A

Project Phase II Report

on

IOT Based Weather Monitoring And Analysis

System Based On GUI Sponserd by Aashay
Measurements Pvt. Ltd.
SUBMITTED TO THE SAVITRIBAI PHULE PUNE UNIVERSITY,
PUNE

IN THE PARTIAL FULFILMENT OF PROJECT PART-II OF THE

REQUIREMENT FOR THE AWARD OF THE DEGREE
OF
BACHELOR OF ELECTRONICS &
TELECOMMUNICATION ENGINEERING
BY
Adesh Dhage B191223018.
Yogesh Pawar B191223070
Rohan Yadav B191223067

Under the guidance of

Prof.Dr. Dipali Shende

Department of Electronics and Telecommunication Engineering

Pimpri Chinchwad College of Engineering and Research,Ravet,
Pune-412101
ACADEMIC YEAR:2023-24
 PIMPRI CHINCHWAD EDUCATION TRUST’S
Pimpri Chinchwad College of Engineering and Research,
Pune-412101

CERTIFICATE
This is certify that the Project Phase-I report entitled
”IOT Based Weather Monitoring And Analysis System Based On GUI”
Sponserd by Aashay Measurements Pvt. Ltd.,
Submitted by
Adesh Dhage (B191223018)
Yogesh Pawar (B191223070)
Rohan Yadav (B191223067)

is a bona fide work carried out by them under the supervision of Prof.Dr. Dipali
Shende and it is approved for the partial fulfillment of the Project Phase I of the require-
ment of Savitribai Phule Pune University,Pune for the award of the degree of Bachelor
of Engineering (Electronics &Telecommunication).The Dissertation work has not been
earlier submitted to any other institute or university for the award of degree.

Prof.Dr. Dipali Shende Prof.Kishore Bhangale

Project Guide Project Co-ordinator

Prof. Dr.Rahul G.Mapari Prof.Dr.H.U.Tiwari

Head of Department Principal

1
Project Completion Certificate

CamScanner

2
Acknowledgement

We are greatly indebted to Head of Electronics & Telecommunication Engineering

and Project Guide Prof. Dr.Dipali Shende for her guidance throughout the course of
this work.It has been an altogether different experience to work with him and we would
like to thank him for his help,suggestions and numerous discussions.

We are also thankful to Mr. Rakesh Nalawade (CTO, Aashay Measurements Pvt.
Ltd)for providing valuable guidance and environment.

We are heartily thankful to Prof. Dr.HarishU.Tiwari (Principal, Pimpri Chinch-

wad College of Engineering & Research,Ravet)for providing research environment;also for
his kind inspiration.

We are gladly taking this opportunity to thank Head of Electronics Telecommuni-

cation Engineering Prof. Dr. Rahul Mapari and project Coordinator Mr. Kishor
Bhangale for their valuable guidance and providing facility during progress of seminar.
Last but not least we are also thankful to all those who help directly or indirectly to
develop this Project work and complete it successfully. With Deep Reverence,

Adesh Dhage
Yogesh Pawar
Rohan Yadav

Date:
Place:

3
Contents

1 Introduction 1
1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Research background . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.3 Problem Statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.4 Objectives of the Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.5 Scope of the Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.6 Overview of Proposed System . . . . . . . . . . . . . . . . . . . . . . . . 5
1.7 Organization of the Report . . . . . . . . . . . . . . . . . . . . . . . . . . 6

2 Literature Review 7
2.1 Literature Survey . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2.2 Gap Identification: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

3 Proposed Methodology: 11
3.1 Block Diagram: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
3.2 System Explanation: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
3.3 Hardware Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
3.3.1 Circuit Diagram: . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
3.3.2 PCB Design: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.3.3 Hardware Components and Specifications . . . . . . . . . . . . . . 15
3.4 Software Setup: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.5 Flowchart and Algorithm: . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3.5.1 Flowchart: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3.5.2 Algorithm: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

4 Results and Discussions 27

4.1 Result and Discussion: . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
4.2 Database Information: . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
4.2.1 SQL Queries: . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

5 Advantage and Applications 36

5.1 Advantages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
5.2 Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

6 Conclusion and Future Scope 40

6.1 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
6.2 Future Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40

7 References 42

4
A Project Outcomes 43
A.1 Plagiarism Report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
A.2 CopyRight . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
A.3 Research Paper . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
A.4 Sponsorship Letter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
A.5 Publication Certificate . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
A.6 Project Competition Paricipation Certificate . . . . . . . . . . . . . . . . 59

5
List of Figures

3.1 System Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

3.2 System Block Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
3.3 Circuit Diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.4 PCB Circuit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.5 BME280 Sensor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
3.6 FC-28 Soil Moisture Module . . . . . . . . . . . . . . . . . . . . . . . . . 16
3.7 MH-RD Rain drop module . . . . . . . . . . . . . . . . . . . . . . . . . . 16
3.8 RS-FXJT-N01 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.9 WS102-Anemometer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3.10 ESP32 Microcontroller . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3.11 Node Red Flow-1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
3.12 Node Red Flow-2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
3.13 Orange tool work flow . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
3.14 Flowchart . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

4.1 Dashboard . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
4.2 Hardware Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

6
 Abstract

The integration of Internet of Things (IoT) technology with weather monitoring sys-
tems has revolutionized how weather data is collected, analyzed, and utilized. This
abstract presents a comprehensive overview of an IoT-based weather monitoring and
analysis system enhanced with a Graphical User Interface (GUI). The system employs
various sensors to gather real-time weather data such as temperature, humidity, air pres-
sure, wind speed, and precipitation. These sensors are connected to a microcontroller or
a single-board computer, which acts as the central processing unit.
The collected data is transmitted to a cloud-based server using wireless communication
protocols such as Wi-Fi or LoRaWAN for storage and analysis. A web-based GUI allows
users to visualize the gathered data in real-time through interactive charts, graphs, and
maps. Additionally, the GUI provides historical data analysis, trend prediction, and
customizable alerts for specific weather conditions.
Furthermore, the system is designed to be scalable and adaptable, allowing for the
integration of additional sensors or functionalities based on specific requirements. The
utilization of IoT technology and a user-friendly GUI enhances accessibility and usability,
making it suitable for various applications including agriculture, transportation, urban
planning, and disaster management. This abstract provides insights into the architecture,
functionality, and potential applications of the proposed IoT-based weather monitoring
and analysis system with GUI.
Weather forecasting is the application of science and technology to predict the state
of the atmosphere at a particular location. The majority of early weather forecasting
techniques were based on pattern identification, or the observation of recurring patterns
in the events. For instance, one may notice that an especially crimson sunset typically
heralded favorable weather the next day. Not all of these forecasts, though, turn out to
be accurate.
Here this system will predict weather based on parameters such as temperature, hu-
midity, wind speed, rain gauge, wind direction, soil temperature, atmospheric pressure,
soil moisture. We will collect the weather data with the help of some sensors and pass it
to the system. System will take this parameter and will predict weather from previous
data in database(dataset). The primary objective of our project is to develop a well-
made and accurate weather forecasting model capable of predicting climate changes over
extended periods and assessing their implications for future generations.
The weather forecasting system forecasts weather based on historical records; thus,
this prediction will prove to be trustworthy. It takes into account elements like temper-
ature, humidity, wind speed, rain gauge, wind direction, soil temperature, atmospheric
pressure, soil moisture. Air traffic, maritime, agriculture, forestry, military, navy, and
other fields may all use this technology.
Chapter 1

Introduction

1.1 Introduction
In recent years, the fusion of Internet of Things (IoT) technology with various domains
has sparked innovation, efficiency, and enhanced decision-making processes. One such
domain experiencing a transformative impact is weather monitoring and analysis. Tradi-
tional weather monitoring systems often rely on manual data collection methods, which
are not only labor-intensive but also prone to inaccuracies and delays. To address these
limitations, IoT-based solutions have emerged, revolutionizing the way weather data is
collected, analyzed, and utilized. IoT-based weather monitoring and analysis system en-
riched with a Graphical User Interface (GUI). By integrating IoT sensors, wireless com-
munication protocols, cloud computing, and user-friendly interfaces, this system offers
real-time monitoring, comprehensive data analysis, and intuitive visualization capabili-
ties.
The proliferation of IoT sensors capable of measuring various meteorological param-
eters such as temperature, humidity, air pressure, wind speed, and precipitation has
facilitated the creation of robust weather monitoring networks. These sensors, often de-
ployed across geographical regions, continuously gather data and transmit it to a central
processing unit, typically a microcontroller or a single-board computer. Wireless com-
munication protocols such as Wi-Fi, LoRaWAN, or cellular networks enable seamless
transmission of data to cloud-based servers. In the cloud, the collected data is stored
securely and made accessible for analysis and visualization. Cloud computing infrastruc-
ture provides the scalability and computational power necessary to handle large volumes
of data and perform complex analytics in real-time.
The integration of a Graphical User Interface (GUI) with the IoT-based weather
monitoring system enhances user interaction and accessibility. Through a web-based
interface, users can visualize the collected weather data using interactive charts, graphs,
and maps. Additionally, the GUI enables historical data analysis, trend prediction, and
the customization of alerts for specific weather conditions. The versatility of the IoT-
based weather monitoring and analysis system extends its applicability to various sectors
including agriculture, transportation, urban planning, and disaster management. By
providing stakeholders with timely and accurate weather information, informed decisions
can be made to optimize operations, enhance safety, and mitigate risks. In summary,
the convergence of IoT technology with weather monitoring and analysis has ushered in

1
a new era of data-driven insights and decision-making. This introduction sets the stage
for exploring the architecture, functionality, and potential applications of the IoT-based
weather monitoring and analysis system with GUI in greater detail.
The Indian economy’s most significant sector is agriculture. Fifty percent of India’s
workforce is employed in the agriculture industry, which generates eighteen percent of the
country’s GDP. The world’s top producer of wheat, rice, pulses, spices, and spice-related
items is India. The weather is a vital natural element for life; depending on its severity,
it may either provide or destroy opportunities for survival. To put it simply, weather
forecasting is the process of predicting the state of the atmosphere in a given location
at a given time utilizing physics principles, technology, and a range of statistical and
empirical methodologies. The weather prediction is incredibly helpful in protecting both
people and property and to protect the crops and it will tell us what to expect in our
atmospheric environment.
Weather prediction data is essential for making quick modifications to daily agricul-
tural activities, which reduces losses due to unfavorable weather conditions and raises the
quantity and quality of agricultural output. aids in seasonal or long-term planning as well
as crop selection according to ideal weather conditions. Seasonal forecasting contributes
to the betterment of farmers’ lives in areas where there is significant fluctuation in yearly
rainfall. Seasonal forecasts play an important role in deciding the eco- nomic policies of
governments, during major drought was forecasted, monetary policy could be relaxed to
maintain growth targets. The unique aspect of our approach is that ability to provide
fast and reliable forecasts while operating on low-cost and resource- efficient computing
systems. These forecasts have practical applications in our daily lives, enabling us to
make informed decisions and adapt effectively to changing weather conditions.

1.2 Research background

The integration of Internet of Things (IoT) technology with weather monitoring systems
marks a substantial advancement in meteorological research and practical applications.
This area of study has seen significant developments, driven by the proliferation of IoT
devices and sensors capable of gathering diverse environmental data. These sensors, char-
acterized by miniaturization, cost reduction, and enhanced energy efficiency, have enabled
the establishment of dense and comprehensive weather monitoring networks. Unlike tra-
ditional methods reliant on manual data collection, IoT-based systems offer real-time data
collection capabilities, ensuring meteorologists and researchers access timely and precise
information for analysis and decision-making. A cornerstone of IoT-based weather mon-
itoring systems is the utilization of wireless communication protocols, including Wi-Fi,
Bluetooth, LoRaWAN, and cellular networks, facilitating seamless data transmission from
sensors to central processing units and cloud-based servers. This wireless connectivity
extends the coverage and accessibility of weather monitoring networks, allowing for data
transmission over long distances and in remote locations. Cloud computing infrastructure
plays a pivotal role in these systems, providing the computational power and scalability
needed to handle the vast volumes of weather data generated by IoT sensors. Leverag-
ing big data analytics techniques, real-time analysis of weather data becomes possible,
enabling the identification of patterns, trends, and actionable insights for various appli-
cations.
Graphical User Interfaces (GUIs) play a vital role in enhancing the usability and ac-

2
cessibility of IoT-based weather monitoring systems. These interfaces provide intuitive
visualization tools such as interactive charts, graphs, and maps, facilitating data interpre-
tation and decision-making. GUIs also support historical data analysis, trend prediction,
and customization of alerts for specific weather conditions. Across diverse sectors such
as agriculture, transportation, urban planning, disaster management, and renewable en-
ergy, IoT-based weather monitoring and analysis systems find applications. By offering
stakeholders real-time and accurate weather information, these systems help optimize
operations, enhance safety, and mitigate risks associated with weather-related events.
Despite the advancements made, challenges such as data security, privacy concerns, sen-
sor calibration, and network reliability persist in IoT-based weather monitoring systems.
Future research directions include the development of advanced sensor technologies, op-
timization of data transmission protocols, integration of machine learning algorithms for
predictive analytics, and enhancement of GUIs for improved user experience. In sum-
mary, the ongoing research in IoT-based weather monitoring and analysis systems with
GUIs holds promise for advancing meteorological research, improving decision-making
processes, and addressing challenges related to weather events.
Internet of Things (IoT) has revolutionized various industries, including weather mon-
itoring and forecasting. IoT-based weather monitoring systems can collect real-time data
from various sensors, process it, and present it to users through a web-based graphical
user interface (GUI). These systems provide accurate and timely weather information,
which can be crucial for various applications, from agriculture to disaster management.
Building an IoT-based weather monitoring and forecasting system with a web GUI is a
complex project that requires hardware, software, and data analysis expertise. Be pre-
pared to continuously monitor and maintain the system to ensure accurate and up-to-date
weather information for users.

