energies

Review
AI-Driven Urban Energy Solutions—From Individuals to
Society: A Review
Kinga Stecuła 1, * , Radosław Wolniak 1 and Wieslaw Wes Grebski 2

1 Faculty of Organization and Management, Silesian University of Technology, 44-100 Gliwice, Poland;
[email protected]
2 Penn State Hazleton, Pennsylvania State University, 76 University Drive, Hazleton, PA 18202-8025, USA;
[email protected]
* Correspondence: [email protected]

Abstract: This paper provides a comprehensive review of solutions based on artificial intelligence
(AI) in the urban energy sector, with a focus on their applications and impacts. The study employed
a literature review methodology to analyze recent research on AI’s role in energy-related solutions,
covering the years 2019 to 2023. The authors classified publications according to their main focus,
resulting in two key areas of AI implementation: residential and individual user applications, and
urban infrastructure integration for society. The objectives of this review of the literature are the
following: O1: to identify trends, emerging technologies, and applications using AI in the energy
field; O2: to provide up-to-date insights into the use of AI in energy-related applications; O3: to
gain a comprehensive understanding of the current state of AI-driven urban energy solutions; O4: to
explore future directions, emerging trends, and challenges in the field of AI-driven energy solutions.
This paper contributes to a deeper understanding of the transformative potential of AI in urban
energy management, providing valuable insights and directions for researchers and practitioners
in the field. Based on the results, it can be claimed that AI connected to energy at homes is used in
the following areas: heating and cooling, lighting, windows and blinds, home devices, and energy
management systems. AI is integrating into urban infrastructure through the following solutions:
enhancement of electric vehicle charging infrastructure, reduction in vehicle emissions, development
of smart grids, and efficient energy storage. What is more, the latest challenges associated with the
Citation: Stecuła, K.; Wolniak, R.; implementation of AI-driven energy solutions include the need to balance resident comfort with
Grebski, W.W. AI-Driven Urban energy efficiency in smart homes, ensuring compatibility and cooperation among various devices,
Energy Solutions—From Individuals preventing unintended energy consumption increases due to constant connectivity, the management
to Society: A Review. Energies 2023, of renewable energy sources, and the coordination of energy consumption.
16, 7988. https://doi.org/10.3390/
en16247988
Keywords: energy; artificial intelligence; AI; smart city; smart home; smart grid; electric vehicle
Academic Editors: David Borge-Diez
and Joao Ferreira

Received: 9 November 2023

1. Introduction
Revised: 28 November 2023
Accepted: 7 December 2023 In a rapidly evolving urban landscape, the challenges of energy consumption, sus-
Published: 9 December 2023 tainability, and efficiency remain critical concerns. The need for energy in cities is still
growing and is related to the growing activities of residents, the need for advanced services,
and the use of technical means in the city infrastructure. The last ones are connected
with the Internet of Things and innovative solutions connected with energy in cities [1–3].
Copyright: © 2023 by the authors. The Internet of Things (IoT) has ushered in a new era of urban connectivity and intelli-
Licensee MDPI, Basel, Switzerland.
gence. In cities around the world, IoT technologies are revolutionizing the way urban
This article is an open access article
infrastructure operates, providing an interconnected web of smart devices and sensors
distributed under the terms and
that collect and share data in real time [4]. These IoT systems allow city planners and
conditions of the Creative Commons
administrators to gain unprecedented insight into various aspects of urban life, from traffic
Attribution (CC BY) license (https://
patterns [5], the use of different sources of energy [6–8] and energy consumption [9–11]
creativecommons.org/licenses/by/
to waste management [12] and air quality [13]. The pursuit of a ubiquitous Internet and
4.0/).

Energies 2023, 16, 7988. https://doi.org/10.3390/en16247988 https://www.mdpi.com/journal/energies

Energies 2023, 16, 7988 2 of 34

the development of urban infrastructure means that the achievements of Industry 4.0 are
being used on an increasingly larger scale, leading to more and more solutions based on
smart elements [14,15] and neural networks [16]. One of the rapidly developing elements
of the fourth industrial revolution is artificial intelligence (AI). Artificial intelligence is a
transformative force fundamentally reshaping the way we live, work, and interact with
our environment. It is important in the context of Industry 4.0 and the broader use of a
ubiquitous Internet not only for entertainment [17] but also for regular everyday life. AI
uses advanced algorithms, machine learning, and data analysis to mimic human cognitive
functions, enabling machines to perceive, reason, and make decisions [18]. In cities, AI
can be used to optimize energy infrastructure and create a more sustainable and resilient
urban future. Artificial intelligence offers many solutions in various areas of human activity,
including those used in the energy sector [19,20]. Some solutions are aimed at an individual
resident, and others at the whole society. Nevertheless, AI-based solutions are the driving
force towards a new intelligent society living in cities called smart cities.
As urban areas continue to expand and evolve, AI has become a critical tool for im-
proving energy management, reducing carbon emissions, and improving the overall quality
of life of urban residents. This paper presents a review of the literature of solutions driven
by artificial intelligence that are or are planned or are in the study stage to be applied in the
urban energy sector. Therefore, this article reviews and discusses the dynamic intersection
of AI-driven solutions and the urban energy sector, providing a comprehensive review
of the profound impact that AI technologies are having on cities around the world. The
paper consists of six sections. Section 2 describes materials and methods. Sections 3 and 4
include AI-driven energy solutions, respectively, among residents/individual users and in
urban infrastructure. Section 5 presents a discussion, and Section 6 summarizes the review.

2. Materials and Methods

The method used in this paper is literature review. The authors analyzed publications
that describe research on AI-driven solutions connected with energy. In the first stage of
the research, the authors used search engines including Web of Science and Scopus. The
authors searched and reviewed published publications over the last five years, that is, from
2019 to 2023, to obtain the latest results of recent research on artificial intelligence and
its application in energy-related solutions. The databases were used to search using two
main keywords, which were ‘artificial intelligence’/‘AI’ and ‘energy’. It is worth noting
that the authors entered the full name of artificial intelligence and its abbreviation into the
search engine as separate searches. Additionally, the third keyword was added to each
search—it was ‘smart city’, ‘smart home’, ‘smart grid’, and ‘electric vehicle’. There were
eight searches in total. The results of the search results in terms of the number of articles
in each of the two databases—Scopus and Web of Science—are presented in Table 1. The
largest number of articles were found in the Scopus database for the keywords ‘artificial
intelligence’, ‘energy’, and ‘smart grid’ keywords.

Table 1. Search results for papers in Scopus and Web of Science depending on keywords.

Number of Number of Papers–Web

Keywords
Papers–Scopus of Science
Artificial intelligence, energy, smart city 127 22
Artificial intelligence, energy, smart home 58 15
Artificial intelligence, energy, smart grid 264 79
Artificial intelligence, energy, electric vehicle 132 30
AI, energy, smart city 26 9
Energies 2023, 16, 7988 3 of 34

Table 1. Cont.

Number of Number of Papers–Web

Keywords
Papers–Scopus of Science
AI, energy, smart home 10 4
AI, energy, smart grid 39 12
AI, energy, electric vehicle 18 4
Source: authors’ own work.

For all of the papers found, the authors read abstracts and then chose all the papers
that really fit the scope of the undertaken review on the set topic. Finally, the authors chose
153 papers that cover the topic of AI-driven urban energy solutions.
These publications were classified into these divisions based on their main focus or
application areas within AI-driven urban energy solutions. Each category represents a
specific aspect of how AI-driven energy solutions are implemented and utilized, including
the following:
• Residential and individual user applications;
• Urban infrastructure integration for individual users and community oriented.
This categorization allows for a more structured analysis and understanding of the var-
ious applications and impacts of AI for urban energy. On the basis of the search results and
analysis, the authors divided the mentioned parts into subsections. The first part includes
five subsections, which include the following: heating and cooling, lighting, windows and
blinds, home devices—refrigerators, and energy management systems. The second part
was divided into four subsections including electric vehicle charging infrastructure, vehicle
emission reduction, smart grid, and energy storages.
The main objectives of the literature review are the following:
• O1: To identify trends, emerging technologies, and applications using artificial intelli-
gence in the energy field;
• O2: To provide up-to-date insights into the use of artificial intelligence in energy-
related applications;
• O3: To gain a comprehensive understanding of the current state of AI-driven urban
energy solutions;
• O4: To explore future directions, emerging trends, and challenges in the field of
AI-driven energy solutions.
The research questions set in the study include:
• R1: What are the key emerging technologies in AI-driven energy solutions for residen-
tial users and society?
• R2: How is artificial intelligence integrated into urban infrastructure to enhance
energy-related solutions?
• R3: What challenges are associated with the implementation of AI-driven solutions in
urban energy management?

3. Residential and Individual User Applications for AI-Driven Urban

Energy Solutions
3.1. Heating and Cooling
The first area that very often appears in research related to energy in residential
buildings is heating and cooling devices, in other words, HVAC (heating ventilation and air
conditioning) systems. These systems play a central role in regulating the indoor climate of
homes, directly impacting both the comfort of inhabitants and energy consumption. Using
artificial intelligence, innovative solutions are emerging to optimize heating and cooling
processes, ensuring that homes remain comfortable while minimizing energy usage and
environmental impact [21].
Energies 2023, 16, 7988 4 of 34

Heating in smart homes refers to the application of advanced technologies, including

