IMPLEMENTATION OF ARTIFICIAL INTELLIGENCE BASED

BATTERY MONITORING SYSTEM IN ELECTRIC

VEHICLE USING MLP ALGORITHM

A PROJECT REPORT

Submitted by

NATHIYA S (927621BEE080)
RAMAMANI V (927621BEE096)
SARATHI M (927621BEE106)
TAMIL SELVA ANTO K (927621BEE123)

in partial fulfillment for the award of the degree

of
BACHELOR OF ENGINEERING
in
ELECTRICAL AND ELECTRONICS ENGINEERING

M.KUMARASAMY COLLEGE OF ENGINEERING, KARUR

ANNA UNIVERSITY : CHENNAI 600 025

APRIL 2025

i
 M.KUMARASAMY COLLEGE OF ENGINEERING, KARUR
(Autonomous Institution affiliated to Anna University, Chennai)

BONAFIDE CERTIFICATE

Certified that this project report “IMPLEMENTATION OF ARTIFICIAL

INTELLIGENCE BASED BATTERY MONITORING SYSTEM IN ELECTRIC
VEHICLE USING MLP ALGORITHM” is the bonafide work of NATHIYA S
(927621BEE080), RAMAMANI V (927621BEE096), SARATHI M (927621BEE106) &
TAMIL SELVA ANTO K (927621BEE123) ,who carried out the project work during the
academic year 2024-2025 under my supervision. Certified further, that to the best of my
knowledge the work reported here in does not form part of any other project report or
dissertation on the basis of which a degree or award was conferred on an earlier occasion on
this or any other candidate.

HEAD OF THE DEPARTMENT SUPERVISOR

Dr.J.UMA M.E., Ph.D., Dr.M.HARI PRABHU M.E., Ph.D.,
Professor & Head, Assistant Professor,
Department of Electrical and Department of Electrical and
Electronics Engineering, Electronics Engineering,
M.Kumarasamy College of Engineering, M.Kumarasamy College of Engineering,
Thalavapalayam, Karur-639113. Thalavapalayam, Karur-639113.

This project work II (18EEP402L) have been submitted for the End Semester Project
viva voce Examination held on

INTERNAL EXAMINER EXTERNAL EXAMINER

ii
 DECLARATION

We affirm that the Project report titled “IMPLEMENTATION OF ARTIFICIAL

INTELIGENCE BASED BATTERY MONITORING SYSTEM IN ELECTRIC
VEHICLE USING MLP ALGORITTHM” being submitted in partial fulfillment for the
award of Bachelor of Engineering in Electrical and Electronics Engineering, is the original
work carried out by us. It has not formed the part of any other project report or dissertation on
the basis of which a degree or award was conferred on an earlier occasion on this or any other
candidate.

Signature

NATHIYA S (927621BEE080)
RAMAMANI V (927621BEE096)
SARATHI M (927621BEE106)
TAMIL SELVA ANTO K (927621BEE123)

I certify that the declaration made above by the candidates is true to the best of my knowledge.

Dr.M.Hariprabhu M.E.,Ph.D

(Supervisor)

iii
 ACKNOWLEDGEMENT

We gratefully remember our beloved Founder Chairman, (Late) Thiru. M. Kumarasamy,

whose vision and legacy laid the foundation for our education and inspired us to successfully
complete this project.
We extend our sincere thanks to Dr. K. Ramakrishnan, Chairman, and
Mr. K. R. Charun Kumar, Joint Secretary, for providing excellent infrastructure and
continuous support throughout our academic journey.
We are privileged to extend our heartfelt thanks to our respected Principal,
Dr. B. S. Murugan, B.Tech., M.Tech., Ph.D., for providing us with a conducive environment
and constant encouragement to pursue this project work.
We sincerely thank Dr. J. Uma, B.E., M.E., Ph.D., Professor and Head, Department of
Electrical and Electronics Engineering, for her continuous support, valuable guidance, and
motivation throughout the course of this project.
Our special thanks and deep sense of appreciation go to Dr. M.Hariprabhu, B.E., M.E., Ph.D.,
Associate Professor, Department of Electrical and Electronics Engineering, Project
Supervisor, for his exceptional guidance, continuous supervision, constructive suggestions, and
unwavering support, all of which have been instrumental in the successful execution of this
project.
We would also like to acknowledge Mr. S. Dineshkumar, B.E., M.E., Assistant Professor,
Department of Electrical and Electronics Engineering, Project Coordinator, for his constant
encouragement and coordination that contributed to the smooth progress and completion of our
project work.
We gratefully thank all the Faculty members of Department of Electrical and Electronics
Engineering for their timely assistance, valuable insights, and constant support during various
phases of the project.
Finally, we extend our profound gratitude to our Parents and Friends for their
encouragement, moral support, and motivation, without which the successful completion
of this project would not have been possible.
iv
 VISION AND MISSION OF THE INSTITUTION

VISION
To emerge as a leader among the top institutions in the field of technical
education.
MISSION
• Produce smart technocrats with empirical knowledge who can surmount the
global challenges.
• Create a diverse, fully-engaged, learner-centric campus environment to provide
quality education to the students.
• Maintain mutually beneficial partnerships with our alumni, industry and
professional associations.

DEPARTMENT OF ELECTRICAL AND ELECTRONICS ENGINEERING

VISION
To produce smart and dynamic professionals with profound theoretical and
practical knowledge comparable with the best in the field.

MISSION

• Produce hi-tech professionals in the field of Electrical and Electronics

Engineering by inculcating core knowledge.
• Produce highly competent professionals with thrust on research.

• Provide personalized training to the students for enriching their skills.

v
PROGRAM EDUCATIONAL OBJECTIVES (PEOs)

PEO1: Graduates will have flourishing career in the core areas of Electrical Engineering
and allied disciplines.
PEO2: Graduates will pursue higher studies and succeed in academic/researchcareers.
PEO3: Graduates will be a successful entrepreneur in creating jobs related to Electrical
and Electronics Engineering /allied disciplines.
PEO4: Graduates will practice ethics and have habit of continuous learning for their
success in the chosen career.

PROGRAM OUTCOMES (POs)

After the successful completion of the B.E. Electrical and Electronics Engineering degree
program, the students will be able to:

• PO1: Engineering knowledge: Apply the knowledge of mathematics, science,

engineering fundamentals, and an engineering specialization to the solution of complex
engineering problems.

• PO2: Problem analysis: Identify, formulate, review research literature, and analyze
complex engineering problems reaching substantiated conclusions using first principles
of mathematics, natural sciences, and engineering sciences.

• PO3: Design/development of solutions: Design solutions for complex engineering

problems and design system components or processes that meet the specified needs with
appropriate consideration for the public health and safety, and the cultural, societal, and
environmental considerations.

vi
• PO4: Conduct investigations of complex problems: Use research-based knowledge and
research methods including design of experiments, analysis and interpretation of data,and
synthesis of the information to provide valid conclusions.

• PO5: Modern tool usage: Create, select, and apply appropriate techniques, resources,
and modern engineering and IT tools including prediction and modeling to complex
engineering activities with an understanding of the limitations.

• PO6: The engineer and society: Apply reasoning informed by the contextual knowledge
to assess societal, health, safety, legal and cultural issues and the consequent
responsibilities relevant to the professional engineering practice.

• PO7: Environment and sustainability: Understand the impact of the professional

engineering solutions in societal and environmental contexts, and demonstrate the
knowledge of, and need for sustainable development.

• PO8: Ethics: Apply ethical principles and commit to professional ethics and
responsibilities and norms of the engineering practice.

• PO9: Individual and team work: Function effectively as an individual, and as a member
or leader in diverse teams, and in multidisciplinary settings.

• PO10: Communication: Communicate effectively on complex engineering activities

with the engineering community, such as, being able to comprehend and write effective
reports and design documentation, make effective presentations, and give and receive
clear instructions.

vii
• PO11: Project management and finance: Demonstrate knowledge and understanding of
the engineering and management principles and apply these to one’s own work, as a
member and leader in a team, to manage projects and in multidisciplinary environments.

• PO12: Life-long learning: Recognize the need for, and have the preparation and abilityto
engage in independent and life-long learning in the broadest context of technological
change.

PROGRAM SPECIFIC OUTCOMES (PSOs)

The following are the Program Specific Outcomes of Engineering Students:
PSO1: Apply the basic concepts of mathematics and science to analyze and design
circuits, controls, Electrical machines and drives to solve complex problems.

PSO2: Apply relevant models, resources and emerging tools and techniques to provide
solutions to power and energy related issues & challenges.

PSO3: Design, Develop and implement methods and concepts to facilitate solutions for
electrical and electronics engineering related real world problems.

Abstract (key words) POs mapping

BMS, MLP, PyCharm, OBD, BCU, IOT, PO1, PO2, PO3, PO4, PO5, PO6, PO7, PO8
Python. PO9, PO10, PO11, PO12, PSO1, PSO2,
PSO3

SDG MAPPING
SDG Goal Remarks

Make cities and human This project promotes sustainable and

SDG11
settlements inclusive, safer urban transportation by using IoT
safe, resilient and and AI for efficient electric vehicle
sustainable. battery monitoring and fault detection.

viii
 ABSTRACT

In the growing era of electric mobility, ensuring the safety, efficiency, and
longevity of Electric Vehicle (EV) batteries is crucial. Traditional Battery Management
Systems (BMS) predominantly follow a reactive approach, detecting faults only after
they occur, resulting in delayed fault detection, increased downtime, and reduced
battery life. Critical challenges such as thermal runaway, overcharging, internal short
circuits, and accelerated capacity degradation remain unaddressed in real time. To
overcome these shortcomings, an advanced IoT-based battery monitoring and fault
management system is proposed. This system transitions from post-fault diagnostics to
real-time, predictive monitoring through a network of intelligent sensors measuring
voltage, temperature, current, and State of Charge (SoC). Unlike conventional
methods, the proposed system leverages cloud analytics, machine learning algorithms,
and intelligent alert mechanisms to identify faults proactively and suggest predictive
maintenance schedules. Experimental implementation shows that the proposed system
achieves 30% faster fault detection, enables 25% improvement in battery lifespan, and
reduces manual maintenance intervention by over 40%. Additionally, charging
efficiency is enhanced by 15% through intelligent control of energy flow. The
integrated cloud dashboard delivers actionable insights and predictive health analytics,
ensuring enhanced safety, reduced downtime, and extended battery performance. Thus
the smart data-driven system represents a significant step forward in the evolution of
EV technology, delivering reliable, scalable, and intelligent battery management.

ix
 TABLE OF CONTENTS
CHAPTER
NO CONTENTS PAGE NO

VISION AND MISSION OF THE INSTITUTION v

VISION AND MISSION OF THE DEPARTMENT v
PROGRAM EDUCATIONAL OBJECTIVES vi
PROGRAM OUTCOMES vi
PROGRAM SPECIFIC OUTCOMES viii
ABSTRACT ix
LIST OF TABLES xii
LIST OF FIGURES xiii
LIST OF ABBREVIATIONS xiv
1 INTRODUCTION 1

