AI-BASED FAULT PREDICTION IN TRANSMISSION LINE BY USING

WIDE NEURAL NETWORK

PROJECT REPORT

Submitted by
MANIKANDA PRABU.C -953221105018
RAKESH. M -953221105307

In partial fulfillment for the award of the degree

Of
BACHELOR OF ENGINEERING
IN
ELECTRICAL AND ELECTRONICS ENGINEERING

UNIVERSITY VOC COLLEGE OF ENGINEERING,

THOOTHUKUDI.

ANNA UNIVERSITY : CHENNAI 600 025

MAY 2025

i
 ANNA UNIVERSITY: CHENNAI 600 025

BONAFIDE CERTIFICATE
Certified that this project report “AI-BASED FAULT PREDICTION IN
TRANSMISSION LINE BY USING WIDE NEURAL NETWORK” is the bonafide work
of
RAKESH.M 953221105307
MANIKANDA PRABU.C 953221105018

who carried out the project work under my supervision.

SIGNATURE SIGNATURE
DR. P. ANITHA M.E., Ph.D., DR. P. ANITHA M.E., Ph.D.,
HEAD OF THE DEPARTMENT PROJECT GUIDE
Department of Electrical And Department of Electrical And
Electronics Engineering. Electronics Engineering.
University VOC College of University VOC College of
Engineering. Engineering.
Thoothukudi – 628008 Thoothukudi - 628008

Submitted for the VIVA-VOCE Examination held on __________________.

INTERNAL EXAMINER EXTERNAL EXAMINER

ii
 ACKNOWLEDGEMENT
It is one of the most difficult tasks in life to choose words to express one’s attitudes gratitude
towards the beneficiaries. We are very much grateful to the ALMIGHTY GOD who helped us
the way throughout the project and who has molded us into what we are today. We wish to
express our heartfelt regards and sincere thanks to our beloved Dean Dr. C. PETER
DEVADOSS M.E., Ph.D., for him constant encouragement during the course of this project
work.

We have immense pleasure in expressing our sincere thanks to our Head of the Department
Dr. P. ANITHA, M.E., Ph.D., Department of Electrical and Electronics Engineering, for her
insightful comments and constructive suggestions to improve the quality of this project work.

We would like to articulate our profound gratitude and indebtedness to our project coordinator
Dr. P. ANITHA, M.E., Ph.D., for his stimulating guidance, continuous encouragement and
supervision throughout the course of this work. It has been an immense pleasure for us to get
an opportunity to work under him and finish the project successfully. Finally, we acknowledge
our indebtedness to our parents, friends, all our teaching and non-teaching faculties and all who
participated enthusiastically with their constructive criticism either directly or indirectly in
accomplishing this project successfully.

iii
 ABSTRACT

This project presents a robust artificial intelligence-based framework for fault

prediction and location in electrical transmission lines using Phasor
Measurement Unit (PMU) data and a Wide Neural Network (WNN). Leveraging
synchronized voltage and current measurements from PMUs, the system
accurately identifies fault types and locates faults in real time within the WSCC
9-bus test system. The WNN model is trained with comprehensive fault
scenarios—including various types and locations—under different operating
conditions. The proposed approach demonstrates high accuracy, noise
resilience, and computational efficiency compared to traditional methods,
enhancing the reliability and speed of fault management in modern power grids.