AI, to control and optimize the heating systems within residential buildings. In smart
homes, heating systems are designed to be more energy efficient, convenient, and adaptable
to individual preferences. AI-driven heating solutions can analyze data from various
sources, such as weather forecasts and occupancy patterns, to make real-time adjustments
to heating systems. For example, if a smart home detects that no one is at home during the
day, it can lower the temperature to save energy and then start warming the house before
the occupants return. These systems can be controlled remotely via smartphones or voice
commands, allowing users to fine-tune their heating preferences. In general, AI-enhanced
heating in smart homes aims to improve comfort while reducing energy consumption and
costs. In the literature, there is different research on the topic of heating in smart homes.
The authors of [21] develop a smart heater system characterized by high performance and
low cost. It functions due to an open-source controller specially programmed with software
that cooperates with the temperature and humidity sensor. The solution function via the
cloud server that presents values measured using special application for smartphones.
Due to this, users are able to remotely control the temperature (and the whole heating
system) and schedule tasks via an application. One of the biggest advantages is that the
solution reduces the risk of fire outburst. According to the authors, this solution provides
safety, positively influences costs, and increases efficiency, performance, and useability. The
authors of the paper [22] propose a different strategy of heating control, emphasizing the
integration of user presence as an additional factor in managing heating within the living
space. They present a flexible heating control system that adapts to diverse occupants
within the heating network. Their method involves predicting heating dynamics using a
multilayer perceptron neural network based on time series data. To manage the heating
controller effectively, they employ a fuzzy inference system utilizing the Takagi–Sugeno
model. In another paper [23], the authors present the IoT prototype that introduces a
smart control approach named the smart token-based scheduling algorithm. It aims at
optimizing energy in buildings’ heating systems. The solution includes, among others,
special hardware, software, networking. Other authors [24] present a novel residential
energy management approach that reduces electricity consumption for space heating and
battery connected to the grid. It achieves this without relying on pricing signals. The
method employs a unique algorithm, integrating seasonal calculations and considering
various factors such as temperature and photovoltaic (PV) generation.
In homes, smart thermostats with the function of real-time sensing are often used.
They help minimize energy consumption and at the same time they maintain user comfort.
There are different types of thermostats, for example, in residential buildings there are
usually managed by reactive and heuristic-driven ones or by more advanced controls [25].
Advanced control systems in residential buildings have typically relied on either model-
based approaches like model predictive control (MPC) [26] or model-free methods such as
reinforcement learning (RL) [27,28] for their development. Today, more and more research
is related to the development and scope of thermostat operation. Marantos et al. [29] in
their research apply a smart thermostat concept which focuses mainly on cost reduction and
deployment flexibility so they can be adopted on a big scale in many buildings and regions.
Its idea integrates supervised and reinforcement learning to address the challenge of
meeting occupants’ thermal comfort requirements while minimizing energy usage. Goman
and Koroloev [30] present the newest achievements on smart thermostats in and smart
buildings. These devices can contribute to energy savings and control. The authors also
describe prospects for the future which are self-learning algorithms for smart thermostats.
According to them, it seems possible to develop a self-learning smart thermostat that is
able to support a big building. Thermostats should follow and analyze user behavior
and effect control at the level of a building and single rooms. Huang et al. [31] describe
the work and function of a self-learning algorithm to predict indoor temperature and
cooling demand from a smart Wi-Fi thermostat in a residential building. The dynamic
model provided for any residence can be applied to guide residents when it comes to
Energies 2023, 16, 7988 5 of 34

energy savings coming from set point schedule switches. Broader practical research is
presented by Duman et al. [32] as the study combines a smart thermostat with a home
energy management system (HEMS) for cost-effective load scheduling, demand response
(DR), and photovoltaic self-consumption. The thermostat uses fuzzy logic to adjust set
points based on electricity prices, solar radiation, and occupancy.
On the other hand, smart homes have cooling systems. Smart cooling systems use
sensors to monitor temperature and humidity levels in various parts of the home and can
automatically adjust the cooling settings to maintain the desired comfort level. These sys-
tems can also be controlled remotely via smartphones or voice-activated devices, allowing
homeowners to customize their cooling preferences even when they are away from home.
Additionally, AI-driven cooling systems can learn user preferences and adapt to daily
routines, making them more energy-efficient by cooling or ventilating specific areas only
when needed. It is worth noting, after [33], the significance of passive adaptive systems
in the context of mode-switching for cooling and heating. Passive adaptive systems stand
out for their ability to respond to environmental cues, such as changes in temperature or
humidity, and autonomously shift between cooling and heating modes. Their capacity
to adapt naturally, without requiring manual intervention, is a notable advantage. This
not only streamlines their operation but also enhances their efficiency by ensuring that
they can adjust to varying conditions, making them a promising choice for energy-efficient
and user-friendly cooling and heating solutions. As we read in the paper of Daneshvar
et al. [34], one of the innovative solutions within this topic is to provide a new cooling
control approach as an element of the smart energy system that can achieve a balance
between thermal comfort and building energy usage through the utilization of sensing
and machine programming technology. To achieve this purpose, an overall form of a
building must be coupled with this smart system, while the energy use with the thermal
comfort cooling of people must be provided based on the special dedicated software. On
the other hand, Nezhad et al. [35] present a new model for home energy management,
considering inverter-based air conditioning and solar panels. The model aims to minimize
daily electricity costs using time-of-use tariffs. It includes fixed and flexible loads and is
formulated as a mixed-integer linear programming problem. The system uses a PV system
and electrical energy storage to handle unpredictable solar power generation and optimize
load management during peak hours. The air conditioning settings are adjusted based on
an indoor-outdoor temperature model to reduce energy consumption and lower bills.
In the area of home heating and cooling, there is a very large potential for the use of
artificial intelligence. For this reason, the authors have developed key directions for the
future usage of AI in this area. Table 2 shows these key directions.

Table 2. Key directions for future usage of AI for heating and cooling in homes.

Direction Description
A focus on developing AI systems that prioritize user comfort and preferences, learning and
Human-Centric AI
adapting to individual user habits, creating personalized and user-centric experiences.
Research into advanced control algorithms, including reinforcement learning and predictive
Advanced Control Algorithms models, to optimize energy consumption and enhance user comfort, adapting to changing user
behavior and environmental conditions.
The development of standardized protocols and interfaces for seamless integration of various
Integration AI-driven systems within smart homes, enabling better synergy and coordination
between systems.
Investment in the development of more sophisticated IoT devices and sensors for enhanced data
IoT and Sensor Technologies collection and improved AI system performance, including better occupancy detection,
environmental monitoring, and energy usage tracking.
Energies 2023, 16, 7988 6 of 34

Table 2. Cont.

Direction Description
Exploring methods to integrate energy storage solutions, such as batteries, with heating and
Energy Storage Integration cooling systems to maximize the utilization of renewable energy sources (RES) and reduce
grid dependence.
Providing AI-driven solutions accessible to a wide range of households, regardless of their size,
Accessibility
location, or economic status, promoting inclusivity and adoption.
Investment in long-term research into the integration of renewable energy sources, such as solar
Sustainability and wind, with residential heating and cooling systems to promote sustainability and reduce the
reliance on fossil fuels.
Source: Authors’ own work based on: [21,23,24,33–39].

3.2. Lighting
Intelligent lighting control in a smart home context involves the use of advanced
technologies and automation to manage and optimize lighting systems. Smart homes
incorporate sensors, IoT devices, and AI to enhance the lighting experience for residents.
These systems can automatically adjust lighting levels based on factors such as occupancy,
natural light availability, and time of day. Like in the case of smart heating, users can
remotely control their lighting via smartphone apps or voice commands. Moreover, in-
telligent lighting control in smart homes contributes to energy efficiency, making homes
more sustainable, and improves the overall living experience by offering convenience
and personalization. Smart lighting refers to lighting technology with an increased level
of functionality such as dimming or remotely control the on/off button to enhance user
comfort and save energy [40].
The authors of the paper [41] propose an intelligent lighting system designed for office
environments, emphasizing cost-efficiency and environmental sustainability. The system is
built around an Arduino microcontroller, infrared inductive sensors, and light sensors, with
communication facilitated through a Wi-Fi network. The system’s key functions include
automatic control of delay, turn-off, and dimming of office lighting fixtures based on sensor
readings, thereby enabling real-time detection and adjustment of the office lighting envi-
ronment. By integrating these components, the system offers low-cost and environmentally
friendly features tailored for office settings. Multiple sensors are strategically placed in
the room and work in conjunction with the Arduino controller to detect various indoor
parameters. The Arduino microcontroller acts as a data hub, receiving, processing, and
transmitting the collected data through the communication network. It further displays
this information on an LCD screen and uploads it to a client data center. One of the notable
features of this intelligent lighting system is its automatic brightness adjustment, allowing
users to achieve optimal lighting conditions at any given time. Moreover, users can re-
motely control lighting switches, adjust brightness, and activate predefined lighting scenes
using a mobile application. This instantaneous feedback mechanism not only conserves
energy when people leave the room but also enhances efficiency and provides users with
real-time, comfortable illumination. In another paper [42], a comprehensive and functional
prototype of an IoT-based lighting control system is introduced, with a primary focus on
balancing natural and artificial lighting while incorporating a dynamic shading system.
It provides a scalable approach to smart system integration within buildings, relying on
sensing and actuating nodes (Arduino-driven) and a central unit (Raspberry Pi-driven). A
dedicated control application is developed, allowing users to interact with the system by
configuring automatic seasonal modes or manual settings. The system is designed to adjust
the required illuminance threshold, with the shading system aligning itself with seasonal
profiles based on bioclimatic design principles. Notably, the control system incorporates
a fuzzy logic solution to ensure fast and responsive control without high computational
demands. The overarching goals of this work encompass the development of a shading
system for internal daylight intensity control, the design and construction of a versatile LED
lighting system, the integration of shading and lighting into a unified prototype, the explo-
Energies 2023, 16, 7988 7 of 34