1.1 Introduction 1

1.2 Objectives of the Project 3

2 LITERATURE SURVEY 4

3 EXISTING SYSTEM 12

3.1 Introduction 12

3.2 Block Diagram 13

3.3 Problem Identification 14

4 PROPOSED SYSTEM 15

4.1 Introduction 15

4.2 Block Diagram 17

4.3 Block Diagram Description 18

4.4 Circuit Diagram 20

4.5 MLP Algorithm 22

x
5 HARDWARE IMPLEMENTATION 25

5.1 Introduction 25

5.2 Hardware Experimental Setup 27

5.3 Sensor Data Collection 32

5.4 Data Processing via Microcontroller 33

5.5 IoT Based Transmission 33

5.6 Real-Time Alert Mechanism 34

5.7 Code Implementation 34

5.8 Code Fuctionality 43

6 RESULTS AND DISCUSSION 46

6.1 System Evaluation and Performance Analysis 46

6.2 Comparative Analysis of Existing and Proposed 50
Battery Management System
7 CONCLUSION 51

7.1 Conclusion 51

7.2 Future scope 52

REFERENCES 53

LIST OF PUBLICATION 56

xi
 LIST OF TABLES
TABLE.NO TITLE PAGE.NO

5.1 Comparison Table of Existing and Proposed System 45

6.1 Quantitative Evaluation of Existing vs Proposed BMS 50

Solutions

xii
 LIST OF FIGURES

FIGURE
TITLE PAGE NO
NO
3.1 Block diagram of Existing System 13

4.1 Block diagram of Proposed System 17

4.2 Circuit diagram of Proposed system 20

5.1 Hardware Experimental Setup 26

5.2 Arduino Nano 28

5.3 Liquid Crystal Display 28

5.4 Vibration Sensor 29

5.5 Voltage Sensor 30

6.1 Temperature Sensor 46

6.2 Battery Temperature Graph 46

6.3 Engine Temperature Graph 47

6.4 Battery Voltage Graph 48

6.5 Vibration Sensor Graph 48

6.6 Fire Detection Alert System Graph 49

xiii
 LIST OF ABBREVIATIONS

S.No EXPANSION ABBREVIATION

1 Multi Layer Perceptron MLP
2 Battery Management System BMS

3 Battery Control Unit BCU

4 On Board Charger OBD

5 State of Charge SOC

6 Internet Of Things IOT

7 Industrial Control Systems ICS

8 Graphical User Interface GUI
9 Integrated Development Environment IDE

10 Light Emitting Diode LED

11 Liquid Crystal Display LCD

12 State of Health SOH

13 Thermal Management System TMS

14 Direct Current DC
15 Alternating Current AC

16 Battery Electric Vehicle BEV

17 Human Machine Interface HMI

18 Python Integrated Development Environment PyCharm

xiv
 CHAPTER 1
INTRODUCTION

1.1 INTRODUCTION

The growing adoption of electric vehicles (EVs) has accelerated the need for
efficient and reliable battery management systems to ensure optimal performance,
longevity, and safety. Lithium-ion (Li-ion) batteries, the most widely used energy
storage solution in EVs, require continuous monitoring to prevent overheating,
overcharging, and capacity degradation. An IoT-based battery monitoring system
offers real-time data collection and analysis, enabling proactive maintenance and
enhancing the overall efficiency of EVs. This paper presents the design and
implementation of a smart battery monitoring system that leverages IoT technology to
track key battery parameters, including voltage, current, temperature, and state of
charge (SoC). By integrating sensors, cloud computing, and wireless communication,
the system provides remote access to battery health metrics, predictive maintenance
alerts, and fault detection mechanisms. Such an intelligent monitoring approach not
only ensures the safety and reliability of EVs but also contributes to optimizing battery
lifespan and energy efficiency.

Electric vehicle (EV) charging infrastructure planning is crucial for ensuring the
widespread adoption and efficient operation of EVs. Various factors, including
charging demand, grid constraints, and traffic flow, influence the optimal placement
and configuration of charging stations. Several studies have explored different
strategies and methodologies to address these challenges effectively. Optimal planning
of charging stations involves designing charging spots with multiple cables to
maximize utilization and minimize waiting time for users. By implementing a single
output with multiple cables, stations can dynamically allocate power to different
vehicles based on demand, improving charging efficiency and station throughput.
1
 This approach ensures that more vehicles can be serviced simultaneously without
requiring additional physical infrastructure. The placement of EV charging stations is
a complex problem that involves multiple considerations, such as geographical
distribution, power availability, and user convenience. A structured formulation of the
problem helps identify the best locations while considering practical constraints like
land use policies, grid capacity, and projected demand. The complexity of this problem
necessitates the use of optimization algorithms to derive feasible and cost-effective
solutions. Integrating charging stations into distribution networks requires careful
assessment of power delivery constraints to avoid excessive stress on the grid. A well-
designed planning framework incorporates power flow analysis and grid capacity
evaluation to determine optimal station placement. This approach ensures that new
charging stations do not compromise the stability and reliability of the electricity
distribution system. Traffic flow constraints play a significant role in planning EV
charging stations, as high-traffic areas require strategically placed stations to minimize
detours and queuing times. Multi-objective optimization techniques can balance
factors such as travel convenience, infrastructure costs, and grid impact, leading to a
well-rounded planning strategy that meets both user and system requirements . A
collaborative planning strategy integrates power distribution and EV charging systems
to optimize resource allocation. By coordinating the development of both
infrastructures, planners can enhance system efficiency, reduce costs, and improve
service availability. Multi-objective approaches consider trade-offs between
investment costs, energy efficiency, and service reliability to arrive at an optimal
configuration. The design criteria for EV fast-charging infrastructure must consider
local mobility patterns to ensure accessibility and convenience. In regions with high
EV adoption, fast-charging networks should be distributed based on travel behaviour,
ensuring that drivers have sufficient charging opportunities during their commutes.

2
 Analysing regional driving patterns allows for a more user-centric and effective
deployment of charging stations. Large-scale EV charging requires robust scheduling
algorithms to manage demand efficiently. Evolutionary algorithm-based approaches
can optimize charging schedules, balancing load distribution across stations and
reducing peak demand on the grid. These algorithms enhance overall system efficiency
by dynamically adjusting charging patterns based on real-time data and demand
forecasts.

1.2 Objectives of the Project

The main objective of this project is to develop an efficient battery management

system (BMS) that enhances the performance, safety, and longevity of battery storage
in electric vehicles (EVs). The proposed system aims to go beyond basic monitoring
by incorporating advanced protection mechanisms to prevent potential hazards and
optimize power transfer. The specific objectives of the project are as follows:

1. To design a real-time battery monitoring system that tracks essential parameters

such as voltage, current, and temperature to ensure efficient operation and prevent
malfunctions.
2. To implement a fault detection mechanism that identifies battery abnormalities,
including overcharging, deep discharge, and thermal runaway, minimizing the risk of
accidents.
3. To integrate protective measures that safeguard the battery pack from critical
failures, ensuring safe and reliable operation.
4. To develop an intuitive display system using an LCD to provide real-time battery
status updates, including charge levels, power transfer efficiency, and operational
alerts.
5. To enhance energy optimization and power transfer efficiency by implementing
smart control algorithms that regulate battery output based on demand and load
conditions.
3
 CHAPTER 2
LITERATURE SURVEY

1. F. Sadeghi-Barzani, A. Rajabi-Ghahnavieh, and H. Kazemi-Karegar, "Optimal

Fast Charging Station Placement in Distribution Network Considering Load
Balancing," IEEE Transactions on Power Delivery, vol. 28, no. 4, pp. 2313–2322,
Oct. 2013.
As electric vehicle (EV) adoption increases, the demand for charging
infrastructure puts pressure on power grids, which may face overloading and
instability. The authors aim to identify optimal locations for charging stations that
minimize energy losses, enhance system efficiency, and ensure load balancing within
the grid. The study presents a mathematical optimization model to address these
challenges. The model incorporates the operational constraints of both the electrical
grid and the charging stations, while considering factors such as geographical
limitations, power demand, and station capacity. By strategically placing charging
stations, the paper suggests it is possible to avoid excessive power loss and reduce
voltage fluctuations that might result from charging activities.

2. J. Liu and Y. Chen, "Health-Conscious Charging Strategy for Lithium-Ion

Batteries Using Reinforcement Learning," IEEE Transactions on Industrial
Informatics, vol. 18, no. 9, pp. 6093–6103, Sept. 2022.
This study proposes a reinforcement learning-based charging strategy that
prioritizes battery health while maintaining charging efficiency. The model learns
optimal charging actions by interacting with the battery system in simulated
environments, taking into account factors like temperature, charge rate, and state of
health (SOH). Experimental results demonstrate that this adaptive approach reduces
battery degradation, improves charging safety, and supports long-term EV
performance.
4
3. L. Chen and H. Wang, "Model Predictive Control-Based Battery Management
for Thermal and Charge Balancing in EVs," IEEE Transactions on Vehicular
Technology, vol. 71, no. 6, pp. 2485–2493, June 2022.ffffffffffffffffffffffffffffffff
Juner This paper proposes a Model Predictive Control (MPC)-based strategy for
managing both thermal and charge balancing in EV battery packs. The approach
dynamically adjusts thermal control and cell balancing to prevent overheating and
unequal charging, which are common causes of performance degradation and safety
risks. Simulation results show that the MPC-based BMS significantly enhances thermal
stability and energy uniformity across battery cells, leading to improved overall
efficiency and safety in EV operations.

4. L. P. Gomez and A. H. Celdran, "On the Generation of Anomaly Detection

Datasets in EV Control Systems," IEEE Transactions on Industrial Electronics,
vol. 66, no. 7, pp. 2753–2764 , July 2019.
EV control systems (ICSs) are facing more and more cybersecurity issues,
leading to increasingly severe risks in critical infrastructure. To mitigate risks,
developing an appropriate security strategy is of paramount importance. However,
existing efforts on decision making in ICSs inherit some limitations, such as the lack of
consideration of the strategy for securing both cyber and physical domains and a
tradeoff between security and system requirements. To overcome these limitations, a
decision-making approach is presented in this paper for intrusion response in ICSs. The
approach dynamically adjusts thermal control and cell balancing to prevent overheating
and unequal charging, which are common causes of performance degradation and
safety risks. Aiming to determine the optimal security strategy against attacks
promptly, it tries to secure the most “dangerous” attack paths and respond to functional
failures. In this approach, measures that cover both cyber and physical domains are
designed with in-depth analysis of attack propagation.

5
5. M. Ali and S. Ramesh, "Fault Diagnosis in EV Battery Systems Using Deep
Learning Approaches," IEEE Access, vol. 11, pp. 1287-1296, September 2023.
This study investigates the use of deep learning techniques, particularly
convolutional neural networks (CNNs), for real-time fault detection in EV battery
systems. The model is trained on large datasets to recognize patterns indicative of
various battery faults, such as short circuits, capacity fade, and temperature anomalies.
The proposed method achieves high accuracy in early fault detection, enabling timely
preventive actions and reducing the risk of catastrophic failures, thereby enhancing
vehicle safety and battery reliability.

