SHUSHIL LAL DAS - Final Project Report - Shushil Lal Das
SHUSHIL LAL DAS - Final Project Report - Shushil Lal Das
SHUSHIL LAL DAS - Final Project Report - Shushil Lal Das
Submitted by
(1613108003 / 16SCSE108003)
of
BACHELOR OF TECHNOLOGY
IN
Professor
i
SCHOOL OF COMPUTING AND SCIENCE AND ENGINEERING
BONAFIDE CERTIFICATE
Date: 18/05/2020
This is to certify that the dissertation entitled “SMART METERING USING IOT
work. It is further certified that this work has not been submitted for similar purpose
anywhere else. His work has been found satisfactory for the partial fulfillment of the
I
DECLARATION
I here by declare that the project work entitled “SMART METERING USING IOT
and AI” is an authenticated work carried out by SHUSHIL LAL DAS. Under the
guidance of Mr. HIMANSHU SHARMA for the partial Fulfillment of the award of
the degree of B.TECH and this work has not been submitted for similar purpose
II
ACKNOWLEDGEMENT
It is high privilege for me to express my deep sense of gratitude to all those faculty
members who helped me in the completion of the project, especially my internal guide
Mr. Himanshu Sharma who was always there at hour of need and whose careful
My special thanks Ms. Garima Pandey and Dr. Santar Pal Singh (Panel Member),
Galgotias University, for helping me in the completion of project work and its report
submission.
Signature:
Date: 15/05/2020
III
ABSTRACT
Keywords: Smart Energy Meter, Wi-Fi Module, Power Theft, Automatic Billing, Real
Time Data Monitoring, GSM,
IV
TABLE OF CONTENTS
Certification……………………………………………………………………….. i
Declaration…………………………………………………………………………..ii
Acknowledgement…………………………………………………………………..iii
Abstract……………………………………………………………………………...iv
Table of Contents……………………………………………………………………vi
Abbreviations……………………………………………………………..................vii
List of Figures…………………………………………………………….................viii
List of Tables………………………………………………………………………...ix
3.1 Connectivity…………………………………………………………. 9
3.2 Control………………………………………………………………. 10
3.3 Data Analysis……………………………………………………….. 11
3.4 Security……………………………………………………………… 12
4.1 NodeMCU/ESP8266………………………………………………… 14
V
4.4 Edge Processing………………………………………………………. 16
4.5 MQTT IoT Protocol………………………………………………….. 17
4.6 AWS IoT Broker……………………………………………………… 18
4.7 AWS Lambda………………………………………………………… 20
4.8 Amazon Kinesis………………………………………………………. 21
CHAPTER:6 OUTPUT
CHAPTER:7 TESTING
CHAPTER:9 CONCLUSION
CHAPTER:10 REFRENCES
VI
ABBREVIATIONS
VII
LIST OF FIGURES
VIII
LIST OF TABLES
IX
CHAPTER-1
INTRODUCTION
1
memory location. This system can also be used to disconnect the power supply of home
when needed.
This paper mainly deals with smart energy meter, which utilizes features of the
embedded system i.e. a combination of hardware and software. The paper discusses
how and what type of work is done by IOT based smart energy meter. Also with the
help of Wi-Fi Modem the consumer can monitor its consumed reading and can set
threshold value through the webpage. If the consumer is not aware of threshold
notification, then meter gets off automatically after that consumer can increment the
threshold value and meter will automatically turn on.
Finally, the overall monthly bill with amount will be sent to the consumer as well as a
service provider of text at first day of every month.
Smart meters are being incorporated into the grid to record and upload electrical and
background data, with specially designed sensors and IoT-enabled devices. They are
deployed in every electricity consumption unit and share their information with local
control centers.It is an electronic device that records consumption of electric energy in
intervals of an hour or less and communicates that information at least daily back to the
utility for monitoring and billing.
Smart meters not only have two-way communication, but are equipped with real-time
sensors that can gather the data on relevant factors, including frequencies used by
different equipment and appliances.Traditional meters only measure the total
consumption, providing no further breakdown of information; smart meters on the other
hand measure site specific information, allowing utilities to introduce different prices
for consumption based on usage during the time of day and according to the season.