1.3 Problem Statement

It is important to exactly determine the weather conditions for effective use of water
resources, crop productivity. Harvesting and pre-planning of water structures. But many
weather stations are not giving an accurate prediction to the farmers.

1.4 Objectives of the Study

The objectives of the proposed works are mentioned as follow:
1.To design and develop an IoT-based weather monitoring system capable of collecting
real-time data on meteorological parameters such as temperature, humidity, air pressure,
wind speed, and precipitation. This involves selecting appropriate sensors, establishing
wireless communication protocols, and integrating the necessary hardware and software
components.
2.To integrate a user-friendly Graphical User Interface (GUI) into the weather monitoring
system. The GUI will enable stakeholders to visualize weather data through interactive
charts, graphs, and maps, facilitating data interpretation and analysis. Additionally,
the GUI will support features such as historical data analysis, trend prediction, and
customizable alerts for specific weather conditions.

3
3.To ensure seamless real-time data transmission from sensors to central processing units
and cloud-based servers. This involves implementing robust wireless communication pro-
tocols and cloud computing infrastructure to handle large volumes of data. Furthermore,
advanced data analytics techniques will be employed to perform real-time analysis of
weather data, identify patterns, trends, and anomalies, and generate actionable insights.
4.To design the system which is scalable and adaptable, allowing for the integration of
additional sensors or functionalities based on specific requirements. This objective entails
ensuring interoperability with existing weather monitoring infrastructure and systems, as
well as accommodating future expansion and technological advancements.
5. To evaluate the effectiveness and user-friendliness of the GUI. Feedback from stake-
holders will be gathered to identify areas for improvement and refinement. User accep-
tance testing will also be conducted to assess stakeholders’ satisfaction and willingness
to adopt the IoT-based weather monitoring system with GUI.
6. To assess the application and impact of the developed system in various sectors such
as agriculture, transportation, urban planning, and disaster management. This involves
evaluating the system’s effectiveness in improving decision-making, enhancing safety, and
optimizing operations in real-world scenarios.

1.5 Scope of the Study

The scope of this study encompasses various aspects of designing, developing, and imple-
menting an IoT-based weather monitoring and analysis system with a Graphical User In-
terface (GUI). Initially, the study will focus on defining the system architecture, outlining
the hardware components, communication protocols, and software interfaces required for
seamless operation. Subsequently, the selection and integration of suitable sensors capa-
ble of measuring meteorological parameters such as temperature, humidity, air pressure,
wind speed, and precipitation will be undertaken, ensuring compatibility and accuracy
within the system. Wireless communication protocols, including Wi-Fi, Bluetooth, or
LoRaWAN, will be explored and implemented to facilitate the transmission of data from
sensors to central processing units and cloud-based servers. Robustness, reliability, and
security of data transmission over wireless networks will be prioritized. Cloud computing
infrastructure will be established to handle the storage and processing of large volumes
of weather data, utilizing scalable and cost-effective solutions for data management.
Furthermore, algorithms for real-time analysis of weather data, including trend iden-
tification, anomaly detection, and predictive analytics, will be developed and integrated
into the system. Interactive visualization tools will be designed within the GUI to present
weather data effectively through charts, graphs, and maps, ensuring ease of interpretation
for stakeholders. Usability testing will be conducted to assess the effectiveness and user-
friendliness of the GUI, with stakeholder feedback informing refinement and improvement
efforts.Scalability and adaptability will be key considerations in the system design, allow-
ing for future expansion and technological advancements while ensuring interoperability
with existing weather monitoring infrastructure and systems. Moreover, robust security
measures will be implemented to safeguard sensitive weather data during transmission
and storage, in compliance with data privacy regulations and standards.
The study will culminate in the evaluation of the application and impact of the devel-
oped system across various sectors such as agriculture, transportation, urban planning,

4
and disaster management. Assessing the system’s effectiveness in improving decision-
making, enhancing safety, and optimizing operations will be central to this evaluation.
Comprehensive documentation and reporting will summarize the study’s findings, con-
clusions, and recommendations, providing valuable insights for future research and im-
plementation endeavors.
The weather has a wide range of social, cultural, and economic effects on society. Crop
productivity may be influenced by weather alone by up to 50 percent. Rainfall is the most
crucial prediction among those that are needed since it determines agricultural output in
a certain area and ultimately the nation’s economy. Planning for moisture conservation
must be done during light monsoon seasons, and during strong monsoon seasons, flood
relief must be decided upon primarily in accordance with local meteorological conditions.
By using accurate weather forecasts, we can lessen the harm brought on by bad weather
conditions, both directly and indirectly. If adequate early forecast information on pest
and disease incidence is provided based on meteorological characteristics, the frequency
of crop losses can be decreased. By monitoring buffer stock activities, it helps to contain
variations in the price of food grains. That implies that during years of excellent monsoon,
prices decrease and the government purchases agricultural goods; during years of poor
monsoon, when prices tend to rise, it may sell some of the commodities it has acquired.
Water use can be scheduled over an area based on the predicted level. The weather
forecasts assist farmers in organizing their agricultural practices.

1.6 Overview of Proposed System

In this project, our primary objective is to create a comprehensive weather monitoring
and prediction system for agricultural purposes.To achieve this, we will follow a struc-
tured approach:
1.Data Collection: Initially, we will gather weather-related data using various sensors.
Specifically, we will use to measure parameters such as temperature, humidity, wind
speed, rain gauge, wind direction, soil temperature, atmospheric pressure, soil moisture
by using specific sensors.
2.Data Transmission: We will employ a ESP32 to upload the collected data to a central
server. This server acts as a repository for the weather data, ensuring its availability for
analysis and prediction.
3.Machine Learning Model: Our next step involves building a machine learning model.
For predication purpose we are using Orange tool which is no code tool.Orange’s Scoring
Sheet widget provides explainable machine learning predictions using a simple scoring
system. Each feature’s influence obtains an integer score, making it easier to understand
and communicate the model, crucial in sectors where transparency is vital.
4.Data Comparison and Prediction: Once the machine learning model is trained, it will
retrieve the live data uploaded to the server. It will then compare this real-time data
with historical data from the sensors. Using the insights gained from the training, the
model will generate accurate weather predictions.
5.Web GUI Display: The final step is to display these predictions in a user-friendly man-
ner through a web-based graphical user interface (GUI). Users can use this user interface
to view the current weather conditions and forecasts, which are generated by the machine
learning model’s analysis of live and historical data.

5
1.7 Organization of the Report
The project report is oraganized as follow:
Chapter 1: Introduction:
In this report, we explore the concept of an accurate weather forecasting with the help
of sensors and machine learning module. We begin by highlighting the importance of
weather forecasting in agriculture and modern farming practices, as they offer increased
efficiency and precision in agricultural operations. The objectives and scope of this report
are outlined, focusing on the specifically enhance the precision and reliability of weather
prediction and their significance in agricultural activities.

Chapter 2: Literature Review:

To provide a comprehensive understanding of accurate weather forecasting, this report
presents a detailed review of existing literature. Specifically, we focus on studies related
to weather forecasting, sensors and machine learning, summa- rizing key findings, ad-
vancements, and challenges identified in previous research. This literature review serves
as a foundation for the current study, allowing us to identify gaps and limitations in the
existing literature and establish the need for further investigation.

Chapter 3: Proposed Methodology:

The research methodology employed in this study involved a combination of data col-
lection methods, experimental setup, and relevant procedures. We carefully selected a
suitable sensor for data collection, considering factors such as wind speed di- rection, soil
temperature moisture, rain, temperature, pressure, reliability and ease of use. Parame-
ters such as accuracy, speed, and effectiveness were evaluated during the implementation
of weather forecasting.

Chapter 4: Results and Discussion:

In this section, we present the quantitative and qualitative results obtained from imple-
menting the IOT based weather forecasting using machine learning module. Utilizing
sensors, data and figures we illustrate the data collected during the experiments and pro-
vide a detailed analysis. The results are interpreted, and their im- plications are discussed
in the field of agriculture and disaster management using data provided by weather sta-
tion and data predicted by the machine learning module.

Chapter 5: Advantages and Applications:

This chapter carries the information about the outcomes followed by advantages and ap-
plications of the presented technology.

Chapter 6: Conclusion and Future Scope:

This chapter concludes the purpose of the study and gives the futuristic capabilities of
the system.

Chapter 7: References:
The report concludes with a comprehensive list of references, employing an appropriate
referencing style to acknowledge and credit the sources cited throughout the report. This
ensures the accuracy and consistency of the reference list, allowing readers to access the
relevant literature for further study and exploration. Or limitations in current knowledge.

6
Chapter 2

Literature Review

2.1 Literature Survey

[1] Gaurav Verma, Pranjul Mittal, Shaista Farheen (Department of Electronics Com-
munication) Dayananda Sagar College of Engineering/ “Real Time Weather Prediction
System Using IOT and Machine Learning”/ IEEE 2020 - Methodology Used ‘Logistic
Regression Model’, Parameters ‘Temp., Humidity, Light Intensity’, Conclusion ‘By us-
ing sensors machine Learning module trained with prerecorded data and predict weather
forecast’. — The emphasis of today is on intelligent technologies, such as machine learn-
ing and the Internet of Things. There are numerous IoT implementations available for
IoT hardware platforms. Among them is the ESP8266 chip. This study creates a real-
time weather prediction system that may be used to forecast the weather in a variety of
settings, including residences, businesses, stadiums, and agricultural settings. The sys-
tem makes use of an LDR light intensity sensor and a DHT11 temperature and humidity
sensor. Using an ESP8266-01 module and NodeMCU, the sensed data from the sensors
is uploaded to a ThingSpeak cloud server.
[2]Dires Negash Fente, Prof. Dheeraj Kumar Singh, (Department of Information tech-
nology, Parul University Parul Institute of Engineering and Technology)/ “Weather Fore-
casting Using Artificial Neural Network”/ IEEE 2018 - Methodology Used ‘LSTM – Long
short-term Memory’, Parameters ‘Temp., Humidity, Light Intensity, Wind Speed, Wind
Direction’, Conclusion ‘LSTM gives Result with High Accuracy’. Since the weather has
a major impact on the agricultural and industrial sectors, accurate weather forecasting
is essential in today’s society. Additionally, it is employed to predict and warm up natu-
ral disasters. Determining the appropriate values for weather parameters and projecting
future weather conditions based on these factors is known as weather forecasting. In this
study, several meteorological parameters were gathered from the National Climate Data
Center, and the neural network was trained for various combinations using the long-short
term memory (LSTM) technique.
[3]Sushmitha Kothapalli, S. G. Totad, (Dept of Computer Science and Engineering) “A
Real-Time Weather Forecasting and Analysis”/ IEEE 2017 - Methodology Used ‘ARIMA
Model’, Parameters ‘Temp., Humidity, Light Intensity, Wind, Rainfall’, Conclusion ‘Val-
ues are Predicted by implementing ARIMA Model using R Studio’. - The endeavor of
meteorologists to anticipate the weather and potential outcomes at a future date is known
as weather forecasting. Temperature, wind, humidity, rainfall, and data set size are the
factors that determine the climatic state parameters. In this case, only the parameters
of temperature and humidity are taken into account for the experimental study. The

