Quaid e Awam University Of Engineering

Science and Technology Nawabshah

FYP-1
IOT BASED SMART ENERGY METER MONITORING
WITH THEFT DETECTION

Group Members:
Muhammad Umer (18EL79) {GL}
Abdul Razaque (18EL76) {AGL}
Muharram Shah (18EL03)
Ghazanfar Ali (18EL106) Supervisor
Prof.Abdul Sattar Saand
 Presnentation Outline
Introduction

Literature Review

Problem Statement

Research Objectives

Research Methodology

Possible Outcomes

References
 Introduction

In this project, smart energy meter will be made using

IoT(Internet of things).
We measures the voltages and current through sensors and
calculate the power consumption and electricity bill, and
also detect the power theft.
Sensor’s data will be transferred to Wi-Fi module-esp32
and this Wi-Fi module will send data to server which will
monitor from every where.
Wi-Fi module-esp32 will be programmed through different
logics.
System functionality based on programming.
 Literature Review
Title Author Year of Remarks
publication
A Novel Energy Theft W. Li, T. Logenthiran, 2016 The author developed an energy theft detection system based
System for Iot Based V.-T. Phan and W. L. on recognition framework dependent on AI idea and factual
Smart Home Woo idea. The main aim is to consolidate different AI methods into
a framework called forecast to predict the amount of energy
used. The following module is the essential dynamic model
that utilizes Simple Moving Average for screening the
framework dependent on the vitality level utilized. The last
module is the optional dynamic model that is utilized for
recognizing the vitality robbery.

Development of Smart W. Li, T. Logenthiran, 2017
Energy Meter Reading V.-T. Phan and W. L.
and Monitoring System Woo
The author has proposed a system capable of monitoring the
system at a certain interval of time and alerting the provider
and customer of how many units consumed by the user. The
power consumptions are monitored automatically with the
help of server.
 Literature Review
Title Author Year of Remarks
publication
Energy Theft Issues for W. Li, T. Logenthiran, 2016 The author described Energy theft as one of the serious issue
Advanced-Metering W. Woo, V. Phan and implemented using smart grid. It is found that the country may
Infrastructure in Smart D. Srinivasan suffer from lot of energy theft. It is estimated that many
Grid companies have faced billions of loss during the energy theft.
To overcome this Automated Metered Infrastructure is used.
Here a tree based model is used. The behavior of automated
metered reading is monitored regularly. The infrastructure is
classified into three different categories.
Smart Home Energy C. Yang, J. Yao, W. 2017 The author built up a energy theft identification framework
Management System Lou and S. Xie dependent on AI idea and factual idea. The dynamic modules
Using Fuzzy Logic for can be arranged into three kinds, the first module is the
Continuous Power multimodal gauging framework which is called expectation.
Supply with Economic The principle point is to join different AI procedures into a
Utilization of Electrical solitary anticipating framework to foresee the measure of
Energy power utilized. The following module is the essential dynamic
model that utilizes Simple Moving Average (SMA) for
filtering the framework dependent on the energy level utilized.
The last module is the auxiliary dynamic model that is utilized
for identifying the power theft.
 Literature Review
Title Author Year of Remarks
publication
A Machine Learning F. L. Quilumba, W.-J. 2015 The creator showed the effectiveness of approach with a total
Decision Support Lee, H. Huang, D. Y. Bayesian Network expectation model and contrast and three
System Improves the Wang and R. L. AI forecast model classifiers: Naive Bayes, Random Forest
Internet of Things Szabados and Decision Tree. Results exhibit that these methodologies
Smart Meter Operations produces factually forecast and that the Decision Support
System will improve the cost effectiveness of the system and
the system upkeep.