Proponents believe that billing customers at a higher rate for peak times will encourage
consumers to adjust their consumption habits to be more responsive to market prices
and could delay the construction of additional generation, thereby controlling sharp rise
in electricity prices. There are some concerns, however, that low income and vulnerable
consumers may not benefit from intraday time-of-use tariffs; besides the inherent ability
2
of smart meters to provide two-way communication and data readings remotely could
result in large layoffs of meter readers. The smart metering business case is broad and
complex, as the technology has the potential to impact the entire electricity system,
from generation investment and dispatch, through network optimisation, all the way to
retail operations and beyond into the home. The most commonly pursued benefits,
however, have tended to be focused on the retail area, particularly the core areas of
meter reads and consumer service support. While the benefits are becoming well
characterized in the retail area, it is clear that many of the potential benefits from
distribution optimisation and capital efficiencies are commonly discounted or ignored.
“The benefit of any smart metering to a utility is that the utility gets a better view of
the customer’s usage of electricity by the customer. While, the benefit to a customer is
that a customer can also get a better view of his usage, through the customer portal and
thereby control the usage of electricity. The customer and the utility can work closely
to implement multiple programs such as time of day billing, demand management
programs etc.,
3
the readings. There is no protection for energy meter tampering. The consumers cannot
monitor the everyday energy consumption or usage. The major drawback of this system
is the management of power consumption is difficult. Traditional meters only measure
the total consumption, providing no further breakdown of information.
1.4 Objectives
The main objective of this project is to design and build a functional Smart Meter
through software implementation, which is to be overlaid on the existing traditional
perimeter infrastructure.
➢ To analyze and control the electrical devices at whatever time from any part of
the world
➢ To enable two-way communication between the meter and the central system
➢ To analyze the data incoming from numerous IoT enriched electric meters
4
➢ To enable the consumer to pay its electricity bills through online payment
system
➢ A service provider can engage its consumer through some real time usage
statistics, comparisons and tips for wise utilization through the consumer portal
In this modern period of Internet of Things, a physical world such as machines and
sensors can be interfaced through the web.The things or items can be modified into
smart things by providing it exclusive identification.
All in all,the primary scope is to implement IoT and AI based Smart meter that not only
enhance the ease with which electricity bills are generated and transmitted, they also
provide better control on one’s electricity consumption by providing real time
consumption data. The consumer can monitor his/her electricity usage and in turn
optimize his/her consumption.
A smart metering using IoT and AI also offers insights into the expenditure on
electricity which proves to be a great incentive for consumers to save electricity and in
turn reduce the pressure on the constantly depleting non-renewable sources of energy.
A network of all the smart metering using IoT and AI connected to a smart grid will
positively influence the methods and modes of electricity generation, transmission and
distribution.
The aim to allow power companies that could only guess what kind of usage was
happening in a particular area, based on the data from the sub-stations supplying power
to that section of the grid to know all the data of a particular household too.
5
CHAPTER-2
LITERATURE SURVEY
According to a paper titled "ARM-based energy management system using smart meter
and Web server"[7],prepared by "carmine Landi, Pietro Merola, Giacomo Lanniello"in
2011,a low cost ARM-based energy management system was designed. It is a part of
distributed system that measures the main power system quantities and collect the
statistics of power consumption and power quality which is able to interface devices
6
for load movement. The device easily access the information and local access the
combination of a smart meter and data communication. Similarly, according to the
paper titled ‘An Approach to Automate Power Meter Reading & Billing
System’(2013)[8], it proposed a solution to implement Automated Power Meter
Reading & Billing System for bringing smart governance in power/energy supply
departments by the use of GPRS/IP, Server based remote meter reading and billing
algorithms to automate the reading and billing process, nTier Architectures based
Model that approach provides real time energy which leaded to more throughput in
power/energy management.
In 2014[9], an approach to implement smart meter using image processing was made.
It described a prototype for Automatic Meter Reading (AMR) system that uses a
Traditional meter, Zig-Bee modules, and a serial camera unit. This camera will take the
photo of meter reading and transmit it to the server PC through Zig-Bee, where that
image undergoes segmentation, recognition process and reading get separated which
would be further used for preparing bill. While in 2015[10], the GSM and Zibee based
Automatic Energy Meter Reading System with Instant Billing was implemented using
Cortex-M3 LPC1768 as important processor to do communication in short distance and
SIM900 to achieve communication function in long distance, using RS-232 link
communication joint to connect the communication between Zigbee and GSM
technology. This system has many significant excellences , such as wireless, low cost,
a little power consumption, great quantity of data transmission , while it has great
extension & security.