ration of Arduino and single-board computers like Raspberry Pi for IoT control, and the
creation of a user-friendly mobile application for configuring seasonal lighting and shading
modes, illuminance thresholds, and manual settings. In another paper [43], the authors
present a system designed to control lighting and electrical loads that takes advantage of
embedded systems equipped with cost-effective wireless communication modules, which
makes it suitable for a distributed and intelligent home automation architecture. The sys-
tem integrates a range of sensors that facilitate the efficient use of electricity by automating
tasks such as turning off lights and electrical devices while maintaining lighting regulation.
Multiple modules communicate wirelessly with a central node, and user interaction is facili-
tated through a mobile application. The development and validation of the system involved
using UML and Petri nets for design and modeling, while the implementation was carried
out in C/C++ for 32-bit microcontrollers. Testing of the prototype demonstrated stable
performance, fast communication, and sufficient coverage for a typical single-family house.
Remarkably, the system’s performance surpasses that of similar solutions found in the sci-
entific community. The paper of Cho et al. [44] introduces a novel concept of lifelog-based
smart lighting control, aiming to personalize lighting environments to match individual
characteristics. While the potential for lifelogs to enable personalized lighting has long been
recognized, the lack of data collection and synthesis methods has been a barrier. In this
study, the authors propose lifelog data collection methods and an analytical approach to
recommend custom lighting environments. They deploy sensors, lighting controllers, and
control interfaces in a mock-up space and connect these to a machine learning server in the
cloud. The platform they establish utilizes emotional, activity, and environmental lifelog
data to create a truly personalized lighting experience. This innovative approach opens
new possibilities for enhancing user comfort and well-being through lighting controls. It is
worth mentioning also the results of the research of Ayan and Turkey [40] on smart LED
bulbs as parts of smart lighting systems. This research presents the influence of using these
bulbs on energy efficiency. The subject of the study was power consumption depending on
different colors emitted by these bulbs. According to the results, the different colors have
different power usage. In addition, case studies including detailed comparison between
(1) halogen, (2) CFL, (3) LED, and smart LED were considered in the context of energy
savings. It was proven that a smart LED consumes the least energy among other bulbs but
solely when dimmed and under remote control.
Different scenario for lights, but also at the same time for alarms, is presented by
Ozeer et al. [45]. The authors describe two key user stories that have been implemented as
part of their study. The first scenario, termed the “Bedtime Scenario,” involves a button
press in the bedroom, signaling the house tenant’s intention to go to bed. Subsequently,
all lights are turned off and the alarm is set. If any motion is detected in the living room
or the kitchen or if the door is opened, an alarm is triggered on a speaker, and the house’s
lamps are illuminated in red to indicate potential intruders or disturbances. The second
scenario, referred to as the “Welcome Home Scenario,” is initiated when the home tenant
arrives at the front door. The Wemo motion sensor reports motion, and, in response, the
lamp at the entrance is turned on, letting the person to unlock the door. When entering
the house, the living room lamp is also turned on and a welcoming sound is played on
the speaker to create a welcoming atmosphere. The paper suggests that these scenarios
are only the beginning and that more complex scenarios can be orchestrated by their
system based on patterns of events sensed and actuated, offering enhanced automation and
convenience for users. Another paper [46] also presents the solution that combines smart
lighting with other functions of a smart home. The framework proposed by the authors
has versatile applications, including automated burglar alarm systems, guest attendance
monitoring, and light switches. With the use of IoT solutions, these systems enable real-
time monitoring and connectivity to central systems for automated burglar alarms. The
monitoring framework is designed as a web application, providing real-time display,
storage, and alerting functions for both local and remote monitoring control. Importantly,
the monitoring system is described as stable and reliable when utilizing the SHA-256
Energies 2023, 16, 7988 8 of 34

authentication method. The system comprises three core components: hardware nodes, a
secure server, and a web application. The IoT node hardware is designed for real-world
testing and receiving IoT data from diverse devices. A dedicated server is established to
monitor the IoT nodes within the system. Lastly, a user-friendly application is developed,
Energies 2023, 16, x FOR PEER REVIEW
accessible via smartphones or web browsers over Wi-Fi, enabling real-time control of the
IoT smart system. Figure 1 shows the block diagram of the proposed smart lighting system.

Figure 1. Block diagram of the smart lightning system—own work based on [46].

Summing up, when

Figure it comes
1. Block to empirical
diagram findings
of the smart related
lightning to specificwork
system—own implementations
based on [46].
of AI models for lighting control within residential settings, the literature [40–42,44–47]
presents the following: sensor-based
Summing adaptive
up, when lighting,
it comes naturalfindings
to empirical light integration,
related topersonal-
specific impleme
ized lighting scenes, dynamic lighting control, voice-controlled lighting systems,
of AI models for lighting control within residential settings, the literature predictive
[40–42,44–
lighting adjustments,
sents and context-aware
the following: lighting. These
sensor-based adaptive examples
lighting, showcase how integration,
natural light AI mod- perso
els can be practically implemented
lighting scenes, dynamicin lighting controls
lighting within
control, residential environments,
voice-controlled lighting systems, pr
offering diverse lighting
functionalities to enhance
adjustments, energy efficiency,
and context-aware user These
lighting. comfort, and theshowcase
examples overall how AI
living experience.can be practically implemented in lighting controls within residential environment
It is also worth explaining
ing diverse in detail one
functionalities of the mentioned
to enhance examples.
energy efficiency, userThe “Bedtime
comfort, and the overa
Scenario” [45] inexperience.
the smart home application for light automation and intrusion detection
involves several software entities
It is also worthand devices working
explaining in detailtogether
one of the tomentioned
facilitate actions based
examples. The “Bedti
on specific triggers:
nario” [45] in the smart home application for light automation and intrusion detec
• Button press: whenseveral
volves a button in the bedroom
software entities andis pressed,
devicesthe “NodeHueSense”
working softwareactions b
together to facilitate
entity reports this event
specific triggers:from the Hue buttons and publishes it on the MQTT bus.
• Event subscription:
• the “Orchestrator”
Button software
press: when a button insubscribes
the bedroom to all events published
is pressed, in the
the “NodeHueSense” s
MQTT Broker. In this scenario, it identifies the button press event from the bedroom.
entity reports this event from the Hue buttons and publishes it on the MQTT b
• Scenario definition: the Orchestrator defines a specific scenario based on this event,
• Event subscription: the “Orchestrator” software subscribes to all events publ
triggering a sequence of actions in response. It sends messages to other software
the MQTT Broker. In this scenario, it identifies the button press event from t
entities according to the predefined scenario.
room.
• Light control: the “NodeHueActuate” entity receives messages from the Orchestrator
• Scenario definition: the Orchestrator defines a specific scenario based on thi
to control the Hue lamps. In the bedtime scenario, it turns off all the lights in the house.
triggering a sequence of actions in response. It sends messages to other softw
• Alarm setting: the scenario triggers an alarm setup, which might involve sending a
ties according to the predefined scenario.
signal to the “SoundPlayer” entity, instructing it to activate the speaker and set an
• Light control: the “NodeHueActuate” entity receives messages from the Orch
alarm sound.
to control the Hue lamps. In the bedtime scenario, it turns off all the lights in th
• Intrusion detection: if any motion is detected in certain areas or if the door is opened,
• Alarm setting: the scenario triggers an alarm setup, which might involve se
the “FibaroAdapter” and “WemotionSense” report these events on the MQTT bus.
The Orchestrator,signal
basedtoonthe “SoundPlayer”
this information, can entity, instructing
trigger an alarmiton to the
activate
speakertheand
speaker and
alarm sound.
illuminate the house’s lamps in red as a signal of potential intruders or disturbances.
• Intrusion detection: if any motion is detected in certain areas or if the door is
This scenario showcases the coordination among various software entities and devices,
the “FibaroAdapter” and “WemotionSense” report these events on the MQTT b
orchestrated by the “Orchestrator” based on specific events, such as the button press,
Orchestrator, based on this information, can trigger an alarm on the speaker a
minate the house’s lamps in red as a signal of potential intruders or disturbanc
This scenario showcases the coordination among various software entities and
orchestrated by the “Orchestrator” based on specific events, such as the button p
automate actions like turning off lights, setting alarms, and responding to potenti
Energies 2023, 16, 7988 9 of 34

to automate actions like turning off lights, setting alarms, and responding to potential
intrusions or disturbances.
For the topic of lighting and AI, the authors have developed the key directions. They
are presented in Table 3.

Table 3. Key directions for future usage of AI for lighting in homes.

Direction Description
Development of lighting solutions aligned with rhythms and individual user
Human-Centric AI preferences (including for example different lighting for bedtime and welcome
home routines).
Designing intuitive mobile applications and voice-activated controls to enable users to
User-Friendly Control Interfaces
customize their lighting environments with ease.
Development of user-friendly web applications and remote-control options to provide
Web Applications and Remote Control
users with convenient access to their smart lighting systems from various devices.
Further development aimed at integration of AI-driven smart lighting systems with
Integration
other smart home devices.
Advancements in AI algorithms that adapt lighting conditions based on user
Personalized Lighting Experiences
preferences, promoting enhanced user comfort and well-being.
Continued investment in sensor and IoT technologies to enhance occupancy detection,
Sensing and IoT Advancements light level adjustments, and environmental monitoring, resulting in more responsive
and energy-efficient lighting control.
Source: Authors’ own work based on: [40–42,44–47].

3.3. Windows and Blinds

Another element of smart homes are smart windows. Smart windows for homes are
an innovative technological solution designed to improve energy efficiency, comfort, and
convenience [48]. These windows are equipped with various features and technologies that
allow them to adapt to changing environmental conditions and user preferences. Special
systems for smart windows offer a way to improve energy efficiency in the construction
sector, in residential or commercial buildings. These systems can dynamically adjust the
spectral properties of window glazing, controlling how it interacts with visible and infrared
light [49]. This adaptability allows better management of solar radiation, resulting in signif-
icant energy savings, especially in regions where cooling is a primary concern. Additionally,
smart windows optimize the use of natural daylight within buildings, improving visual
comfort. In the literature, there is not so much research on smart windows; however, there
are some studies on electrochromic materials used in smart windows. Electrochromic glass
operates as an electric battery with thin films of specific materials [49]. The level of optical
transparency in this glass is similar to the charge level in a battery. An essential element in
electrochromic devices is the electrolyte, which conducts ions while insulating electrons.
Electrolytes in these systems can be in liquid, gel, or solid form. Liquid electrolytes can
be prone to leakage or evaporation if they contain solvents. Common ions transported
through the electrolyte when an external voltage is applied are typically hydrogen and
lithium, with occasional use of sodium. The electrochromic materials in these devices are
predominantly transition-metal oxides and organic compounds. This material and function
of smart windows is studied, for example by Ke et al. [50], Wang et al. [51], and Zhang
et al. [52]. Some papers discuss smart hydrogel windows, such as [53–56]. In the literature,
there are also review papers on the topic of smart windows, such as Aburas et al. [57],
Nundy [58], and Tällberg et al. [59].
In terms of smart windows, it is justified to mention smart blinds. Smart blinds are
window coverings that can be remotely controlled, typically through a smartphone app or
voice commands. They offer convenience by allowing users to adjust the blinds’ position,
either opening or closing them, without needing to be physically present. Smart blinds can
integrate with smart home systems, enabling automated schedules based on time of day or
sunlight conditions. Some models are also capable of blocking out light to enhance privacy,
regulate room temperature, and save energy. They provide a modern and efficient solution
Energies 2023, 16, 7988 10 of 34

for light control in homes. In practice, however, when it comes to recent years, there are
even fewer articles in the literature about intelligent blinds than articles about intelligent
windows. In an example, in the paper of Jung et al. [60], smart windows combined with
photovoltaics blinds and a ventilation system was proposed. Similar research on PV blinds
are discussed in [61,62].
The key directions developed by the authors for usage of AI for blinds and windows
are presented in Table 4.

Table 4. Key directions for future usage of AI for blinds and windows in homes.