Using Smart Meter T.-C. Chiu, Y.-Y. Shih, 2017 The creator tended to the endeavors associated with
Data to Improve the A.-C. Pang and C.-W. improving the framework level intraday load anticipating by
Accuracy of Intraday Pai applying bunching to recognize gatherings of clients with
Load Forecasting comparative burden utilization designs from shrewd meters
Considering Customer before performing load gauging. Not at all like conventional
Behavior Similarities accumulated framework level burden gauging, the AMI
information acquaints a crisp point of view with the manner in
which burden anticipating is performed, going from transient
burden determining to long haul load estimating at the
framework level, territorial level, feeder level, or even down
to the shopper level.
Problem Statement: Energy Distribution Transmission Energy Load Peak
Year Sale Losses Losses Sent out Factor Demand

In our power system losses are the (GWh) G.R (GWh) (%) (GWh) (%) (GWh) (%) (MW)

main issue, especially at 2013-14 3684 1162 23.2 167 3.33 5014 64 892

distribution network where's the 2014-15 3845 4.3 1187 22.8 172 3.31 5204 64 928

high power losses (19 to 23% 2015-16 4012 4.4 1213 22.5 177 3.28 5402 64 965

according to HESCO report shown 2016-17 4187 4.4 1239 22.1 182 3.25 5608 64 1004

in fig#1). These loses exist due to 2017-18 4371 4.4 1264 21.7 188 3.22 5822 64 1044

lack of proper maintenance and 2018-19 4563 4.4 1289 21.3 193 3.19 6045 64 1087

element of theft. Theft is possible 2019-20 4764 4.4 1315 20.9 199 3.17 6277 63 1130

2020-21 4975 4.4 1340 20.6 205 3.14 6519 63 1176

through two ways (one is
2021-22 5195 4.4 1365 20.2 211 3.11 6770 63 1224
consumer direct hooking of phase 2022-23 5426 4.4 1389 19.8 217 3.08 7031 63 1273
through distributor and second one 2023-24 5668 4.5 1413 19.3 223 3.05 7303 63 1325
is tempering in energy meter). Ave. Growth
(2014-2024)
Theft is main issue and we want 4.40% 3.83% 4.04%

to overcome as early as possible. Fig#1 HESCO report

 Research Objectives
The proposed system is designed based on the fuzzy logic.

The system is capable to measure the parameters voltages and current.

The system is capable of continuously monitoring and notifying number of units consumed by the
customer.

The energy consumption are automatically calculated and the bill is generated using Internet of
Things.

The ultimate goal is to consume the energy If illegal alteration occurs, mobile alert is sent to the
customers.

System also protect the home appliances during variation of voltages.

 Research Methodology

Detailed study about working process of software and hardware model.

Detailed study about ESP-32 Wi-Fi module and various elements of project.

Working on code of Arduino.

Working on response of sensors.

Design and Testing hardware model and troubleshooting errors.

.
Performance analysis of hardware model.

Results and observations about hardware model.

Working Principle
 Possible Outcomes
To reduce wastage of energy.

Prevent electricity shortage during dry seasons.

Make every customer a self-interested guardian of the power (energy) supply.

Real time bill monitoring.

Time reduced receiving bill.

Reduce the theft element from system.

 References
1.W. Li, T. Logenthiran, V.-T. Phan and W. L. Woo, “A novel energy theft system for iot based smart home", Region
10 Conference (TENCON) 2016 IEEE, pp. 3386-3389, 2016

2.C. Yang, J. Yao, W. Lou and S. Xi “development of smart energy meter reading and monitoring
system", Transactions on Environment and Electrical Engineering, vol. 2, no. 2, pp. 64-71, 2017.

3.W. Li, T. Logenthiran, W. Woo, V. Phan and D. Srinivasan, “energy theft issues for advanced-metering
infrastructure in smart grid", IEEE World Congress on Computational Intelligence, pp. 1-8, 2016.

4.C. Yang, J. Yao, W. Lou and S. Xie, “smart home energy management system using fuzzy logic for continuous
power supply with economic utilization of electrical energy", IEEE Internet of Things Journal, 2017.

5.F. L. Quilumba, W.-J. Lee, H. Huang, D. Y. Wang and R. L. Szabados, “A machine learning decision support
system improves the internet of things smart meter operations", IEEE Transactions on Smart Grid, vol. 6, no. 2, pp.
911-918, 2015
6.T.-C. Chiu, Y.-Y. Shih, A.-C. Pang and C.-W. Pai, “using smart meter data to improve the accuracy of intraday load
forecasting considering customer behavior similarities", IEEE Internet of Things Journal, vol. 4, no. 2, pp. 374-383,
2017