With the paper published in 2016 titled ‘Privacy-friendly Forecasting for the Smart Grid
using Homomorphic Encryption and the Group Method of Data Handling’[11], it
concerned with enhancing the privacy of the smart meter readings in the setting of
forecast prediction. In this paper , the author showed the method to compute the forecast
prediction such that the supplier does not learn any individual consumer usage
information by using the Fan-Vercauteren some what homomorphic encryption
scheme, typical prediction algorithms based on artificial neural networks that require
the computation of an activation function, show that Ivakhnenko’s group method of
data handling would be suitable for homomorphic computation.
7
On September 2017, the authors “Mrs Sandhya Shinde, Mr.Yogesh Yadav,Miss. Bharti
Sontakke,Miss. Pratiksha Zapake” presented a paper titled “IoT Based Smart Energy
Meter”[12]in which smart meter was designed based on a Arduino and implementation
of energy meter using IoT concept.The Arduino checks the main meter and sub meter
reading to check the theft status. It provides the consumers to track their energy
consumption. It allows the supplier to take control over the consumers and can
disconnect the connection if a consumer fails to pay bill.
In 2018, the authors named “Amrita Singh, Ravi Gupta” presented a paper titled “Iot
based smart energy meter”[13] in which the traditional meter was replaced by a
metering module which consists metering IC and micro controller that scans the energy
meter automatically after every month and transmit to the consumer and propose
service provider system. The meter gets off automatically if the threshold value of
consumption provided by consumers on their webpage is crossed.
In 2018, a research article was published by the authors “R.Asha, R.Aruna, J.Divya,
K.Balasaranya” titled “Smart Energy Meter for Advanced Metering and Billing Alert
Framework”[14] in which Evolutionary Algorithms(EAs) (Binary Particle Swarm
Optimization (BPSO), Genetic Algorithm (GA) and Cuckoo search) based DSM model
for scheduling the appliances of residential users was presented to optimally consume
grid and renewable energy sources. It was purposed significantly to reduce the
electricity bill and high peaks.
8
CHAPTER-3
PROBLEM STATEMENT
Traditionally, the electricity meters are installed on consumer’s premises and the
consumption information is collected by meter-readers on their fortnightly or monthly
visits to the premises. This method of gauging electricity consumption has the following
disadvantages:
3.1 Connectivity
In India, the current electricity billing system is completely manual. The electric meters
are situated in the houses, offices and factories etc. The energy meter reading is
collected by meter readers on their fortnightly or monthly visits to the premises. This
system has disadvantage of appointing meter reader to take the meter reading, effects
consumer privacy etc. In this direction this project undertakes the meter reading without
human intervention.
3.2 Control
9
bill is used at time when extreme weather conditions occur and meters to be read are
not easily accessible to the reading -so it is problematic for consumer and supplier.
Therefore, we don’t have any control over the usage of total electricity as the bill is
generated at the end of the month and the consumer don’t have data about power
consumption.Suppose when you consume 200 unit but you have bill payment of 500
unit . Since due to lack of data and control ,the consumer cannot impose any action
against this type of fraud. Since, smart metering helps to know and control your
electricity by the application to check your current data and bill.There is evidence of
your power consumption.
While smart meters helps in data storage, stores electricity consumed every half of an
hour, data is automatically transmitted to the metering company, digital data of energy
consumption and TOU are provided in near real time, Automated outage detection
enable distribution companies to restore power quicker than traditional electricity
metering, connections and disconnections are faster, because they are managed
remotely. Traditional electricity meters only show the quantity of electricity consumed
since meter reader last read the meter.
Therefore, in traditional electric meter we only get the data of the of the consumer on
the monthly basis and cannot get the data based on the daily usage of the consumer.
10
3.4 Security
Fig 1.1- Difference in Traditional Metering System and Smart Metering System
most used method. 80% of global power theft is by direct tapping from the line. The
consumer taps into a power line from a point ahead of the energy meter. This energy
consumption is unmeasured and obtained with or without switches.
The various other methods of electrical power theft are:
3.4.1 Bypassing the energy meter
In this method the input terminal and output terminal of energy meter is short circuited
preventing the energy from registration in energy meter.
3.4.2 Injecting foreign element into the energy meter
Meters are manipulated via remote by installing a circuit inside the meter so that the
meter can be slowed down anytime. This kind of modification can avoid external
inspection attempts because the meter is always correct unless the remote is turned on.
3.4.3 Physical obstruction
This type of interfering is done to electromechanical meters with the rotating element.
Foreign material is place inside the meter to obstruct the free moment of the disc. Lower
rotating desk signals less energy consumption.