7
DHT11 sensor, which aids in determining the temperature and humidity values of a cer-
tain region or location, provides the temperature and humidity data that is gathered.
Using an Ethernet shield to facilitate online data uploading, the raspberry pi is utilized
to save the gathered data on the cloud.
[4]Shivam Tandon, Abhishek Patel, Pawan Kumar Singh (School of Computing Science
and Engineering Galgotias University Uttar Pardesh, India/ “Weather Prediction Us-
ing Machine Learning Algorithms”/ IEEE 2020 - Methodology Used ‘Random Forest,
Decision Tree, MLP Classifier, Linear Regression’, Parameters ‘Temp., Humidity, Light
Intensity, Wind, Rainfall, Visibility’, Conclusion ‘By using Machine Learning Module
gives Accurate Result’. — Climate plays a crucial role in numerous important produc-
tion industries, such as agriculture. These days, there is a strong emphasis on climate
change, which is why historical weather forecasts are becoming less accurate and more ac-
curate. For this reason, miles play a crucial role in customizing and adjusting the weather
forecast model. Both the nation’s financial system and people’s lives are impacted by
those forecasts. A forested area utilized for weather forecasting is part of an information
and statistics analysis algorithm system. One of the strongest natural barriers we face
in life is the weather; we must consider factors such as humidity, temperature, and other
forms of protection. astounding. Our artwork aims to format weather predictions in an
efficient manner.
[5]Tulsi Pawan Fowdur, Rosun Mohammad Nassir-Ud-Diin Ibn Nazir (Department of
Electrical and Electronic Engineering, University of Mauritius, Reduit, Mauritius)/ “A
real-time collaborative machine learning based weather forecasting system with multiple
predictor locations” – Weather forecasting is an essential application in meteorology and
has been one of the world’s most technically and technologically difficult topics. As the
devastating repercussions of climate change continue to unfold, accurate localized short-
term weather prediction has become more vital than ever. In this research, we present
a collaborative machine learning-based real-time weather forecasting system that uses
data from several places to predict the weather for a single site. Five machine learning
algorithms were utilized in this work, and tests were conducted in four distinct sites in
Mauritius to forecast weather.Temperature, wind speed, wind direction, pressure, hu-
midity, and cloudiness are all variables to consider. The Open Weather API was used to
collect the weather data.
[6]A THESIS Submitted by - KALA A (Faculty of Information Communication Engi-
neering ANNA UNIVERSITY CHENNAI)/ “Hybrid Weather Forecasting Models Based
On Deep Learning And Mode Decompositions Methods”/ Shodhganga 2022 - Method-
ology Used ‘Artificial Neural Network, hybrid model based on long-short term memory
the artificial algae algorithm (LSTM Model)’, Result Component ‘Result Shows that hy-
brid performance model perform a better prediction in comparison with LSTM.’, Future
Scope ‘It could be extended to forecast other events such as drought and flood’.
[7]A THESIS Submitted by - Uma Sharma (Faculty of Mathematics Computing Ba-
nasthali Vidyapith Rajasthan, India)/ “Smart Weather Forecasting Using Deep Learn-
ing”/ Shodhganga 2022- Methodology Used ‘Bi-LSTM Model for historical weather data.
Based on ARIMA Model and Deep Learning Model CRNN’, Result Component ‘Increas-
ing the size of training data sample cloud result in better prediction and deep learning
produce accurate result.’, Future Scope ‘Combining different local Weather station data
regional weather data for an accurate prediction’.
[8]A THESIS Submitted by - VASAVI RAVURI (Research Development Jawaharlal
Nehr Technological University)/ “Optimized Deep LSTM Based Ensemble Classifier for

8
Weather Forecasting”/ Shodhganga 2023 - Methodology Used ‘Use of big data Clustering
using Spark. Prediction weather with MFATA – based deep LSTM using time series.’,
Result Component ‘It produce very accurate result depending on size of data efficiency.’,
Future Scope ‘Interfacing IOT Sensors increase more efficiency using data mining tech’.

2.2 Gap Identification:

In identifying gaps for an IoT-based weather monitoring and analysis system integrated
with a graphical user interface (GUI), several key areas emerge. Firstly, while exist-
ing research predominantly focuses on the technical aspects of IoT and data analysis,
there appears to be a gap in the exploration of user-friendly GUI design tailored explic-
itly for weather monitoring systems. This gap suggests a need for studies that delve
into intuitive visualization techniques for weather data, customizable dashboards, and
navigational structures accommodating users with varying levels of technical proficiency.
Additionally, there seems to be a gap in research concerning the seamless integration of
diverse IoT devices and sensors into a cohesive system. This integration aspect might
benefit from investigations into communication protocols, device management strategies,
and scalability measures to facilitate the addition or removal of sensors as required.
Furthermore, another gap lies in the real-time visualization of weather data through
the GUI. While real-time data collection is prevalent in IoT-based systems, there appears
to be a lack of research focusing on dynamic visualization techniques that update visual-
izations in real-time as new data streams in. This gap presents an opportunity to explore
methods for providing alerts on significant weather events and ensuring the accuracy and
reliability of the displayed data. Moreover, customization and personalization options for
users or specific applications seem underexplored in existing research. This area might
benefit from studies investigating user-driven customization features, personalized alert
systems, and adaptable GUI layouts to cater to diverse use cases and environments.
Lastly, the aspect of accessibility and inclusivity in GUI design for IoT-based weather
monitoring systems emerges as a potential gap. While accessibility features are crucial
for ensuring that the system is usable by individuals with disabilities or diverse needs,
existing research appears to lack focus in this area. Addressing this gap could involve ex-
ploring design strategies to make the GUI compatible with screen readers, incorporating
high-contrast options, and supporting alternative input methods to enhance accessibil-
ity. By addressing these identified gaps, future research endeavors can contribute to the
development of more user-friendly, efficient, and inclusive IoT-based weather monitoring
and analysis systems integrated with GUIs.
Accurate weather forecasting during abrupt changes in weather patterns has been
recognized as a research gap. The primary drawbacks listed in Table 1 are that when a
change in the weather is seen within the learnt time sequence, the proposed baseline and
derived models anticipate temperature incorrectly. Meteorological research is gradually
indicating that changes in one variable that affects another frequently cause changes
in the weather. When there is a rapid shift in the weather, weather forecasters can
use the interplay of these mutually correlated weather features to properly anticipate
a particular weather characteristic. Therefore, our goal is to create a spatial feature
attention mechanism that can concurrently learn the connections between input features
in lengthy sequences and reliably predict the target feature.

9
 The majority of NMHSs use a typical forecasting approach in which forecasters use
numerical weather prediction (NWP) output as guidance to produce text-based, sensible
weather-element forecast products (such as maximum/minimum temperature and cloud
cover). Usually, the procedure is labor-intensive, schedule-driven, and product-focused.
The hydrometeorological predictions and alerts issued by NMHS have grown significantly
more precise and precise during the past ten years due to scientific and technical advance-
ments. Customer and partner demands for the National Weather Service (NWS) included
the need for precise forecasts in gridded, digital, and visual forms as computer technology
and high-speed dissemination channels (like the Internet) advanced. The amount of addi-
tional information that may be provided to the user community is limited by traditional
NWS text forecast products. The idea of digital database forecasting gives you the flexi-
bility to satisfy partner and client requests for more precise, in-depth hydrometeorological
forecasts. One of the most interesting chances to combine PWS forecast distribution and
service delivery which best benefits the user community is provided by digital database
forecasting.

10
Chapter 3

Proposed Methodology:

The Internet of Things denotes the connection of devices, machines, and sensors to the
Internet. An IoT system comprises four basic building blocks: sensors, processors, gate-
ways, and applications. This article will thoroughly discuss what each component of the
IoT architecture represents.

Figure 3.1: System Architecture

The architecture of IoT components:

1.Sensors convert a non-electrical input to an electrical signal. (ex: pressure, humidity,
temperature, etc).
2.Processors are the brain, the main part of the IoT system. They process the raw data
captured by the sensors and extract valuable information. Examples of processors are
microcontrollers and microcomputers.(e.g.ESP32)
3.Gateways are the combination of hardware and software used to connect one network
to another. Gateways are responsible for bridging sensor nodes with the external Internet
or World Wide Web. The figure below depicts how using gateways works.(mqtt, node
red)
4.Applications provide a user interface and effective utilization of the data collected.
(grafana)

3.1 Block Diagram:

A weather monitoring and analysis system with a graphical user interface (GUI) relies
on interconnected modules. It starts with data collection from various sources, followed
by processing and storage in a database. The GUI facilitates user interaction and data

11
visualization, while an analysis module predicts weather patterns. Reporting and alerting
functionalities keep users informed, while integration and security measures ensure seam-
less operation and data protection. Overall, these components work together to provide
users with accurate weather insights through an intuitive interface.

Figure 3.2: System Block Diagram

3.2 System Explanation:

The system depicted in the block diagram is a sophisticated cloud-based predictive mod-
eling infrastructure. It leverages a network of sensors connected through an ESP32 mi-
crocontroller to gather real-time data, which is then routed to a Node-RED gateway for
initial processing. Subsequently, the processed data is transmitted to an AWS server, a
prominent cloud computing platform renowned for its robustness and scalability. Within
the AWS environment, the data is stored securely and subjected to analysis by a predic-
tion model tailored to the specific application domain. This prediction model, whether
based on statistical algorithms or machine learning techniques, scrutinizes the data to
discern patterns and correlations, ultimately generating forecasts of future events. These
forecasts, considered actionable insights, are then made accessible to users through an
intuitive web-based graphical user interface (GUI), offering a user-friendly platform for
interaction and visualization.

12
 Components of the above System:
1.Sensors: Varied sensors, such as BME280,Rain Gauge, Wind speed, Wind Direction,
Soil moisture sensors are strategically deployed based on the application requirements.
These sensors serve as the primary data sources, capturing real-world environmental
parameters.
2.ESP32 Microcontroller: The ESP32 microcontroller orchestrates the data collection
process, interfacing with the sensors and facilitating the transmission of collected data to
downstream components within the system architecture.
3.Gateway (Node-RED): Node-RED acts as an intermediary hub for data aggregation
and preliminary processing. It receives the raw sensor data from the ESP32, performs
necessary pre-processing tasks such as data cleaning or normalization, and forwards the
refined data streams to the AWS server.
4.Server (AWS): AWS serves as the backbone infrastructure for data storage, compu-
tation, and deployment of machine learning models. The AWS server securely stores the
incoming data streams, orchestrates the execution of the prediction model, and hosts the
necessary computational resources to handle the forecasting workload.
5.Prediction Model: For predication purpose we are using Orange tool which is no code
tool.Orange’s Scoring Sheet widget provides explainable machine learning predictions
using a simple scoring system. Each feature’s influence obtains an integer score, making
it easier to understand and communicate the model, crucial in sectors where transparency
is vital.
6.Web GUI: The web-based graphical user interface (GUI) provides an intuitive plat-
form for users to interact with the system and visualize the generated forecasts. Through
the GUI, users can explore historical data trends, customize prediction parameters, and
gain valuable insights into anticipated future events, empowering informed decision-
making.

3.3 Hardware Setup

3.3.1 Circuit Diagram:
Here by using Thinkercad online simulation tool we have designed the above circuit di-
agram which interface the three sensors with ESP32 controller to collect the weather
data.To measure the temperature, humidity and pressure we have used BME280 sensor
which is connected to the Vin, GPIO21, GPIO22 and GND pin of ESP32 microcon-
troller.The Rain Gauge sensor is used to measure the rain level which gives an analog
output, it is connected to the analog pin A0 of ESP.In additional, the ESP32 analog
pin A0 is connected to the soil moisture sensor which accurately measures the level of
moisture presenent in soil.

13
 Figure 3.3: Circuit Diagram

3.3.2 PCB Design:

Figure 3.4: PCB Circuit

A printed circuit board (PCB) is a flat board with conductive pathways etched onto
it, facilitating connections between electronic components. It consists of layers of non-
conductive substrate material, typically fiberglass, bonded with copper foil. Components
like ICs, resistors, and capacitors are mounted onto the PCB using soldering techniques.
The design process involves schematic capture, circuit design, and layout optimization.

14
PCB fabrication includes substrate preparation, copper etching, and protective coating
application. Components are assembled using automated pick-and-place machines or
manual soldering. Testing ensures functionality and quality control measures maintain
consistency. PCBs are integral to electronics across industries, enabling compact and
reliable circuit integration. They serve applications in consumer electronics, telecommu-
nications, automotive, aerospace, medical devices, and industrial automation. Overall,
PCBs are essential for connecting and supporting electronic components, contributing to
the functionality and reliability of electronic devices and systems.
In our circuit we have integrated three distinct sensors with an ESP32 controller for
the comprehensive weather forecasting. The first sensor a BME280 it is I2c based sensor
which is responsible for measuring temperature, pressure and humidity which is connected
to the Vin, GPIO21, GPIO22, GND of Esp32 controller. as well as Rain Gauge sensor is
used to measure the rain level which gives an analog output, it is connected to the analog
pin A0 of ESP.In additional, the ESP32 analog pin A0 is connected to the soil moisture
sensor which accurately measures the level of moisture presenent in soil.

3.3.3 Hardware Components and Specifications

1 BME280 Sensor:-

Figure 3.5: BME280 Sensor

• It is primarily used for temperature compensation of the pressure and humidity

sensors, and can also be used for estimating ambient temperature.
• Operating Voltage: 5V
• Interface:I2C (up to 3.4MHz) ; SPI (up to 10 MHz)
• Resolution: Temperature: +-1°C ; Humidity: +-3%; Pressure: +-1Pa
• Based on GY-BME280 chip

15
2 Soil moisture sensor:-

Figure 3.6: FC-28 Soil Moisture Module

• Operating voltage: DC 3.3V - 5V

• Output voltage signal: 0 4.2V

• Current: 35mA

3 Rain Gauge sensor:-

Figure 3.7: MH-RD Rain drop module

• Operating voltage: +5V

16
• Provide both digital and analog output

• TTL Compatible

• Adjustable sensitivity

• Bolt holes for easy installation

• Compatible with Arduino

4 Wind Direction Sensor:-

Figure 3.8: RS-FXJT-N01

• Power supply: 5 V

• Operating temperature: -20℃ +60℃,0%RH 80%RH.