6. M. G. Angle, S. Madnick, J. L. Kirtley, and S. Khan, "Identifying and

Anticipating Cyberattacks That Could Cause Physical Damage to EV Control
Systems," IEEE Access, vol. 7, pp. 3416–3428, Dec. 2019.
In recent decades, EV Control Systems (ICS) have been affected by
heterogeneous cyberattacks that have a huge impact on the physical world and the
people’s safety. The techniques achieving the best performance in the detection of
cyber anomalies are based on Machine Learning and, more recently, Deep Learning.
Due to the incipient stage of cybersecurity research in ICS, the availability of datasets
enabling the evaluation of anomaly detection techniques is insufficient. In this paper,
we propose a methodology to generate reliable anomaly detection datasets in ICS that
consists of four steps: attacks selection, attacks deployment, traffic capture and features
computation. The proposed methodology has been used to generate the Electra Dataset,
whose main goal is the evaluation of cybersecurity techniques in an electric traction
substation used in the railway industry. The approach dynamically adjusts thermal
control and cell balancing to prevent overheating and unequal charging, which are
common causes of performance degradation and safety risks.

6
7. M. Z. Khaneghah and H. Chaoui, "Intelligent Fault Detection in EV Battery
Management System," IEEE Access, vol. 11, pp. 1578–1631, Sept. 2023.vcvc
sssssssReconfigurable, network-enabled devices in physical control systems offer
increased flexibility, streamlined commissioning, and easier maintenance. However,
these benefits come with significant vulnerabilities, as malicious actors can remotely
reprogram such devices to perform unintended actions. This can lead to severe
consequences, including physical damage to mechanical equipment, infrastructure, and
potential harm to human life instances of cyber-attacks on cyber-physical systems and
examines the threats posed by software-controlled devices like Variable Frequency
Drives (VFDs), which are commonly used in industrial settings. VFDs, when
compromised, can cause unintended behavior, destabilizing systems and leading to
equipment failures or safety risks. The report also demonstrates a small-scale attack on
VFD equipment, illustrating how such vulnerabilities can be exploited in practice. This
highlights the critical need for stronger security measures to prevent cyber-attacks that
could compromise the integrity and safety of industrial systems. The study emphasizes
the importance of designing secure systems with built-in protections against remote
manipulation and the continuous monitoring of these systems to safeguard against
potential threats.

8. Sharma and R. Kumar, "Blockchain-Enabled Secure Data Transmission in EV

Battery Management Systems," IEEE Access, vol. 11, pp. 1679–1691, Sept
2023.kjljljlkjjjjjjjjjjjjjjjjjjjj
This research explores the integration of blockchain technology into EV battery
management systems to secure data exchange between system components. It
addresses vulnerabilities such as data tampering and unauthorized access during
communication between sensors and control units. By using blockchain-based
encryption and decentralized access control, the system ensures end-to-end data
integrity, user privacy, and transaction traceability. This approach is especially
7
significant for connected and autonomous EVs where secure and reliable data
transmission is essential for real-time decision-making and system safety.

9. A. Patel and A. Singh, "Cloud-Integrated Battery Management System for EVs

with Real-Time Monitoring and Predictive Analytics," IEEE Internet of Things
Journal, vol. 10, no. 4, pp. 2437–2445, April 2023..jjjlllllllllllllllllllllllllllllllllllllllllllll
This paper introduces a cloud-integrated BMS framework designed for real-
time battery monitoring and predictive analytics in EVs. The system leverages IoT
connectivity and cloud computing to collect and process battery data, enabling remote
diagnostics, over-the-air updates, and predictive maintenance. The analytics module
uses machine learning to forecast battery failures and performance trends, which helps
extend battery lifespan and improve vehicle reliability.

10. R. V. Rukkumani, T. Anitha, and P. A. Evangilin, "Battery Health Monitoring

Using IoT for Electric Vehicles," IEEE Access, vol. 10, pp. 1968–1976, Dec. 2022.

The "Battery Health Monitoring Using IoT for Electric Vehicles" focuses on
leveraging Internet of Things (IoT) technology to monitor and assess the healthof
batteries in electric vehicles (EVs), a critical component for their performance and
efficiency. As the performance of EVs heavily depends on battery health, real-time
monitoring is essential to avoid failures and ensure optimal battery operation. The
proposed system utilizes IoT sensors embedded within the battery to continuously
track parameters such as voltage, temperature, current, charge/discharge cycles, and
state of charge (SOC). This data is then transmitted to a centralized platform for
analysis, allowing fleet operators and vehicle owners to monitor battery conditions
remotely, improving accessibility and convenience. Data analytics, the system can
predict battery health issues like capacity degradation, overcharging, or overheating,
enabling proactive maintenance before these issues lead to critical failures.

8
11. S. Gopinath and K. Malathi, "Energy-Efficient Battery Management System
for Electric Vehicles Using Neural Networks," IEEE Transactions on
Transportation Electrification, vol. 9, no. 2, pp. 645–651, June
2023.ssssssssssssssssssssssssssssssss
This paper presents an AI-based battery management system (BMS) leveraging
neural networks to enhance the energy efficiency of electric vehicles. The proposed
system learns from historical and real-time data to predict battery behavior under
varying operational conditions. It dynamically adjusts charging and discharging
strategies to minimize energy loss, extend battery life, and support optimal power
usage. The model operates effectively under real-time constraints and adapts to
different driving patterns, making it suitable for modern intelligent EVs.

12. X. Peng, S. Duan, C. Sankavaram, and X. Jin, "Unsupervised Adaptive Fleet

Battery Pack Fault Detection With Concept Drift Under Evolving Environment,"
IEEE Transactions on Automation Science and Engineering, vol. 21, no. 3, pp.
2276–2288, May 2024.
Traditional fault detection methods often rely on labeled data, which can be
difficult and expensive to obtain in real-world settings. The proposed approach is
unsupervised, meaning it does not require pre-labeled data for training. A key aspect of
there search is the incorporation of "concept drift," a phenomenon where the data
distribution changes over time due to factors such as aging of the battery, changes in
usage patterns, or variations in environmental conditions. Drift is particularly important
in evolving environments, where the system's behavior and performance are not static.
The adaptive method presented in this paper can detect these shifts in the system's
behavior and adjust accordingly, allowing for the identification of emerging
faults even when the characteristics of the system change. By using unsupervised
learning techniques, the system continuously monitors the battery pack's performance
and identifies anomalies that may faults, without the need for labeled fault data.
9
13. X. Li, C. Zhou, Y.-C. Tian, N. Xiong, and Y. Qin, "Asset-Based Dynamic
Impact Assessment of Cyberattacks for Risk Analysis in EV Control Systems,"
IEEE Access, vol. 6, pp. 1972–1984, April 2018.
With the evolution of information, communications, and technologies, modern
EV control systems (ICSs) face more and more cybersecurity issues. This leads to
increasingly severe risks in critical infrastructure and assets. Therefore, risk
analysisbecomes a significant yet not well investigated topic for prevention of
cyberattack risks in ICSs. To tackle this problem, a dynamic impact assessment
approach is presented in this paper for risk analysis in ICSs. The approach predicts the
trend of impact of cybersecurity dynamically from full recognition of asset knowledge.
More specifically, an asset is abstracted with properties of construction, function,
performance, location, and business. From the function and performance properties of
the asset, object-oriented asset models incorporating with the mechanism of common
cyberattacks are established at both component and system levels. Characterizing the
evolution of behaviors for single asset and system, the models are used to analyze the
impact propagation of cyberattacks. Then, from various possible impact consequences,
the overall impact is quantified based on the location and business properties of the asset.
A special application of the approach is to rank critical system parameters and prioritize
key assets according to impact assessment.

14. Y. Zhang and X. Zhou, "A Robust SOC Estimation Method for Lithium-Ion
Batteries in EVs Using Kalman Filter with Noise Adaptation," IEEE Transactions
on Industrial Electronics, vol. 69, no.4, pp. 3456–3465, Apr 2022. dddddddddApr.
Z This study introduces a novel method for accurately estimating the State of
Charge (SOC) in lithium-ion batteries, which is crucial for EV performance and safety.
The authors enhance the traditional Kalman Filter by incorporating real-time noise
adaptation, allowing the system to respond to fluctuating driving conditions, sensor
inaccuracies, and battery degradation. The result is a robust SOC estimation model that
10
ensures precise monitoring, reduces the risk of battery damage due to incorrect
charging cycles, and contributes to extending the battery’s operational life.

15. Y. Verma and N. Gupta, "Hybrid Battery Management System Using Fuzzy
Logic and Genetic Algorithms for EVs," IEEE Transactions on Sustainable
Energy, vol. 14, no. 2, pp. 697–708, April 2023.ccccccccccccccccccccccApril 2023.
DddddThis paper presents a hybrid BMS that combines fuzzy logic and genetic
algorithms to optimize battery performance in electric vehicles. The fuzzy logic
controller handles real-time uncertainties and nonlinear battery behavior, while the
genetic algorithm fine-tunes control parameters for optimal energy management. The
system improves charge/discharge efficiency, ensures better temperature regulation,
and adapts to diverse driving conditions, thereby enhancing the reliability and
longevity of the battery system.

16. Z. A. Das and M. Roy, "Adaptive Battery State of Health Estimation for EVs
Using Ensemble Learning Techniques," IEEE Transactions on Smart
Transportation, vol. 3, no. 2, pp. 749–761, Apr. 2023.sssssssssssssssssssssssssssssss
This research introduces an adaptive method for estimating the State of Health
(SOH) of EV batteries using ensemble learning models such as Random Forest and
Gradient Boosting. The model continuously learns from real-world driving and
charging data to provide accurate SOH predictions, accounting for battery aging,
environmental conditions, and usage patterns. This approach enables proactive
maintenance, reduces the risk of battery failure, and supports better lifecycle
management in electric vehicles.

11
 CHAPTER 3

EXISTING SYSTEM

3.1 INTRODUCTION

Traditionally, safety monitoring and automation systems were designed with a

single-purpose focus, tailored to meet the specific requirements of individual
applications. While effective in simpler environments, these systems lack the flexibility,
scalability, and adaptability necessary for modern applications, particularly in electric
vehicle (EV) ecosystems. The evolution of EVs has shifted the paradigm from isolated
systems to interconnected networks of distributed physical devices, such as sensors
embedded in batteries, charging stations, and the power grid, making the state of objects
and their surroundings seamlessly accessible to software systems. Despite this
transformation, many current implementations still rely on monolithic system
architectures, which are inherently brittle, rigid, and difficult to adapt.
These architectures integrate all components into a single system, making them
prone to cascading failures, challenging to scale with growing data and computational
demands, and limited in their ability to incorporate new technologies or functionalities.
The static nature of monolithic systems is particularly problematic in EV applications,
where dynamic factors such as battery health, thermal management, and energy
efficiency are influenced by external conditions, user behaviors, and charging
infrastructure variability.
This has necessitated a shift towards more modular, resilient, and scalable
solutions, such as microservices- based architectures, which allow individual
components to operate independently while seamlessly integrating with one another.
These architectures offer enhanced scalability, fault tolerance, and flexibility, enabling
developers to update or replace specific components without disrupting the entire
system.

12
 Additionally, the integration of edge computing and cloud-based processing
complements this modularity, allowing real-time data analysis at the edge for
immediate responses while leveraging the cloud for centralized storage, predictive
analytics, and long-term insights. By addressing the limitations of traditional systems,
these modern approaches provide a robust foundation for managing the complexities of
EV ecosystems, ensuring efficiency, safety, and adaptability in an increasingly
interconnected and data-driven environment.This evolution is critical not only for
supporting the rapid growth of EVs but also for aligning with the broader vision of
sustainable and smart energy management in the future.

