International Journal of Trend in Scientific Research and Development (IJTSRD)

Volume 6 Issue 5, July-August 2022 Available Online: www.ijtsrd.com e-ISSN: 2456 – 6470

Artificial Intelligence in Smart Grid

Matthew N. O. Sadiku1, Uwakwe C. Chukwu2, Abayomi Ajayi-Majebi3, Sarhan M. Musa1
1
Roy G. Perry College of Engineering, Prairie View A&M University, Prairie View, TX, USA
2
Department of Engineering Technology, South Carolina State University, Orangeburg, SC, USA
3
Department of Manufacturing Engineering, Central State University, Wilberforce, OH, USA

ABSTRACT How to cite this paper: Matthew N. O.

The smart grid is an electrical power grid that is integrated with an Sadiku | Uwakwe C. Chukwu | Abayomi
AI-enabled, two-way communication network providing energy and Ajayi-Majebi | Sarhan M. Musa
information. It is a technology that enables instantaneous feedback "Artificial Intelligence in Smart Grid"
Published in
from various sensors and devices on the operation of the power grid.
International Journal
Although AI is relatively new, it is poised to revolutionize the way
of Trend in
we produce, transmit, and consume energy. AI will constitute the Scientific Research
brain of future smart grid. The power sector has started to use AI and and Development
related technologies for communication between smart grids, smart (ijtsrd), ISSN: 2456-
meters, and Internet of things devices. This paper presents some 6470, Volume-6 | IJTSRD50563
applications of AI in smart grid. Issue-5, August
2022, pp.770-775, URL:
KEYWORDS: smart grid, artificial intelligence, artificial intelligence www.ijtsrd.com/papers/ijtsrd50563.pdf
in smart grid
Copyright © 2022 by author (s) and
International Journal of Trend in
Scientific Research and Development
Journal. This is an
Open Access article
distributed under the
terms of the Creative Commons
Attribution License (CC BY 4.0)
(http://creativecommons.org/licenses/by/4.0)

INTRODUCTION
We all depend on energy to do things. Electricity efficiency, changing supply and demand patterns.
has been one of the most important and the most These challenges are more acute in developing
widely used forms of energy since the 19th century. nations.
The electric power has changed our society. The
The word “smart” in smart grid refers to the notion of
electric industry essentially consists of three primary
a power grid with intelligence. The main objective of
functional areas: generation, transmission, and
the smart grid is to bring reliability, flexibility,
distribution. This entire structure is popularly known
efficiency, and robustness to the power system. Smart
as the “grid.” Thus, the power grid is a dynamic
grid does this by introducing two-way data
power system that delivers electric power from a
communications into the power grid. Thus, the smart
generation system through transmission and
grid consists of the power infrastructure and
distribution systems to end-users. The concept of
communication infrastructure, which correspond to
smart grid is shown in Figure 1 [1].
the flow of power and information respectively [3].
The power infrastructure consists of a vast network of For efficiency, a smart grid infrastructure should also
power plants, transmission lines, and distribution include distributed generation and AI.
centers (comprising roughly 5,800 power plants and
OVERVIEW ON ARTIFICIAL INTELLIGENCE
over 2.7 million miles of power lines). This
The term “artificial intelligence” (AI) was first used
deteriorating structure supports one-way power flow
at a Dartmouth College conference in 1956. AI is
from centralized generation to end customers and is
now one of the most important global issues of the
yet to receive a modern overhaul [2]. The energy
21st century. AI is the branch of computer science
sector worldwide faces some challenges related to
that deals with designing intelligent computer systems
rising demand of increasing global population,
that mimic human intelligence, e.g. visual perception,
integration with various distributed components,