• Output signal: RS485

• Measuring range: 0-360°

17
5 Wind Speed Sensor:-

Figure 3.9: WS102-Anemometer

• The mode of its output signal:4 to 20mA

• supply voltage: DC 12-24V.

• Power consumption: MAX0.3W.

• Start wind speed0.4-0.8m/s.

• Effective measurement range:0-30m/s

• Transmission distance (m):¡ 1000 m

6 ESP32 Microcontroller:-

Figure 3.10: ESP32 Microcontroller

18
• Ultra-low-power management.

• Version: 1.1

• 4 MB Flash.

• Current: 80 mA.

• Frequency: 2.4 GHz.

• Data Rate: 54 Mbps

• Supply voltage: 2.2 v 3.6 v

3.4 Software Setup:

The software setup for our IoT-based Weather Monitoring and Analysis System
with GUI is carefully orchestrated to ensure seamless data collection, processing,
and visualization. Initially, weather data comprising temperature, humidity, at-
mospheric pressure, wind speed, wind direction, and soil moisture is captured by
sensors distributed across different locations. This data is then transmitted to an
ESP microcontroller, acting as a central hub for aggregation and processing. Uti-
lizing Wi-Fi connectivity, the ESP microcontroller uploads the prepared data to
the AWS Cloud infrastructure. To facilitate communication between the microcon-
troller and the cloud, we employ Node-RED as a gateway, leveraging the MQTT
protocol for efficient messaging.
Within the AWS Cloud, the received weather data is stored and processed using
services such as AWS IoT Core and AWS Lambda. Here, historical weather data is
utilized to train machine learning models with Orange, enabling predictive analysis.
These models forecast upcoming weather conditions based on the processed data,
enhancing the system’s predictive capabilities. Meanwhile, the graphical user inter-
face (GUI) is developed using Grafana, offering users a visually intuitive platform
to interact with the weather data and predictions. The GUI displays live weather
data from the sensors and presents forecasts generated by the machine learning
models in a comprehensible manner. Users can explore real-time weather trends,
compare different parameters, and access predictive forecasts seamlessly through
the GUI.
This software setup is designed with customization and scalability in mind. It
can accommodate additional sensors, machine learning algorithms, or cloud ser-
vices with ease, ensuring adaptability to evolving requirements. The GUI interface
can be tailored to meet specific user preferences and application needs, enhanc-
ing usability and accessibility. In summary, our software setup integrates sensor
data collection, cloud integration, machine learning, and graphical visualization to
deliver accurate and user-friendly weather monitoring and prediction capabilities
suitable for a wide range of applications across industries.

19
• AWS (Amazon Web Services) Server:-
AWS (Amazon Web Services) is a comprehensive cloud computing platform pro-
vided by Amazon. It offers a wide range of services that include computing power,
storage solutions, networking, databases, analytics, machine learning, and more.
Here’s an explanation of the components you mentioned in your setup:

• Instance: An instance in AWS refers to a virtual server that you can launch within
the AWS cloud. It can be configured with various operating systems, applications,
and networking settings. In your case, you’ve created an instance named ”nodered.”

• Node-RED: Node-RED is an open-source programming tool for wiring together

hardware devices, APIs, and online services. It provides a visual flow-based pro-
gramming interface, making it easy to create IoT (Internet of Things) applications
and workflows.

• Grafana: Grafana is an open-source analytics and visualization platform. It allows

you to create customizable dashboards for monitoring and analyzing data from
various sources, including databases, APIs, and streaming analytics platforms.

• MQTT (Message Queuing Telemetry Transport): MQTT is a lightweight

messaging protocol designed for low-bandwidth, high-latency, or unreliable net-
works. It’s commonly used for IoT applications to facilitate communication between
devices and servers. Port 1883 is the default port for MQTT.

• MySQL: MySQL is a popular open-source relational database management system

(RDBMS). It’s commonly used for storing and managing structured data in web
applications. You’ve installed MySQL on your AWS instance to store data.

• Ports: Ports are endpoints used by networking protocols to enable communication

between different applications or services. Each service typically uses a specific port
for communication. Here are the ports you’ve mentioned:
1.Port 22: SSH (Secure Shell) - Used for secure remote access to your AWS in-
stance.
2.Port 80: HTTP (Hypertext Transfer Protocol) - Used for web traffic, typically
for accessing web servers.
3.Port 1880: Node-RED - Node-RED uses this port for its web-based user inter-
face.
4.Port 1883: MQTT - Used for MQTT communication.
5.Port 3000: Grafana - Grafana uses this port for its web-based user interface.
By installing and configuring these services on your AWS instance, you’ve created
a platform for building IoT applications, storing data, visualizing analytics, and
managing your infrastructure in the cloud.

20
• Node Red:-
We have created following node red flows:

Figure 3.11: Node Red Flow-1

Figure 3.12: Node Red Flow-2

21
• Flow Structure: The flow consists of five separate flows labeled ”Flow 1” through
”Flow 5”. Each flow likely represents a specific functionality or processing logic
related to sensor data.

• Inject Nodes: Each flow starts with an ”inject” node, which injects a message
into the flow at regular intervals. These inject nodes are probably used to simulate
sensor readings or trigger specific actions.

• Template Nodes: After the inject node, the message is passed through a ”tem-
plate” node. This node is likely used to format the message payload, possibly to
prepare it for further processing or transmission.

• Specific Sensor Nodes: Following the template node, there are nodes labeled
with sensor names such as ”ESP/TEMP”, ”ESP/HUM”, and ”ESP/PRES”. These
nodes likely represent processing or handling specific to each type of sensor data.

• Debug Nodes: Debug nodes are used to display the message data in the Node-
RED debug sidebar. They help visualize the message payload and its properties at
different points in the flow for debugging and monitoring purposes.

• Link Nodes: Link in and link out nodes are used to connect flows together,
allowing for the triggering of one flow from another. They provide a mechanism for
organizing and structuring complex flows.

• Function Nodes: Function nodes allow for custom JavaScript code to be written
to process message data. In ”Flow 1”, there is a function node labeled ”function
4”, which likely contains custom JavaScript code to perform specific operations on
the message data.

• Catch Nodes: Catch nodes are used to catch errors that occur within the flow.
They help handle and manage errors gracefully, ensuring the smooth execution of
the flow. Overall, the flow appears to be designed to process sensor data, format it
appropriately, and perform specific actions or processing based on the data received.
Each flow likely handles a different aspect of the overall functionality, contributing
to the overall processing and handling of sensor data within the Node-RED envi-
ronment.

• Prediction Model:
In our system for the accurate weather prediction purpose we have used the Oran-
age tool. Orange is an exceptionally versatile open-source platform designed to
streamline a plethora of tasks in data science, including data visualization, machine
learning, data mining, and general data analysis. Developed primarily in Python,
Orange offers an intuitive graphical interface, effectively democratizing the field by
enabling users with varying levels of programming expertise to conduct complex
analyses. Its user-friendly drag-and-drop interface, coupled with a rich library of
widgets, empowers researchers, students, and professionals across diverse domains
to harness the power of machine learning without extensive coding requirements.

22
 Figure 3.13: Orange tool work flow

The Figure 3.13 shows a workflow created using Orange, a visual programming tool
for data science and machine learning. This particular workflow is designed for
weather prediction using linear regression. Here’s a breakdown of the components
and their roles in the workflow:
File (Input): This is where your data file is imported into the Orange environment.
The data likely includes various features related to weather, such as temperature,
humidity, wind speed, etc.
Raw Data: This widget displays the raw data imported from the file. It’s useful
for initial inspection of the dataset.
Feature Statistics: This widget calculates and displays statistics for each feature
in the dataset. It helps in understanding the distribution and properties of the
data.
Box Plot: This widget visualizes the distribution of numerical features in the
dataset using box plots. It’s useful for identifying outliers and understanding the
spread of the data.
Scatter Plot: This widget creates scatter plots to visualize relationships between
different features. It’s particularly useful for visualizing correlations between vari-
ables.
Linear Regression: This widget is used to create a linear regression model. It
takes the input data and outputs a model that can predict a target variable based
on the input features.
Test and Score: This widget evaluates the performance of the linear regression

23
model. It uses cross-validation or other evaluation methods to score the model’s
accuracy and other metrics.
Coefficients: This widget displays the coefficients of the linear regression model.
These coefficients indicate the importance and impact of each feature in predicting
the target variable.
New Case (Input): This widget is used to input new data cases that the model
hasn’t seen before. It’s for making predictions on new instances.
Predictions: This widget generates predictions for the new cases based on the
linear regression model. It outputs the predicted values for the target variable.
Temp Prediction: This appears to be another output widget that specifically
handles the predicted temperature values, possibly providing a more detailed view
or further processing of the predictions.
Output: This widget collects and displays the final output, which includes predic-
tions and possibly other processed data.
This workflow involves importing weather data, exploring and visualizing the data,
training a linear regression model, evaluating the model, and using the model to
make predictions on new data cases. The Orange tool’s visual approach makes it
easier to understand and manage the various steps in the machine learning process.

3.5 Flowchart and Algorithm:

3.5.1 Flowchart:
1.Start: This is the starting point of the process. It indicates the beginning of the
workflow.
2.Collect Weather Data Using Sensor: In this step, weather data is gathered
using a sensor. This sensor could be any device capable of measuring weather pa-
rameters such as temperature, humidity, pressure, etc.
3.Transfer Data to Microcontroller (ESP32): The collected weather data
is then transferred to a microcontroller, specifically an ESP32 in this case. The
ESP32 is a widely used microcontroller with built-in Wi-Fi and Bluetooth capabil-
ities, making it suitable for IoT applications.
4.Transmit Data to AWS Server via Gateway: The weather data collected
by the microcontroller is transmitted to an AWS (Amazon Web Services) server
using a gateway. This gateway acts as an intermediary between the microcontroller
and the server, facilitating the transfer of data over the internet.
5.Weather Condition Prediction: Upon receiving the weather data, the server
performs weather condition prediction. This could involve using machine learn-
ing algorithms, statistical models, or other forecasting techniques to predict future
weather conditions based on the collected data.
6.Display Real-time and Predicted Data on GUI: The predicted weather
data, along with real-time data if applicable, is then displayed on a Graphical User
Interface (GUI). This GUI could be a web interface, a mobile app, or any other
user-friendly interface that allows users to view the weather forecast and other rel-
evant information.

24
 Figure 3.14: Flowchart

7.Stop/End: This is the end point of the process. It indicates the completion
of the workflow. Overall, this flowchart represents a system for collecting weather
data, transmitting it to a server for prediction, and displaying the results on a user
interface. It demonstrates the various steps involved in the process and the flow of
data between different components of the system.

3.5.2 Algorithm:
Algorithm for Weather Monitoring and analysi System:

• Start: Begin the weather monitoring and prediction process.

25
• Collect Weather Data Using Sensor: Utilize weather sensors to gather data
on parameters such as temperature, humidity, pressure, etc.

• Transfer Data to Microcontroller (ESP32): - Send the collected weather data

to an ESP32 microcontroller, leveraging its Wi-Fi and Bluetooth capabilities for
data processing.

• Transmit Data to AWS Server via Gateway: Establish communication be-

tween the microcontroller and an AWS server using a gateway, facilitating data
transfer over the internet.

• Weather Condition Prediction: Upon receiving the weather data, the AWS
server performs weather condition prediction.Employ machine learning algorithms,
statistical models, or other forecasting techniques to predict future weather condi-
tions based on the collected data.

• Display Real-time and Predicted Data on GUI: Present the predicted weather
data, along with real-time data if available, on a Graphical User Interface (GUI).
Design the GUI to be user-friendly, allowing users to easily view the weather forecast
and other relevant information.

• Stop/End: Conclude the weather monitoring and prediction process.