Direction Description
Advancements in AI-based technologies to improve the energy efficiency of smart
Energy Efficiency windows and blinds, enabling better management of solar radiation and enhancing
insulation, especially in regions with varying climate conditions.
Development of intuitive user interfaces, such as smartphone applications and
User-Friendly Control Interfaces voice-activated commands, to improve user convenience and control over smart
blinds and windows.
Enhancing remote control features for smart blinds, ensuring that users can adjust
Web Applications and Remote Control them even when they are away from home, contributing to energy savings
and security.
Research on integration of smart windows and blinds with broader smart home
Integration
systems to allow synchronized automation
Ongoing research on electrochromic materials for smart windows, exploring their
Advancements in Materials properties, durability, and environmental impact, with the aim of making them more
accessible and effective.
Research into the integration of photovoltaic blinds with smart window systems to
PV
harness solar energy and improve energy sustainability in buildings.
Source: Authors’ own work based on: [48,49,57–62].

3.4. Home Devices–Refrigerators

According to Wang et al. [63], daily, a significant amount of food is needlessly dis-
carded due to prolonged storage in refrigerators, resulting in environmental strain. Re-
frigerators themselves are energy-intensive appliances, and minimizing refrigerator door
openings could substantially contribute to environmental preservation. Hence, the authors
in their research use the (r,n)-threshold SIS scheme to develop picture-sharing technology
and introduce a series of smart refrigerator designs. This incorporates food identifica-
tion technology, enabling users to promptly identify available ingredients on the smart
fridge’s display. This innovative approach minimizes food waste and conserves energy
consumption. Other authors [64] in their research proposed a hardware upgrade for smart
home refrigerators. They designed a Wi-Fi-enabled main control board that maintains
compatibility with existing components. They implemented a simple learning algorithm on
the microcontroller to optimize system efficiency while ensuring food safety temperatures
or user preferences. Additionally, they introduced a wireless sensor node to provide accu-
rate food temperature data for precise monitoring and control. The results demonstrate
a significant efficiency improvement, with an enhancement in the refrigeration cycle and
an improvement in the defrost cycle. Another paper [65] analyzed the energy efficiency
of IoT-controlled refrigerators. The authors explored the impact of different temperature
hysteresis bands and the presence of internal products on the energy consumption of house-
hold refrigerators using an IoT-controlled system. It reveals that larger hysteresis bands
lead to increased energy use when the fridge is empty, but the opposite is true when there
is an internal product (more thermal mass). By choosing the right hysteresis band based on
the product’s characteristics, potential energy savings of up to 20% are achievable. Achiev-
ing this efficiency requires a real-time adaptation scheme that identifies the refrigerator’s
underlying dynamics and control, as described in the paper. For larger-scale refrigeration
systems, such as those in superstores, employing IoT as demonstrated in the paper enables
candidate algorithms for demand-side management (DSM) and real-time hysteresis band
Energies 2023, 16, 7988 11 of 34

adjustments to accommodate product mass changes. Some other authors from the study
from this year (2023) conducted a SmartFridge project as a study toward environmental
sustainability and the economy. In their paper [66], they evaluated a refrigerator model
in terms of its energy consumption, considering factors such as indoor temperature and
moisture. The results of their study show that as the indoor temperature increases, energy
consumption increases. Moisture has a minor impact on the energy use of the refrigerator,
according to simulation data. The choice of interior temperature also significantly affects
energy consumption. Table 5 shows the key directions for AI-driven refrigerators.

Table 5. Key directions for future usage of AI for refrigerators.

Direction Description
Developing energy-efficient algorithms and control boards to optimize refrigerator
performance, reduce energy consumption, and lower environmental impact.
AI can optimize the cooling cycles of the refrigerator based on factors such as the
Energy Efficiency
outside temperature, usage patterns, and the contents of the refrigerator. This
dynamic adjustment ensures that the refrigerator does not work harder than necessary,
reducing energy consumption.
Exploring IoT-controlled refrigerators to optimize energy use based on the presence of
IoT-Enabled Temperature Control
internal products.
Smart refrigerators use AI to monitor their own performance and detect early signs of
malfunctions or maintenance issues. By identifying problems in advance, they can
Predictive Maintenance
schedule repairs or maintenance during low-demand periods, preventing sudden
breakdowns that might lead to energy waste
Smart refrigerators can be integrated into larger home energy management systems,
Integration allowing homeowners to coordinate the operation of various smart appliances, heating
and cooling systems, and lighting to maximize energy efficiency throughout the home.
PV Research on the development of solar PV-powered refrigerators.
Source: Authors’ own work based on [63–69].

3.5. Energy Management Systems

In residential buildings, AI-driven systems are becoming increasingly important for
managing energy consumption. As homes consume more electricity and incorporate
distributed energy sources, there is a growing focus on optimizing the expenses related to
purchasing electricity for household users. There are different systems and approaches to
energy management in homes described in the literature. Some authors name these systems
a smart energy management system (SEMS), some call it a home energy management
system (HEMS), and others just call it an energy management system (EMS), as it is
described below. Figure 2 shows an example of a home energy management system [70].
The aim of the system with various household appliances is to optimize their energy usage.
This system is connected to a service provider through a bidirectional communication
network, enabling the exchange of pricing and appliance energy consumption data. The
EMS receives hourly pricing information from the service provider and adjusts the energy
consumption of each appliance in response to these prices. Typically, household electric
appliances are categorized into three main types based on their characteristics and priorities:
non-shiftable, shiftable, and controllable loads.
The previously mentioned paper of Duman et al. [32] also present the study of a
home energy management system (HEMS). The HEMS schedules time-shiftable loads,
including battery storage and electric vehicles, while considering bi-directional power
flow and battery degradation. It is worth noting that electric cars become more and more
popular when it comes to research [71–73], and, therefore, homes also have started to
have charging stations for them. Simulation results indicate a cost reduction under time-
of-use and a reduction in air conditioning (AC) costs. Homeowners input occupancy,
AC settings, appliance usage windows, hot water schedule, and electrical vehicle (EV)
charging preferences. The utility sends energy prices, demand response, and weather
forecasts. A smart thermostat adjusts AC settings based on solar radiation, electricity
Energies 2023, 16, 7988 12 of 34

prices,
Energies 2023, 16, x FOR PEER REVIEW and occupancy. The home energy management system optimizes load scheduling,
12 of 34
considering bidirectional power flow and battery health to minimize costs and enhance
self-consumption.

Figure 2. An example of home energy management system—own work based on [70].

Figure 2. An example of home energy management system—own work based on [70].
Ma et al. [74] describe a HEMS model designed to optimize the energy usage in homes.
The previously mentioned paper of Duman et al. [32] also present the study of a home
It takes into account various factors including household devices, distributed energy
energy
sources, management
energy storage, system and (HEMS).
electric The HEMS
vehicle schedules
charging. HEMS time-shiftable loads, including
is a self-regulating system
battery
that canstorage
adapt and electricin
to changes vehicles,
electricitywhile considering
prices bi-directional
and consumption. power flow
It controls and bat-
the energy for
tery degradation.
different devices suchIt is worth
as air noting that electric
conditioners, watercars become
heaters, more machines,
washing and more popular when
electric vehicle
itchargers,
comes toand research
others. [71–73],
Chauhan and,et therefore, homes also
al. [75] present haveon
research started
the costto have charging
reduction usingsta-a
tions for them. Simulation results indicate a cost reduction under
smart energy management system. SEMS reduces energy waste, cost, and electricity bills time-of-use and a reduc-
tion
withoutin aircompromising
conditioning (AC) user costs.
comfort.Homeowners
To test SEMS, inputa occupancy, AC settings,
residential microgrid appliance
system with
usage
smartwindows,
appliances, hotphotovoltaic
water schedule, and and
panels, electrical
energy vehicle
storage(EV)is charging
evaluated preferences.
under flat The and
utility sends energy prices, demand response, and weather forecasts.
real-time pricing. The results indicate energy savings and a reduction in electricity costs. A smart thermostat
adjusts
The SEMS’sAC settings
innovative based on solar radiation,
scheduling and cost-drivenelectricity prices,management
appliance and occupancy. The home
demonstrate its
energy management system optimizes load scheduling, considering
novelty and effectiveness. It must be noted that PV installations are increasingly popular bidirectional power
flow
among andindividuals
battery health to minimize
[76,77]. The authors costsofand
[78]enhance
presentself-consumption.
the study on a smart energy system
that Mauseset al. [74] describe
machine learning a HEMS model designed
algorithms. to optimizecan
These techniques theautonomously
energy usage inregulate homes.
Itheating
takes andinto hotaccount
water various
systems,factors
which are including household
major energy consumers devices, distributed This
in households. energynot
sources,
only helps energy
reducestorage,
energy and electric
usage, butvehicle charging.
also enhances HEMSfor
comfort is aresidents.
self-regulating
With the system that
growing
can adapt to
adoption of changes
renewable in electricity
energy sourcesprices inand consumption.
homes, coordinatingIt controls
energy theconsumption
energy for differ- with
ent devices such as air conditioners, water heaters,
production becomes crucial for additional savings and reducing peak loads. washing machines, electric vehicle
The authors
chargers,
propose the anddevelopment
others. Chauhan of a et al. [75]
deep present research
reinforcement learningon the
(DRL) costalgorithm
reductionfor using a smart
controlling
energy
indoormanagement
and hot water system. SEMS reduces
temperatures energyenergy
to optimize waste,consumption
cost, and electricity bills without
by leveraging solar
compromising
energy production. user comfort. To test SEMS,
Additionally, they alsoa residential
introducemicrogrid
a methodsystem with smartsetting
for dynamically appli-
ances,
indoorphotovoltaic
temperature panels, and energy
preferences, storage
offering is evaluated
greater under
flexibility and flatenergy
and real-time
savings.pricing. Other
authors
The results[79] introduce
indicate energy a novel approach
savings that combines
and a reduction three distinct
in electricity costs. artificial
The SEMS’s intelligence
innova-
techniques
tive scheduling to address energy demand
and cost-driven applianceplanning
management in smart homes. its
demonstrate Designed
novelty as and a effec-
multi-
objectiveItscheduling
tiveness. must be noted problem,
that PV this method seeks
installations a balance between
are increasingly popularenergyamongcost and user
individuals
comfort.
[76,77]. The Byauthors
utilizing ofan elitist
[78] non-dominated
present the study onsorting
a smartgenetic
energyalgorithm
system that II, ituses
incorporates
machine
demand-side
learning management
algorithms. These that considers
techniques canfactors like electricity
autonomously regulatepriceheating
variations,
and equipment
hot water
priority, which
systems, operational
are major cycles,
energyandconsumers
energy storage. Additionally,
in households. This not it includes
only helps a distributed
reduce en-
generation
ergy usage, forecast
but also using
enhances support
comfortvector
forregression for the
residents. With next
the day. Itadoption
growing is worth mentioning
of renewa-
that there are also many review articles from 2023
ble energy sources in homes, coordinating energy consumption with production (the newest) on the topic of energy
becomes
management systems in homes, for example [80–83]. Table
crucial for additional savings and reducing peak loads. The authors propose the develop- 6 shows the key directions for
AI-driven
ment refrigerators.
of a deep reinforcement learning (DRL) algorithm for controlling indoor and hot water
temperatures to optimize energy consumption by leveraging solar energy production. Ad-
ditionally, they also introduce a method for dynamically setting indoor temperature prefer-
ences, offering greater flexibility and energy savings. Other authors [79] introduce a novel
approach that combines three distinct artificial intelligence techniques to address energy
Energies 2023, 16, 7988 13 of 34

Table 6. Key directions of usage of AI for home energy management systems.