@ IJTSRD | Unique Paper ID – IJTSRD50563 | Volume – 6 | Issue – 5 | July-August 2022 Page 770
 International Journal of Trend in Scientific Research and Development @ www.ijtsrd.com eISSN: 2456-6470
speech recognition, decision-making, and language learning architectures are able to process
translation. The ability of machines to process natural hierarchies of increasingly abstract features,
language, to learn, to plan makes it possible for new making them especially useful for purposes like
tasks to be performed by intelligent systems. The speech and image recognition and natural
main purpose of AI is to mimic the cognitive function language processing. Deep learning networks can
of human beings and perform activities that would deal with complex non-linear problems.
typically be performed by a human being. Without
Natural Language Processors: For AI to be
being taught by humans, machines use their own
useful to us humans, it needs to be able to
experience to solve a problem. communicate with us in our language. Computer
AI is stand-alone independent electronic entity that programs can translate or interpret language as it
functions much like human expert. Today, AI is is spoken by normal people.
integrated into our daily lives in several forms, such
Robots: These are computer-based programmable
as personal assistants, automated mass transportation, machines that have physical manipulators and
aviation, computer gaming, facial recognition at
sensors. Sensors can monitor temperature,
passport control, voice recognition on virtual humidity, pressure, time, record data, and make
assistants, driverless cars, companion robots, etc. AI critical decisions in some cases. Robots have
is not a single technology but a range of moved from science fiction to your local hospital.
computational models and algorithms.
In jobs with repetitive and monotonous functions
Some forms of AI that are most commonly used in they might even completely replace humans.
electrical and computer engineering include the Robotics and autonomous systems are regarded as
following [4,5]: the fourth industrial revolution.
Expert systems: They solve problems with an
These AI tools are illustrated in Figure 2 [6]. Each AI
inference engine that draws from a knowledge
tool has its own advantages. Using a combination of
base equipped with information about a
these models, rather than a single model, is
specialized domain, mainly in the form of if-then
recommended. AI systems are designed to make
rules. Expert systems are the earliest and most
decisions using real-time data. They have the ability
extensive, the most active and most fruitful area. to learn and adapt as they make decisions.
Fuzzy logic: This makes it possible to create rules
AI IN SMART GRID
for how machines respond to inputs that account
Like other industries, the power sector has been
for a continuum of possible conditions, rather
inundated by AI-enabled technologies. To support the
than straightforward binary.
existing systems and to extend the flexibility and
Neural networks: These are specific types of applicability of smart grids, AI has been naturally
machine learning systems that consist of artificial adapted. As a result, large regional grids will be
synapses designed to imitate the structure and replaced by microgrids that manage local energy
function of brains. They are similar to the human demand.
brain. They are made up of artificial neurons, take
Researchers at Argonne National Laboratory (the first
in multiple inputs, and produce a single output.
US national laboratory) are developing new ways to
The network observes and learns as the synapses
extract insights from the massive data on the electric
transmit data to one another, processing
grid, with the intent of ensuring greater reliability,
information as it passes through multiple layers.
resilience, and efficiency. They are working on
Machine learning: This includes a broad range optimization models that use machine learning, to
of algorithms and statistical models that make it simulate the electric system and the severity of various
possible for systems to find patterns, draw problems [7].
inferences, and learn to perform tasks without
Some of the adapted AI techniques in the smart grid
specific instructions. Machine learning is a include [8]:
process that involves the application of AI to
Managing the grid users and controllers
automatically perform a specific task without
explicitly programming it. ML techniques may System based operation strategies for the grid
result in data insights that increase production Power supply optimization
efficiency. Today, artificial intelligence is narrow Consensus-based intelligent distribution
and mainly based on machine learning. techniques
Deep learning: This is a form of machine Machine learning and deep learning enabled
learning based on artificial neural networks. Deep costing mechanisms

