power electronics and drives from Anna
ests include the IoT, cloud computing, artificial University, Chennai, and the Doctor of Philosophy
intelligence, industrial automation, power elec- degree from Anna University, in 2015. She is
tronics, industrial drives, and renewable energy currently working as an Associate Professor with
applications. the School of Electrical Engineering, VIT, Vellore,
Tamil Nadu. She has been engaged in research
work for the past 13 years in the area of the
Internet of Things, smart grid, power electronics, cyber physical systems,
and renewable energy systems. She has also organized many national and
R. MANIKANDAN received the B.E. degree international workshops/symposiums/conferences in collaboration with top
in electrical and electronics engineering from leading universities. She awarded Gold Medal for the Achievement of
Bharathidasan University, in 2003, and the M.E. University First Rank.
degree in VLSI design from Anna University,
India, in 2008. He is currently pursuing the Ph.D.
degree with the Department of Energy and Power
Electronics, Vellore Institute of Technology (VIT),
Vellore, India. His research interests include fault
diagnosis, remain useful lifetime (RUL) estima-
tion of power converters, and applications of
machine learning algorithms.

KETAN KOTECHA is currently an Administrator

and a Teacher of deep learning. He has exper-
tise and experience in cutting-edge research and
projects in AI and deep learning for the last
25 years. He has published more than 100 widely SUBRAMANIYASWAMY VAIRAVASUN-
in several excellent peer-reviewed journals on DARAM received the B.E. degree in computer
various topics ranging from cutting edge AI, science and engineering from Bharathidasan Uni-
education policies, teaching-learning practices, versity, India, the M.Tech. degree in information
and AI for all. He has published three patents technology from Sathyabama University, India,
and delivered keynote speeches at various national and the Ph.D. degree from Anna University,
and international forums, including at the Machine Intelligence Laboratory, India. He continued the extension work with the
USA; IIT Bombay under the World Bank Project; the International Indian Department of Science and Technology support as
Science Festival organized by the Department of Science and Technology, a Young Scientist Award Holder. He is currently
Government of India; and many more. His research interests include artificial working as a Professor with SASTRA Deemed
intelligence, computer algorithms, machine learning, and deep learning. University, Thanjavur, India. In total, he has 18 years of experience in
He was a recipient of the two SPARC projects worth INR 166 lakhs from academia. He has contributed more than 160 articles and chapters for many
the MHRD Government of India in AI in collaboration with Arizona State high-quality Scopus and SCI/SCIE indexed journals and books. He is serving
University, USA, and The University of Queensland, Australia. He was as a research supervisor, and he is also a visiting expert to various universities
also a recipient of numerous prestigious awards, like the Erasmus+ Faculty in India. His technical competencies lie in recommender systems, social
Mobility Grant to Poland; the DUO-India Professors Fellowship for research networks, the Internet of Things, information security, and big data analytics.
in responsible AI in collaboration with Brunel University, U.K.; the LEAP He is on the reviewer board of several international journals and has been a
Grant at Cambridge University, U.K.; the UKIERI Grant with Aston program committee member for several international/national conferences
University, U.K.; and a Grant from the Royal Academy of Engineering, and workshops. He also serves as a guest editor for various special issues of
U.K., under Newton Bhabha Fund. He is an Associate Editor of IEEE ACCESS reputed international journals.
journal.

VOLUME 10, 2022 78267