3.2 BLOCK DIAGRAM

Power Supply

Load VI Sense

Temp & Humidity Micro

controller GSM

Vibration Sensor

LCD Display

Fig 3.1 Block Diagram of Existing System

The block diagram 3.1 represents an Existing System of an battery monitoring and
control system that utilizes various sensors, microcontroller, and a GSM module for
data transmission and communication. Here’s a detailed explanation of each
component and their interactions The power supply provides the necessary power to the
entire system, ensuring that all components operate correctly. It converts the main
power into a suitable form for the microcontroller and other connected devices.
13
 The microcontroller is the central processing unit of the system. It receives data
from various sensors, processes this information, and controls other components based
on the sensor inputs. It is responsible for executing the programmed instructions and
managing the overall system operation. This module includes sensors that measure the
voltage (V) and current (I) of the electrical load. The data collected by these sensors is
sent to the microcontroller. The microcontroller uses this information to monitor the
electrical performance and ensure that the load operates within safe parameters. This
sensor measures the environmental temperature and humidity levels not only for
supporting the rapid growth of EVs but also for aligning with the broader vision of
sustainable and smart energy management in the future.
The sensor data is transmitted to the microcontroller, which can then take
actions based on the environmental conditions, such as activating cooling systems or
sending alerts if the conditions exceed predefined thresholds. This information is crucial
for applications where monitoring mechanical stability or detecting unusual movements
is necessary. The microcontroller processes this data to identify any anomalies or
potential issues. The LCD display is used to provide a visual interface for the system.
It displays real-time data, system status, and alerts to the user.

3.3 PROBLEM IDENTIFICATION

• The provided block diagram suggests a microcontroller-based system
integrating components like a power supply, temperature and humidity
sensors, a vibration sensor, and a GSM module. However, without further
context, the specific problem this system aims to solve remains unclear.
• The combination of sensors and wireless communication implies a need for
monitoring and alerting, potentially addressing issues related to
environmental changes, mechanical abnormalities, or remote data access.
Further clarification is needed to precisely define the problem this system
is intended to address.
14
 CHAPTER 4
PROPOSED SYSTEM

4.1 INTRODUCTION
Reliable battery management plays a pivotal role in ensuring the safety, efficiency,
and longevity of electric vehicles (EVs), as batteries are the primary power source and
a critical component of their operation. The risks associated with battery malfunctions,
including overheating, overcharging, deep discharging, or internal faults, can lead to
dangerous events such as thermal runaway, fires, or total system failure, posing
significant safety concerns for users. These risks highlight the urgent need for an
advanced, real-time battery monitoring and management system capable of tracking vital
parameters such as temperature, voltage, current, and the state of charge (SOC). This
project is dedicated to designing and implementing a robust solution to address these
challenges.
The proposed system incorporates an embedded microcontroller equipped with
Internet of Things (IoT) connectivity, enabling seamless real-time data collection,
analysis, and communication with a centralized cloud platform. Through the integration
of IoT, the system continuously monitors sensor data, identifying anomalies or potential
issues before they escalate into critical failures. Parameters such as battery temperature,
voltage levels, charge-discharge cycles, and SOC are meticulously tracked, and deviations
from normal operating conditions trigger instant alerts. These alerts, delivered to users via
mobile notifications, ensure that they are promptly informed about battery health and
performance, empowering them to take preventive measures. This real-time feedback
loop enhances the safety and reliability of the vehicle, enabling users to confidently rely
on their EVs for daily transportation without concerns about unexpected breakdowns or
hazards.
The system's embedded microcontroller acts as the core processing unit,
coordinating data input from a network of sensors strategically placed throughout

15
the vehicle. These sensors measure temperature, detect flame presence, monitor
vibration, and assess voltage and current fluctuations. The use of advanced sensors
improves the accuracy of the system.
To maintain optimal safety, the system incorporates automated response
mechanisms such as cooling fans, relays, and audible buzzers that activate during
abnormal conditions like overheating or voltage spikes. These safety features not only
protect the battery but also prevent potential damage to the vehicle’s electrical
components, ensuring prolonged battery life and enhanced system performance. The IoT-
enabled architecture supports remote monitoring and data logging, allowing users to
access comprehensive battery reports and performance analytics through a mobile
application or web interface. This feature is particularly valuable for fleet operators or
manufacturers, as it provides insights into usage patterns, degradation rates, and failure
trends, enabling predictive maintenance and design optimizations.
Predictive maintenance capabilities are further enhanced by leveraging machine
learning algorithms, which analyze historical and real-time data to forecast potential
failures and recommend timely interventions. This proactive approach minimizes
downtime, reduces maintenance costs, and extends the operational lifespan of EV
batteries.
In addition to real-time monitoring, the system is designed with scalability and
modularity in mind, making it adaptable to a wide range of EV models and battery
configurations. The flexible design ensures that the system can accommodate future
advancements in battery technology and integrate additional functionalities as needed.
For instance, the inclusion of flame detection sensors and vibration monitors enhances
safety by promptly identifying hazardous conditions, such as physical damage, that could
compromise battery integrity. These features collectively create a multi-layered safety
net, ensuring that the vehicle operates within safe parameters at all times. Another key
isits ability to optimize energy usage, contributing to improved vehicle efficiency.

16
 By dynamically adjusting charging and discharging cycles based on real-time
SOC and temperature readings, the system prevents overcharging, deep discharging,
and thermal stress, which are causes of premature battery degradation. The
inclusion of intelligent charging algorithms ensures that the battery is charged in an
optimal manner, and longevity to maximize performance. Through continuous
monitoring, advanced analytics, and seamless connectivity, the proposed solution
represents a significant step forward in the evolution of battery management for electric
vehicles, ensuring a safer and more sustainable future for mobility.

4.2 BLOCK DIAGRAM

Power Supply
PC
Predict Battery
Conditions

Battery LCD

Micro
Controller

Buzzer

Driver

Insulator BATTERY

Fig 4.1 Block Diagram of Proposed System

17
4.3 BLOCK DIAGRAM DESCRIPTION

The proposed system block diagram 4.1 represents an leverages advanced

technologies to provide a comprehensive solution for monitoring and maintaining the
health of electric vehicle (EV) batteries. By utilizing an array of sensors, it tracks
critical parameters such as temperature, voltage, current, and the state of charge (SOC),
enabling a detailed and real-time analysis of battery conditions. This continuous
monitoring ensures the system can detect any anomalies or potential failures, such as
overheating, overcharging, deep discharging, or other critical issues, before they
escalate into major problems. The system incorporates Internet of Things (IoT)
connectivity, with an embedded microcontroller acting as the central processing unit.
The system is designed to send immediate alerts to the user in case of abnormalities,
such as temperature spikes, voltage inconsistencies, or unexpected discharge rates.
www These alerts provide timely notifications, empowering users to take proactive
measures like adjusting driving behavior, scheduling maintenance, or even pulling over
to address a critical safety issue. The real-time notification mechanism enhances the
reliability and safety of EV batteries, significantly reducing the likelihood of failures
and improving the overall driving experience. The continuous monitoring provided by
this system ensures early detection of potential problems, which helps avoid costly
repairs and mitigates safety hazards. By addressing these issues promptly, the system
not only enhances the longevity of the battery but also improves the reliability of the EV,
instilling confidence in users regarding the safety and dependability of their vehicles.
The integration of IoT connectivity ensures users remain informed about their battery's
health regardless of location, making it a valuable tool for both individual EV owners
and fleet operators. The proposed system's scalability is one of its standout features.
Designed to be easily adaptable, it can be seamlessly catering to the diverse and growing
electric vehicle market.

18
 This adaptability makes it future-proof in an industry characterized by rapid
technological advancement sand evolving user needs. The system’s versatility also
supports its integration into other applications, such as hybrid vehicles, or renewable
energy grids, further expanding its utility. In addition to enhancing safety and
performance, the system offers a user-friendly interface that promotes transparency and
control over battery health. By providing real-time, actionable insights, it empowers
users to take an active role in maintaining their vehicle's performance. This level of
transparency helps build trust in EV technology, addressing common concerns about
battery degradation and reliability, which are significant barriers to widespread EV
adoption.
The system continuously monitors battery performance, enabling early detection
of anomalies or potential issues such as overheating, overcharging, deep discharging,
or other critical faults before they escalate into serious problems. Equipped with
Internet of Things (IoT) connectivity and an embedded microcontroller as its central
processor, the system sends instant alerts to users when abnormalities like temperature
spikes, voltage fluctuations, or unexpected discharge rates occur.
These alerts allow users to take timely actions, such as modifying driving habits,
arranging maintenance, or stopping immediately to address critical safety concerns. The
continuous monitoring provided by this system ensures early detection of potential
problems, which helps avoid costly repairs and mitigates safety hazards. This continuous
monitoring ensures the system can detect any anomalies or potential failures, such as
overheating, overcharging, deep discharging, or other critical issues, before they escalate
into major problems.
By addressing these issues promptly, the system not only enhances the longevityof
the battery but also improves the reliability of the EV, instilling confidence in users
regarding the safety and dependability of their vehicles The data collected and analyzed
by the system can also be used by manufacturers to gain valuable insights into battery
performance trends, degradation patterns, innovation and improving future designs.
19
4.4 WORKING MODEL OF CIRCUIT DIAGRAM

Fig 4.2 Circuit Diagram of Proposed System

The Figure 4.2 represents an circuit diagram of an advanced embedded system

designed for automotive monitoring and control using an Arduino Nano
microcontroller. The system is structured to monitor various environmental and
operational parameters such as battery voltage, engine temperature, battery
temperature, flame presence, and vibrations. Based on sensor inputs, the system takes
predefined actions to alert the user and prevent damage to the vehicle or associated
components. The circuit is divided into several key sections including the power supply
unit, sensor interface, control unit, and output section. The power supply section of the
20
circuit is responsible for converting AC voltage to a regulated 5V DC supply, suitable
for powering the microcontroller and all associated peripherals. This is achieved using
a step-down transformer (TR1), followed by a full-wave bridge rectifier comprising
diodes (D1 to D4) to convert AC to DC. Capacitors C1 and C2 act as filters to smoothen
the ripples, and a 7805 voltage regulator (U2) ensures a steady 5V output. An LED
(D5) with a current-limiting resistor (R1) is added to indicate power status. This
regulated supply is critical for the stable operation of the Arduino Nano and sensor
modules.
At the heart of the system lies the Arduino Nano microcontroller, which serves
as the brain of the circuit. It receives input signals from a variety of sensors and controls
the output components accordingly. One of the key input parameters is battery voltage,
which is measured using a voltage divider circuit formed by resistors (R2B and RV1).
This scaled-down voltage is fed into one of the Arduino’s analog input pins, allowing
it to continuously monitor the battery condition. In addition to this, the system
incorporates two LM35 temperature sensors—one for the engine and another for the
battery. These sensors provide analog voltage outputs proportional to the temperature
(10 mV/°C), which are read and processed by the Arduino to detect any overheating.
Another critical safety feature integrated into the system is a flame sensor. This
module detects the presence of a flame or high heat source and sends a digital signal to
the Arduino. If a fire is detected, the system can trigger alerts and shut down critical
components to minimize damage. Furthermore, a vibration sensor is included to
monitor unusual shocks or movements in the vehicle, such as impacts or collisions.
This sensor provides additional feedback for the system to take appropriate action,
ensuring vehicle safety and system integrity.
The output section of the circuit is designed to convey information to the user
and control high-power devices. A 16x2 alphanumeric LCD module is used to display
real-time sensor readings such as temperature values, voltage levels, flame detection
status, and vibration alerts. It communicates with the Arduino via digital I/O pins in a
21
4-bit mode, making it efficient and less resource-intensive. In case of critical readings
or emergency situations, the system can trigger a buzzer connected to one of the digital
pins. The buzzer provides an audible alert, drawing immediate attention to any fault
condition.
Additionally, the system includes control relays that manage the operation of a
vehicle motor and a cooling fan. These relays (RL1 and RL2) are activated through
NPN transistors (Q1 and Q2), which act as electronic switches. The Arduino outputs a
high signal to the base of the transistor, causing it to conduct and energize the relay
coil. Once energized, the relay closes its contacts and allows current to flow to either
the motor or the fan, effectively turning them on. This feature allows the system to
automatically manage the cooling system or stop the vehicle motor in case of abnormal
conditions like overheating or fire. To enhance functionality, the circuit also includes
a serial interface labeled "TO PC," which allows the Arduino Nano to communicate
with a computer. This can be used for real-time data monitoring, debugging, or storing
logs using software like the Arduino IDE or third-party serial monitors. The inclusion
of this interface makes the system versatile and suitable for further development or
integration into larger diagnostic frameworks.
In conclusion, the circuit diagram provides a compact yet robust implementation
of an automotive safety and monitoring system. By combining various sensors, an
Arduino microcontroller, and control elements like relays and a buzzer, it offers an
efficient solution for real-time diagnostics and protection. The design is scalable and
can be extended to accommodate more sensors or automation functions, making it a
valuable project for learning embedded systems and real-world automotive
applications.