@ IJTSRD | Unique Paper ID – IJTSRD50563 | Volume – 6 | Issue – 5 | July-August 2022 Page 771
 International Journal of Trend in Scientific Research and Development @ www.ijtsrd.com eISSN: 2456-6470
Intelligent energy storage systems understand and optimize user’s behavior and
Intelligent voltage profile regulation techniques manage energy consumption. In a smart
using smart algorithms networked home, the networked devices react to
prices on the electricity market and adapt to
Integrating privacy into the smart grid.
household usage accordingly. By monitoring the
APPLICATIONS OF AI IN SMART GRID energy consumption pattern of individuals and
Artificial Intelligence is everywhere. It is the fastest businesses, AI companies can offer solutions to
growing branch of the high-tech industry. AI’s vast optimize usage, save electricity, and reduce costs.
potential has motivated several initiatives by utilities, For example, SmartTthermostat Nest adapts
government agencies, and academia. The US temperatures according to user behavior to reduce
Department of Energy (DOE) recently established a energy consumption.
new Artificial Intelligence and Technology Office
(AITO) to coordinate the agency’s AI development, Electricity Trading: In electricity trading, AI
delivery, and adoption [6]. In September 2017, the helps improve forecasts. Machine Learning and
DOE funded researchers at Stanford University to use Neural Networks play an important role in
artificial intelligence to improve grid stability. The improving forecasts in the energy industry.
UK’s National Grid collaborated with DeepMind to Failure management: Without regular checks on
add AI technology to the country’s electricity system. power equipment, equipment failures are
common. Using AI to observe equipment and
The German government sees AI as a key strategy for
mastering some of our greatest challenges such as detect failures before they happen can save
money, time, and lives.
climate change and pollution.
Energy Storage: The Smart grid with energy
AI is becoming more and more important in the
storage will continuously collect massive data to
energy industry. Typical areas of application are
make timely decisions on how best to allocate
autonomous grids, smart meters, energy consumption,
energy resources. Combined with other
electricity trading, failure management, and energy
technologies such as big data, the cloud and the
storage. These applications are discussed as follows
Internet of things (IoT), energy storage with AI
[9-11].
can play an important role in power grid
Autonomous Grid: In the US, the DOE is management. A smart grid with energy storage is
developing an autonomous grid using AI. With able to use energy sources in the most efficient
the power grids now collecting energy from way by better integrating renewable resources.
different sources and the increasing
These applications are simply a taste of what is
decentralization of the grids, operating the grids
ultimately possible. There are many more
has become more complex. This requires
applications of AI in smart grid or energy industry
analyzing massive data. Artificial Intelligence
helps process this data as quickly and efficiently such as energy management, managing electric
vehicles, network planning, fraud detection, load
as possible, thereby bringing stability and
efficiency. forecasting, stability analysis, security assessment,
stability assessment, fault diagnosis, fault prediction,
Smart Meters: A smart meter is a high-tech and stability control in smart grids. Figure 3 shows
meter that measures electricity consumption and some of the applications of AI in energy industry
provides additional information to the utility [10].
company unlike the conventional, analog meter.
Smart meters (SMs) are essentially digital meters BENEFITS
that read remotely over a secure wireless network. The application of AI-based technologies to the
power grid cuts energy waste, facilitates the use of
They are an important component of the smart
grid system. They will be able to constantly clean and renewable energy sources, and improve the
planning, operation, and control of the power
monitor demand and supply of customers. These
systems. The technologies can also help improve
smart meters process information that can be
power management, efficiency, stability. resilience.
related to a person and be privacy sensitive. With
and transparency, and increase the use of renewable
smart metering, one can monitor every appliance,
energy sources. Smart grid facilitates large amounts
providing the homeowner with a comprehensive
of renewable energy integration.
picture of their energy usage [12].
Energy Consumption: In addition to making the A major benefit of AI is the ability for the customers
power grid smart, flexible, and autonomous, AI and the grid to be connected directly, creating win-
algorithms help utilities and energy companies win situation. The price of solar has come down