Direction Description
Development of advanced AI algorithms that can optimize energy usage in homes.
Energy Efficiency This includes predicting energy demand, dynamically adjusting energy sources,
and prioritizing energy consumption based on user preferences and real-time data.
Development of AI-driven systems for real-time monitoring of energy
consumption and production within homes. This can involve the use of sensors
Real-Time Energy Monitoring
and IoT devices to gather data and AI algorithms to provide insights
and recommendations.
The use of AI to predict maintenance needs for home energy systems, such as
Predictive Maintenance
heating, cooling, and renewable energy installations.
The development of ways in which AI can facilitate demand response
mechanisms, enabling homes to interact with the broader energy grid more
Demand Response and Grid Integration
intelligently. AI can help homes respond to grid signals and optimize energy
consumption during peak and off-peak hours.
The development of ways in which AI can facilitate the integration of diverse
Energy Source Integration energy sources, such as solar panels, wind turbines, and battery storage, into home
energy systems.
Research on how AI can effectively integrate with human behavior in homes. This
Human-Behavior Integration involves understanding how occupants interact with energy systems and
developing AI solutions that adapt to users’ energy-related habits and preferences.
Source: Authors’ own work based on [32,70,74,75,78–80,84–86].

4. Urban Infrastructure Integration of AI-Driven Energy Solutions

4.1. Electric Vehicle Charging Infrastructure
Artificial intelligence plays an increasingly important role in electric vehicle charging
infrastructure to enhance its efficiency, reliability, and user experience [87]. AI algorithms
can analyze various factors such as the grid load, energy prices, and individual user
preferences to create optimal charging schedules for EVs [88]. This ensures that charging
occurs during periods of lower energy demand or when renewable energy sources are
abundant, reducing costs and the environmental impact [89]. Also, AI can facilitate demand
response programs, allowing EV owners to participate in load-shifting initiatives. During
peak demand periods, AI can coordinate with users to temporarily reduce their charging
rates, alleviating stress on the grid and helping to prevent blackouts [90].
El Husseini et al. [91] discuss the integration of blockchain and AI technologies as a
solution to address these challenges. It suggests that combining these technologies can lead
to a more secure, efficient, and decentralized charging ecosystem. They discuss a couple
of use cases where AI and blockchain technologies complement each other to enhance
the charging infrastructure for EVs [88,92]. These use cases likely illustrate scenarios
where technologies work together to improve security and optimize charging schedules.
According to this research, it is intended to help stakeholders identify potential directions
and implementations for better charging systems for EVs. This implies that the paper aims
to inform decision-makers about the possibilities and advantages of integrating AI and
blockchain in the EV charging infrastructure [93].
Chaihoie et al. [94] points out that to prepare an appropriate predictive model for
charger planning, AI usage is very useful. They described the “predict-then-optimize”
approach, where AI is used to predict the EV charging demand. This prediction is made
using a multi-relation graph convolutional network (GCN)-based encoder–decoder deep
architecture. This predictive model allows for more data-driven planning and allocation of
resources. In the describe approach, AI is also utilized to optimize the competitive resource
allocation strategy for charger planning. This likely involves determining where to place
charging stations and how to distribute them effectively to meet anticipated demand. AI
can be used to address the optimal size of EV chargers, determining the number of chargers
that each service provider should deploy in various areas of the city.
Energies 2023, 16, 7988 14 of 34

Another complex analysis on AI usage in context of electric vehicle charging infras-

tructure was described by Qin and Folly [95]. They pointed out that AI, particularly deep
learning methods like LSTM and GRUs, can be useful for forecasting EV charging and
discharging patterns. The paper highlights the importance of accuracy in forecasting mod-
els, considering the stochastic and unpredictable nature of EV charging patterns. It also
mentions the use of hybrid and ensemble techniques to improve forecast accuracy.
Dynamic pricing strategies are crucial to influence EV owners’ charging and discharg-
ing behaviors. The paper discusses the challenges of existing dynamic pricing models,
such as undervaluing or overvaluing stored battery power. It suggests the need for pricing
models that reflect real-time power system conditions and balance the interests of system
operators and EV owners. The AI can be used to achieve this dynamic pricing approach.
Also, the AI can be useful in the development phase of vehicle-to-grid (V2G). The
development of V2G can be categorized into three phases [96]. In the first phase, the EV
charging load is a small proportion of the power grids, mainly using uncontrolled and
controlled charging strategies. The second phase sees an increased EV charging load with
the high penetration of EVs, necessitating smart charging/discharging control strategies
and aggregator coordination. The third phase envisions a mature state in which many EVs
provide ancillary services to power grids [97].
Another possible application of AI in vehicles charging infrastructure was described
by Mosayebi et al. [98]. The authors describe the need for improved charging infrastructure
for electric vehicles as a result of their increasing numbers worldwide. It introduces the
concept of a smart extreme fast portable charger (SEFPC) for EVs with multiple input
sources, including the power grid and renewable energy sources such as an energy storage
system (ESS). The SEFPC is designed to optimize the charging process by considering
available power sources and the condition of the EV battery to save energy and time. A
machine learning algorithm, based on IT, specifically a model-free sliding mode controller,
is applied to determine the optimal charging operation mode based on the state of the
battery and power source conditions. This approach aims to enhance battery life and
overall system efficiency. The text concludes by mentioning real-time results obtained
using the OPAL-RT platform to validate the effectiveness and feasibility of the SEFPC and
the model-free sliding mode controller.
The very important problem in the case of charging infrastructure is connected with
the importance of effectively placing charging stations to support the growth of EVs and en-
hance the traffic network’s efficiency. Existing research often focuses on EV users’ mileage
anxiety but overlooks their strategic and competitive charging behaviors. According to
Lazari and Chassiakos et al. [99], to address this issue, the concept of charging cost for an EV
user can be introduced, considering factors such as the cost of traveling to access charging
stations and the cost of queuing at charging stations. The problem can be formulated as
the charging station placement problem (CSPP), initially as a bilevel optimization problem.
It then leverages the equilibrium of the EV charging game to convert the problem into a
single-level optimization task, proposing the “Optimizing eleCtric vEhicle chArging sta-
tioN” (OCEAN) algorithm for optimal charging station allocation. Recognizing OCEAN’s
scalability limitations, a heuristic algorithm based on AI called OCEAN can be used with
continuous variables to handle large-scale real-world scenarios. The results of extensive
experiments demonstrate that their approach significantly outperforms baseline methods
in addressing the competitive and strategic charging behaviors of EV users.
Artificial intelligence profoundly impacts EV charging infrastructure, optimizing
charging schedules based on grid load, energy prices, and user preferences. Studies
propose integrating AI with the blockchain to enhance the security, efficiency, and decen-
tralization of EV charging ecosystems. Predictive AI models aid in anticipating EV charging
demand, facilitating data-driven planning, resource allocation, and optimal sizing of charg-
ing stations. AI-driven deep learning methods like LSTM and GRUs enhance accuracy
in forecasting EV charging patterns, which is crucial for dynamic pricing strategies and
vehicle-to-grid development. Additionally, the concept of a smart extremely fast portable
Energies 2023, 16, 7988 15 of 34

charger utilizes AI to optimize charging considering various input sources and EV battery
conditions. In Table 7, there is a comparison of the advantages and disadvantages of AI
usage in electric vehicle charging infrastructure.

Table 7. Advantages and disadvantages of AI usage in electric vehicle charging infrastructure.

Advantages Disadvantages
Efficient charging scheduling to reduce grid
Initial setup and integration costs can be high.
strain during peak hours.
Accurate range prediction for improved Dependence on AI technology, which may
trip planning. have downtime or errors.
Privacy concerns related to data collection and
Reduced grid congestion and load balancing.
monitoring of user behavior.
Smart charging infrastructure for a better Potential job displacement in traditional
user experience. charging station maintenance.
Concerns about cybersecurity and
Energy cost optimization for cost savings.
data protection.
Need for continuous updates and maintenance
Battery management to extend battery life.
of AI systems.
Predictive maintenance to reduce the Possible resistance or skepticism from users
downtime of charging stations. unfamiliar with AI technology.
Environmental impact and sustainability
Adaptive charging rates for efficient charging.
concerns related to energy sources.
Improved user experience with real-time Challenges of integration in existing
information and remote management. infrastructure and grid systems.
Grid integration for V2G services and Complexity in regulating and standardizing AI
grid stability. usage in the industry.
Source: Authors’ own work on basis: [87–89,91,95,98,100–108].

The most important direction for the future usage of artificial intelligence in electric
vehicle charging infrastructure are summarized in Table 8.

Table 8. Key directions for future usage of AI in electric vehicle charging infrastructure.

Direction Description
Integrate AI with the smart grid to balance energy supply and demand, optimize
Smart Grid Integration
charging schedules, and support bidirectional charging for grid stability.
Use AI to identify optimal locations for new charging stations based on traffic patterns,
Dynamic Charging Station Placement
EV adoption rates, and local energy infrastructure.
Using artificial intelligence for predictive maintenance of charging stations to reduce
Predictive Maintenance downtime and ensure reliable service, including monitoring components such as
connectors and power electronics.
Develop AI-driven apps and services that offer personalized charging
User-Centric Charging Services
recommendations, payment solutions, and real-time station availability information.
Energy Management and Implement AI to manage energy costs, ensuring that charging stations use electricity
Cost Optimization at the most cost-effective times while considering renewable energy sources.
Enable V2G capabilities with AI to allow EVs to feed surplus energy back to the grid,
Vehicle-to-Grid (V2G) Integration
reducing peak demand and earning rewards for vehicle owners.
Create AI-powered solutions for fleet managers to optimize charging schedules,
Fleet Charging Solutions
monitor vehicle health, and reduce operational costs of electric vehicle fleets.
Establish AI-driven standards that ensure interoperability between different charging
Interoperability and Standardization
networks, vehicle models, and manufacturers, promoting EV adoption.
Improve DC fast charging technology with AI to manage high-power charging,
AI-Enhanced DC Fast Charging
battery safety, and thermal management for shorter charging times.
Incorporate energy storage systems at charging stations and use AI to manage energy
Energy Storage Integration
flow, enhancing the resilience of the charging station and grid support.
Energies 2023, 16, 7988 16 of 34

Table 8. Cont.