26
Chapter 4

Results and Discussions

4.1 Result and Discussion:

Our project focuses on the comprehensive monitoring and forecasting of various
weather parameters crucial for environmental analysis and prediction. We employ a
sophisticated system comprising an array of sensors to meticulously collect real-time
data on key meteorological factors, including temperature, humidity, atmospheric
pressure, wind speed, wind direction, soil moisture, and precipitation levels through
a rain gauge.
To ensure seamless data acquisition and processing, we utilize cutting-edge tech-
nology, with the ESP32 microcontroller serving as the central hub. This microcon-
troller efficiently aggregates the sensor readings and leverages its integrated WIFI
module to transmit this data to an Amazon Web Services (AWS) server for storage
and analysis.
Once the data is securely transferred to the AWS platform, we employ advanced
data analysis techniques, particularly leveraging the capabilities of Orange, a ver-
satile machine learning tool. Through Orange’s powerful algorithms and predictive
modeling capabilities, we extrapolate insights from the accumulated data to forecast
upcoming weather conditions with a high degree of accuracy.
One of the primary deliverables of our project is the creation of an intuitive and
user-friendly interface to visualize both real-time and forecasted weather data. To
achieve this, we employ Grafana, a leading open-source analytics and monitoring
platform. The Grafana-based graphical user interface (GUI) provides users with a
dynamic dashboard showcasing live updates of temperature, humidity, atmospheric
pressure, wind speed, wind direction, soil moisture, and precipitation levels.
Furthermore, the GUI offers a predictive element by displaying forecasted values
of temperature, humidity, and atmospheric pressure. This feature empowers users
to anticipate future weather patterns and make informed decisions based on the
projected meteorological conditions.
Our project seamlessly integrates state-of-the-art sensor technology, microcontroller
systems, cloud computing infrastructure, machine learning algorithms, and visual-
ization tools to create a robust and insightful weather monitoring and forecasting

27
 system. By providing both real-time and predictive weather data in an accessi-
ble format, our solution aims to enhance decision-making processes across various
sectors reliant on accurate weather information.

Figure 4.1: Dashboard

We have successfully designed this user friendly Graphical User Interface (GUI)
using Grafana.We Display the temperature, humidity, atmospheric pressure, wind
speed and ind direction as well as soil moisture on the GUI. As well as we display
the predicted values of temperature, humidity and atmospheric pressure of the next
day on the GUI.

• Hardware Setup of the system: The given image shows the prototype of our
system. We have successfully designed the given portable system.As per the given
image we have installed the BME2800 sensor for temperature, humidity and atmo-
spheric pressure measurement. And We have installed FC-28 Soil moisture module
for measurement of moisturizer level of the soil. For the Rain measurement we
have installed the MH-RD rain drop module. For wind direction we used RS-
FXJT-NO1 sensor and for the measurement of wind speed we have installed the
WS-102 Anemometer sensor. All the above sensors integrate with the ESP32 mi-
crocontroller which take the weather parameters values from the sensors and upload
the values on the server by using communication gateway as Nodered and MTQT
protocol.

28
 Figure 4.2: Hardware Setup

4.2 Database Information:

we gather crucial environmental data including temperature, humidity, atmospheric
pressure, wind speed, wind direction, and soil moisture employing an array of sen-
sors deployed across our designated locations. To effectively organize and analyze
this extensive dataset, we rely on a MySQL database hosted on the AWS cloud
platform.The core component of our database schema is the WeatherData table,
designed to accommodate the collected weather data. This table consists of several
fields including timestamp, temperature, humidity, pressure, wind speed, wind di-

29
rection, and soil moisture. Each entry in the WeatherData table corresponds to a
specific timestamp, facilitating chronological organization and detailed analysis of
the collected data.
Data transmission from the sensors to our AWS server occurs seamlessly through
secure communication protocols such as MQTT, ensuring real-time data upload
and integrity. Once received, the data is promptly stored in the WeatherData table
within our MySQL database. This structured storage allows for efficient retrieval
and manipulation of the data for various purposes including analysis, visualization,
and reporting.Through SQL queries and statistical functions, authorized users can
access the database to retrieve specific weather parameters for a defined time period
or location. Common analyses conducted on the data include calculating averages,
identifying trends over time, and correlating different weather parameters to derive
meaningful insights.To maintain the integrity and security of our database, access
control measures are implemented to restrict database access to authorized per-
sonnel only. User roles and permissions are defined to regulate data access and
manipulation. Furthermore, data encryption protocols such as SSL/TLS are em-
ployed to safeguard data transmission between sensors and the AWS server, ensuring
confidentiality and preventing unauthorized access or tampering.

30
4.2.1 SQL Queries:

CREATE DATABASE db_iot CHARACTER SET = 'utf8' COLLATE = 'utf8_general_ci';

------------------------------------------------------------------------------------

USE db_iot;

CREATE TABLE tbl_temp (

temp_id INT UNSIGNED NOT NULL AUTO_INCREMENT,

temp_value FLOAT DEFAULT 0.00,

date TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP,

PRIMARY KEY (temp_id)

);

SELECT * from tbl_temp;

INSERT INTO tbl_temp (temp_value) VALUES (28.8)

httpsarduinogetstarted.comtutorialsarduino-mysql

------------------------------------------------------

USE db_iot;

CREATE TABLE tbl_hum (

hum_id INT UNSIGNED NOT NULL AUTO_INCREMENT,

hum_value FLOAT DEFAULT 0.00,

date TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP,

PRIMARY KEY (hum_id)

);

SELECT * from tbl_hum;

31
INSERT INTO tbl_hum (hum_value) VALUES (28.8);

------------------------------------------------------

USE db_iot;

CREATE TABLE tbl_pres (

pres_id INT UNSIGNED NOT NULL AUTO_INCREMENT,

pres_value FLOAT DEFAULT 0.00,

date TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP,

PRIMARY KEY (pres_id)

);

SELECT * from tbl_pres;

INSERT INTO tbl_pres(pres_value) VALUES (943.8);

-------------------------------------------------------

USE db_iot;

CREATE TABLE tbl_rainlvl (

rainlvl_id INT UNSIGNED NOT NULL AUTO_INCREMENT,

rainlvl_value INT DEFAULT 0,

date TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP,

PRIMARY KEY (rainlvl_id)

);

SELECT * from tbl_rainlvl;

INSERT INTO tbl_rainlvl(rainlvl_value) VALUES (25);

------------------------------------------------------------

32
USE db_iot;

CREATE TABLE tbl_sml (

sml_id INT UNSIGNED NOT NULL AUTO_INCREMENT,

sml_value INT DEFAULT 0,

date TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP,

PRIMARY KEY (sml_id)

);

SELECT * from tbl_sml;

INSERT INTO tbl_sml(sml_value) VALUES (30);

------------------------------------------------------------

USE db_iot;

CREATE TABLE tbl_wdir (

wdir_id INT UNSIGNED NOT NULL AUTO_INCREMENT,

wdir_value INT DEFAULT 0,

date TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP,

PRIMARY KEY (wdir_id)

);

SELECT * from tbl_wdir;

INSERT INTO tbl_wdir(wdir_value) VALUES (30);

--------------------------------------------------------------

USE db_iot;

CREATE TABLE tbl_wspeed (

33
 wspeed_id INT UNSIGNED NOT NULL AUTO_INCREMENT,

wspeed_value INT DEFAULT 0,

date TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP,

PRIMARY KEY (wspeed_id)

);

SELECT * from tbl_wspeed;

INSERT INTO tbl_wspeed(wspeed_value) VALUES (30);

--------------------------------------------------------

USE db_iot;

CREATE TABLE tbl_lwt (

lwt_id INT UNSIGNED NOT NULL AUTO_INCREMENT,

lwt_value INT DEFAULT 0,

date TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP,

PRIMARY KEY (lwt_id)

);

SELECT * from tbl_lwt;

INSERT INTO tbl_lwt(lwt_value) VALUES (30);

--------------------------------------------------------

USE db_iot;

CREATE TABLE tbl_lwt1 (

lwt1_id INT UNSIGNED NOT NULL AUTO_INCREMENT,

lwt1_value INT DEFAULT 0,

date TIMESTAMP NOT NULL DEFAULT CURRENT_TIMESTAMP,

34
 PRIMARY KEY (lwt1_id)

);

SELECT * from tbl_lwt1;

INSERT INTO tbl_lwt1(lwt1_value) VALUES (30);

-------------------------------------------------------

35
 Chapter 5

Advantage and Applications

5.1 Advantages
The advantages of the proposed system can be summarized as follow:

• Accurate and Real-time Data: One of the primary advantages of this project is
its ability to collect accurate and real-time weather data from various sensors. By
continuously monitoring parameters such as temperature, humidity, atmospheric
pressure, and wind speed/direction, the system provides up-to-date information
about current weather conditions.

• Comprehensive Weather Analysis: The integration of multiple sensors allows

for a comprehensive analysis of weather conditions. Beyond basic parameters like
temperature and humidity, the system also measures atmospheric pressure, rain
gauge, and soil moisture. This holistic approach provides users with a more com-
plete understanding of their local weather environment.

• Predictive Weather Forecasting: Utilizing machine learning algorithms, such

as those available in Orange, enhances the system’s capability to predict future
weather conditions. By analyzing historical data and current trends, the system
can generate accurate forecasts, enabling users to plan activities and make informed
decisions based on predicted weather outcomes.

• User-friendly Interface: The graphical user interface (GUI) makes it easy for
users to interact with the weather monitoring system. Through intuitive visuals
and easy-to-understand displays, users can access live weather data and predic-
tions without requiring specialized technical knowledge. This accessibility fosters
widespread adoption and usage of the system.

• Customizable Alerts and Notifications: The system can be configured to pro-

vide customized alerts and notifications based on specific weather conditions. Users
can set thresholds for parameters like temperature, rainfall, or wind speed, and re-
ceive alerts when these thresholds are exceeded. This feature enhances safety and
preparedness by warning users of potential weather-related hazards.

• Remote Monitoring and Accessibility: With the integration of IoT technology,

users can remotely monitor weather conditions from anywhere with an internet

36
 connection. Whether accessing the GUI from a desktop computer, smartphone,
or tablet, users have convenient access to real-time weather data and forecasts,
enhancing their situational awareness.

• Support for Decision-making: By providing accurate data and predictive in-

sights, the system supports decision-making across various sectors and industries.
Farmers can optimize irrigation schedules based on soil moisture levels, outdoor
event planners can anticipate weather changes, and emergency responders can pre-
pare for extreme weather events, ultimately improving efficiency and safety.

• Environmental Monitoring and Sustainability: Beyond individual benefits,

the weather monitoring system contributes to environmental monitoring and sus-
tainability efforts. By collecting data on weather patterns and environmental pa-
rameters, the system facilitates research into climate change, ecological impacts,
and resource management, empowering initiatives aimed at preserving the planet.

• Scalability and Adaptability: The modular design of the system allows for
scalability and adaptability to different environments and applications. Whether
deployed in urban areas, rural landscapes, or industrial settings, the system can be
customized to meet specific requirements and expand to accommodate additional
sensors or functionalities as needed.

• Data-driven Insights and Innovation: Through the collection and analysis

of large volumes of weather data, the system generates valuable insights that drive
innovation and continuous improvement. Researchers, meteorologists, and data sci-
entists can leverage this data to develop new models, algorithms, and applications,
advancing the field of weather forecasting and analysis.

37
 5.2 Applications
The proposed system can be employed at variety of applications such as :

• Agriculture and Farming: Farmers can utilize the system to optimize irriga-
tion schedules based on soil moisture levels and weather forecasts. By monitoring
weather conditions and soil moisture in real-time, farmers can improve crop yields,
conserve water resources, and mitigate risks associated with extreme weather events.

• Urban Planning and Infrastructure Management: City planners can leverage

the system to make informed decisions regarding urban infrastructure and develop-
ment projects. By analyzing weather data and predictions, city officials can imple-
ment measures to mitigate the impact of weather-related hazards such as flooding,
heatwaves, or storms, and design resilient infrastructure systems.

• Energy Management: Energy companies can use the system to optimize en-
ergy production and distribution based on weather patterns. For example, wind
farm operators can adjust turbine operations according to wind speed and direc-
tion forecasts, while solar power plant managers can anticipate fluctuations in solar
radiation to optimize energy generation.

• Transportation and Logistics: Transportation agencies and logistics companies

can benefit from real-time weather monitoring to improve safety and efficiency. By
integrating weather data into route planning and scheduling systems, transportation
operators can anticipate weather-related disruptions, optimize travel routes, and
minimize delays in logistics operations.

• Emergency Response and Disaster Management: Emergency responders can

utilize the system to prepare for and respond to weather-related emergencies more
effectively. By monitoring weather conditions and predictions, emergency manage-
ment agencies can issue timely warnings, evacuate at-risk areas, and coordinate
disaster response efforts to mitigate the impact of severe weather events.

• Tourism and Outdoor Recreation: Tourism operators and outdoor recreation

facilities can use the system to provide visitors with up-to-date weather information
and safety recommendations. By integrating weather data into tourism websites or
mobile apps, businesses can enhance visitor experiences, promote outdoor activities
during favorable weather conditions, and ensure visitor safety during inclement
weather.

• Environmental Monitoring and Conservation: Environmental organizations

can leverage the system to monitor and assess the impact of weather on ecosys-
tems and natural habitats. By collecting data on weather patterns, soil moisture,
and other environmental parameters, conservationists can identify areas at risk of
drought, erosion, or habitat loss and implement targeted conservation measures to
protect biodiversity.

• Research and Education:Academic institutions and research organizations can

use the system as a valuable tool for studying climate trends, conducting research
on weather-related phenomena, and educating students about meteorology and en-
vironmental science. By providing access to real-time weather data and analysis

38
 tools, the system enhances research capabilities and facilitates hands-on learning
experiences for students.