iv
 TABLE OF CONTENTS
PAGE
CHAPTER TITLE
NO.
ABSTRACT iv

LIST OF FIGURES vii

LIST OF TABLES ix

ABBREVATIONS xi

1 INTRODUCTION 1

1.1 BACKGROUND 1

1.2 MOTIVATION 2

1.3 PROBLEM STATEMENT 2

1.4 OBJECTIVES 2

1.5 SCOPE OF THE PROJECT 2

2 LITERATURE REVIEW 3

2.1 INTRODUCTION 3

2.2 CONVENTIONAL FAULT LOCATION TECHNIQUES 3

2.3 ARTIFICIAL INTELLIGENCE IN FAULT DIAGNOSIS 4

2.4 WIDE NEURAL NETWORKS (WNNS) IN POWER

6
SYSTEMS

2.5 SUMMARY OF FINDINGS 6

3 SYSTEM AND TECHNIQUES 7

3.1 INTRODUCTION 7

3.2 WSCC 9-BUS SYSTEM OVERVIEW 7

3.3 LOAD FLOW ANALYSIS 9

3.4 TYPES OF TRANSMISSION LINE FAULTS

12
CONSIDERED

v
 PAGE
CHAPTER TITLE
NO.
3.5 DATA ACQUISITION AND PREPROCESSING 13

3.6 SUMMARY 13

4 METHODOLOGY 14

4.1 INTRODUCTION 14

4.2 OVERALL WORKFLOW 14

4.3 SIMULATION OF FAULT SCENARIOS 14

4.4 WIDE NEURAL NETWORK (WNN) ARCHITECTURE 15

4.4.1 WHY WIDE NEURAL NETWORK? 15

4.4.2 MODEL STRUCTURE 15

4.5 PERFORMANCE METRICS 16

4.6 RETRIEVING DATA 17

4.7 DATA PREPARATION 18

4.7.1 DATA TRANSFORMATION SIMULATION 19

4.8 DATA EXPLORATION 20

4.9 MODEL DEVELOPMENT 24

4.9.1 PREDICTION NETWORK ARCHITECTURE 24

4.9.2 CLASSIFICATION NETWORK

24
ARCHITECTURE

4.9.2.1 DATA INTEGRATION 28

4.9.2.2 NETWORK DESIGN PROGRAM 29

4.9.2.3 MODEL DEVELOPMENT (CLASSIFICATION) 30

4.10 FAULT IDENTIFICATION PROGRAM 33

5 RESULTS AND DISCUSSION 35

vi
 PAGE
CHAPTER TITLE
NO.
5.1 PERFORMANCE EVALUATION 40

6 CONCLUSION AND FUTURE SCOPE 41

REFERENCES 42

vii
 LIST OF FIGURES

FIGURE NO TITLE PAGE NO

2.1 SIMPLIFIED LINE MODEL FOR 3

GENERAL ONE END
METHOD
2.2 TRAVEL BASED FAULT 4
SYSTEM
2.3 ARCHITECTURE OF NEURAL 5
NETWORK
2.4 SUPPORT VECTOR MACHINE 5
IN FAULT DETECTION
3.1 SIMULATION MODEL OF 8
9-BUS SYSTEM
3.2 TYPES OF FAULT IN 12
TRANSMISSION LINE
4.1 WNN NETWORK 15
4.2 ANN WORKFLOW 16
4.3 RETRIEVING DATA 17
4.4 SIMULATION SIM-INPUT 18
SIM-OUTPUT
4.5 NORMAL CONDITION 22
[VOLTAGE & CURRENT]
4.6 ABNORMAL CONDITION 23
[VOLTAGE & CURRENT]
4.7 PREDICTION NETWORK LAYER 24
4.8 FUNCTION OF ANN 25
4.9 RESULT OF DATASET TRAINING 26
4.10 NEURAL NETWORK PREDICTION 26
4.11 CLASSIFICATION NETWORK 27
LAYER
4.12 RESULT OF CLASSIFICATION 31
DATASET [FAULT LINE]

viii
FIGURE NO TITLE PAGE NO

4.13 CONFUSION MATRIX OF 31

FAULT LINE DATASET
4.14 RESULT OF CLASSIFICATION 32
DATASET [FAULT TYPE]
4.15 CONFUSION MATRIX OF 32
FAULT TYPE DATASET
4.16 FAULT IDENTIFICATION NETWORK 33