Direction Description
Implement AI algorithms for adaptive load management that balance energy
Adaptive Load Management
distribution among multiple charging stations, minimizing grid strain.
Analyze user behavior with AI to understand charging patterns, preferences, and peak
User Behavior Analytics
usage times to optimize station planning and energy management.
Use AI for real-time monitoring of the electric grid’s health, identifying vulnerabilities
Real-time Grid Health Monitoring
and proactively addressing issues to ensure charging station reliability.
Develop AI models to assess the environmental impact of EV charging infrastructure
Environmental Impact Assessment
and inform decisions regarding its expansion and sustainability.
Enhance cybersecurity with AI to protect charging stations from hacking and fraud,
Security and Fraud Detection
protecting user data and financial transactions.
Utilize AI for educational campaigns and awareness initiatives to inform the public
Education and Awareness Initiatives
about the benefits of EVs and the accessibility of the charging infrastructure.
Source: Authors’ own work based on: [87–89,91,95,96,98,100–109].

4.2. Vehicle Emission Reduction

One of the promising technological advances in this effort is the application of artificial
intelligence to mitigate vehicle emissions. AI is revolutionizing the automotive industry
by enhancing the efficiency of conventional vehicles, accelerating the adoption of electric
and hybrid vehicles, and optimizing traffic management. This two-page essay explores the
multifaceted role of AI in vehicle emissions reduction [110].
One of the primary ways that AI contributes to vehicle emission reduction is by opti-
mizing the performance of traditional internal combustion engines. AI algorithms are inte-
grated into the engine control systems, enabling real-time monitoring and the adjustment
of parameters to minimize emissions while maintaining efficiency [111]. These AI-driven
systems take into account factors such as engine temperature, load, and fuel–air mixture
to ensure that combustion is as clean and efficient as possible. By constantly adapting to
changing driving conditions, these AI systems can significantly reduce harmful emissions.
According to Zhao et al. [112], the research of electric vehicles equipped with artificial
intelligence has the capacity to significantly mitigate air pollution and carbon emissions. AI
assistance enables these vehicles to operate more efficiently and make real-time decisions,
contributing to a cleaner environment. By optimizing energy consumption and reducing
the carbon footprint, AI-assisted electric vehicles offer a promising solution to combat the
environmental challenges associated with conventional vehicles.
The rapid growth of electric and hybrid vehicles is a key strategy for reducing emis-
sions [113]. AI plays an important role in the development and operation of these cleaner
alternatives. For instance, AI is used to manage the power distribution in hybrid vehicles,
deciding when to use electric or gasoline power based on driving conditions. It optimizes
battery performance, expanding the range of electric vehicles [114].
Abduljabbar et al. [115] state that artificial intelligence is seen as a well-suited solution
to address the complex challenges faced by transportation systems, including growing
travel demands, rising CO2 emissions, safety issues, and environmental degradation. These
challenges are a direct result of the continuous expansion of traffic, both in rural and
urban areas, driven by population growth, especially in developing countries. For instance,
in Australia, the cost of congestion is projected to rise to 53.3 billion as the population
increases to 30 million by 2031. In Melbourne, Australia, alone, over 640 km of arterial
roads experience congestion during peak hours, leading to an annual CO2 emission of
2.9 tons.
AI-driven eco-driving assistants provide real-time feedback to drivers on how to
optimize their driving habits for better fuel efficiency and lower emissions [116]. These
systems analyze data from various vehicle sensors, including engine performance, speed,
and fuel consumption, to advise drivers on the most fuel-efficient speeds, optimal gear
shifting points, and efficient acceleration and braking patterns [117]. By following these
recommendations, drivers can significantly reduce their carbon footprint, and over time,
Energies 2023, 16, 7988 17 of 34

this has a collective impact on emissions reduction and ensures a seamless transition
between electric and internal combustion modes [118]. These assistants also incorporate
route optimization, suggesting the most efficient routes to reach a destination. By avoiding
traffic congestion and stop-and-go driving, vehicles can operate more efficiently, resulting
in reduced emissions [119].
Delnevo et al. [120] explored the integration of big data and machine learning to
forecast when the friction brake will be activated. The objective is to enhance energy
efficiency in electric vehicles, raise driver awareness, and alleviate concerns related to
‘range anxiety.’ Subsequently, the in-vehicle human–machine interface can take advantage
of these real-time predictions to provide drivers with more precise and comprehensive
insights into their braking habits, ultimately promoting eco-friendly driving practices.
Also, an AI-based solution can be useful in optimizing traffic flow and reducing
congestion, which, in turn, can lead to emissions reduction. AI-powered traffic management
systems use real-time data from various sources, including traffic cameras, sensors, and
smartphones, to analyze traffic patterns [121]. These systems can adjust traffic signals,
suggest alternate routes, and even implement dynamic toll pricing to reduce congestion
during peak hours [122].
In the 21st century, numerous researchers [110,113,122–124] are striving to establish a
more reliable transportation system that minimizes its impact on people and the environ-
ment, while remaining cost-effective and efficient through the application of AI techniques.
AI holds significant promise for enhancing various aspects of transportation, including
road infrastructure, driver assistance, road user experience, and vehicle operation.
Summing up, artificial intelligence demonstrates a pivotal role in curbing vehicle
emissions by optimizing traditional combustion engines in real time. These AI-integrated
systems meticulously adjust parameters, like engine temperature and fuel–air mixtures, to
substantially minimize harmful emissions while ensuring efficiency, thereby showcasing
promising results in emission reduction. Additionally, AI plays a vital role in enhancing
electric and hybrid vehicles’ efficiency by managing power distribution, extending battery
range, and offering real-time decision-making capabilities, thereby significantly reducing
the environmental impact. AI-enabled eco-driving assistants provide personalized feedback
to drivers, optimizing driving habits for fuel efficiency and lower emissions by advising
on optimal speed, gear shifting points, and efficient acceleration patterns. Furthermore,
AI-based traffic management systems leverage real-time data to optimize traffic flow,
reduce congestion, and subsequently cut down on emissions during peak hours. These
implementations underscore AI’s effectiveness in mitigating emissions and optimizing
transportation systems.
Some key directions for the future usage of artificial intelligence in vehicle emission
reduction are presented in Table 9.

Table 9. Key directions for future usage of AI in vehicle emission reduction.

Direction Description
Develop AI systems that provide real-time monitoring and reporting of vehicle
Real-Time Emission Monitoring emissions. These systems can enable immediate corrective actions and help regulatory
agencies enforce emission standards effectively.
Implement AI algorithms that predict emissions based on driving conditions, enabling
Predictive Emission Control proactive emission reduction strategies. This can include adaptive engine control and
route optimization.
Expand AI-powered fleet management solutions to optimize the operation of large
Enhanced Fleet Management vehicle fleets, including route planning, load balancing, and eco-driving coaching for
commercial vehicles.
Energies 2023, 16, 7988 18 of 34

Table 9. Cont.

Direction Description
Advance the use of AI in autonomous vehicles to optimize driving patterns, minimize
Autonomous Vehicles and
idle, and enhance communication between vehicles and traffic management systems
Emission Reduction
for emission reduction.
Develop AI systems that improve electric vehicle range predictions, taking into
Electric Vehicle Range Optimization account factors such as weather, terrain, and driving habits. This can reduce “range
anxiety” and promote electric vehicle adoption.
Create AI-driven platforms that integrate various modes of transportation (e.g., public
Integrated Transportation Ecosystem transit, ridesharing, electric scooters) to provide seamless, efficient, and eco-friendly
travel options.
Utilize AI to design incentive programs for eco-friendly driving, such as discounted
Emission Reduction Incentives
tolls or insurance rates for low-emission vehicles and eco-driving practices.
Enhance AI-powered air quality monitoring systems in urban areas and provide
Air Quality Monitoring and Alerts
real-time alerts and recommendations to residents and policymakers.
Utilize AI for urban planning and infrastructure development, considering the impact
Green Infrastructure Planning on vehicle emissions. This can include optimizing traffic flow, promoting public
transportation, and expanding electric vehicle charging networks.
Emission Reduction Continue developing AI tools for robust emission testing and compliance verification,
Regulation Compliance ensuring that vehicles meet stringent environmental standards and regulations.
Apply AI in the manufacturing process to reduce the carbon footprint of vehicle
Energy-Efficient Manufacturing production. AI can optimize supply chains, minimize waste, and improve
energy efficiency.
Develop comprehensive AI models that consider the environmental impact of a
Lifecycle Carbon Footprint Analysis vehicle’s entire lifecycle, from manufacturing and operation to disposal, helping
consumers make informed decisions.
Utilize AI for personalized public awareness campaigns and eco-driving education,
Public Awareness and Education
helping individuals understand their role in reducing emissions.
Foster international collaboration to establish global AI standards and best practices
Global Collaboration and Standards
for vehicle emissions reduction, allowing consistency in technology implementation.
Source: Authors’ own work based on: [110,113,114,116–119,121,124–127].

4.3. Smart Grid

Artificial intelligence is a very useful solution in enhancing the efficiency, reliability,
and sustainability of smart grids, which are modernized electrical grids that use digital
technology to monitor and manage electricity generation, distribution, and consumption.
AI is applied in various ways within smart grids to optimize operations, improve energy
management, and enhance overall grid performance [128].
The utilization of AI in the smart grid offers a digital framework that harnesses
advanced technological capabilities. AI strategies within the smart grid encompass various
aspects such as power management, automation of the power system, analysis of energy
usage trends, and the detection of faults [129]. The ultimate objective of an intelligent grid is
to substitute manual procedures with AI-driven solutions, resulting in enhanced efficiency,
stability, and cost savings [130]. This covers every facet of an electrical network [128]
including power generation [131], energy transmission [132], power conversion [133],
electricity distribution [134], and energy consumption [135].
AI can analyze data from sensors and other sources to predict when grid equipment,
such as transformers or circuit breakers, might fail. This proactive approach to maintenance
reduces downtime and prevents costly outages [136]. AI can forecast electricity demand
and adapt the grid’s operation accordingly. It can communicate with smart appliances,
thermostats, and electric vehicles to optimize energy consumption during periods of high
demand or low supply [137], reducing peak load and managing grid stress [138].
Omitaomu and Niu [139] have described main artificial intelligence techniques which
can be used in smart grids. The first important method is load forecasting. The increasing
complexity of load forecasting is due to the integration of renewable energy sources in
smart grids. Load forecasting is categorized into three levels: short-term load forecasting
(STLF), mid-term load forecasting (MTLF), and long-term load forecasting (LTLF). Various
Energies 2023, 16, 7988 19 of 34

AI techniques, including deep learning, are explored to enhance forecasting accuracy.