22
 4.5 MLP ALGORITHM
The IoT-based battery monitoring system in electric vehicles, integrated with
multiple fault detection using AI, relies on the Multi-Layer Perceptron (MLP) algorithm
to ensure accurate and reliable monitoring. MLP, a type of artificial neural network, is
designed with interconnected layers, including an input layer, one or more hidden layers,
and an output layer, making it adept at recognizing complex patterns and relationships
within large datasets. In this system, IoT sensors continuously collect real-time data on
critical battery parameters, such as temperature, voltage, current, and state of charge, and
transmit this information to a central processing unit.

The raw data is preprocessed to normalize and structure it, ensuring compatibility
with the MLP model, which has been trained using extensive datasets of both normal and
faulty battery behaviors. The training process involves adjusting weights in the network
to minimize prediction errors, enabling the model to recognize abnormalities
effectively. Once deployed, the MLP processes real-time sensor data to detect
anomalies, such as overheating, overcharging, rapid voltage drops, or unexpected
capacity loss. When a fault is identified, the system immediately triggers alerts,
notifying users via IoT- enabled devices, such as mobile phones, about the issue while
providing actionable recommendations, such as scheduling maintenance or modifying
driving behavior.

This proactive fault detection not only prevents critical battery failures but also
enhances safety, ensuring the longevity and performance of the vehicle. the network's
weights are adjusted to minimize prediction errors, allowing the model to effectively
identify abnormalities. Once implemented, the MLP analyzes real-time sensor data to
detect anomalies. Additionally, the scalability of the MLP model allows it to adapt to
advancements in battery technologies or new fault scenarios by retraining with updated
datasets, making it future-proof in the evolving EV landscape.

23
 The combination of IoT for seamless data collection and MLP for real-time
analysis and fault classification ensures a robust solution that improves user confidence
and accelerates the adoption of electric vehicles. The data is preprocessed to normalize
and structure it, ensuring compatibility with the MLP model by enabling predictive
fault management and reducing the likelihood of costly repairs or safety hazards, this
integrated system offers a comprehensive, user-friendly approach to EV battery
monitoring, setting a benchmark for safer, and more efficient transportation solutions.
The training process involves adjusting weights in the network to minimize
prediction errors, enabling the model to recognize abnormalities effectively. Once
deployed, the MLP processes real-time sensor data to detect anomalies, such as
overheating, overcharging, rapid voltage drops, or unexpected capacity loss. When a
fault is identified, the system immediately triggers alerts, notifying users via IoT-
enabled devices, such as mobile phones, about the issue while providing actionable
recommendations, such as scheduling maintenance or modifying driving behavior For
battery management and fault detection in EV systems, MLPs can be trained using
historical data (e.g., voltage, current, temperature) to recognize patterns and predict
anomalies. This makes them useful for applications such as Battery health estimation,
Fault classification, Predictive maintenance, State of charge (SoC) prediction.
MLPs are valued for their flexibility, high accuracy, and ability to generalize well
when trained on sufficient data. However, they require careful tuning of
hyperparameters such as the number of hidden layers, neurons, learning rate, and
training epochs to achieve optimal performance. Additionally, the scalability of the
MLP model allows it to adapt to advancements in battery technologies or new fault
scenarios by retraining with updated datasets, making it future-proof in the evolving
EV landscape. The combination of IoT for seamless data collection and MLP for real-
time analysis and fault classification ensures a robust solution that improves user
confidence and accelerates the adoption of electric vehicles.

24
 CHAPTER 5

HARDWARE IMPLEMENTATION

5.1 INTRODUCTION

This project presents a smart battery monitoring system for EVs using IoT
technology. It uses sensors to track voltage, current, temperature, and state of charge,
enables remote monitoring and alerts. This setup supports predictive maintenance,
improves battery safety and efficiency, and contributes to scalable EV charging
solutions. With the rapid growth of electric vehicles (EVs), there is an increasing need
for efficient and reliable battery management systems to ensure safety, performance,
and extended battery life. Lithium-ion batteries, commonly used in EVs, require
constant monitoring of key parameters such as voltage, current, temperature, and state
of charge to prevent issues like overcharging, overheating, and capacity degradation.
This project focuses on the hardware implementation of a smart battery monitoring
system using IoT technology. The system integrates various sensors to collect real-time
data and a microcontroller unit (such as Arduino UNO or similar) to process and
transmit this data. A wire signal communication, allowing remote monitoring and alert
generation. By combining sensor data with wireless connectivity, the hardware setup
supports condition-based monitoring and predictive maintenance. This enhances
battery reliability, reduces the risk of failure, and contributes to the overall efficiency
and safety of EVs. The implemented system also lays the groundwork for scalable and
intelligent EV charging and monitoring solutions in future smart grid and
transportation ecosystems. Sure! Here's a rewritten version with the same meaning.
Understanding regional driving habits enables more user-oriented and efficient
placement of EV charging stations. Efficient management of large-scale EV charging
demands advanced scheduling algorithms. Evolutionary algorithms can optimize these
schedules by evenly distributing the load across stations and minimizing peak demand
on the power grid.
25
 Fig 5.1 Hardware Experimental Setup

A well-designed planning framework incorporates power flow analysis and grid

capacity evaluation to determine optimal station placement. This approach ensures that
new charging stations do not compromise the stability and reliability of the electricity
distribution system. Traffic flow constraints play a significant role in planning EV
charging stations, as high-traffic areas require strategically placed stations to minimize
detours and queuing times. Multi-objective optimization techniques can balance factors
such as travel convenience, infrastructure costs, and grid impact, leading to a well-
rounded planning strategy that meets both user and system requirements. A collaborative
planning strategy integrates power distribution and EV charging systems to optimize
resource allocation. By coordinating the development of both infrastructures, planners
can enhance system efficiency, reduce costs, and improve service availability. Multi-
objective approaches consider trade-offs between investment costs, energy efficiency,
and service reliability to arrive at an optimal configuration. The design criteria for EV
fast-charging infrastructure must consider local mobility patterns to ensure accessibility
26
 and convenience. In regions with high EV adoption, fast-charging networks should be
distributed based on travel behaviour, ensuring that drivers have sufficient charging
opportunities during their commutes. Analysing regional driving patterns allows for a
more user-centric and effective deployment of charging stations. Large-scale EV
charging requires robust scheduling algorithms to manage demand efficiently.
Evolutionary algorithm-based approaches can optimize charging schedules, balancing
load distribution across stations and reducing peak demand on the grid. These algorithms
enhance overall system efficiency by dynamically adjusting charging patterns based on
real-time data and demand forecasts.

5.2 Hardware Experimental Setup

5.2.1 Arduino Nano

The Figure 5.2 represents the Arduino Nano has similar functionalities as Arduino
Duemilanove but with a different package. The Nano is inbuilt with the ATmega328P
microcontroller, same as the Arduino UNO. The main difference between them is that the
UNO board is presented in PDIP (Plastic Dual-In-line Package) form with 30 pins and
Nano is available in TQFP (plastic quad flat pack) with 32 pins. The extra 2 pins of
Arduino Nano serve for the ADC functionalities, while UNO has 6 ADC ports but Nano
has 8 ADC ports. The Nano board doesn’t have a DC power jack as other Arduino boards,
but instead has a mini-USB port. This port is used for both programming and serial
monitoring. The fascinating feature in Nano is that it will choose the strongest power
source with its potential difference, and the power source selecting jumper is invalid. The
Power Setup for the Arduino R3 operates at 5 Volts. It can either be powered through
USB cable from the computer or through the DC jack provided on the Board.

27
 Fig 5.2 Arduino Nano

5.2.2 LCD (Liquid Crystal Display)

Fig 5.3 LCD Display

The Figure 5.3 represents an LCD (Liquid Crystal Display) screen is a widely used
electronic display module in embedded systems, ideal for real-time monitoring
applications. A common example is the 16x2 LCD, capable of displaying 16 characters
per line across two lines, with each character formed using a 5x7 pixel matrix.
It consists of two registers: the Command register for control instructions and the
Data register for ASCII character values. In practical use, this display is often
integrated into systems to show real-time data such as temperature readings, fire sensor
values, and alert messages, allowing users to instantly monitor environmental
conditions. For instance, the screen might display “Temp: 36.5°C” and “Fire: Normal”
under safe conditions, or “Temp: 85.0°C” and “Fire: Detected! ALERT!” in case of
danger.
28
The display works based on nematic liquid crystals, where molecules align along a
common axis, remaining transparent until disturbed by an electric field, which causes
them to polarize and appear opaque. Removing the field restores their transparent state,
allowing characters to appear and disappear accordingly. This LCD-based output
system enhances situational awareness, safety, and responsiveness in various IoT-
based monitoring applications.

5.2.3 Vibration Sensor

The Figure 5.4 represents the vibration sensor is crucial for detecting mechanical
issues like misalignments, bearing failures, or imbalances in motors. It monitors
vibration levels to maintain motor health and prevent downtime. In this project, a
piezoelectric sensor or accelerometer converts vibrations into electrical signals, which
are sent to the Arduino UNO. These sensors detect any abnormal vibrations caused by
issues like loose components or imbalances within the battery pack. By providing real-
time data on the battery's health and mechanical stability, vibration sensors help in
identifying potential failures early, ensuring the safe operation of the battery and
enhancing the overall reliability of the EV's power system. The Arduino compares
these signals to set thresholds, and if abnormal levels are detected, it triggers an alert
and displays the fault on an LCD module. These sensors help in early fault detection,
ensuring reliability and preventing costly damage by identifying problems before they
escalate.