• Healthcare and Public Health: Healthcare providers and public health agen-
cies can utilize the system to monitor weather-related health risks and implement
preventive measures. By analyzing weather data and predictions, health authorities
can anticipate outbreaks of weather-related illnesses such as heatstroke, respiratory
infections, or vector-borne diseases and implement public health interventions to
protect vulnerable populations.

• Smart Home Automation: Homeowners can integrate the system into smart
home automation systems to enhance comfort, energy efficiency, and safety. By
incorporating weather data into smart thermostats, irrigation systems, and home
security devices, homeowners can adjust indoor climate settings, automate watering
schedules, and receive alerts about potential weather-related hazards such as storms
or freezing temperatures.

39
 Chapter 6

Conclusion and Future Scope

6.1 Conclusion
In conclusion, the IoT-based weather monitoring and analysis system using a graph-
ical user interface (GUI) project offers a comprehensive solution for gathering, ana-
lyzing, and presenting weather data. Through a network of sensors, including those
measuring temperature, humidity, atmospheric pressure, rain gauge, wind speed,
wind direction, and soil moisture, the system collects real-time environmental data.
Leveraging the Orange machine learning tool, the collected data is then processed
and analyzed to predict weather conditions. The graphical user interface provides
users with intuitive access to both real-time and predicted weather data, facilitating
informed decision-making and enhancing situational awareness. This project not
only demonstrates the power of IoT technology in environmental monitoring but
also highlights the potential of machine learning for accurate weather forecasting.
Overall, the system serves as a valuable tool for various applications, including
agriculture, urban planning, disaster management, and more, by providing action-
able insights into weather patterns and trends.In essence, the IoT-based weather
monitoring and analysis system with a GUI project exemplifies the convergence
of cutting-edge technology and practical utility. Through its comprehensive data
collection, advanced analysis capabilities, and user-friendly interface, the system
stands as a testament to the potential of IoT and machine learning in environ-
mental monitoring. As such, it holds promise for revolutionizing decision-making
processes and driving innovation across a multitude of applications and industries.

6.2 Future Scope

The future scope for an IoT-based weather monitoring and analysis system using a
GUI:

• Enhanced Sensor Integration: In the future, the system can incorporate ad-
vanced sensors to gather more comprehensive weather data. This might include
sensors for UV radiation, air quality (such as CO2 levels), and even pollens count
for allergy forecasting. The integration of such sensors will provide users with a
more holistic understanding of their environment.

40
• Real-time Data Visualization: While the project already features a GUI for
displaying live weather data, future advancements can focus on improving the visu-
alization capabilities. Implementing 3D weather maps or interactive graphs can pro-
vide users with a more intuitive and engaging experience when monitoring weather
conditions.
• Predictive Analytics: Integration with machine learning algorithms, such as
those available in Orange, can be expanded to include more sophisticated predic-
tive analytics. For instance, incorporating deep learning models can improve the
accuracy of weather forecasting, enabling the system to provide more reliable pre-
dictions of future weather conditions.
• Geographical Expansion: Currently, the system collects weather data from a
specific location. However, in the future, it can be scaled up to cover larger geo-
graphical areas. This expansion can involve deploying multiple sensor nodes across
different locations, allowing users to monitor weather conditions in various regions
simultaneously.
• Mobile Application Integration: Developing a mobile application companion
to the GUI can enhance the accessibility of the weather monitoring system. Users
can access real-time weather data and forecasts on their smartphones or tablets,
enabling them to stay informed about changing weather conditions while on the go.
• Data Analytics and Insights: Beyond basic weather monitoring, the system can
offer advanced data analytics capabilities. Users can leverage the collected weather
data to gain insights into long-term climate trends, seasonal patterns, and potential
correlations with other environmental factors. This can be particularly valuable for
researchers, urban planners, and agriculture professionals.
• Integration with Smart Home Systems: Integration with smart home systems,
such as Google Home or Amazon Alexa, can enable users to interact with the
weather monitoring system using voice commands. For example, users can inquire
about current weather conditions or receive weather alerts simply by speaking to
their smart devices.
• Customization and Personalization: Introducing customization features to the
GUI will allow users to tailor the displayed weather information according to their
preferences. Users can choose which weather parameters they want to prioritize,
set personalized alerts for specific weather events, and customize the layout of the
GUI to suit their needs.
• Community Engagement and Crowdsourcing: Implementing features that
encourage community engagement and crowdsourcing of weather data can enrich
the system’s capabilities. Users can contribute their own weather observations
through the GUI or mobile app, helping to improve the accuracy of local weather
forecasts and fostering a sense of collaboration among users.
• Resilience and Disaster Preparedness: Finally, the system can be further
developed to support resilience and disaster preparedness efforts. This can involve
integrating early warning systems for extreme weather events, such as hurricanes or
floods, and providing users with actionable insights to mitigate risks and enhance
their preparedness.

41
Chapter 7

References

1. 1.Gaurav Verma, Pranjul Mittal, Shaista Farheen (Department of Electron-

ics Communication) Dayananda Sagar College of Engineering/ “Real Time
Weather Prediction System Using IOT and Machine Learning”/
IEEE 2020 https://ieeexplore.ieee.org/document/9182766
2. 2.Dires Negash Fente, Prof. Dheeraj Kumar Singh, (Department of Informa-
tion technology, Parul University Parul Institute of Engineering and Technol-
ogy)/ “Weather Forecasting Using Artificial Neural Network”/
IEEE 2018 https://ieeexplore.ieee.org/document/8473167
3. 3.Sushmitha Kothapalli, S. G. Totad, (Dept of Computer Science and Engi-
neering) “A Real-Time Weather Forecasting and Analysis”/
IEEE 2017 https://ieeexplore.ieee.org/document/8391974
4. 4.Shivam Tandon, Abhishek Patel, Pawan Kumar Singh (School of Computing
Science and Engineering Galgotias University Uttar Pardesh, India/ “Weather
Prediction Using Machine Learning Algorithms”/
IEEE 2020 https://ieeexplore.ieee.org/document/9862337
5. 5.A THESIS Submitted by - KALA A (Faculty of Information Communication
Engineering ANNA UNIVERSITY CHENNAI)/ “Hybrid Weather Forecasting
Models Based On Deep Learning And Mode Decompositions Methods”/
Shodhganga 2022 http://hdl.handle.net/10603/477722
6. 6.Tulsi Pawan Fowdur, Rosun Mohammad Nassir-Ud-Diin Ibn Nazir (Depart-
ment of Electrical and Electronic Engineering, University of Mauritius, Reduit,
Mauritius)/ “A real-time collaborative machine learning based weather fore-
casting system with multiple predictor locations”/ journal homepage: www.sciencedirect.com
7. 7.A THESIS Submitted by - Uma Sharma (Faculty of Mathematics Com-
puting Banasthali Vidyapith Rajasthan, India)/ “Smart Weather Forecasting
Using Deep Learning”/ Shodhganga 2022 http://hdl.handle.net/10603/406576
8. 8.A THESIS Submitted by - VASAVI RAVURI (Research Development Jawa-
harlal Nehru Technological University)/ “Optimized Deep LSTM Based En-
semble Classifier for Weather Forecasting”/
Shodhganga 2023 https://shodhganga.inflibnet.ac.in/handle/10603/467870

42
Appendix A

Project Outcomes

A.1 Plagiarism Report

43
44
A.2 CopyRight

fnukad/Dated:07/12/2023
1. iathdj.k la[;k/Registration Number : L-137228/2023
2. vkosnd dk uke] irk rFkk jk"Vªh;rk : YOGESH YUVRAJ PAWAR , AT- VEKHANDWADI, POST -
Name, address and nationality of the applicant TARALE, TAL- PATAN, DIST- SATARA-415014
INDIAN
ADESH MADHUKAR DHAGE , KALAS BK, TAL – AKOLE,
DIST- AHMEDNAGAR-422601
INDIAN
ROHAN KISAN YADAV , PADMINI NIWAS ,MATHURA
COLONY , NAKHATE WASTI , RAHATANI-411017
INDIAN
DIPALI SHENDE , “SAFALYA”, PLOT NO. 27, SR. NO.
32/2K,NEAR TAMBE SCHOOL, RAHATNI ROAD, PIMPRI
PUNE 17-411017
INDIAN
3. —fr ds çfrfyI;f/kdkj esa vkosnd ds fgr dh ç—fr : AUTHOR
Nature of the applicant's interest in the copyright of the work
4. —fr dk oxZ vkSj o.kZu : LITERARY/ DRAMATIC WORK THE PROJECT IOT BASED
Class and description of the work WEATHER MONITORING AND ANALYSIS USING GUI,
GIVES A WELL MADE AND ACCURATE WEATHER
FORECASTING AND CONTINUOUSLY CHANGING
WEATHER CONDITIONS REPORTS.
5. —fr dk 'kh"kZd : INTERNET OF THINGS BASED WEATHER MONITORING
Title of the work AND ANALYSIS USING GRAPHICAL USER INTERFACE

6. —fr dh Hkk"kk : ENGLISH

Language of the work
7. jpf;rk dk uke] irk vkSj jk"Vªh;rk rFkk ;fn jpf;rk dh e`R;q gks xbZ gS] : YOGESH YUVRAJ PAWAR , AT- VEKHANDWADI, POST -
rks e`R;q dh frfFk TARALE, TAL- PATAN, DIST- SATARA-415014
Name, address and nationality of the author and if the author is INDIAN
deceased, date of his decease
ADESH MADHUKAR DHAGE , KALAS BK, TAL – AKOLE,
DIST- AHMEDNAGAR-422601
INDIAN
ROHAN KISAN YADAV , PADMINI NIWAS ,MATHURA
COLONY , NAKHATE WASTI , RAHATANI-411017
INDIAN
DIPALI SHENDE , “SAFALYA”, PLOT NO. 27, SR. NO.
32/2K,NEAR TAMBE SCHOOL, RAHATNI ROAD, PIMPRI
PUNE 17-411017
INDIAN

8. —fr çdkf'kr gS ;k vçdkf'kr : UNPUBLISHED

Whether the work is published or unpublished

9. çFke çdk'ku dk o"kZ vkSj ns'k rFkk çdk'kd dk uke] irk vkSj jk"Vªh;rk : N.A.
Year and country of first publication and name, address and
nationality of the publisher
10. ckn ds çdk'kuksa ds o"kZ vkSj ns'k] ;fn dksbZ gksa] vkSj çdk'kdksa ds uke] irs : N.A.
vkSj jk"Vªh;rk,¡
Years and countries of subsequent publications, if any, and names,
addresses and nationalities of the publishers

11. —fr esa çfrfyI;f/kdkj lfgr fofHkUu vf/kdkjksa ds Lokfe;ksa ds uke] irs vkSj : YOGESH YUVRAJ PAWAR , AT- VEKHANDWADI, POST -
jk"Vªh;rk,a vkSj leuqns'ku vkSj vuqKfIr;ksa ds fooj.k ds lkFk çR;sd ds TARALE, TAL- PATAN, DIST- SATARA-415014
vf/kdkj dk foLrkj] ;fn dksbZ gksA INDIAN
Names, addresses and nationalities of the owners of various rights ADESH MADHUKAR DHAGE , KALAS BK, TAL – AKOLE,
comprising the copyright in the work and the extent of rights held DIST- AHMEDNAGAR-422601
by each, together with particulars of assignments and licences, if INDIAN
any ROHAN KISAN YADAV , PADMINI NIWAS ,MATHURA
COLONY , NAKHATE WASTI , RAHATANI-411017
INDIAN
DIPALI SHENDE , “SAFALYA”, PLOT NO. 27, SR. NO.
32/2K,NEAR TAMBE SCHOOL, RAHATNI ROAD, PIMPRI
PUNE 17-411017
INDIAN

12. vU; O;fä;ksa ds uke] irs vkSj jk"Vªh;rk,a] ;fn dksbZ gksa] tks çfrfyI;f/kdkj : YOGESH YUVRAJ PAWAR , AT- VEKHANDWADI, POST -
okys vf/kdkjksa dks leuqnsf'kr djus ;k vuqKfIr nsus ds fy, vf/k—r gksa TARALE, TAL- PATAN, DIST- SATARA-415014
Names, addresses and nationalities of other persons, if any, INDIAN
authorised to assign or licence of rights comprising the copyright ADESH MADHUKAR DHAGE , KALAS BK, TAL – AKOLE,
DIST- AHMEDNAGAR-422601
INDIAN
ROHAN KISAN YADAV , PADMINI NIWAS ,MATHURA
COLONY , NAKHATE WASTI , RAHATANI-411017
INDIAN
DIPALI SHENDE , “SAFALYA”, PLOT NO. 27, SR. NO.
32/2K,NEAR TAMBE SCHOOL, RAHATNI ROAD, PIMPRI
PUNE 17-411017
INDIAN

13. ;fn —fr ,d 'dykRed —fr' gS] rks —fr ij vf/kdkj j[kus okys O;fä dk : N.A.
uke] irk vkSj jk"Vªh;rk lfgr ewy —fr dk LFkkuA ¼,d okLrqf'kYi —fr
ds ekeys esa —fr iwjh gksus dk o"kZ Hkh fn[kk;k tkuk pkfg,½
If the work is an 'Artistic work', the location of the original work,
including name, address and nationality of the person in possession
of the work. (In the case of an architectural work, the year of
completion of the work should also be shown).

