ESTIMATING ENERGY CONSUMPTION BASED ON
SMART ENERGY METER USING WIRELESS NETWORK
K.Deepa.
St.Joseph College Of Engineering,Chennai119
ABSTRACT
This paper presents a Smart power metering system or issue is captivated by countless profits. In
India and other countries prove that smart metering is technically practicable. Main issues are the
real value of the payback, the outlay involved, the distribution of overheads and gross settlement of
smart metering between markets parties involved. The fundamental necessities of human beings is
electricity which is commonly used for domestic, industrial and agricultural purposes. Accurate
prediction/forecasts of energy demands(load) is crucials. The progress in technology about electrical
distribution network is a non-stop process. To reduce the power consumption in a house or at
individual site remains to be a difficult problem. This is done using Smart Energy Meter (SME) , a
common form of smart grid technology, are digital meters that replace the old analog meters used in
homes to record electrical usage. SEM is an electric device having energy meter chip for measuring
the electrical energy consumed and a wireless protocol for communication. It presents a automatic
metering and billing system. This meter can work as either post-paid or pre-paid. It avoids the human
intervention, provides accurate billing and minimize the power consumption in a house. In EB Server
Section, Easily we will monitor the home section data and control the load using Smart Metering
system we can avoid Wrong Power Usages and etc. we can view the values remotely using the IOT in
the web server and can be controlled.
I. INTRODUCTION:
Internet of Things (IoT) is an
interconnection of several devices,
networks, technologies and human
resources to achieve a common goal.
There is a variety of IoT based
applications that are being used in
different sectors and have succeeded in
providing huge benefits to the user. Data
Analytics has a significant role to play
in the growth and success of IoT
applications and investments. Electricity
has now become a part of our daily life
and one cannot think of a world without
electricity. It is an important part of
homes & industries. Almost all the
devices at homes, businesses and
industries are running because of
electricity .
In developing countries like India,
power theft is one of the most prevalent

issues which not only cause economic
losses but also irregular supply of
electricity. So, it has become important
for the government to manage
electricity and estimate it properly.
With the evolution of IoT, the
conventional use of energy meter can be
made as smart energy.
A smart meter is a new kind of
electricity meter that can digitally send
meter readings to your energy supplier
for more accurate energy bills. Smart
meters come with in home displays, so
consumer can better understand their
energy usage on daily basis. The energy
consumption will be daily monitored
and on comparison with historical datas,
the energy theft is identified. The
proposed system will reduce the human
intervention. Experiments are carried
out using available datasets and results
are expected to a increase in accuracy
when compared to other methodologies.
This also helps provider and consumer
to have a transparent and balanced
energy billing system.This methodology
can be extended further

II. EXISTING SYSTEM :
In Existing System, the possibility to
use the network of SMs as a sensors
network for the grid monitoring has
been explored and validated through a
dedicated experimental set-up. In fact,
SMs of the next generation, compared to
devices already installed in some
countries provide several parameters,
like active and reactive power, line
frequency, voltage dips and total
harmonic distortion, which can be used
by DSO to monitor the status of the
network. A network composed of 48
SMs has been deployed over the LV
grid of A2A in the city of Brescia, Italy.
Each SM is connected to the MDC by
means of a performing broadband power
line communication network. The
results of the monitoring, performed
over 2 months, highlight the potential
capabilities of a large scale monitoring
system based on the use of a network of
second generation SMs. Using the SM
network has been verified that the
voltage is below the 5 % of the nominal
value only the 3 % of the time in section
of the distribution grid under analysis,
despite the large presence of distributed
energy resources. In addition, the SMs

network identifies inversion of the
energy flow in part of the distribution
grid due to an excess of PV energy
production compared to customers
consumption. we can view the values
remotely using the IOT in the web
server and can be controlled.

III. PROBLEM STATEMENT:
The Electricity Board (EB) have got
used to manual process. Going to each
and every house and generating the bill
is a laborious task and requires lot of
time(Fig 1). If the consumer is not
available , the billing process will be
pending and human operator again
needs to revisit. It becomes difficult
especially in rainy season. If the
consumer did not pay the bill, the
operator needs to go their houses to
disconnect the power supply. This is
time consuming and difficult to handle.
To avoid human intervention in billing
process, automatic billing is done to
reduce the manpower and power
consumption(Fig 2).