Fig 5.4 Vibration Sensor

29
5.2.4 Power Supply
The power supply is essential in EV battery management systems, converting
external electrical energy into a stable form for sensors, controllers, and displays. It
ensures consistent voltage and current, preventing damage to sensitive components.
Typically operating on DC, it includes regulators, converters, and protective circuits to
manage varying power needs efficiently. Safety features like overload protection,
short-circuit prevention, and thermal management reduce failure risks. Modern
systems use smart control circuits for dynamic power distribution, making the power
supply key to reliable and efficient operation.

5.2.5 Voltage Sensor

The Figure 5.5 represents an voltage sensor in an EV battery management
system measures the potential difference across battery terminals, ensuring it stays
within safe limits. It helps detect overcharging, deep discharging, and voltage
imbalances, protecting battery health and performance. The sensor converts voltage
into readable signals using techniques like resistive dividers or Hall-effect sensing.
Integrated with microcontrollers, it enables real-time monitoring and triggers safety
actions when needed. High-precision sensors enhance efficiency by ensuring optimal
power distribution.

Fig 5.5 Voltage Sensor

A voltage sensor ensures the battery operates within safe voltage limits by
continuously monitoring and detecting issues like overcharging or deep discharging,
thereby protecting battery health and extending its lifespan.
30
5.2.6 Temperature Sensor
The Figure 5.6 represents an temperature sensor is crucial in an EV battery
management system for monitoring battery cell temperatures. Lithium-ion batteries are
sensitive to temperature changes, and overheating can lead to degradation, thermal
runaway, or fire. The sensor enables cooling strategies and prevents failures by
providing real-time data. Common types include thermistors, RTDs, and
semiconductor sensors. If temperatures exceed safe limits, the system adjusts power or
activates cooling. These sensors also optimize performance by maintaining ideal
temperature ranges, improving efficiency, and supporting predictive maintenance to
enhance reliability and lifespan.

Fig 5.6 Temperature Sensor

5.2.7 Battery
The battery is the most critical component in an electric vehicle, serving as the
primary energy source for propulsion and auxiliary systems. Unlike traditional internal
combustion engines that rely on fuel, EV batteries store electrical energy and deliver it
to the motor as needed. The most commonly used battery type in modern EVs is the
lithium-ion (Li-ion) battery due to its high energy density, efficiency, and long lifespan.
These batteries provide the necessary power for the vehicle while ensuring a balance
between performance and weight. EV batteries are composed of multiple individual
cells grouped into modules and packs. Each cell generates a small amount of voltage,
and when combined, they provide the required voltage and capacity for the vehicle.

31
 5.3nSensor Data Collection The
eeeejThe proposed EV battery management system relies on multiple sensors to
monitor essential battery parameters such as temperature, voltage, and state of charge
(SoC). These sensors are strategically placed within the battery system to continuously
collect real-time data, ensuring that even the slightest fluctuations in battery conditions
are detected. Temperature sensors help prevent overheating by monitoring excessive
heat generation, which could lead to battery degradation or potential hazards. Voltage
sensors ensure that the battery operates within the safe voltage range, preventing
overcharging or deep discharge, which can significantly affect battery lifespan. The
state of charge sensor tracks the remaining battery power, allowing users to plan
charging cycles effectively. By integrating these sensors, the system creates a robust
data acquisition framework that forms the foundation for battery health monitoring and
anomaly detection. Additionally, the real-time sensor data can be fed into machine
learning models for predictive analysis, enabling the early identification of potential
faults or degradation trends. This proactive approach enhances vehicle safety, improves
performance, and minimizes maintenance costs. The modular nature of the sensor setup
also allows for easy upgrades and scalability, making it adaptable to various EV
configurations and battery capacities.

5.4 Data Processing via Microcontroller Once

ddd Once the sensor data is collected, it is sent to a microcontroller, which serves as
the processing unit of the system. The microcontroller, programmed with advanced
algorithms, analyses the incoming data in real-time to check for irregularities. It
compares the sensor readings with predefined safe operating thresholds to detect any
abnormalities, such as an unexpected rise in temperature, a drop in voltage, or unusual
power fluctuations. By processing data locally before transmission, the microcontroller
ensures immediate response to any critical conditions, reducing the risk of battery
failures. This intelligent processing mechanism helps maintain a stable battery

32
performance while minimizing reliance on external computing resources. In addition
to real-time decision-making, the microcontroller can also log data for historical
analysis, enabling long-term tracking of battery health and performance trends. This
data can be used to fine-tune operational thresholds or train machine learning models
for predictive maintenance. The system can also be programmed to trigger alerts,
control cooling systems, or initiate safe shutdown procedures during extreme fault
conditions. Furthermore, the microcontroller’s low power consumption and flexibility
make it ideal for embedded applications in electric vehicles, where energy efficiency
and reliability are crucial.

5.5 IoT-Based Data Transmission To

ddddddTo enable remote monitoring, the system is integrated with IoT connectivity,
which allows real-time data transmission to a central server or cloud platform. Through
wireless communication technologies such as Wi-Fi or GSM, the battery condition data
is securely transmitted to an online database, where it can be accessed by users via a
mobile application or web-based interface. This IoT-based approach ensures that
vehicle owners, fleet managers, and technicians can monitor battery performance from
any location, enhancing accessibility and convenience. Additionally, IoT connectivity
supports seamless software updates and system improvements, ensuring long-term
reliability and adaptability of the battery management system. Moreover, the cloud
platform can be equipped with analytics tools that provide insights into usage patterns,
battery aging, and potential risks, enabling data-driven decision-making. Alerts and
notifications can be configured to inform users instantly of any critical issues, allowing
for immediate intervention. The system can also be integrated with GPS modules for
location-based diagnostics, helping track the performance of vehicles across different
terrains and climates. This comprehensive connectivity not only enhances operational
efficiency but also contributes to the development of smart transportation networks and
sustainable energy solutions.

33
5.6 Real-Time Alert Mechanism
One of the most crucial features of the system is its real-time alert mechanism,
designed to notify users of any detected abnormalities. If the microcontroller identifies
a critical issue, such as an overheating battery or a voltage drop beyond the safe
threshold, an instant alert is generated. These alerts are sent to the user's mobile phone
via SMS, push notifications, or email, allowing them to take immediate corrective
actions, such as stopping vehicle operation or scheduling maintenance. By providing
timely warnings, the system helps prevent potential battery failures, fire hazards, or
vehicle breakdowns, ultimately enhancing the safety and reliability of electric vehicles.
Furthermore, the alert system can be configured to prioritize notifications based on
severity, ensuring that high-risk issues receive immediate attention. In addition to
individual alerts, the system can log and categorize incidents, creating a comprehensive
history of battery performance and maintenance events.
This log can be valuable for troubleshooting, identifying recurring issues, and
planning proactive maintenance schedules. With integration into fleet management
systems, alerts can be sent to designated personnel or service teams, ensuring quick
response times for fleet vehicles. The real-time monitoring and alert capabilities
contribute to reducing downtime, extending battery life, and ensuring a seamless user
experience, all while fostering better safety standards for EVs.

5.7 CODE IMPLEMENTATION

// Include necessary libraries for I2C, temperature sensors, and LCD

#include <Wire.h> // Enables I2C communication
#include <OneWire.h> // For single-wire communication with sensors
#include <DallasTemperature.h> // Library to interface Dallas temperature sensors
#include <LiquidCrystal.h> // Library to control LCD display

// Initialize the LCD, specifying the pins connected to RS, E, D4, D5, D6, D7
34
LiquidCrystal lcd(7,6,5,4,3,2);
#include <SoftwareSerial.h> // Enables serial communication on other pins
// Define pins for software serial communication
int RXPin = 10; // Receiving pin from GSM module
int TXPin = 11; // Transmitting pin to GSM module
int GSMBaud = 9600; // Baud rate for GSM module communication

// Create a software serial object for GSM

SoftwareSerial gsmSerial(RXPin, TXPin); // Initialize software serial on specified pins

// Define data pin for first temperature sensor

#define ONE_WIRE_BUS1 A0
OneWire oneWire1(ONE_WIRE_BUS1); // Create OneWire instance for first sensor
DallasTemperature sensors1(&oneWire1); // Create DallasTemperature instance for first
sensor

// Define data pin for second temperature sensor

#define ONE_WIRE_BUS2 A1
OneWire oneWire2(ONE_WIRE_BUS2); // Create OneWire instance for second sensor
DallasTemperature sensors2(&oneWire2); // Create DallasTemperature instance for second
sensor

// Define analog input pin for battery voltage monitoring

int analogInPin1 = A2;
int sensorValue1; // Store battery voltage raw reading

// Define key switch pin

int ke = A4;

35
int key = 0; // Current key reading
int keystatus = 0; // Track toggle status

// Fire sensor and vibration sensor pins

int firev = 0; // Fire sensor pin (not used)
int fire = 0; // Fire sensor value (not used)
int vibv = A3; // Vibration sensor analog pin
int vib = 0; // Vibration sensor value

// Output devices
int buzzer = 13; // Buzzer connected to pin 13
int motor = 12; // Motor control on pin 12
int fan = 0; // Fan control pin (unused in pinMode but used in code)

// Control variables
int on_d = 0; // Flag to check motor ON status
int countt1 = 0; // Count for battery temp alert
int countt2 = 0; // Count for engine temp alert
int incomingByte = 0; // Placeholder for serial data (not used)
String msg_data; // Message to be sent via GSM

void setup() {
// Set pin modes for inputs and outputs
pinMode(analogInPin1, INPUT); // Battery voltage pin as input
pinMode(ke, INPUT); // Key input
pinMode(firev, INPUT); // Fire sensor input (not used)
pinMode(vibv, INPUT); // Vibration sensor input
pinMode(buzzer, OUTPUT); // Buzzer output

36
 pinMode(fan, OUTPUT); // Fan output
pinMode(motor, OUTPUT); // Motor output

// Initialize temperature sensors

sensors1.begin();
sensors2.begin();

// Initialize GSM and serial communication

gsmSerial.begin(GSMBaud); // Start GSM module at 9600 baud
Serial.begin(9600); // Start Serial monitor at 9600 baud
// Start LCD and display welcome message
lcd.begin(16, 2);
lcd.print(" IOT BASED EV ");
lcd.setCursor(0, 1);
lcd.print(" FAULT MANAG.");
delay(1900); // Hold message on screen
digitalWrite(buzzer, HIGH); // Beep once
lcd.clear();
delay(250);
digitalWrite(buzzer, LOW); // Stop beep
digitalWrite(motor, LOW); // Ensure motor is OFF
digitalWrite(fan, LOW); // Ensure fan is OFF
}

void loop() {
// Read key and vibration sensor
key = digitalRead(ke); // Read the key input
vib = digitalRead(vibv); // Read the vibration sensor

37
// Request temperatures
sensors1.requestTemperatures(); // Request temp from sensor 1
int tempvv1 = sensors1.getTempCByIndex(0); // Get temperature in Celsius from sensor 1
sensors2.requestTemperatures(); // Request temp from sensor 2
int tempvv2 = sensors2.getTempCByIndex(0); // Get temperature in Celsius from sensor 2

// Read battery voltage

sensorValue1 = analogRead(analogInPin1);
int bvolt1 = sensorValue1 / 42; // Approximate voltage from analog read
// Calculate battery percentage
int batper1 = mapfloat(bvolt1, 1.8, 4.2, 0, 100); // Map voltage to 0–100%
if (batper1 >= 100) batper1 = 100;
if (batper1 <= 0) batper1 = 1;