Registrar of Copyrights
45
46
A.3 Research Paper

IOT Based Weather Monitoring & Analysis

Using GUI
Adesh Dhage Yogesh Pawar
Department of E&TC, Pimpri Chinchwad College of Department of E&TC, Pimpri Chinchwad College of
Engineering and Research, Ravet, Engineering and Research, Ravet,
Pune, India Pune, India
adesh.dhage_entc21@pccoer.in yogesh.pawar_etc2020@pccoer.in

Rohan Yadav Dipali Shende

Department of E&TC, Pimpri Chinchwad College of Department of E&TC, Pimpri Chinchwad College of
Engineering and Research, Ravet, Engineering and Research, Ravet,
Pune, India Pune, India
rohan.yadav_entc21@pccoer.in dipali.shende@pccoer.in

Abstract— Weather forecasting is the application of science unfavorable weather conditions and raises the quantity and
and technology to predict the state of the atmosphere at a quality of agricultural output. aids in seasonal or long-term
particular location. The majority of early weather forecasting planning as well as crop selection according to ideal weather
techniques were based on pattern identification, or the conditions. Seasonal forecasting contributes to the
observation of recurring patterns in the events. For instance, betterment of farmers lives in areas where there is
one may notice that an especially crimson sunset typically significant fluctuation in yearly rainfall. Seasonal forecasts
heralded favorable weather the next day. Not all of these play an important role in deciding the eco- nomic policies of
forecasts, though, turn out to be accurate. Here this system will governments, during major drought was forecasted,
predict weather based on parameters such as temperature,
monetary policy could be relaxed to maintain growth
humidity, wind speed, rain gauge, wind direction, soil
temperature, atmospheric pressure, soil moisture. We will
targets. The unique aspect of our approach is that ability to
collect the weather data with the help of some sensors and pass provide fast and reliable forecasts while operating on low-
it to the system. System will take this parameter and will cost and resource- efficient computing systems. These
predict weather from previous data in database(dataset). The forecasts have practical applications in our daily lives,
primary objective of our project is to develop a well- made and enabling us to make informed decisions and adapt
accurate weather forecasting model capable of predicting effectively to changing weather conditions. This paper
climate changes over extended periods and assessing their presents the review on the various methods of Weather
implications for future generations. The weather forecasting Forecasting. It focuses on the accurate weather prediction
system forecasts weather based on historical records; thus, this using advance IOT[1] and Machine Learning model[5][6]
prediction will prove to be trustworthy. It takes into account which predicts the Realtime weather conditions at max
elements like temperature, humidity, wind speed, rain gauge, accuracy and this data is displayed on the dashboard.
wind direction, soil temperature, atmospheric pressure, soil
moisture. Air traffic, maritime, agriculture, forestry, military, II. LITERATURE REVIEW
navy, and other fields may all use this technology.

Keywords— BME280, Rain Gauge Sensor, Soil 1] Gaurav Verma, Pranjul Mittal, Shaista Farheen
Moisture Sensor, Wind Direction Sensor, Wind Speed (Department of Electronics Communication) Dayananda
Sensor, ESP32 Controller, Linear Regression, AWS, Sagar College of Engineering/ “Real Time Weather
Graphical User Interface. Prediction System Using IOT and Machine Learning”/ IEEE
2020 - Methodology Used ‘Logistic Regression Model’,
I. INTRODUCTION (HEADING 1) Parameters ‘Temp., Humidity, Light Intensity’, Conclusion
According The Indian economy's most significant sector ‘By using sensors machine Learning module trained with
is agriculture. Fifty percent of India's workforce is employed prerecorded data and predict weather forecast’. — The
in the agriculture industry, which generates eighteen percent emphasis of today is on intelligent technologies, such as
of the country's GDP. The world's top producer of wheat, machine learning and the Internet of Things. There are
rice, pulses, spices, and spice-related items is India. The numerous IoT implementations available for IoT hardware
weather is a vital natural element for life; depending on its platforms. Among them is the ESP8266 chip. This study
severity, it may either provide or destroy opportunities for creates a real-time weather prediction system that may be
survival. To put it simply, weather forecasting is the process used to forecast the weather in a variety of settings, including
of predicting the state of the atmosphere in a given location residences, businesses, stadiums, and agricultural settings.
at a given time utilizing physics principles, technology, and The system makes use of an LDR light intensity sensor and a
a range of statistical and empirical methodologies. The DHT11 temperature and humidity sensor. Using an
weather prediction is incredibly helpful in protecting both ESP8266-01 module and NodeMCU, the sensed data from
people and property and to protect the crops and it will tell the sensors is uploaded to a ThingSpeak cloud server.
us what to expect in our atmospheric environment. Weather
prediction data is essential for making quick modifications 2] Dires Negash Fente, Prof. Dheeraj Kumar Singh,
to daily agricultural activities, which reduces losses due to (Department of Information technology, Parul University

XXX-X-XXXX-XXXX-X/XX/$XX.00 ©20XX IEEE

47
Parul Institute of Engineering and Technology)/ “Weather 4] Tulsi Pawan Fowdur, Rosun Mohammad Nassir-Ud-
Fore- casting Using Artificial Neural Network”/ IEEE 2018 Diin Ibn Nazir (Department of Electrical and Electronic
- Methodology Used ‘LSTM – Long short-term Memory’, Engineering, University of Mauritius, Reduit, Mauritius)/
Parameters ‘Temp., Humidity, Light Intensity, Wind Speed, “A real-time collaborative machine learning based weather
Wind Direction’, Conclusion ‘LSTM gives Result with forecasting system with multiple predictor locations” –
High Accuracy’. Since the weather has a major impact on Weather forecasting is an essential application in
the agricultural and industrial sectors, accurate weather meteorology and has been one of the world's most
forecasting is essential in today's society. Additionally, it is technically and technologically difficult topics. As the
employed to predict and warm up natural disasters. devastating repercussions of climate change continue to
Determining the appropriate values for weather parameters unfold, accurate localized short-term weather prediction has
and projecting future weather conditions based on these become more vital than ever. In this research, we present a
factors is known as weather forecasting. In this study, collaborative machine learning-based real-time weather
several meteorological parameters were gathered from the forecasting system that uses data from several places to
National Climate Data Center, and the neural network was predict the weather for a single site. Five machine learning
trained for various combinations using the long-short term algorithms were utilized in this work, and tests were
memory (LSTM) technique. conducted in four distinct sites in Mauritius to forecast
weather. Temperature, wind speed, wind direction, pressure,
humidity, and cloudiness are all variables to consider. The
3] Sushmitha Kothapalli, S. G. Totad, (Dept of Open Weather API was used to collect the weather data.
Computer Science and Engineering) “A Real-Time Weather
Forecasting and Analysis”/ IEEE 2017 - Methodology Used
‘ARIMA Model’, Parameters ‘Temp., Humidity, Light 5] A THESIS Submitted by - KALA A (Faculty of
Intensity, Wind, Rainfall’, Conclusion ‘Val- ues are Information Communication Engineering ANNA
Predicted by implementing ARIMA Model using R Studio’. UNIVERSITY CHENNAI)/ “Hybrid Weather Forecasting
- The endeavor of meteorologists to anticipate the weather Models Based On Deep Learning And Mode
and potential outcomes at a future date is known as weather Decompositions Methods”/ Shodhganga 2022 - Method-
forecasting. Temperature, wind, humidity, rainfall, and data ology Used ‘Artificial Neural Network, hybrid model based
set size are the factors that determine the climatic state on long-short term memory the artificial algae algorithm
parameters. In this case, only the parameters of temperature (LSTM Model)’, Result Component ‘Result Shows that
and humidity are taken into account for the experimental hybrid performance model perform a better prediction in
study. The DHT11 sensor, which aids in determining the comparison with LSTM.’, Future Scope ‘It could be
temperature and humidity values of a certain region or extended to forecast other events such as drought and flood’
location, provides the temperature and humidity data that is & The suggested hybrid models may be expanded to predict
gathered. Using an Ethernet shield to facilitate online data additional occurrences like floods and droughts. It is
uploading, the raspberry pi is utilized to save the gathered possible to create a hybrid model for drought forecasting
data on the cloud.)Shivam Tandon, Abhishek Patel, Pawan that blends linear and nonlinear models by treating the
Kumar Singh (School of Computing Scienceand frequency and intensity of droughts as random events.
Engineering Galgotias University Uttar Pardesh, India/
“Weather Prediction Us- ing Machine Learning 6] A THESIS Submitted by - Uma Sharma (Faculty of
Algorithms”/ IEEE 2020 - Methodology Used ‘Random Mathematics Computing Banasthali Vidyapith Rajasthan,
Forest, Decision Tree, MLP Classifier, Linear Regression’, India)/ “Smart Weather Forecasting Using Deep Learning”/
Parameters ‘Temp., Humidity, Light Intensity, Wind, Shodhganga 2022- Methodology Used ‘Bi-LSTM Model for
Rainfall, Visibility’, Conclusion ‘By using Machine historical weather data. Based on ARIMA Model and Deep
Learning Module gives Accurate Result’. — Climate plays a Learning Model CRNN’, Result Component ‘Increasing the
crucial role in numerous important production industries, size of training data sample cloud result in better prediction
such as agriculture. These days, there is a strong emphasis and deep learning produce accurate result.’, Future Scope
on climate change, which is why historical weather forecasts ‘Combining different local Weather station data regional
are becoming less accurate and more accurate. For this weather data for an accurate prediction’ and Seven distinct
reason, miles play a crucial role in customizing and surface weather parameters are used in this study. Higher
adjusting the weather forecast model. Both the nation's input counts would most likely produce better outcomes.
financial system and people's lives are impacted by those But doing so will make the model more complicated and
forecasts. A forested area utilized for weather forecasting is need estimating a lot more parameters. Furthermore, the
part of an information and statistics analysis algorithm suggested model is trained using 70% of the entire
system. One of the strongest natural barriers we face in life meteorological data collected during a mere 2792 days. A
is the weather; we must consider factors such as humidity, deeper learning network may be predicted more accurately
temperature, and other forms of protection. astounding. Our by expanding the training data sample. More data may be
artwork aims to format weather predictions in an efficient used to refine the developed model and improve its
manner. performance. By adding more meteorological
characteristics, such as soil temperature, soil moisture,
snow, solar radiation balance, and pressure at various levels,
it could be possible to enhance the forecast.

48
 7] A THESIS Submitted by - VASAVI RAVURI III. GAP IDENTIFICATION
(Research Development Jawaharlal Nehr Technological
University)/ “Optimized Deep LSTM Based Ensemble Sr. No Journal Name Gap/Limitations
Classifier for Weather Forecasting”/ Shodhganga 2023 - 1. Real time weather prediction By using sensors & machine
Methodology Used ‘Use of big data Clustering using Spark. system using IOT and Machine Learning module trained with
Prediction weather with MFATA – based deep LSTM using Learning prerecorded data and predict
time series.’, Result Component ‘It produce very accurate weather forecast but it not
result depending on size of data efficiency.’, Future Scope gives Realtime forecast.
‘Interfacing IOT Sensors increase more efficiency using
data mining tech’ and The goal of this effort is to anticipate 2. weather forecasting using LSTM gives Result with
weather properly in situations where there is a sudden shift Artificial Neural Network High Accuracy but by using
linear regression it gives high
in the pattern of the weather. We applied the idea of accuracy.
reciprocal connection between meteorological variables to
solve this issue. In order to predict a single output feature, 3. Realtime weather forecasting Values are Predicted by
and Analysis implementing ARIMA Model
we presented our innovative SFA-LSTM model in this
using R Studio.
study. It has an integrated spatial feature attention
mechanism that allows it to capture long-term dependencies
and spatial feature correlations of multivariate input time 4. A real-time collaborative By using Machine Learning
series. When abrupt changes in input sequences are seen, the machine learning based Module gives Accurate
weather forecasting system Result but It can trained
spatial feature attention mechanism accurately predicts the
with multiple predictor better to achieve max
future by understanding the quantitative mutual effect of locations accuracy.
input characteristics on target features. Simultaneous shifts
noticed in following weather variables that are mutually
5. Hybrid weather Fore casting It could be extended to
connected can be used to determine the extent of a weather
models based in deep learning forecast other events such as
feature shift. The weight of each geographic feature's effect and mode decomposition drought and flood.
from multiple weather variables on the target variable may methods
be ascertained by using multivariate weather variables to
forecast a single target weather characteristic. Accurately 6. smart weather forecasting using Combining different local
forecasting future weather over extended periods of time is deep learning Weather station data &
aided by capturing these relationships during model regional weather data for an
learning. The suggested model was constructed using an accurate prediction
encoder-decoder architecture, in which spatial relations are
learned in the decoder phase and temporal dependencies in 7. Optimized Deep LSTM Based Interfacing IOT Sensors
the data are learned in the encoder phase using LSTM Ensemble Classifier for increase more efficiency
layers. SFA-LSTM is seen to provide correct spatial feature Weather Forecasting using data mining tech.
interpretability, outperforming the performance of the state-
of-the-art models.