11
3.4.5 ESD attack on electronic meter
This type of interfering is done on electronic meter to make either temporary damage
or permanent damage.
12
CHAPTER-4
PURPOSED MODEL
In this project an integrated Internet of Things architecture for smart meter networks
are proposed. It discusses the communication protocol, the data format, the data
gathering procedure, and the decision system based on big data treatment. Real
measurements show the benefits of the proposed IoT architecture for both the customers
and the utilities. In this model, the traditional meter is reused and redesigned which is
embedded with CT, VT, thermositor and other sensor for self performance.
13
4.1 NodeMCU/ESP8266
14
Smart meter using AI and IoT uses GSM module to build connection between
consumers and producers. The consumers get the notifications regarding their power
consumption statistics and different wise tips for controlling energy consumption.
The Kalman filter is an efficient recursive filter that estimates the internal state of a
linear dynamic system from a series of noisy measurements. It is used in real‐time
Global Positioning System (GPS). Kalman filter consists of two distinct processes,
namely, the prediction process and the measurement process. Both processes are
combined and operated in a recursive manner to achieve optimal Kalman filtering
process.
In traditional power system studies, power system state can be estimated by combining
Kalman filters with a hypothetically “true model and available measurements. In the
ideal case, the hypothetical system model has accurate noise statistical characteristics
incorporated into it. However,in reality, noise statistics can never be really known
15
exactly. Apart from errors arising in the modeling phase, the measurements are also
riddled with errors due to device failure or even worse, malicious data attacks. To
facilitate illegal activities like energy theft, attachers construct bad data which evades
bad data detection mechanisms in power systems. This can have devastating effects like
false dispatch in the distribution process and device breakdown during power
generation .
This filter identifies -device failures, unusual disturbances, and malicious data attacks
. Kalman Filter is a dynamic state estimation method which is mainly used in this paper
for noise variation estimation.
The Kalman filter is an advanced type of filter which is used to filter the measurement
noise and provide the optimal estimation of a dynamic system’s state. It is recursive in
nature so that new measurements can be processed as they arrive. Kalman filter
minimizes the MSE (Mean square error) of estimated parameters. An Extended Kalman
Filter,based on Taylor series expansion around a nominal va l u e which is taken as the
previous estimate in this case needs to be designed. The state transition matrix F is
given by the Jacobian vector function ,about state and the noise scaling matrix τ is given
by the Jacobian vector function (,)about state w. Since the process dynamics are
continuous while the measurements are usually discrete in nature, a hybrid continuous-
discrete EKF model is developed. The EKF equations of discrete time cannot be used
directly and thus continuous time EKF equations have to be derived. Also, since the
measurements are discrete in nature, a hybrid of both is developed and described
below(repetitive data).An observable, non-linear dynamical system, with the
continuous process dynamics and discrete measurement of dynamics is explained
by:Here ∈ℜshows the n-dimensional state vector of the system, f (.) Dx →ℜn is a finite
non-linear mapping of system states to system inputs, z∈⊂ℜdenotes the p-dimensional
system measurement, h(.) :⊂ℜ→ℜis a non-linear mapping of system states to output,
∈ℜdenotes the continuous process noise scaling matrix ,∈⊂ℜdenotes the w-
dimensional random process noise and ∈⊂ℜdenotes the v-dimensional random
measurement noise.
16
Edge processing refers to the execution of aggregation, data manipulation, bandwidth
reduction and other logic directly on an IoT sensor or device. The idea is to put basic
computation as close as possible to the physical system, making the IoT device as
“smart” as possible.
The functions of edge processing used in Smart Metering using AI and IoT:
17
Fig 1.5- MQTT Broker
The AWS IoT message broker implementation is based on MQTT version 3.1.1, but it
deviates from the specification as follows:
⚫ AWS IoT Core supports MQTT Quality of Service (QoS) levels 0 and 1 only. AWS
IoT Core does not support publishing or subscribing with QoS level 2. When QoS
level 2 is requested, the AWS IoT message broker does not send a PUBACK or
SUBACK.
⚫ In AWS IoT Core, subscribing to a topic with QoS level 0 means a message is
delivered zero or more times. A message might be delivered more than once.
Messages delivered more than once might be sent with a different packet ID. In
these cases, the DUP flag is not set.
⚫ When responding to a connection request, the message broker sends a CONNACK
message. This message contains a flag to indicate if the connection is resuming a
previous session.