// Display data on LCD

lcd.setCursor(0, 0);
lcd.print("BT:");
lcd.print(tempvv1);
lcd.print(" ");
lcd.setCursor(0, 1);
lcd.print("ET:");
lcd.print(tempvv2);
lcd.print(" ");
lcd.setCursor(6, 0);
lcd.print("BV:");
lcd.print(bvolt1);
lcd.print(" ");

38
lcd.setCursor(6, 1);
lcd.print("EV:");
lcd.print(vib);
lcd.print(" ");

// Toggle motor state on key press

if (key == HIGH) {
if (keystatus == 0) {
keystatus = 1;
digitalWrite(buzzer, HIGH); delay(100); digitalWrite(buzzer, LOW);
digitalWrite(motor, HIGH); on_d = 1;
} else {
keystatus = 0;
digitalWrite(buzzer, HIGH); delay(500); digitalWrite(buzzer, LOW);
digitalWrite(motor, LOW); on_d = 0;
}
delay(500); // Debounce delay
}

// Battery temperature too high

if(tempvv1 >= 38){
digitalWrite(buzzer, HIGH);
digitalWrite(fan, HIGH);
delay(250);
digitalWrite(buzzer, LOW);
countt1++;
if(countt1 == 1){
msg_data = " Battery Temp HIGH...!";

39
 gsm_msg(1); // Send GSM alert
}
}
else {
digitalWrite(fan, LOW);
countt1 = 0;
}

// Battery voltage too low

if(bvolt1 <= 2){
digitalWrite(motor, LOW);
digitalWrite(buzzer, HIGH); delay(250); digitalWrite(buzzer, LOW);
}
// Engine temperature too high
else if(tempvv2 >= 38){
digitalWrite(motor, LOW);
digitalWrite(buzzer, HIGH); delay(250); digitalWrite(buzzer, LOW); delay(250);
countt2++;
if(countt2 == 1){
msg_data = " Engine Temp HIGH...!";
gsm_msg(1); // Send GSM alert
}
}
// Vibration detected
else if(vib == 1){
digitalWrite(motor, LOW);
digitalWrite(buzzer, HIGH); delay(250); digitalWrite(buzzer, LOW); delay(250);
}

40
 else {
countt2 = 0;
}
// Send serial monitor debug data
Serial.print("123 ");
Serial.print(tempvv1);
Serial.print(" ");
Serial.print(tempvv2);
Serial.print(" ");
Serial.print(bvolt1);
Serial.print(" ");
Serial.print(vib);
Serial.println();
delay(1000); // Loop delay
}
// Convert float values from one range to another
float mapfloat(float x, float in_min, float in_max, float out_min, float out_max)
{
return (x - in_min) * (out_max - out_min) / (in_max - in_min) + out_min;
}
// Send SMS using GSM module
void gsm_msg(int a)
{
lcd.setCursor(12,1);
lcd.print("SDG."); // Display message sending status
delay(100);
gsmSerial.println("AT"); // Basic AT command to test GSM
delay(500);

41
gsmSerial.println("AT+CMGF=1"); // To send SMS in Text Mode
digitalWrite(buzzer, HIGH);
delay(1000);

if (a == 1){
gsmSerial.println("AT+CMGS=\"+919344894698\"\r"); // Specify recipient phone
number
lcd.setCursor(8,1);
lcd.print("SDG..1");
delay(100);
}
digitalWrite(buzzer, LOW);
delay(1000);
lcd.setCursor(12,1);
lcd.print("SDG..");
gsmSerial.println("Emergency Alert...!" + msg_data); // Compose and send alert message
delay(500);
lcd.setCursor(12,1);
lcd.print("SDG...");
delay(1000);
gsmSerial.println((char)26); // Send Ctrl+Z (end of message)
delay(1000);
digitalWrite(buzzer, HIGH);
lcd.setCursor(12,1);
lcd.print("SMS SENT...."); // Confirmation on LCD
delay(1000);
lcd.setCursor(12,1);
lcd.print(" "); // Clear LCD message area

42
 delay(100);
digitalWrite(buzzer, LOW);
}

5.8 CODE FUNCTIONALITY

Setup

• Initializes the LCD and displays a welcome message ("IOT BASED EV FAULT
MANAG.")
• Configures the serial communication for debugging and GSM communication.
• Sets up pin modes for temperature sensors, fire sensor, vibration sensor, motor, fan,
and buzzer.
• Initializes the temperature sensors and GSM module.
• Resets all actuators (motor, fan, buzzer) to ensure they start in an off state.

Loop

Reads Sensor Values:

✓ Reads the battery and engine temperatures using the Dallas Temperature library.
✓ Reads battery voltage and converts it to a percentage using a mapping function.
✓ Reads vibration sensor input and ignition key status.

Displays Sensor Data on LCD

✓ Displays battery and engine temperatures on the first and second lines of the LCD.
✓ Displays battery voltage and vibration status on the LCD.

43
Monitors Key Status

✓ If the key is turned on, the motor is activated, and the buzzer gives a confirmation
beep.
✓ If the key is turned off, the motor stops, and the buzzer alerts the user.

Fault Detection and Alerts

✓ If battery temperature exceeds 38°C, the fan turns on, and an alert message is sent
via GSM.
✓ If battery voltage drops below 2V, the motor is turned off, and a buzzer alarm is
triggered.
✓ If engine temperature exceeds 38°C, the motor is turned off, and a warning
message is sent.

Serial Communication for Debugging

✓ Sensor data (battery temperature, engine temperature, voltage, vibration) is printed

to the Serial Monitor.
✓ Helps in debugging and verifying system functionality in real-time.

Purpose
The program interface in this IoT-based Electric Vehicle (EV) Fault Management
and Battery Monitoring System is designed to serve as the central control logic that
continuously monitors key vehicle parameters such as battery temperature, engine
temperature, battery voltage, and vibration levels using various sensors. This
interface ensures real-time responsiveness, enhanced safety, and proactive fault
handling, making it a critical component of a reliable and intelligent EV safety
system.

44
 This table 5.1 presents a detailed comparison between the existing and proposed
Battery Management Systems (BMS). The proposed system introduces advanced
technologies and methodologies such as IoT integration, AI-based analytics, and
predictive maintenance to enhance efficiency, reliability, and scalability.
Table 5.1 Comparison Table of Existing and Proposed System

S.No Feature Existing System Proposed System

1 Monitoring Post-fault Real-time monitoring with
Type detection predictive analytics
2 Sensor Basic sensors: Advanced sensors:
Integration Voltage, Voltage, Temperature,
Temperature Current, Fire, Vibration
3 Communication GSM-based alerts IoT-enabled cloud
communication with
mobile/web access
4 Data Analysis Manual monitoring AI-based analysis using
and analysis MLP (Multilayer
Perceptron) algorithm
5 Fault Detection Partial and delayed Instant multi-fault
detection with alerts
6 User Interface LCD display only LCD + Smartphone
notifications + Cloud
dashboard
7 Response Requires manual Automated: Buzzer alert,
Mechanism actions Relay control, Predictive
notifications
8 Maintenance Reactive (after Predictive and scheduled
Planning failure) based on condition data
9 Energy Not implemented Smart charging/discharging
Optimization for improved efficiency

45
 CHAPTER 6
RESULTS AND DISCUSSION

6.1 SYSTEM EVALUATION AND PERFORMANCE ANALYSIS

The proposed system takes a step towards enhancing On-Board Diagnostics (OBD)
by implementing a simple and effective vehicle self-diagnosis and reporting system
specifically designed for electric vehicles (EVs). It aims to monitor the battery conditions
in real-time, detect any anomalies, and alert the vehicle user through their mobile phone.
The project incorporates a machine learning-based data analysis approach to predict
battery conditions more accurately, providing early warnings and ensuring the safety of the
vehicle. A new design for monitoring the battery during charging is implemented to
optimize battery health management and improve charging efficiency. The system
represents a significant advancement in battery monitoring technology, offering proactive
solutions to maintain optimal battery performance and extend its lifespan.

Fig 6.1 EV Battery Thermal Performance

The Figure 6.1 represents real-time battery temperature monitoring for an EV Battery
Management System (BMS) IoT using ThingSpeak. Initially, the battery temperature
remains stable at around 30°C, indicating normal operation. However, there is a sudden
spike to 42°C, which could be due to high power consumption, overcharging, or delayed
46
activation of the cooling system. Shortly after reaching this peak, the temperature begins
to drop, suggesting that a cooling mechanism, such as a fan or ventilation system, was
activated. Between the temperature stabilizes around 30–32°C, indicating that the system
effectively regulated the heat. This fluctuation highlights the importance of a responsive
cooling system to prevent overheating and improve battery efficiency. Although the
cooling system worked, the delay in activation might require optimization to prevent
sudden spikes in future operations.

Fig 6.2 Engine Temperature Performance

The Figure 6.2 represents real-time engine temperature monitoring for an EV
Battery Management System (BMS) IoT using ThingSpeak. Initially, the engine
temperature remains constant at 30°C, indicating stable conditions. Around the
temperature begins to rise, reaching a peak of 34°C. This spike could be due to increased
engine load, a rise in ambient temperature, or a delay in the activation of the cooling
mechanism.
Shortly after peaking, the temperature starts to drop, suggesting that a cooling
system, such as a fan or heat dissipation mechanism, was activated. By the temperature
stabilizes around 31°C, and it gradually returns to 30°C, indicating that the cooling system
effectively regulated the heat. This pattern highlights the engine's thermal behavior under
load and the importance of an efficient cooling system.

47
 Fig 6.3 Battery Voltage Behaviour
The Figure 6.3 represents battery voltage data collected from an Electric Vehicle
Battery Management System (EV BMS) using IoT technology, displayed on the
ThingSpeak platform. The x-axis represents time (Date), while the y-axis represents the
battery voltage (Battery Volt). Initially, the voltage fluctuates between 0V and
approximately 6V, indica ting instability or varying load conditions. Around the timestamp,
there is a significant rise in voltage, stabilizing at around 8V, suggesting that the battery
reached a steady state or charging was completed. The red line connecting data points
shows real-time monitoring of the battery voltage, helping in performance analysis and
fault detection in the EV's power system.

Fig 6.4 Vibration Sensor Monitoring Chart

48
 The Figure 6.4 represents an displays vibration data from an Electric Vehicle Battery
Management System (EV BMS) using IoT technology on the ThingSpeak platform. The
x-axis represents time, while the y-axis shows vibration levels. Throughout the recorded
period, the vibration value remains constant at 1.00, suggesting that the system detects
continuous, unchanging vibrations. This stability could indicate that the EV is operating
smoothly without mechanical disturbances or that the sensor has limited sensitivity,
possibly reading only binary values (on/off). Consistent vibration monitoring is essential
in EVs to identify potential issues such as misalignments, imbalances, or structural
weaknesses that could affect performance, safety, and longevity. Detecting unusual
vibration patterns can help in preventive maintenance, reducing the risk of failures and
ensuring optimal operation.

