0% found this document useful (0 votes)
30 views21 pages

Seminar Report

The seminar report reviews the application of AI algorithms for integrating renewable energy sources into India's national grid, highlighting the challenges and advancements in the field. It discusses the importance of machine learning techniques in optimizing energy management and addressing issues such as power quality and reliability. The report emphasizes the Indian government's commitment to increasing renewable energy capacity and the role of AI in enhancing the efficiency of power systems.

Uploaded by

sukritsingh085
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
30 views21 pages

Seminar Report

The seminar report reviews the application of AI algorithms for integrating renewable energy sources into India's national grid, highlighting the challenges and advancements in the field. It discusses the importance of machine learning techniques in optimizing energy management and addressing issues such as power quality and reliability. The report emphasizes the Indian government's commitment to increasing renewable energy capacity and the role of AI in enhancing the efficiency of power systems.

Uploaded by

sukritsingh085
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 21

A

SEMINAR REPORT
ON
Application of AI algorithms for Grid integration of
Renewable energy sources in the National Grid- Indian
Government perspective- A Review
Submitted in partial fulfillment of the requirements for the award of degree of

BACHELOR OF TECHNOLOGY

in

ELECTRICAL ENGINEERING

Submitted By: Supervised By:


Shivanshu Chattopadhyay(2023UEE1065) Dr. Akhilesh Mathur
Assistant Professor

DEPARTMENT OF ELECTRICAL ENGINEERING


MALAVIYA NATIONAL INSTITUTE OF TECHNOLOGY JAIPUR, INDIA
2023-2024

i
MALAVIYA NATIONAL INSTITUTE OF TECHNOLOGY

JAIPUR

CERTIFICATE

This is to certify that the seminar entitled “Application of AI algorithms for


Grid integration of Renewable energy sources in the National Grid Indian
Government perspective- A Review submitted” by Mr. Shivanshu
Chattopadhyay(2023UEE1065) at Malaviya National Institute of Technology
Jaipur towards partial fulfillment of the requirements for the award of the
degree of Bachelor of Technology in Electrical Engineering at Department of
Electrical Engineering has been carried out by him under my supervision.

Dr. AKHILESH MATHUR


Assistant Professor
Department of Electrical Engineering
MNIT Jaipur
Jaipur -302017, India

Place: Jaipur
Date:

ii
ACKNOWLEDGEMENTS

I am thankful to Dr. Akhilesh Mathur sir and his PhD. Scholar Mr. Ghanshyam
Meena sir for their invaluable guidance, encouragement, and support throughout
the course of this project. Their expertise and insightful feedback have been
instrumental in shaping the direction of my work, helping me navigate
challenges, and broadening my understanding of the subject. I deeply appreciate
their patience, dedication, and willingness to share his knowledge, which has
greatly enriched my learning experience. This seminar would not have been
possible without their mentorship, and I am truly grateful for their unwavering
support.

I am also thankful to Dr. Prerna Jain Madam for assigning us this project,
providing the opportunity to explore this topic and develop my understanding.

Date: Shivanshu Chattopadhyay

2023UEE1065

iii
ABSTRACT
The global energy scenario faces challenges such as supply-demand imbalance,
greenhouse gas emissions, and frequent blackouts. As a result, there is a
growing focus on renewable energy sources like wind and solar, and advanced
microgrid technology. Traditional power grids are becoming obsolete due to
protection and interconnection issues. Integrating renewable energy resources
has benefits like inexhaustibility, green energy utilization, recycling, storage
potential, and low maintenance, indicating a strong economy. AI and deep
learning can optimize energy source selection by analysing seasonal cost
variations through cloud storage, leading to substantial savings. This approach
can be applied to both large and small power grids.

iv
LIST OF CONTENTS
Pag
TOPIC e
No.

Certificate ii
Acknowledgements iii
Abstract iv
List of Contents v
List of Abbreviations vi

Chapter-1 Introduction 1

Chapter-2 Relevant Studies 3

Chapter-3 Features of Power Quality 7

Chapter-4 Challenges Identified for increasing reliability of power


in network 9

Chapter-5 Power grid operations using ML technique 11

Chapter-6 Conclusion 13

References 14

v
LIST OF ABBREVIATIONS
Abbreviation Full form
DER Distributed Energy Resources
RER Renewable Energy Resources
ML Machine Learning
ANN Artificial Neural Network
FACTS Flexible AC Transmission Systems

vi
II. INTRODUCTION
In the growing 21st century, rapid modernization, and fast developmental
changes of the countries around the globe, there has been approximately
exponential increase in demand of the energy. The prevalent and traditional
sources of energy namely petroleum, diesel, coal, etc. are not able to meet
todays’ power demand and energy crisis. As a result, there is a growing focus on
renewable energy sources (RER) such as solar, wind, hydro, and geothermal
energy. These sustainable alternatives can help reduce our reliance on fossil
fuels and address the energy crisis, paving the way for a stable and sustainable
energy future.