18
⚫ When a client subscribes to a topic, there might be a delay between the time the
message broker sends a SUBACK and the time the client starts receiving new
matching messages.
⚫ The MQTT specification provides a provision for the publisher to request that the
broker retain the last message sent to a topic and send it to all future topic
subscribers. AWS IoT Core does not support retained messages. If a request is
made to retain messages, the connection is disconnected.
⚫ The message broker uses the client ID to identify each client. The client ID is
passed in from the client to the message broker as part of the MQTT payload. Two
clients with the same client ID cannot be connected concurrently to the message
broker. When a client connects to the message broker using a client ID that another
client is using, the new client connection is accepted and the previously connected
client is disconnected.
⚫ On rare occasions, the message broker might resend the same logical PUBLISH
message with a different packet ID.
⚫ The message broker does not guarantee the order in which messages and ACK are
received.
The following functions of MQTT protocol are used in Smart Metering using AI and
IoT:
• Pub/Sub Model
19
Fig 1.6- AWS IoT Broker
AWS Lambda is a serverless computer service that runs your code in response to events
and automatically manages the underlying compute resources for you. You can use
AWS Lambda to extend other AWS services with custom logic, or create your own
back-end services that operate at AWS scale, performance, and security. AWS Lambda
can automatically run code in response to multiple events, such as HTTP requests
via Amazon API Gateway, modifications to objects in Amazon S3 buckets, table
updates in Amazon DynamoDB, and state transitions in AWS Step Functions.
Lambda runs your code on high-availability compute infrastructure and performs all
the administration of the compute resources, including server and operating system
maintenance, capacity provisioning and automatic scaling, code and security patch
deployment, and code monitoring and logging. All you need to do is supply the code.
The code you run on AWS Lambda is called a “Lambda function.” After you create
your Lambda function it is always ready to run as soon as it is triggered, similar to a
formula in a spreadsheet. Each function includes your code as well as some associated
configuration information, including the function name and resource requirements.
Lambda functions are “stateless,” with no affinity to the underlying infrastructure, so
that Lambda can rapidly launch as many copies of the function as needed to scale to the
rate of incoming events.
After you upload your code to AWS Lambda, you can associate your function with
specific AWS resources (e.g. a particular Amazon S3 bucket, Amazon DynamoDB
table, Amazon Kinesis stream, or Amazon SNS notification). Then, when the resource
20
changes, Lambda will execute your function and manage the compute resources as
needed in order to keep up with incoming requests. The followings are the functions:
1. No servers to manage
AWS Lambda automatically runs your code without requiring you to provision or
manage servers. Just write the code and upload it to Lambda.
2. Continuous scaling
With AWS Lambda, you are charged for every 100ms your code executes and the
number of times your code is triggered. You pay only for the compute time you
consume.
4. Consistent performance
With AWS Lambda, you can optimize your code execution time by choosing the right
memory size for your function. You can also enable Provisioned Concurrency to keep
your functions initialized and hyper-ready to respond within double digit milliseconds.
Amazon Kinesis makes it easy to collect, process, and analyze real-time, streaming data
so you can get timely insights and react quickly to new information. Amazon Kinesis
offers key capabilities to cost-effectively process streaming data at any scale, along
with the flexibility to choose the tools that best suit the requirements of your
application. With Amazon Kinesis, you can ingest real-time data such as video, audio,
application logs, website clickstreams, and IoT telemetry data for machine learning,
analytics, and other applications. Amazon Kinesis enables you to process and analyze
21
data as it arrives and respond instantly instead of having to wait until all your data is
collected before the processing can begin.
Benefits:-
1. Real-time
Amazon Kinesis enables you to ingest, buffer, and process streaming data in real-time,
so you can derive insights in seconds or minutes instead of hours or days.
2. Fully managed
Amazon Kinesis is fully managed and runs your streaming applications without
requiring you to manage any infrastructure.
3. Scalable
Amazon Kinesis can handle any amount of streaming data and process data from
hundreds of thousands of sources with very low latencies.
4. OTA updates
22
CHAPTER-5
SOFTWARE ENVIRONMENT
Our product-service system shall have hierarchical order of monitoring system. For this
purpose, we shall have admin management website for high level official in the entire
city which will be embedded with IOT based Smart Metering for monitoring real time
electricity consuming through graphically represented in mobile application through
consumer login and registered and check their real time data consuming.