Another area where AI techniques can be useful is power grid stability assessment. Power
grid stability, comprising transient stability, frequency stability, small-signal stability, and
voltage stability are crucial to ensure the reliability and security of the power system.
Traditional stability assessment models are complex and computationally intensive. Data-
driven AI methods are useful for stability analysis, leveraging technologies such as phasor
measurement units (PMUs) and wide area measurement systems (WAMSs).
Another potential of AI in smart grids is the usage of AI methods for fault detection in
power systems. Various techniques, such as extreme learning machines (ELMs), support
vector machines (SVMs), and ensemble models, are employed to detect and locate faults in
power grids, including high-impedance faults in micro grids and line trip faults [140].
The AI can also be useful in the case of smart grid security. AI technologies, such as
artificial neural networks (ANNs), support vector machines (SVMs), and reinforcement
learning (RL), are employed to enhance smart grid security by detecting and preventing
cyberattacks [128,141].
AI systems continuously monitor the grid for abnormal conditions or disturbances [142].
They can quickly identify and respond to issues such as power outages [142], equipment
failures [143], or cybersecurity threats [128]. Its use can help in integrating variable energy
sources like solar and wind into the grid by predicting their output based on weather con-
ditions and adjusting grid operations to accommodate fluctuations in generation [144,145].
There are many key directions for the future usage of artificial intelligence in smart
grids. The authors present them in Table 10.

Table 10. Key directions for future usage of AI in smart grids.

Direction Description
To realize the vision of a fully self-learning smart grid, integrating AI with cloud
computing is pivotal. This integration brings several benefits including increased
security and robustness, and a reduction in downtime due to outages. The cloud acts
Integration with Cloud Computing
as a reservoir of data and computational power, allowing smart grids to process
information efficiently, adapt quickly to changing conditions, and make
well-informed decisions.
Fog computing introduces a paradigm shift by processing raw data locally, rather than
transmitting it to distant cloud servers. This approach offers several advantages such
as energy efficiency, scalability, and flexibility. Using on-demand computing resources,
Fog Computing fog computing aligns perfectly with the demands of a modern smart grid. Preliminary
research indicates its potential role in enhancing the reliability and performance of
smart grids, particularly as the volume of data generated in these systems continues
to escalate.
Smart grid analysis faces a persistent challenge: the scarcity of labeled data. To
overcome this obstacle, researchers are turning to transfer learning, a technique that
reduces the reliance on large volumes of training data. Recent years have witnessed a
Transfer Learning
surge in interest in deep transfer learning tasks. These approaches hold great promise
and could have widespread applications within smart grid systems, enabling them to
adapt and learn even with limited data.
In the era of fog computing and the evolution of 5G networks, predicting consumer
behavior has become a critical task in managing power systems. Understanding and
learning the patterns of consumer power consumption can significantly contribute to
Consumer Behavior Prediction
demand-side management. With the assistance of AI, smart grids can anticipate and
respond to changes in energy consumption patterns, promoting efficient demand
response initiatives.
Source: The authors’ own work on the basis of: [99,128–130,132,134,135,137–139,141–157].

According to Seyd and Bong [128], AI techniques have revolutionized the energy
market by providing efficient solutions for real-time demand response and decision-making.
This enables grid operators to optimize all aspects of the power grid, from relay switching
to large generator controls, and mitigate unwanted harmonics through sensor networks.
Energies 2023, 16, 7988 20 of 34

Those techniques play a crucial role in coordinating distributed energy resources, enhancing
the acceptability of renewable energy sources, and increasing grid reliability. It allows
for the efficient management of distributed generation and storage capacity, automatic
regulation and optimization, bidirectional energy flow, and the integration of plug-in
hybrid electric vehicles.
Distributed grid management requires real-time optimization for large-scale systems
with renewable generators and controllable loads. AI techniques, such as consensus-
based distributed computational intelligence, offer solutions to address the challenges of
rapidly changing conditions, computation, and communication bottlenecks [148]. AI has
driven the development of decentralized and intelligent controllers, improving processing
speed, reliability, and efficacy. These controllers distribute operations among distributed
units, reducing the burden on centralized controllers and improving the resilience of the
system [158].
The traditional approach of using supervisory control and data acquisition (SCADA)
systems has become impractical due to the complexity of modern grids. AI-driven dis-
tributed load balancing algorithms have emerged as effective solutions to optimize loads
in distributed systems. AI and blockchain technologies have played a significant role in
enhancing security and data management in smart grids, particularly in the context of
distributed data storage and local energy trading [146].
According Sulaiman et al. [138], the integration of artificial intelligence into the smart
grid presents significant opportunities and challenges. AI can enhance grid security by
continuously monitoring, analyzing, and predicting potential threats and vulnerabilities.
It enables proactive responses to security incidents, automates decision-making, and pro-
motes collaboration among various infrastructure components in a smart city. However,
there are several challenges that need to be addressed.
According to Zambrano and Giraldo [154], predictive models based on AI for renew-
able energy hold the promise of revealing valuable glimpses into the expected energy
enhancements in the near future. Ruhnau et al. [155] believe that combining various ap-
proaches can refine these forecasts by making the most of the disparities in individual
prediction models. These approaches encompass both standalone and integrated tech-
nologies that generate predictions based on distinct time series data derived from specific
sources such as weather stations, wind turbines, or solar panels [109]. To enhance forecast
precision, the incorporation of information from nearby areas to the location of interest has
become increasingly popular, particularly in recent years [116].
In Table 11, there is a comparison of the advantages and disadvantages of AI usage in
smart grids.

Table 11. Advantages and disadvantages of AI usage in smart grids.

Advantages Disadvantages
Data Security Concerns
Improved Grid Efficiency
AI systems may be susceptible to cyberattacks, potentially
AI can optimize energy distribution and reduce energy waste,
compromising the security and privacy of grid data
leading to improved grid efficiency.
and operations.
Enhanced Reliability Initial Implementation Costs
AI enables predictive maintenance and self-healing capabilities, Integrating AI into smart grids can require substantial
reducing downtime and improving grid reliability. investments in infrastructure, technology, and expertise.
Real-time Monitoring
Complexity and Maintenance
AI allows for real-time monitoring and analysis of grid
AI systems can be complex to implement and maintain,
performance, enabling quick responses to fluctuations
requiring skilled personnel and ongoing updates.
and outages.
Demand Response Resource Intensive:
AI can predict and respond to changes in energy demand, AI systems may demand significant computational resources,
facilitating efficient demand-side management. potentially increasing operational costs.
Energies 2023, 16, 7988 21 of 34

Table 11. Cont.

Advantages Disadvantages
Renewable Integration Data Privacy Concerns
AI aids in the integration of renewable energy sources by The collection and analysis of large amounts of data can raise
optimizing their output and storage. concerns about consumer data privacy.
Grid Resilience Algorithm Bias
AI can adapt to unexpected events and disasters, contributing to AI algorithms can exhibit bias based on the training data,
grid resilience and disaster recovery. potentially leading to unfair or inequitable outcomes.
Reduced Environmental Impact Lack of Human Oversight
AI can minimize environmental impact by optimizing energy Excessive reliance on AI may reduce human oversight and
usage and promoting sustainable practices. decision-making, potentially introducing risks.
Source: The authors’ own work based on: [88,92,99,104,128–130,132,134,135,137,139,142–145,148–160].

AI has enabled the emergence of “prosumers”, allowing domestic energy users to

both produce and consume electricity and share it with others. This shift from centralized,
fossil-fueled generation to a decentralized, intelligent system enhances economic benefits
for consumers, fostering energy sharing and trade [160].
Sami [153] described how to use AI in prosumers management. He pointed out that
machine learning within the realm of artificial intelligence has the capability to assess and
anticipate energy demand patterns and categorize irregular energy usage. By leveraging
data collected through smart meters and subjecting it to AI analysis and data mining, it
becomes feasible to discern various customer segments’ electricity consumption behaviors.
Subsequently, this data can be employed to enhance statistical precision, facilitating the
targeted delivery of advertisements and services. Fluctuations in the environment, such as
variations in weather conditions, alterations in electrical appliance usage, and changes in
consumer behavior, can impact the accuracy of anomaly detection results. Consequently,
it is imperative to emphasize potential adverse aspects within the power grid that could
influence the equitable distribution of power among consumers. The analysis of energy
consumption is intrinsically linked to human characteristics, which can be addressed by
extracting or taxonomy features. In this context, the development of deep learning models,
particularly multilayered hidden neural networks, augments the predictive performance of
energy demand and consumption.
Rodgers et al. [147], in his study, underscores the significance of AI in smart grids,
aligning goals with global sustainability objectives, emphasizing the role of ICT, and
outlining practical requirements for smart grids. The study delves into the decision-making
processes of experts and their knowledge transfer apparatus. It highlights the importance of
information and communication technology (ICT) and AI usage in facilitating knowledge
transfer for a greener environment. The authors have identified three key goals for smart
grids: universal access to electricity, environmental protection, and efficiency. These
goals align with global sustainability objectives, such as those set by the United Nations
Conference on Sustainable Development (Rio + 20). The AI-based solutions can be useful
to realize them.
The AI can be also used to enable smart grid stability prediction. This possible us-
age was described by Ucar [161]. He proposes an enhanced model using explainable AI
and feature engineering for predicting the stability of the smart grid (SG). This model ap-
proaches the problem with both classification and regression, offering a holistic perspective
on existing studies and proposing a novel structure to address their limitations. The GBM
(gradient boosting machine) and deep learning models are introduced as effective tools for
prediction, despite their drawbacks. The flexibility and practicality of GBMs make them
valuable tools for model design and customization. The text concludes by emphasizing the
importance of combining data analytics with smart grid research for future studies.
Summing up, artificial intelligence plays a pivotal role in optimizing the performance
of smart grids by efficiently managing various grid operations and energy consumption.
These AI-driven systems utilize load-forecasting techniques and predictive analysis to
Energies 2023, 16, 7988 22 of 34

enhance stability, minimize downtime, and proactively maintain grid equipment, prevent-
ing costly outages. Moreover, AI facilitates real-time adjustments in energy consumption
by communicating with smart appliances and electric vehicles, ensuring optimization
during peak demand periods, and consequently reducing stress on the grid. AI’s inte-
gration within the energy market enables grid operators to coordinate distributed energy
resources efficiently, enhancing grid reliability and managing distributed generation and
storage capacity effectively. Furthermore, AI-driven decentralized controllers enhance
system resilience and processing speed, optimizing operations among distributed units,
and reducing the dependency on centralized controllers. Moreover, AI’s role extends to
improving grid security by continuously monitoring and predicting potential threats, au-
tomating decision-making, and fostering collaboration among infrastructure components
in smart cities.