Wind and solar energy generation is continuously increasingly day by day by


day and becoming one of the prominent sources of electrical energy for
household and industrial purposes. The main advantage of offshore wind energy
is their consistency in the generation of the electric power despite stronger and
more consistent wind gusts that too in steady rate [1]. For the energy efficient
system and reliable operation of the system, every homeland has incorporated
different grid codes. In the recent years, Machine Learning Techniques like
Artificial Neural Networks (ANN), Decision Tree (DT), Support vector
machines (SVM), have been used widely, in accordance to effective decision
making and control actions in the secured and stable operations of the power
system [2].

The modernization of power system networks through renewable energy


integration, liberalized markets, and advanced monitoring and communication
infrastructures has led to more resilient and efficient grids. However, these
advancements also introduce challenges such as cyberattacks, voltage
instability, and power quality disturbances, necessitating efficient detection and
mitigation strategies. Machine learning techniques (MLTs), including artificial
neural networks (ANN), decision trees (DT), and support vector machines
(SVM), have gained prominence in addressing these concerns by enabling
effective decision-making and control for secure power system operations. This
paper presents a comprehensive review of recent MLT applications for power
system security and stability, with a focus on cyberattack detection, power
quality disturbance studies, and dynamic security assessments. Additionally,
reinforcement learning (RL) and deep reinforcement learning (DRL) approaches

1
for transient stability assessment are discussed, highlighting methodologies,
achievements, limitations, and potential directions for future research

2
II. RELEVANT STUDIES
We have made a comparison of what advancements in renewable energy have
been made in other countries versus what has been going in India.

(These studies have been done across the globe done in past decade)

1.United States of America: The study of various countries is [2]: Funded by


USA Department of energy, the NERL, approx. #5% energy penetration of
wind, solar and CSP, 20% wind for the major part of energy demand.

Energy distribution from various sources in USA [9]

2.China: According to the report published by China in year 2015, the


installation capacity of renewable energy is much high and the same is carried
by Energy research institute, NDRC.

3.Europe: In the near future, the European super grid is considered possible
super grid, connecting the European countries as well as the regions
neighbouring the European continent, including North Africa and Middle East.

The European super grid

3
4.The Great Britain: In 2016, August, the Government of British has indicated
green signal for its biggest offshore wind projects, known as Hornsea Project. It
is big enough to power 1.6 million households [3].

The Hornsea Project

5.INDIA: (is not mentioned in the Research Paper)

As of 2021, India's national policy for renewable energy operations aims to


achieve a target of 450 GW of renewable energy capacity by 2030, with a
further commitment made at COP26 to reach 500 GW of non-fossil fuel-based
capacity by the same date. This policy includes initiatives to encourage solar
and wind power generation, promote domestic manufacturing of renewable
energy components, and facilitate grid integration of renewable energy sources.

Key points about the 2021 Indian renewable energy policy:

 Ambitious target: The central goal is to significantly increase renewable


energy capacity, aiming for 450 GW by 2030.

 Policy focusses on solar and wind: The policy actively promotes the
development of solar and wind power projects.

 Domestic manufacturing push: Incentives are provided to encourage


domestic production of solar photovoltaic cells and modules.

4
 Grid integration: Measures are in place to ensure smooth integration of
renewable energy into the existing power grid.

 Investment attraction: The government is actively seeking private


investment in the renewable energy sector.

Source of information:

 Ministry of New and Renewable Energy (MNRE), Government of India

 International Renewable Energy Agency (IRENA)

 NITI Aayog: https://iea.blob.core.windows.net/assets/7b6bf9e6-4d69-


466c-8069-bdd26b3e9ed1/RenewablesIntegrationinIndia2021.pd:

5
(News Paper article published on Dainik Bhaskar which shows Indian
Government’s aim to promote renewable energy)

6
(NITI Aayog report of 2021)

7
III. FEATURES OF POWER QUALITY
Following technical aspects are important as listed below, which are considered
for environmental studies from electrical point of view.

1.Harmonics: The elimination of harmonics is necessary as they pose a threat to


electrical networks’ capability to operate in reliable and consistent way [4]. The
harmonics distortion can occur in the system by added efficiency losses to the
system, unexpected resonances, and overload at the non-linear load end,
malfunction of the motors or generators. The adverse effects of harmonics on
the system are power factor loss, conductor losses, and skin effect, triple
harmonics, and resonances, losses on the side of motor, generator, and
transformer.