@ IJTSRD | Unique Paper ID – IJTSRD50563 | Volume – 6 | Issue – 5 | July-August 2022 Page 772
 International Journal of Trend in Scientific Research and Development @ www.ijtsrd.com eISSN: 2456-6470
recently years to bolster the cost-effectiveness of efficiency. Significant investments in the
renewables. This can lead to a more efficient market infrastructure will be needed to help smart grids fully
and more cost-effective electricity production. It is take off [14].
realistic to expect the smart grid system to lower New generation of energy networks will make
electricity bills and prevent catastrophic blackouts. AI efficient use of renewable energy sources, support
can help use less energy to accomplish more. It is real time and efficient demand response, as well as
also helping compress and analyze the massive the large-scale deployment of electric vehicles (EVs)
amounts of data produced by energy industry. To [15]. To provide a low-cost and flexible solution for
curb data access by private companies, the European the grid-wide information exchange, wireless
Union (EU) Commission has developed four basic communications technology is expected to play the
ethical principles for AIs: AI should respect human key role in the emerging smart grid applications.
autonomy, avoid social harm, be fair, and be
explainable. CONCLUSION
The smart grid is the developmental trend of power
CHALLENGES systems. It has attracted much attention all over the
Smart grid faces a wide range of challenges, such as world. It provides a platform for clean, sustainable,
extreme weather, imperfections in the available efficient and reliable energy generation, delivery, and
infrastructure, reliability issues, equipment failures, consumption. It is inevitable that the smart grid will
gigantic customer base, decentralized generation, and become part of our society. The global energy
decarbonizing the global economy. A major challenge demand is expected to increase steadily in the future.
is the rise of distributed generation, where individual Therefore, future generations should realize that AI
customers generate and use their own electricity from and energy are not mutually exclusive career paths.
renewable sources, such as wind and solar. The For more information about artificial intelligence in
current power system was not designed to agriculture, one should consult the books in [16,17].
accommodate this diversification in energy sources
and fluctuating supplies of renewable energy. Industry REFERENCES
leaders in the AI energy grid industry are aware of [1] V. Robu, “ Why artificial intelligence could be
these challenges and must address them. The use of key to future-proofing the grid,”
AI in smart grid is not without risks. One risk has to https://robohub.org/why-artificial-intelligence-
do with the privacy of customer data collected by AI could-be-key-to-future-proofing-the-grid/
systems. [2] F. Wolfe, “How artificial intelligence will
THE FUTURE OF AI IN SMART GRID revolutionize the energy industry,”
Artificial intelligence holds significant potential http://sitn.hms.harvard.edu/flash/2017/artificial-
across a wide array of sectors. It is expanding its intelligence-will-revolutionize-energy-industry/
scope over tasks traditionally performed by humans. [3] M.N.O. Sadiku, and S.M. Musa and S. R.
Its applications in the energy industry are helping to Nelatury, “Smart grid – An introduction,”
make the grid “smarter” or more responsive, thereby International Journal of Electrical Engineering
develop the smart grid of the future. AI has been & Technology, vol. 7, no.1, Jan-Feb, 2016, pp.
regarded as the brain behind the future smart grid, 45-49.
enabling the real-time optimization and automation of
distribution planning and operation decision. The [4] M. N. O. Sadiku, Y. Zhou, and S. M. Musa,
smart grid is a dynamic system that continues to “Natural language processing in healthcare,”
evolve as technologies are tested and perfected [13]. International Journal of Advanced Research in
Computer Science and Software Engineering,
Although the use of AI in the smart grid faces some vol. 8, no. 5, May 2018, pp. 39-42.
challenges such as insufficient reliability, imperfect
infrastructure, and lack of special algorithm for power [5] “Applications of AI and machine learning in
industry, AI is a powerful tool to push smart grid into electrical and computer engineering,” July,
the new generation of power systems. AI supports 2020, https://online.egr.msu.edu/articles/ai-
and optimizes electric networks around the world, machine-learning-electrical-computer-
pushing the concept closer towards a global adoption. engineeringapplications/#:~:text=Machine%20l
It should be added to all levels in the energy grids, in earning%20and%20electrical%20engineering,c
order to enhance their development. an%20%E2%80%9Csee%E2%80%9D%20the
%20environment.
Although the smart is not quite here yet, it is slowly
becoming a reality. It is on its way to usher the energy [6] “Hype and hope: Artificial intelligence’s role in
industry into a new era of reliability, availability, and the power sector,” February 2020,