The technical frameworks, we use are NodeJS for server hosting, Python for machine
learning and AI algorithms along with web front end languages like HTML, CSS,
Bootstrap, Javascript & JQuery for the system.
Similarly, the database used under our system is MongoDB (NoSQL). In the same
manner the feature of IoT is obtained using Raspberry Pi coded with python having
relay connected to its output pin(s). And create react app for the implementation of the
system.
23
Available Scripts
yarn start
Runs the app in the development mode.
Open http://localhost:3000 to view it in the browser.
yarn test
Launches the test runner in the interactive watch mode.
See the section about running tests for more information.
yarn build
Builds the app for production to the build folder.
It correctly bundles React in production mode and optimizes the build for the best
performance.
The build is minified and the filenames include the hashes.
Your app is ready to be deployed!
24
CHAPTER-6
OUTPUT
25
6.2 Mobile Registration Screen
26
Fig 1.9- Cost Diagram and Data Diagram
27
6.4.2 Energy meters location
28
6.4.4 Payments of the User
29
CHAPTER-7
TESTING
30
CHAPTER-8
BUSINESS MODEL
31
8.2 Cost structure
4. Case 500 5
32
8.2.2 Cost for Redesigned Traditional Meter
SN Task Cost
33
5. Penetration Testings 3 Lakh
Total 20 Lakh
34
8. AWS Kinesis Data Streams Real time IoT data capture $200
9. AWS Kinesis Data Firehose Real time IoT data filter $200
10. AWS Kinesis Data Analytics Real time IoT data analysis $800
35
18. Amazon Cognito User authentication $2000
20 lakh monthly for 1 crore smart set (user and meter) i.e. 20 paisa per set. Also the
GSM cost per month on data seems to be almost Rs. 2 maximum per month per device
can be further reduced via partnership with network provides like Airtel, Vodafone, JIO
etc.
36
CHAPTER-9
CONCLUSION
Hence, this paper includes the proposed model of a smart meter using AI and IoT to
achieve the required goals of fleet management, real time data analysis and seamless
visualization. Old meter is reused and redesigned such that re-modelling into the old
module saves cost and new meter is designed for industrial application which is
embedded with CT, VT, thermistor and other sensor for safe performance.
Normal meters are converted to digital smart meters in which data is send securely to
reliable cloud platform and are visualized and processed for smart consumption in order
to overcome lack of connectivity between consumers and producers, manual control
and reading of electricity data, lack of result oriented energy data analysis and theft
crisis.
37
REFRENCES
1. U.S. Patent 3,842,208 (Sensor Monitoring Device)
2. Milani, Kate (10 November 2003). "Inventor reveals the names behind numbers".
www.bizjournals.com. Retrieved 2019-09-05.
4. U.S. Patent 4,241,237 and U.S. Patent 4,455,453 and Canadian Patent # 1,155,243
(Apparatus and Method for Remote Sensor Monitoring, Metering and Control).
7. Landi C., Dept. d.Ing., dell inf. Secondauniv. di Napoli,Aversa Italy, Morela, P.;
lannilo G,, “ARM based energy management system using smart meter and web
server,", IEEE Instrumentation and Measurement Technology conference
Binjiang,, pp.1-5, May-2011.
10. M., Revidi, Suri Reddy and Kishore A.,“ GSM & Zigbee Based Automatic
Energy Meter Reading System with Instant Billing ”, International Journal Of
Scientific Engineering and Technology Research, ISSN 2319-8885 Vol.04,
Issue.51, Pages No 11091-11097, December-2015.
38
11. W. Bos1 Joppe, Castryck Wouter, Iliashenko Ilia and Vercauteren Frederik, “
Privacy-friendly Forecasting for the Smart Grid using Homomorphic Encryption
and the Group Method of Data Handling ”, IARC, 2016.
12. Shinde Sandhya, Yadav Yogesh, Sontakke Bharti and Zapake Pratiksha, “ IoT
Based Smart Energy Meter ”, International Journal of Trend in Scientific
Research and Development (IJTSRD), ISSN No: 2456 - 6470, Volume - 1, Issue -
6, Page No. 1151, September- October, 2017.
13. https://www.ijarnd.com/manuscripts/v3i3/V3I3-1256.pdf
14. http://ijesc.org/upload/130e4b7ebec86677e4b06f012c3c8d41.Smart%20Energy%
20Meter%20for%20Advanced%20Metering%20and%20Billing%20Alert%20Fra
mework.pdf
15. https://en.wikipedia.org/wiki/Smart_meter#Overview
39