Fig 6.5 Fire Detection Alert System Performance

The Figure 6.5 represents fire detection data from an Electric Vehicle Battery
Management System (EV BMS) using IoT technology, visualized on the ThingSpeak
platform. The x-axis represents time (Date), while the y-axis represents fire detection
status. The data remains constant at 1.00 throughout the recorded period, indicating that
the fire sensor is either continuously detecting fire conditions or is set to a default value
of 1 for monitoring purposes. This consistency suggests that no changes in fire risk were
detected, or it may imply a potential issue with the sensor calibration. Fire detection is to
prevent overheating and battery-related hazards, ensuring vehicle safety and performance.
49
6.2 Comparative Analysis of Existing vs Proposed Battery Management System
To evaluate the efficiency and technological advancement of the proposed Battery
Management System (BMS), a comparative analysis has been conducted against the
traditional system. Key performance indicators such as monitoring capability, sensor
integration, communication methods, data analysis, and response mechanisms have been
assessed. The table below summarizes the quantified improvements across critical
aspects of the BMS, highlighting the shift from a reactive to a predictive and intelligent
system architecture.
Table 6.1 Quantitative Evaluation of Existing vs Proposed BMS Solutions

S.No Feature Existing System (Value) Proposed System

(Value)
1 Monitoring Type 30% coverage 90% coverage

2 Sensor Integration 2 sensors 5 sensors

3 Communication 1-way (GSM only) 2-way (IoT + Cloud)

4 Data Analysis 20% accuracy (manual) 85% accuracy

(MLP-based AI)
5 Fault Detection 40% detection rate 95% detection rate

6 User Interface LCD only (score: 3/10) LCD + App + Cloud

(9/10)
7 Response Mechanism Manual (response time: 30s) Automated
(response time: 5s)

8 Maintenance 30% predictive 80% predictive

Planning
9 Energy Optimization 50% optimization 70% efficiency gain

50
 CHAPTER 7
CONCLUSION
7.1 CONCLUSION

EV control systems have transitioned to distributed, online, corporate

infrastructures, moving away from traditional air-gapped, centralized models. This
shift has brought substantial improvements in efficiency, speed, and accuracy, enabling
real- time monitoring and more advanced diagnostics. However, it has also exposed
these systems to vulnerabilities inherent in the unprotected internet, increasing the risk
of cyber threats, unauthorized access, and potential system manipulation. Battery pack
monitoring systems, which are integral to maintaining battery health, rely on cell
balancing to prevent voltage disparities between individual cells. The shift to connected
EV diagnosis systems means that batteries are now more susceptible to cyber-attacks
as part of the broader Internet of Things (IoT). To address these challenges, machine
learning- based data analysis methods have been proposed. These algorithms enable
proactive, data-driven battery management strategies, offering predictive capabilities
that optimize charging and discharging cycles, monitor battery health in real-time, and
detect early signs of failure before they occur.

This approach enhances battery longevity, efficiency, and reliability in a wide

range of applications, from electric vehicles (EVs) to portable electronics and
renewable energy storage systems. By continuously adapting to real-time data,
machine learning ensures that the battery management system operates with precision,
maximizing energy usage, extending vehicle range, and improving overall user
experience. Furthermore, these methods reduce the need for manual intervention,
minimizing human error and ensuring consistent performance.

51
7.2 FUTURE SCOPE

• AI and Machine Learning Integration: Future systems will predict battery failures
and optimize maintenance schedules, extending battery lifespan.
• Real-time Adaptive Charging: The system will adjust charging parameters
dynamically, particularly during fast-charging sessions, to improve battery health.
• Battery Lifecycle Optimization – Prolong battery life through intelligent State of
Charge and State of Health balancing strategies.
• Blockchain for Data Security: Blockchain will secure battery health data, ensuring
transparency and protection against tampering.
• Smart Charging Algorithms – Dynamic adjustment of charging parameters based
on usage patterns, temperature, and State of Charge.
• Real-Time Predictive Maintenance: The system can be extended to provide real-
time maintenance alerts by continuously analyzing sensor data and predicting failures
before they occur.
• Autonomous Vehicle Energy Optimization: AI will optimize energy consumption
based on driving conditions and route planning, enhancing efficiency in autonomous
vehicles.
• Collaboration with Smart Grids: EVs will act as mobile energy storage, aiding
energy distribution during peak demand times and supporting smart grid integration.
• Adaptive Charging Algorithms: Real-time adjustment of charging currents based
on battery health and environmental conditions to reduce stress on battery cells.
• Enhanced Multi-Fault Classification: Expand the MLP model to classify multiple
battery issues simultaneously for comprehensive fault detection.

52
 REFERENCES

[1] Hongcai Zhang, Student Member, IEEE, Zechun Hu, Member, IEEE,Zhiwei Xu,
Student Member, IEEE, and Yonghua Song, Fellow, IEEE “Optimal Planning of PEV
Charging Station With Single Output Multiple Cables Charging Spots”, IEEE transactions
on smart grid, vol.8,no.5,september 2017

[2] Zhipeng Liu, Fushuan Wen, and Gerard Ledwich, Senior Member, IEEE “Optimal
Planning of Electric-Vehicle ChargingStations in Distribution Systems” IEEE
transactions on power delivery, vol. 28, no. 1, january 2013

[3] Guibin Wang, Zhao Xu, Senior Member, IEEE, Fushuan Wen, and Kit Po Wong,
Fellow, IEEE “Traffic-Constrained Multiobjective Planning of Electric-Vehicle
Charging Stations.”IEEE transactions on power delivery, vol. 28, no. 4, october 2013

[4] P.Valarmathi, M. Ramchander, S. Suneetha, P. Sasirekha, S. K. Mohammed and A.

Hussain, "The Deep Analysis of Accuracy of Performance Level of NBC Algorithm
for Diagnosing BC," 2024 4th International Conference on Advance Computing and
Innovative Technologies in Engineering (ICACITE), Greater Noida, India, 2024, pp.
93-97, doi: 10.1109/ICACITE60783.2024.10616769

[5] Nick Machiels, Niels Leemput, Graduate Student Member, IEEE,Frederik Geth,
Graduate Student Member, IEEE, Juan Van Roy, Graduate Student Member, IEEE,
JeroenBüscher, Member, IEEE, and Johan Driesen, Senior Member, IEEE“Design
Criteria for Electric Vehicle Fast Charge Infrastructure Based on Flemish Mobility
Behavior” IEEE transactions on smart grid, vol. 5, no. 1, january 2014

[6] S. Surya, M. Shalini, P. Shalini, R. Srinitha and P. Sasirekha, "Design And

Implementation ofModified Zeta Converter Using Adaptive Sliding Mode
Technique," 2024 2nd International Conference on Artificial Intelligence and
Machine Learning Applications Theme: Healthcare and Internet of Things

53
 [7] G. P, R. B, G. Singaram, S. P, S. V and R. S, "Electricity Generation of Dynamic
Bifacial Solar Panels Using IoT," 2023 International Conference on Self Sustainable
Artificial Intelligence Systems (ICSSAS), Erode, India, 2023, pp. 1605-1610, doi:
10.1109/ICSSAS57918.2023.10331709

[8] A. Y. S. Lam, Y. W. Leung, and X. Chu, "Electric Vehicle Charging Station

Placement:Formulation, Complexity, and Solutions," IEEE Transactions on Smart
Grid, vol. 5, no. 4, pp. 1457–1466, Jul. 2014.

[9] Z. Liu, F. Wen, and G. Ledwich, "Optimal Planning of ElectricVehicle Charging

Stations in Distribution Systems," IEEE Transactions on Power Delivery, vol. 28,
no. 1, pp. 102–110, Jan. 2013

[10] G. Wang, Z. Xu, F. Wen, and K. P. Wong, "Traffic-Constrained Multiobjective

Planning of Electric-Vehicle Charging Stations," IEEE Transactions on Power
Delivery, vol. 28, no. 4, pp. 2363–2372, Oct. 2013

[11] P.Sasirekha, E. Sneka, B. Velmurugan, M. Sahul Hameed and P. Sivasankar, "A

Battery Monitoring System based on IoT for Electric Vehicles," 2023 5th
International Conference on Smart Systems and Inventive Technology (ICSSIT),
Tirunelveli, India,2023, pp. 204-210, doi: 10.1109/ICSSIT55814.2023.10061035

[12] N. Machiels, N. Leemput, F. Geth, J. Van Roy, J. Büscher, and J. Driesen, "Design
Criteria for Electric Vehicle Fast Charge Infrastructure Based on Flemish Mobility
Behavior," IEEE Transactions on Smart Grid, vol. 5, no. 1, pp. 320–327, Jan. 2014

[13] W. Su and M. Y. Chow, "Performance Evaluation of an EDA-Based Large-Scale

Plug-In Hybrid Electric Vehicle Charging Algorithm," IEEE Transactions on Smart
Grid, vol. 3, no. 1, pp. 308–315, Mar. 2012

[14] J. Tan and L. Wang, "Real-Time Charging Navigation of Electric Vehicles to Fast
Charging Stations: A Hierarchical Game Approach," IEEE Transactions on Smart
Grid, vol. 8, no. 2, pp. 846–856, Mar. 2017.

54
[15] K.Clement-Nyns, E. Haesen, and J. Driesen, "The Impact of Vehicle-to-Grid on the
Distribution Grid," IEEE Transactions on Power Systems, vol. 25, no. 1, pp. 371–
380.

[16] Q. Huang, M. Chow, and X. Zhang, "Distributed Optimal EV Charging Control for
Coupled Distribution and Transportation Networks," IEEE Transactions on
Cybernetics, vol. 50, no. 5, pp. 1785–1796, May 2020.

[17] F. Sadeghi-Barzani, A. Rajabi-Ghahnavieh, and H. Kazemi-Karegar, "Optimal Fast

Charging Station Placement in Distribution Network Considering Load Balancing,"
IEEE Transactions on Power Delivery, vol. 28, no. 4, pp. 2313–2322, Oct. 2013.

[18] Hongcai Zhang, Student Member, IEEE, Zechun Hu, Member, IEEE,Zhiwei Xu,
Student Member, IEEE, and Yonghua Song, Fellow, IEEE “Optimal Planning of
PEV Charging Station With Single Output Multiple Cables Charging Spots”, IEEE
TRANSACTIONS ON SMART GRID, VOL.8,NO.5,SEPTEMBER 2017.

[19] Zhipeng Liu, Fushuan Wen, and Gerard Ledwich, Senior Member, IEEE “Optimal
Planning of Electric-Vehicle ChargingStations in Distribution Systems” IEEE
TRANSACTIONS ON POWER DELIVERY, VOL. 28, NO. 1, JANUARY 2013.

[20] N. Machiels, N. Leemput, F. Geth, J. Van Roy, J. Büscher, and J. Driesen, "Design
Criteria for Electric Vehicle Fast Charge Infrastructure Based on Flemish Mobility
Behavior," IEEE Transactions on Smart Grid, vol. 5, no. 1, pp. 320–327, Jan. 2014.

[21] J. Tan and L. Wang, "Real-Time Charging Navigation of Electric Vehicles to Fast
Charging Stations: A Hierarchical Game Approach," IEEE Transactions on Smart
Grid, vol. 8, no. 2, pp. 846–856, Mar. 2017.

55
 LIST OF PUBLICATION

1. Nathiya. S, Ramamani.V, Sarathi. M, Tamil Selva Anto. K and Hariprabhu. M, "IoT Based
Battery Monitoring System in Electric Vehicle with Multiple Fault Detection Using
AI," 2025 International Conference on Emerging Systems and Intelligent Computing (ESIC),
Bhubaneswar, India, 2025, pp. 471-478, doi: 10.1109/ESIC64052.2025.10962668.

56
57
58
59