Fig. 4. Measurement scheme for supra harmonics [6]

2. Power Imbalance and Voltage Fluctuations: During the operation of


microgrids, switching can cause power imbalance and voltage inconstancy in
the system. Thus, resulting into slow dynamic response and low inertia of the
microgrid. To reduce this many FACTS (Flexible Alternating
Current Transmission System) devices are used such as STATCOM, PID
controllers, D- STATCOM with FACTS technology.

8
Voltage fluctuation in classic elaborative way is variation in power system in
which the voltage magnitude or the phase angle differences between them are
not equal [4]. There are various factors contributing to it such as unbalanced
loading, uneven distribution of single-phase and three-phase load, incorrect
tapping of transformer, and unequal impedance in system. Power balance can be
termed, as electricity consumption is nearly equal to the electricity production.

Power quality refers to the stability and consistency of electrical power supplied
to equipment, ensuring voltage, frequency, and waveform remain within
acceptable limits [4]. There are many features in these systems comprising a
large number of semiconductor equipment, stochastic nature of the renewable
energy resources generation and bi-directional power flows. The assurance of
the required power quality indicator is a big challenge [5].

3.Production Simulation: Today many countries are utilizing time-series


production simulation method to understand and study the renewable energy
development scenario. The various constraint that are kept in mind are
equilibrium between system load and generator output grid operation [3].

9
IV. CHALLENGES IDENTIFIED FOR INCREASING
RELIABILITY OF POWER IN NETWORK
1. Protection Challenges: When we consider power system components, the
system is more prone to fault occurrences that affects the reliability of the
system as well as stability and safety. Traditional fault detection devices fail to
detect the fault of lower magnitude, nonlinear PV characteristics, transition
between days to night. Because of structural consideration and control, fault
characteristics are quite different. A reference is of installation of protection
devices to reduce potential damage for PV system where the area is prone to
lightning strikes. Along with high impedance faults, there are various other
faults namely physical, environmental, electrical failures, etc. [6].

2. Electricity Market Challenges: The trading in the electricity markets needs


to be swapped off instantaneously. The drawback of renewable energy
resources is the power generation skepticism in it that leads to create hindrance
to the productive involvement in markets specifically in the short-term
electricity markets [6]. The inadequateness of appropriate market skeleton and
innovative incentive packages impede renewable energy resources trading in the
electricity market. The platform comprising both energy resources like the
renewable and non-renewable energy resources can be marked as the tool to
participate in the electricity market with increasing flexibility in the market
scenario.

3. Use of Modern Technology Like Artificial Intelligence and Machine


Learning: In the recent years, more use of modern technology in everyday life
has become common. Use of technology has its own flaws. They are prone to
cyberattacks, phishing attack, and SQL injection attack [5]. The most possible
and common reason is consumers’ data leakage in which sensitive energy can
be revealed. It can lead to burglary and series of unpredictable events. Another
challenge is network congestion, leading to poor quality of service, blocking of
new connection. If the system faces cyberattacks or firewall breach, it can lead
to even blackouts causing haphazard, eventually leading to breakdown of the
network [7].

10
4.Environmental Challenge: The microgrid is considered green energy as it
uses renewable sources of energy. However, it has adverse impact on the
neighboring surrounding. It requires considerable project area, plenitude
measure of water during formation, production, cleaning of the system; it
releases hazardous materials that can cause damage to the surrounding
environment [6]. Although PV is considered to be, green energy, the waste due
to it millions of tons. The alternatives for PV waste management are landfill,
incineration, and the other is reusing the PV products that can be done by
repairing them but have low efficiency as compared to the former. Both the
landfill technique and recycling have their pros and cons.

11
V. POWER GRID OPERATIONS USING ML TECHNIQUES
The machine learning algorithm works on the fact that economics is balanced
which includes cost of energy generation per KWh, cost of transport of power
from remote distance, variable demand during day or seasons, etc. Our attempt
of artificial intelligence and machine learning algorithms is to reduce the cost of
energy to the consumers.

The novel algorithm using supervised learning which works on two major
types: Regression and Classification consisting of regression viz. Linear
regression, Regression tree, Non-linear regression, Bayesian linear regression,
Polynomial Regression and Classification viz. Random Forest, Decision tree,
Logistic regression, Support Vector Machine, have been found helpful.

As per the new national energy policy defined by Indian Government, the
government seek to address the issues of energy development including its
production, distribution and consumption; there are different priorities of
decision outlined by national renewable energy policy, including reducing
impact on the environment, increasing the energy efficiency, sustainable
development with improved effectiveness and competitiveness [10]. The Indian
government has undertaken various policy measures for increasing share of
renewable energy in national energy mix.

Machine learning can help foretell many allowable factors affecting renewable
energy. This results in renewables becoming more reliable, affordable, and
desirable. Renewable energy is optimistic on its own, but machine learning
expands its potential. (Advantages)Machine learning models can be used to
predict specific properties of new materials without the need for costly
characterization; they can generate new material structures with desired
properties; they can understand patterns in renewable energy usage and
generation; and they can help to inform energy policy by optimizing energy.