Fig.1. Monthly (4-week) energy consumption plots of Fig.2. Daily energy consumption plots of
Three normal and abnormal household electricity use households

Fig 3: Piecewise Regression Model Fig 4:Comparison Of ARMA&ARIMA:

IV. LITERATURE SURVEY:
Kasun Amarasinghe., Daniel L.
Marino and Milos
Manic1Investigates the
effectiveness of using Convolutional
NeuralNetworks (CNN) for
performing energy load forecasting
at individual building level.
Methodology uses convolutions on
historical loads. Weather data to find
the effectiveness of deep learning
algorithms in load forecasting.
Yi Wang, Qixin Chen., Tao Hong
and Chongqing Kang2 It
conducts an application-oriented
review of smart meter data analytics.
It follows three stages of analytics,
namely, descriptive, predictive and
prescriptive analytics. To provide a
complete picture and deep insights
into this area of big data issue,
developments of machine learning,
novel business model, energy
system transition, and data privacy
and Security.
Ramyar Rashed Mohassel., Alan
Fung., Farah Mohammadi and
Kaamran Raahemifar3Advanced
Metering Infrastructure provides a
needs improvement in the areas of
communication, data analysis and
control schemes. Smart home
system with RaspberryPi and
NodeMCU as the backend that not
only serves as home automation and
merely a switch replacement, but to
also record and report important
things to the owner of the house.
Further smart home development
with better security and more
functions.V.
Preethi and G. Harish4 SME
helps user in identifying the usages
between authorized and
unauthorized users which helps in
controlling the power theft.
Communication between
user/household and substation is
done using Zigbee. GSM network is
used for sending SMS to the local
authorities.
Yasirli Amri and Mukhammad
Andri Setiawan5 Used to
calculate the power that has been
used by electronic devices so that

homeowners can save money in
electricity bills. This greatly helps
the homeowner to monitor the house
even when he is not at home. Smart
home system with RaspberryPi and
NodeMCU as the backend that not
only serves as home automation and
merely a switch replacement, but to
also record and report important
things to the owner of the house.
Further smart home development
with better security and more
functions.
Maninderpal Singh and Er.Varun
Sanduja6IOT technology the
government person can find the
dishonest user, it can make the
assignment of the agents
impracticable to steal the electricity,
the power theft. This analysis work
has been implemented to find the
dishonest use. Energy meter
communicate with raspberry pi
through GPIO pins. GPIO pins fetch
the effective data from energy meter
and it send effective data to the
raspberry pi, then connect wifi
module with raspberry pi. The
implementation smart meter
automatically cut electricity when
any one tried to theft and it also
monitor the electricity consumption
through smart phone and smart
meter that sends status if any fault
occurred in transmission line.
Darshan Iyer N and Dr. K A
Radhakrishna Rao7 Ease of
accessing information for consumer
from energy meter through IoT.
Theft detection at consumer end in
real time. LCD displays energy
consumption units and temperature.
Disconnection of service from
remote server. IoT and PLC based
meter reading system is designed to
continuously monitor the meter
reading and service provider can
disconnect the power source
whenever the customer does not pay
the monthly bill and also it
eliminates the human involvement,
delivers effective meter reading,
prevent the billing mistake. IoT
energy meter consumption is
accessed using Wi-Fi and it will help
consumers to avoid unwanted use of

electricity. The performance of the
system can be enhanced by
connecting all household electrical
appliances to IoT.
Pooja D Talwar , Prof. S B
Kulkarni8-To generate bill
automatically by checking the
electricity unit’s consumption in a
house and in a way to reduce the
manual labor. The calculations are
performed automatically and the bill
is updated on the internet by using a
network of Internet of Things. The
bill amount can be checked by the
owner anywhere globally. Wifi
ESP8266 is a low cost chip and
microcontroller. Displaying the
information about the energy
consumed in terms of units, about
the bill and if any theft occurs that
will be displayed in the website. The
main improvement for the future is
going to make energy meter
readings, tampering identification
techniques, and connection and
disconnection and also the pre
information providing to the users
all is going to happen on wifi
internet.
Mr. Rakeshkumar D. Modi , Mr.
Rakesh P. Sukhadia9 -The design
can be eliminate the man power
involvement to maintain the
electricity. The consumers of
electricity need to pay as per the
utilization of electricity on schedule,
somehow consumers fail to pay, the
transmission of electricity can be
turned off from the distant server
automatically. Energy meter
provides provision to the consumers
that they can monitor the energy
consumption in units by using web
page. The smart electricity energy
meter consists Microcontroller, LCD
display, theft detection unit,
temperature sensor, PLC (Power
Line Communication)modem and
Wi-Fi module. Theft detection and
supplier can disconnect service to
the consumers in the event of meter
tempering or un authorized use of
electricity. It eliminates the man
power required for meter reading.