IV. GENERALIZED METHODOLOGY • Machine Learning Model: Our next step involves
building a machine learning model, specifically employing
linear regression. To train this model, we will utilize
In this project, our primary objective is to create a historical weather data from the past five years. This process
comprehensive weather monitoring and prediction system allows the model to master and understand the relationships
for agricultural purposes. To achieve this, we will follow a between various weather parameters.
structured approach according to flowchart:
• Data Comparison and Prediction: Once the
• Data Collection: Initially, we will gather weather- machine learning model is trained, it will retrieve the live
related data using various sensors. Specifically, we will use data uploaded to the server. It will then compare this real-
to measure parameters such as temperature, humidity, wind time data with historical data from the sensors. Using the
speed, rain gauge, wind direction, soil temperature, insights gained from the training, the model will generate
atmospheric pressure, soil moisture by using specific accurate weather predictions.
sensors.
• Web GUI Display: The final step is to display these
• Data Transmission: We will employ a ESP32 to predictions in a user-friendly manner through a web-based
upload the collected data to a central server. This server acts graphical user interface (GUI). Users can use this user
as a repository for the weather data, ensuring its availability interface to view the current weather conditions and
for analysis and prediction. forecasts, which are generated by the machine learning
model’s analysis of live and historical data.

49
 2.Rain Gauge Sensor

A clever and reasonably priced rain detection gadget is

the rain drop sensor module.[1] It consists of a control board
and a rain detecting pad. The control board receives these
signals and has the ability to binarize them, while the
sensitive sensor pad detects the presence of any water on it.
The automotive sector is one of the main applications for the
rain drop module. When rain is detected, it may be used to
monitor the weather and request that shutters or windows be
closed. This is a handbook to assist in creating our project.

Fig. 1. Block Diagram

Fig. 4. Rain Gauge

Sensor

3.Wind Speed Sensor

One device that measures wind speed is the three cups
type Wind Speed Sensor Voltage Type (0-5V)
Anemometer Kit. It is made up of the circuit module, wind
cup, and shell. The internal drive integrates photovoltaic
modules, an industrial microcomputer processor, a current
generator, electric current and other components. The
sensor shell and wind cups are made of an aluminum alloy
that is carefully cast using a unique mold technique,
resulting in extremely tiny tolerances. The sensor has a
high strength, is waterproof, resistant to corrosion, and has
an internal circuit that has been protected. The surface
precision is also quite good. The cable's plug is a military
plug with strong anticorrosive and erosion-prevention
capabilities. In the event that pertinent standards are
applied to guarantee the correctness of the wind speed
acquisition, it can guarantee the instrument employed for
Fig. 2. Flowchart
an extended period of time. Military-grade A material
makes up the circuit PCB, ensuring both the quality and
stability of the electrical qualities. All electronic
components are industrial chips, which ensure that the host
V. Hardware Components & Software Specification can function correctly at temperatures between -20°C and
50°C and humidity levels between 35% and 85%
1.BME280 Sensor (condensation). Overall, these chips have incredibly
You may use BME280 (datasheet, Adafruit)[1] dependable electromagnetic interference resistance.
humidity, pressure, and temperature sensors with ESP32
Home by using the bme280 sensor platform. The sensor is
utilized in forced mode, which involves taking a
measurement and then putting it to sleep until the next
measurement. For this sensor to function, the I2C has to be
configured in your setup.

Fig. 5. Wind Speed Sensor

Fig. 3. BME280 Sensor

50
4.Wind Direction Sensor 7.Machine Learning Algorithm
The FST200-202 wind direction sensor is utilized in
several domains, including meteorology, environment, Linear Regression is a supervised learning algorithm that
ropeway, greenhouse, and breeding, to determine the falls under the category of machine learning. Its primary
direction of wind. Alloy materials with surface treatment are objective is to perform regression tasks by modeling a target
used to make the major structural components. The product prediction value using independent variables. This algorithm
resists corrosion and is waterproof well. Since all of the is commonly employed to determine the relationship
internal and rotational components are sealed, water, salt between variables and make accurate forecasts. The various
spray, and sand dust entry may be successfully avoided. regression models available differ based on the type of
relationship they consider between dependent and
independent variables, as well as the number of independent
variables utilized.

Fig. 6. Wind Direction

Sensor

5.Soil Moisture Sensor

The two probes that make up the soil moisture sensor are
used to determine the volume of water in the soil. In order to
assess the moisture content, the two probes let current flow
through the soil and then obtain the resistance value. There Fig. 9. Linear Regression Graph
will be less resistance when there is more water since the
soil will conduct more electricity. As a result, there will be
8.AWS Server
more moisture present. Since dry earth doesn't carry
electricity well, when because the soil will conduct less
Amazon.com offers Amazon Web Services (AWS), a
electricity when there is less water present, there will be
prominent cloud computing platform that delivers various
greater resistance. As a result, there will be less moisture.
services such as computing power, storage solutions,
networking, databases, machine learning, and more. These
services are accessible through a secure and scalable cloud
infrastructure. AWS enables users to efficiently deploy,
oversee, and expand applications and services worldwide,
with costs based on actual resource consumption. The
platform's flexibility, reliability, and scalability make it a
preferred option for startups, corporations, and
Fig. 7. Soil Moisture
Sensor
governmental organizations seeking to enhance and expand
their digital projects.

7.ESP32 Controller 8.GUI

The ESP32 line of low-cost, low-power system-on-a-chip To develop a dashboard that presents real-time weather
microcontrollers has dual-mode Bluetooth and integrated conditions, Grafana and MySQL can be utilized as essential
Wi-Fi. The ESP32 series incorporates built-in antenna elements of the solution. The initial step involves
switches, power amplifier, low-noise receive amplifier, configuring MySQL as the database to house weather-
filters, and power-management modules. It uses either a related information, such as temperature, humidity, wind
Tensilica Xtensa LX6 microprocessor in both dual-core and speed, and precipitation metrics. By integrating a weather
single-core variations, Xtensa LX7 dual-core API, the most recent data can be regularly retrieved and
microprocessor, or a single-core RISC-V microprocessor. stored in the MySQL database. Subsequently, Grafana
The Shanghai-based Chinese business Espress if Systems should be installed and configured to establish a connection
designed and developed ESP32, which is produced by with the MySQL database as a data source. Leveraging
TSMC utilizing their 40 nm technology. It is the ESP8266 Grafana's user-friendly interface, users can create
microcontroller's replacement. personalized dashboards that exhibit weather metrics
through diverse visualization panels like graphs, tables, and
gauges. By crafting SQL queries within Grafana, specific
weather data can be extracted, and the dashboard layout can
be tailored to effectively present information. Additionally,
users have the option to configure alerts that notify them of
Fig. 8. ESP32 critical weather conditions. It is imperative to continuously
assess and enhance the dashboard to ensure its precision and
user-friendliness, facilitating the acquisition of valuable
insights into real-time weather patterns and trends.

51
 IV. CONCLUSION V. REFERENCES
In the development of this IoT-based weather prediction [1] Gaurav Verma, Pranjul Mittal, Shaista Farheen (Department of
initiative, we begin by selecting appropriate sensors and a Electronics & Communication) Dayananda Sagar College of
Engineering/ “Real Time Weather Prediction System Using IOT and
microcontroller for efficient data collection, ensuring Machine Learning”/ IEEE 2020
connectivity to the internet for smooth data transmission. https://ieeexplore.ieee.org/document/9182766
The gathered data is then forwarded to a cloud platform for [2] Dires Negash Fente, Prof. Dheeraj Kumar Singh, (Department of
processing, encompassing tasks such as data cleansing and Information technology, Parul University Parul Institute of
feature extraction. Employing machine learning Engineering and Technology)/ “Weather Forecasting Using Artificial
methodologies, we train models to forecast real-time Neural Network”/ IEEE 2018
https://ieeexplore.ieee.org/document/8473167
weather conditions. The system incorporates a user-friendly
[3] Sushmitha Kothapalli, S. G. Totad, (Dept of Computer Science and
interface for displaying both current and predicted weather Engineering) “A Real-Time Weather Forecasting and Analysis”/
details. An integrated alert system is designed to notify users IEEE 2017 https://ieeexplore.ieee.org/document/8391974
of extreme weather events, offering valuable early warnings [4] Shivam Tandon, Abhishek Patel, Pawan Kumar Singh (School of
for potential climate shifts. Prior to implementation, Computing Science and Engineering Galgotias University Uttar
rigorous testing is conducted to verify the system's Pardesh, India/ “Weather Prediction Using Machine Learning
reliability and efficacy, particularly in the context of Algorithms”/ IEEE 2020
https://ieeexplore.ieee.org/document/9862337.
safeguarding against environmental challenges, notably in
[5] Tulsi Pawan Fowdur, Rosun Mohammad Nassir-Ud-Diin Ibn Nazir
agriculture. Additionally, the GUI simplifies data (Department of Electrical and Electronic Engineering, University of
interpretation, making it user-friendly for individuals with Mauritius, Reduit, Mauritius)/ “A real-time collaborative machine
varying technical expertise. In conclusion, this innovative learning based weather forecasting system with multiple predictor
solution contributes to efficient weather monitoring, locations”/ journal homepage: www.sciencedirect.com/journal/array.
analysis, and decision-making without the risk. [6] A THESIS Submitted by - KALA A (Faculty of Information &
Communication Engineering ANNA UNIVERSITY CHENNAI)/
“Hybrid Weather Forecasting Models Based On Deep Learning And
Mode Decompositions Methods”/ Shodhganga 2022
http://hdl.handle.net/10603/477722.
[7] A THESIS Submitted by - VASAVI RAVURI (Research
Development Jawaharlal Nehr Technological University)/ “Optimized
Deep LSTM Based Ensemble Classifier for Weather Forecasting”/
Shodhganga 2023 http://hdl.handle.net/10603/477722.

52
53
A.4 Sponsorship Letter

CamScanner

54
55
A.5 Publication Certificate

56
57
58
A.6 Project Competition Paricipation Certificate
Pimpri Chinchwad Education Trust’s
PIMPRI CHINCHWAD COLLEGE OF ENGINEERING & RESEARCH
An ISO 9001:2015 Certified Institute / Computer Engineering / Electronics & Telecommunication Engineering / Civil
Engineering / Mechanical Engineering Programs are NBA Accredited, NAAC A++ Accredited
(Approved by Govt. of Maharashtra, Affiliated to SPPU, Approved by AICTE< DTE Code – 6822)

Agro TechFest- 2K24

Organized by
Department of Electronics and Telecommunication

Certificate of Appreciations
This is to certify that ADESH DHAGE has successfully participated in Agro
TechFest-2K24 (National Level Project Competition) organized on 19 April,
2024 by Department of Electronics & Telecommunication.

Mr. K. B. Bhangale Dr. R. G. Mapari Dr. H. U. Tiwari

Coordinator HOD Principal

59
 Pimpri Chinchwad Education Trust’s
PIMPRI CHINCHWAD COLLEGE OF ENGINEERING & RESEARCH
An ISO 9001:2015 Certified Institute / Computer Engineering / Electronics & Telecommunication Engineering / Civil
Engineering / Mechanical Engineering Programs are NBA Accredited, NAAC A++ Accredited
(Approved by Govt. of Maharashtra, Affiliated to SPPU, Approved by AICTE< DTE Code – 6822)

Agro TechFest- 2K24

Organized by
Department of Electronics and Telecommunication

Certificate of Appreciations
This is to certify that ROHAN YADAV has successfully participated in Agro
TechFest-2K24 (National Level Project Competition) organized on 19 April,
2024 by Department of Electronics & Telecommunication.

Mr. K. B. Bhangale Dr. R. G. Mapari Dr. H. U. Tiwari

Coordinator HOD Principal

60
 Pimpri Chinchwad Education Trust’s
PIMPRI CHINCHWAD COLLEGE OF ENGINEERING & RESEARCH
An ISO 9001:2015 Certified Institute / Computer Engineering / Electronics & Telecommunication Engineering / Civil
Engineering / Mechanical Engineering Programs are NBA Accredited, NAAC A++ Accredited
(Approved by Govt. of Maharashtra, Affiliated to SPPU, Approved by AICTE< DTE Code – 6822)

Agro TechFest- 2K24

Organized by
Department of Electronics and Telecommunication

Certificate of Appreciations
This is to certify that YOGESH PAWAR has successfully participated in Agro
TechFest-2K24 (National Level Project Competition) organized on 19 April,
2024 by Department of Electronics & Telecommunication.

Mr. K. B. Bhangale Dr. R. G. Mapari Dr. H. U. Tiwari

Coordinator HOD Principal

61