Renewable energy is increasingly being used to replace electricity demand


because it has been regenerated and reused several times. The estimation rate of
renewable energy is difficult as it is considered to be an intermediate and
unpredictable natural resource. The hybrid machine learning technique is
precisely employed to predict the energy level of natural resources. The hybrid
machine learning is the combination of Multilayer Perceptron (MLP), Support
Vector Regression (SVR) and Cat Boost algorithm that increases the
performance and predictability of renewable energy consumption [8]. The
12
operation of the grids depend mainly on predicting the electricity demand so
that the generation of the energy can be addressed to meet the demand and to
dodge the overload on power system. In present scenario, it is mandatory to
foretell both the electricity demand as well as the distributed generation to get
the better overview as to how much power needs to be produced to meet the
demands [9].

What is Machine Learning?

Machine-learning techniques defines the relations between input data and


output data with or without mathematical forms of problems, with the available
data available on cloud storage.

Present Scenario: Currently development of algorithms is under process, with


modifications of novel algorithms present. However, in practical situations,
considering the dynamic demand (of load) on the grid, ML algorithms may have
limitations, but it will have higher accuracy, more reliability, and less manual
interference, avoiding misinterpretation is reduced with ML algorithms.

13
VI. CONCLUSION
The consumption of non-renewable energy sources like coal, petroleum
products for generations together is leading to release of harmful by-products
like carbon dioxide, nitrogen oxide, carbon monoxide. The escalation of these
harmful by products can be done by use of renewable energy resources like
wind, solar, etc. [11]. The prerequisites of the reliable, secure power, stable,
safer transformation of modern Microgrids into much strong, robust, and smart
dynamic power delivery unique, smart, dynamic power delivery network with
the use of modern technology such as Artificial Intelligence or DL still needs
practical research.

In order to achieve the reliable renewable energy, the power system must be a
multi supply system that includes traditional power supplies, sources of
renewable energy and energy storage power supply. By the use of AI and ML,
various factors related to energy sources can be analyzed accurately such as
environment management, uncertainty analysis, thermal energy, economic and
investment. Deep learning models shall find more usages in future time.

14
REFERENCES
1. Syed Wajahat Ali, Muhammad Sadiq, Yacine Terriche, Syed Ahmad Raza Naqvi, Le
Quang Nhat Hoang, Muhammad Umair Hassan, Mustafa Alrayah Hassan, Guangya yang,
chun-lien Su, joseph M. Guerrero, “Offshore wind farm-grid integration: A review on
infrastructure, challenges and Grid solutions,”July 20, 2021, DOI:
10.1109/ACCESS.2021.3098705
2. Oyeniki akeem Alimi, Adnan M. Abu-Mahfouz, “A review of machine learning
approaches to power system security and stability,” Open Access Journal, June 19, 2020,
DOI: 10.1109/ACCESS.2020.3003568
3. Yan Li, Yougning Chi, Xuefeng Wang, Xinshou Tian, Jiao Jianqing, “Practises and
challenges on Planning with large scale Renewable energy grid integration”, 3rd IEEE
Conference on Energy Internet and Energy System Integration, November 8-10, 2019,
China.
4. Muhammad Hammad Saeed, Wang Fangzong, Basheer Ahmad Kalwar, Sajid Iqbal, “A
review on Microgrids’ challenges and perspectives,” open access journal, December 27,
2021, DOI: 10.1109/ACCESS.2021.3135083.
5. Andrey V. Shalukho, Ivan a. Lipuzhin, Alexander A. Voroshilov, “Power quality in
microgrids with distributed generation,” 2019 International Ural conference on Electrical
Power Engineering, May 31.
6. MD Shafiullah, Shakir D. Ahmed, “Grid integration challenges and solution strategies for
solar PV systems: A Review” open access journal, May 19, 2022, DOI:
10.1109/ACCESS.2022.3174555
7. Nzsmus Sakib, Eklas Hossain, Sheikh Iqbal ahamed, “a qualitative study on the United
States Internet of Energy: A step towards computational sustainability,”,April 7, 2020,
DOI: 10.1109/ACCESS.2020.2986317
8. Rasha M. ab El-Aziz, “Renewable power source energy consumption by hybrid machine
learning model,” Alexandria Engineering Journal, DOI: 10.1016/j.aej.2022.03.019.
9. Salahuddin A. Azad, Saleh A. Wasimi, Fariza Sabrina, “Transformation of Smart grids
using Machine learning,” November 2019.
10. Indian National Policy for renewal energy operations - 2021
11. Komal R. Waghmare, Dr. Surendra Bhosale, “Impact of renewable energy resources and
energy storage device on transmission system,”

15

You might also like