@ IJTSRD | Unique Paper ID – IJTSRD50563 | Volume – 6 | Issue – 5 | July-August 2022 Page 773
 International Journal of Trend in Scientific Research and Development @ www.ijtsrd.com eISSN: 2456-6470
https://www.powermag.com/hype-and-hope- [12] M. N. O. Sadiku, S.M. Musa, A. Omotoso, and
artificial-intelligences-role-in-the-power-sector/ A.E. Shadare, “A primer on smart meters,”
[7] C. Nunez, “Artificial intelligence can make the International Journal of Trend in Research and
U.S. electric grid smarter,” June 2019, Development, vol. 5, no. 4, 2018, pp. 65-67.
https://www.tdworld.com/grid- [13] C. Frye, “4 Ways artificial intelligence is
innovations/smartgrid/article/20972769/artificia powering the energy industry,” November
l-intelligence-can-make-the-us-electric-grid- 2018, https://www.kolabtree.com/blog/4-ways-
smarter artificial-intelligence-is-powering-the-energy-
[8] V. Sooriarachchi, “Prospect of adapting industry/
artificial intelligence in smart grids for [14] M. Sulikowski, “Smart grid - AI at the service
developing countries,” of the power distribution network,” June 2019,
https://smartgrid.ieee.org/newsletters/november https://naturaily.com/blog/smart-grid-ai-in-
-2020/prospect-of-adapting-artificial- power-distribution-
intelligence-in-smart-grids-for-developing- network#:~:text=The%20term%20%E2%80%9
countries Csmart%20grid%E2%80%9D%20describes,fas
[9] “5 Ways the energy industry is using artificial ter%20restoration%20after%20power%20black
intelligence,” March 2018, outs.
https://www.cbinsights.com/research/artificial- [15] N. Bassiliades and G. Chalkiadakis, “Artificial
intelligence-energy-industry/ intelligence techniques for the smart grid,”
[10] “What is artificial intelligence in the energy Advances in Building Energy Research, vol. 12,
industry?” https://www.next- no. 1, 2018, pp. 1-2,
kraftwerke.com/knowledge/artificial- [16] L.. A. Kumar, L. S. Jayashree, and R.
intelligence Manimegalai (eds.), Proceedings of
[11] S. Bilodeau, “Artificial intelligence in a ‘no International Conference on Artificial
choice but to get it smart’ energy industry!” Intelligence, Smart Grid and Smart City
April 2019, Applications. Springer, 2020.
https://towardsdatascience.com/artificial- [17] S. F. Bush, Smart Grid: Communication-
intelligence-in-a-no-choice-but-to-get-it-smart- Enabled Intelligence For The Electric Power
energy-industry-1bd1396a87f8 Grid. John Wiley & Sons, 2014.

Figure 1 The concept of smart grid [1].

@ IJTSRD | Unique Paper ID – IJTSRD50563 | Volume – 6 | Issue – 5 | July-August 2022 Page 774
 International Journal of Trend in Scientific Research and Development @ www.ijtsrd.com eISSN: 2456-6470

Figure 2 Artificial intelligence (AI) encapsulates several concepts including natural language
processing (NLP), deep learning (DL), and neural networks (NN) [6].

Figure 3 Some applications of AI in energy industry [10]

@ IJTSRD | Unique Paper ID – IJTSRD50563 | Volume – 6 | Issue – 5 | July-August 2022 Page 775