Manasi Giridhar and
S.Kayalvizhi10 Monitoring the
systems in smart grid and collect the
information of electricity uses then
establish communication with the
consumers which can be useful for
providers as well as consumer by
using message queue telemetry
transport protocol. In additionally it
provides real time pricing and
monitored usage information to the
consumers. Current and voltage
sensors are used for measuring the
power consumption .Those readings
are sent to the consumer as well as
Electricity Board by using ZigBee
and message Queue Telemetry
Transport (MQTT) protocol. MQTT
is a network protocol which is used
transfer data between publisher and
subscriber. It is a publisher and
subscriber protocol. IoT energy
meter consumption is accessed using
Wi-Fi and it will help consumers to
avoid unwanted use of electricity.
The performance of the system can
be enhanced by connecting all
household electrical appliances to
IoT.
Ning Lu.,Pengwei Du Xinxin Guo
and Frank L. Greitzer11
Defining each individual load profile
.In home energy management
systems (HEMS), daily energy
consumption patterns can be an
important variable to monitoring and
triggering customer actions. A 15-
minute meter and weather data set
collected by researchers at Pacific
Northwest National Laboratory
(PNNL) . For customers, however,
daily monitoring for security and
energy consumption(Fig 2) is better
than monthly monitoring because
customers have exclusive
information about their electricity
usage. Focuses on studying the data
correlation among multiple data
sources from both the distribution
and transmission power grids.
Energy consumption (Fig 1) varies
daily due to weather, occupancy,
and different customer consumption
patterns, daily information alone is
of little value to a utility unless an
abnormality in monthly energy
consumption is detected.

Xiufeng Liu., Lukasz Golab and
Ihab F. Ilyas12 SMAS, our smart
meter analytics system, which
demonstrates the actionable insight
that consumers and utilities can
obtain from smart meter data .
SMAS, our smart meter analytics
system, which demonstrates the
actionable insight that consumers
and utilities can obtain from smart
meter data . The slope (Fig 3) of the
90th percentile line corresponding to
high temperature is the cooling
gradient, and the slope of the line
corresponding to low temperature is
the heating gradient. Furthermore,
the height of the 10th percentile
lines at their lowest point is the base
load, which corresponds to load due
to appliances that are always on,
such as a refrigerator.
Savita Pawar ,Dr. B. F.
Momin13The smart meters main
functionality is measuring, capturing
and transforming data (Table 1)
related to usage or consumption of
electricity, gas or water and events
such as meter status and power
quality . Smart meters and smart
grid will be the dominant part of
Internet of Thing (IoT) which
integrate various views of peoples’
requirements and services to satisfy
them and therefore many analytical
challenges arises such as real time
analytics.
Dr.P.Mathiyalagan.,Ms.A.Shanmu
gapriya.,Geethu.A.V14Electricity
consumption data of a smart meter
are used at a sampling rate of one
minute. The streaming data is loaded
into HDFS in a hive table, which is
further exported into R in order to
perform predictive analysis and load
profile analysis. The data are
aggregated into daily, weekly,
monthly and quarterly series, and
used ARMA and ARIMA (Fig
4)model to predict the future
electricity consumption.
Arunesh Kumar Singh.,
Ibraheem., S. Khatoon., Md.
Muazzam and D. K.
Chaturvedi15 it can be inferred
that demand forecasting techniques
based on soft computing methods

are gaining major advantages for
effective use. There is also a clear
move towards hybrid methods,
which combine two or more
techniques such as regression,
multiple regression, exponential
smoothing, iterative reweighted
leastsquares, adaptive load
forecasting, stochastic time
seriesautoregressive, ARMA model,
ARIMA model etc. This paper
presents a review of electricity
demand forecasting techniques.

Table 1 :Techniques and tools and its Application in Smart Meter Data
Analytics

Techniques and Tools Applications
SOM Clustering load curve , Load forecasting,
Load Profiling
SVM Electricity theft detection and Appliance
type recognition.
FL Intelligent support system, Automated
decision making.
BN and HMM Appliance identification, load
disaggregation, supply demand analysis.

V. RESEARCH METHODOLOGY
The proposed system has
two section namely, Home Section
and Electrical Base Station. The
communication between these two is
done by wireless network. It
monitors the load and calculates the
power consumed exactly by the user
at a given time. Energy utilized and
the corresponding current, voltage,
power and amount will be displayed
on LCD continuously and
communicated to the base station.
An SMS containing monthly bill
along with due date is sent to
respective meter owner using GSM.
IOT Gateway Device with wifi is
used to transit the power consumed
to the Electricity Board website. In
case of bill not paid , the automatic
power cut is done through Relay

Driver Circuit. The system can act
as either pre- paid or post-paid.In
pre-paid mode , the power
consumption(in watts) with respect
to time that corresponding amount is
deducted from the total amount and
is displayed on the LCD
continuously. If the recharged
amount reduces to half of the total
amount a buzzer is activated and
continuously alerts the consumer.In
post-paid mode, it simply displays
the power consumption and the
corresponding amount on LCD
continuously.