ix
 LIST OF TABLES

TABLE NO TITLE PAGE NO

3.1 LOAD FLOW ANALYSIS 9

4.1 DATA EXPLORATION 20
UNDER NO FAULT CONDITION
[VOLTAGE]
4.2 DATA EXPLORATION 20
UNDER NO FAULT CONDITION
[CURRENT]
4.3 ASYMMETRICAL FAULT 21
CONDITION
[AG FAULT-VOLTAGE]
4.4 ASYMMETRICAL FAULT 21
CONDITION
[AG FAULT-CURRENT]
4.5 SYMMETRICAL FAULT 21
CONDITION
[ABCG FAULT-VOLTAGE]
4.6 SYMMETRICAL FAULT 22
CONDITION
[ABCG FAULT-CURRENT]
5.1 ACTUAL AND PREDICTED 35
FAULT LOCATION
[AG FAULT AT LINE 1-2]
5.2 ACTUAL AND PREDICTED 36
FAULT LOCATION
[ABG FAULT AT LINE 1-2]
5.3 ACTUAL AND PREDICTED 36
FAULT LOCATION
[ABCG FAULT AT LINE 1-2]

x
TABLE NO TITLE PAGE NO
5.4 ACTUAL AND PREDICTED 37
FAULT LOCATION
[BG FAULT AT LINE 5-6]
5.5 ACTUAL AND PREDICTED 37
FAULT LOCATION
[BCG FAULT AT LINE 5-6]
5.6 ACTUAL AND PREDICTED 38
FAULT LOCATION
[ABCG FAULT AT LINE 5-6]
5.7 ACTUAL AND PREDICTED 38
FAULT LOCATION
[CG FAULT AT LINE 3-4]
5.8 ACTUAL AND PREDICTED 39
FAULT LOCATION
[CAG FAULT AT LINE 3-4]
5.9 ACTUAL AND PREDICTED 39
FAULT LOCATION
[ABCG FAULT AT LINE 3-4]

xi
 ABBREVATIONS

AI ARTIFICIAL INTELLIGENCE
ANFIS ADAPTIVE NEURO-FUZZY INFERENCE SYSTEM
ANN ARTIFICIAL NEURAL NETWORKS
DGNN DEEP GRAPH NEURAL NETWORK
DL DEEP LEARNING
DT DECISION TREE
DTT DIGITAL TWIN TECHNOLOGY
DWT- DISCRETE WAVELET TRANSFORM-ADAPTIVE NEURO-FUZZY
ANFIS INFERENCE SYSTEM
EmHW EMBRYONIC HARDWARE
FD FAULT DISTANCE
FF-ANNC FIREFLY ALGORITHM-TRAINED ANN CONTROLLER
FIA FAULT INCEPTION ANGLE
FL FAULT LOCATION
FR FAULT RESISTANCE
GPS GLOBAL POSITIONING SYSTEM
LIMITED-MEMORY BROYDEN, FLETCHER, GOLDFARB, AND
L-BFGS
SHANNO
L-G LINE-TO-GROUND FAULT
LL LINE-TO-LINE FAULT
LL-G DOUBLE LINE-TO-GROUND FAULT
LLL TRIPLE LINE FAULT
LLL-G TRIPLE LINE-TO-GROUND FAULT
MAE MEAN ABSOLUTE ERROR
MATLAB MATRIX LABORATORY
ML MACHINE LEARNING
MSE MEAN SQUARED ERROR

xii
OPP OPTIMAL PMU PLACEMENT
PMU PHASOR MEASUREMENT UNIT
PQ POWER QUALITY
ReLU RECTIFIED LINEAR UNIT
RMSE ROOT MEAN SQUARED ERROR
SG SMART GRIDS
SOM SELF-ORGANIZING MAP
SPV SOLAR PHOTOVOLTAIC
SVM SUPPORT VECTOR MACHINE
TL TRANSMISSION LINE
TLBO TEACHING-LEARNING-BASED OPTIMIZATION
WAMS WIDE AREA MONITORING SYSTEM
WNN WIDE NEURAL NETWORK
WSCC WESTERN SYSTEM COORDINATING COUNCIL

xiii
xiv