Fig 5: Implementation diagram

VI. CONCLUSION:
The Electrical distribution network is a
non-stop process in today technology.
An advanced metering road and rail
network put forward the leeway for
auxiliary energy allied services such as
demand side management and
consciousness of virtual power plants.
The potential of smart metering relies
profoundly on the policy and
decisiveness of the legislative bodies
mixed up. Energy savings and an
improved security of supply are the
major drivers and deems in smart
metering as huge targets of a nation.
The Smart metering system will
monitor the consumed power in
particular home and transmitted via
IOT Gateway Device. It avoids human
intervention in billing process , Lack of
details in meter cards. Delay in meter
reading by assessors, Failure to upload
meter readings on main server even if
done promptly and assessors wrongly
notifying the date of assessment in the
meter cards
VII. REFERENCE:
1. Kasun Amarasinghe, Milos
Manic” Deep Neural Networks for
Energy Load Forecasting”, : 2017
POWER
SUPPLY
(SMPS)
ENERGY
METER
MICRO
CONTRO
LLER
E
RELA
Y
IOT
GATEWAY
DEVICE
SMS
LCD
DISPL
GSM
NETWORK
(COMMUNIC
ATION
BASE
MOBI

IEEE 26th International Symposium
on Industrial Electronics (ISIE).
2. Yi Wang , Qixin Chen
“Review of Smart Meter Data
Analytics: applications,
Methodologies, and
Challenges”,:IEEETransactions on
Smart Grid ( Early Access ) in 22nd
march 2018.
3. Ramyar Rashed Mohassel, Alan S.
Fung”A survey on AdvancedMetering
Infrastructure”, Published in: 2014
IEEE 27th Canadian Conference on
Electrical and Computer Engineering
(CCECE).
4. Mofeed Turky Rashid” Design and
Implementation of Smart Electrical
Power Meter System”,Iraq J. Electrical
and Electronic Engineering Vol.10
No.1 , 2014.
5. Yasirli Amri and Mukhammad
Andri Setiawan”Improving Smart
Home Concept with the Internet of
Things Concept Using RaspberryPi
and NodeMCU”, Published under
licence by IOP Publishing Ltd IOP
Conference Series: Materials Science
and Engineering,2018.
6. Vani.H, S. M. Varun Kumar,
Chetan Sastry,
Shrinidhi.W,”Minimizing Electricity
Theft by Internet of Things,
International Journal of Latest
Technology in Engineering,
Management & Applied Science
(IJLTEMAS), Volume VII, Issue V,
May 2018
7. Darshan Iyer N, Dr. K A
Radhakrishna Rao”IoT Based
Electricity Energy Meter Reading,
Theft Detection and Disconnection
using PLC modem and Power
optimization”, International Journal of
Advanced Research in
Electrical,Electronics and
Instrumentation Engineering(An ISO
3297: 2007 Certified Organization)
Issue 7, July 2015.
8. Pooja D Talwar “Iot based
energy meter reading”,
International Journal of Recent
Trends in Engineering & Research
(IJRTER)Volume 02, Issue 06;
June – 2016
9. Mr. Rakeshkumar D. Modi
“A review on iot based
smart electricity energy meter”,
International Journal For
Technological Research In
Engineering Volume 4, Issue 1,
September-2016.
10. Arati Kurde “A review on iot
based smart electricity energy
meter”, National Conference on
“Internet of Things: Towards a
Smart Future” & “Recent Trends in
Electronics&Communication”(IOT
TSF-2016).11. Ning Lu” Smart
Meter Data Analysis” Conference:
Conference: Transmission and
Distribution Conference and
Exposition (T&D), 2012 IEEE
PES.
12. Xiufeng Liu “SMAS: A
Smart Meter Data Analytics
System “Conference: IICDE,in
april 2014.
13. Savita Pawar ; B. F. Momin
“Smart Electricity Meter Data
Analytics: A Brief Review
Published 2017 in 2017 IEEE
Region 10 Symposium
(TENSYMP).
14. P. Mathiyalagan ; A.
Shanmugapriya ; A. V. Geethu ”
Smart Meter Data Analytics
using R and Hadoop”, : 2017 IEEE
International Conference on (EIT).