4.4. Energy Storages

With the global shift toward renewable energy sources like solar and wind, the need
for efficient and reliable energy storage solutions has become increasingly critical. AI plays
an important role in addressing the challenges associated with energy storage, making it
smarter, more cost-effective, and environmentally friendly [162].
Energy storage technology has a role to play in enhancing the capabilities for utilizing
new energy sources, ensuring the reliable and cost-effective power systems operation,
and advancing the extensive adoption of renewable energy technologies [163]. Various
fresh innovations, concepts, methodologies, and technologies have been introduced in this
domain, stemming from disciplines such as materials science, knowledge management,
electrical engineering, control systems, and artificial intelligence [164].
AI algorithms are being used to enhance the performance of energy storage systems,
particularly lithium-ion batteries. By continuously monitoring and analyzing data from
these batteries, AI can optimize their charging and discharging cycles, extending their
lifespan and improving their efficiency [133]. This not only reduces maintenance costs,
but also reduces the environmental impact of battery disposal. This solution is employed
to predict potential issues in energy storage systems before they lead to costly break-
downs [165]. Through real-time data analysis and machine learning models, AI can detect
anomalies in system behavior, enabling operators to perform timely maintenance and
prevent unexpected downtime [166].
Energy storage systems equipped with AI can respond rapidly to fluctuations in the
grid. When the supply of renewable energy is inconsistent, AI can instantly adjust the flow
of stored energy, stabilizing the grid and ensuring a consistent power supply [167].
In the literature, many techniques of AI usage in energy storage can be found. Ahmed
and Abdallia [168] proposed hybrid differential evolution optimization of AI. The efficiency
of the proposed controller is confirmed in an electrical grid that includes a synchronous gen-
erator, a photovoltaic power source, and a battery energy storage system. The controller’s
parameters are adaptively tuned in real-time by training the artificial neural network (ANN)
with datasets generated during the optimization phase of both controllers using the hybrid
differential evolution optimization method under varying levels of disturbance, ranging
from low to high. Athari and Ardehali [169] used the fuzzy logic controller-based approach.
The membership features of the fuzzy logic controller (FLC) are tailored to reduce opera-
tional costs in green energy hybrid systems. This reduction is achieved by utilizing weekly
and periodic data predictions for factors such as water availability, electricity demand, and
environmental conditions like wind speed, sunlight, and air temperature. This optimiza-
tion process employs algorithms inspired by frog-spring shuffling. It is worth noting that
accurate accounting of power grid costs plays a significant role in enhancing the efficiency
of energy storage components for the hybrid renewable energy systems (HRESs) when
connected to the grid. This efficiency improvement is achieved because the configured
weekly and periodic FLCs help minimize the operating hours of fuel cells and gas-based
generators while reducing state-of-charge (SOC) variability in the battery stack [168].
Energies 2023, 16, 7988 23 of 34

Zahedi and Ardehali [170] described the situation when a novel energy storage system
(ESS) control system employing a multi-agent setup was implemented for a 100-megawatt
system. The system’s control performance was verified through simulation analysis and
practical testing. The AI-driven solution based on hierarchical control was described
by Yunhao et al. [171]. By employing balance regulation, the simulated impedance is
dynamically adjusted to eliminate the impact of inaccurate line impedance on the precision
of the current distribution. Subsequently, each power storage unit can fine-tune its current
based on state-of-charge (SoC) balance control, taking into account its capacity and charging
status. This helps reduce SoC discrepancies and facilitates a gradual state of charge (SoC)
balance during both charging and discharging operations.
Summing up, artificial intelligence is revolutionizing energy storage solutions by opti-
mizing the performance and longevity of storage systems. In energy storage, AI algorithms
continuously analyze and fine-tune the charging and discharging cycles, notably enhancing
the efficiency of lithium-ion batteries. By leveraging real-time data and predictive analytics,
AI predicts potential system issues, enabling proactive maintenance, reducing downtime,
and mitigating the environmental impact associated with battery disposal. These AI-driven
solutions in energy storage effectively stabilize the grid by swiftly responding to fluc-
tuations in renewable energy supply, ensuring consistent power flow and minimizing
interruptions. Additionally, diverse AI-based approaches, such as hybrid optimization
and fuzzy logic controllers, significantly improve system efficiency and reduce operational
costs in hybrid renewable energy systems. These advancements underscore AI’s role in
enhancing the reliability, efficiency, and sustainability of energy storage systems, offering
promising avenues for smarter and more eco-friendly energy management.
In Table 12, there is a comparison of the advantages and disadvantages of AI usage in
energy storage management.

Table 12. Key directions for future usage of AI in energy storages.

Direction Description
Implement AI to optimize the management of energy storage systems, maximizing
Optimized Energy Management
their efficiency and overall performance.
Develop AI solutions that facilitate seamless integration of energy storage with power
Grid Integration
grids, enhancing grid stability and ensuring reliable power supply.
Utilize AI to advance the development of new battery technologies, making them
Advanced Battery Technologies
more efficient, longer lasting, and cost-effective.
Employ AI for predictive maintenance of energy storage systems to reduce downtime
Predictive Maintenance
and extend the lifespan of storage devices.
Enhance AI algorithms to seamlessly integrate energy storage with renewable energy
Renewable Energy Synergy sources such as solar and wind, enabling more efficient and stable renewable
energy utilization.
Develop AI-driven solutions for managing decentralized energy storage resources,
Decentralized Storage
including microgrids and distributed storage systems, improving grid resilience.
Strengthen cybersecurity measures to protect energy storage systems and ensure data
Cybersecurity and Data Privacy
privacy when handling sensitive grid information through AI technologies.
Use AI to optimize energy consumption patterns in homes, businesses, and industries,
Energy Consumption Optimization
ensuring efficient use of stored energy and reducing energy waste.
Develop AI-powered solutions that promote environmental sustainability by
Environmental Sustainability
minimizing the environmental impact of energy storage systems.
Collaborate with policymakers to ensure that AI-based energy storage systems comply
Regulatory Compliance with regulations and standards while promoting responsible and ethical AI use in the
energy sector.
Source: Authors’ own work on basis: [133,157,163,164,167,168,171–173].

5. Discussion–Challenges
Based on a review of the literature and an analysis of solutions based on artificial
intelligence and connected to energy both at home and in cities, the authors identified and
described the challenges in the given areas. This section is divided into two subsections
Energies 2023, 16, 7988 24 of 34

corresponding to fields from Sections 3 and 4, and in each of these fields the authors have
identified the most important challenges that resulted from a review of the latest research
on the topic.

5.1. Field of Residential and Individual Users

In the first field, the following challenges were identified:
• Providing the desired level of resident comfort while minimizing energy consumption
in smart homes;
• Providing cooperation of different devices in smart homes as they are often produced
by different manufacturers and use various communication protocols;
• Ensuring that energy management systems in homes effectively reduce energy con-
sumption and not inadvertently increase it through constant connectivity and de-
vice usage;
• Using machine learning and deep reinforcement learning to manage appliances, dis-
tributed energy sources, and electric vehicle charging in smart homes;
• Providing the stability and availability of energy sources for smart homes as smart
homes become more dependent on renewable energy sources;
• Coordinating energy consumption in smart homes with the growing adoption of
renewable energy sources;
• Optimizing AI-driven HVAC systems by making real-time adjustments based on data
from sources such as weather forecasts and occupancy patterns;
• Developing AI-driven home energy management systems that adapt to changes in
electricity prices and consumption;
• Providing the compatibility of smart lighting with various smart home platforms;
• Developing cost-effective and environmentally sustainable lighting systems;
• Developing and implementing new materials in smart windows to control natural
light and use it to obtain the optimal temperature inside building;
• Creating intelligent control systems for smart blinds that adapt to changing sunlight
and temperature to ensure optimal light management and energy efficiency in homes;
• Designing hardware upgrades for refrigerators to improve efficiency while ensuring
food safety;
• Developing self-learning smart thermostats for more extensive applications;
• Getting users to adopt energy-efficient behaviors and make the most of smart home
features is vital—educating and motivating users is vital.

5.2. Field of Urban Infrastructure

In the other field, the following challenges were identified:
• Implementing of AI in the charging infrastructure can come with high upfront costs
for hardware, software, and integration;
• Dependency of the charging infrastructure on AI systems, which can occasionally
experience downtime or errors, potentially inconveniencing EV owners;
• Collecting of user data and behavior monitoring for optimal charging can raise privacy
concerns, necessitating robust data protection measures;
• Shifting to AI-powered charging infrastructure may lead to traditional charging station
maintenance jobs being displaced;
• Protecting the charging infrastructure from hacking and fraud is a significant concern,
requiring strong cybersecurity measures;
• Requirements of AI systems ongoing updates and maintenance to ensure their effec-
tiveness and security;
• Resistance of some users to adopt AI technology to charge their EVs;
• Integrating AI into existing infrastructure and grid systems can be complex and
require standardization;
• Implementing AI solutions for vehicle emission reduction may involve integration
costs and technology adoption challenges;
Energies 2023, 16, 7988 25 of 34

• Ensuring that vehicles meet stringent emission standards and regulations requires the
development of robust AI tools;
• Convincing drivers to adopt eco-friendly driving practices may be a challenge;
• Fostering international collaboration and standards for AI in vehicle emission reduc-
tion can be complex;
• Using of AI in smart grids and energy storage systems introduces cybersecurity risks,
potentially compromising the security and privacy of grid data and operations;
• Integrating AI into smart grids and energy storage systems can require substantial
investments in infrastructure, technology, and expertise;
• Implementation complexity and maintenance of AI-driven systems, requiring skilled
personnel and ongoing updates to keep them operational;
• Requiring substantial computational resources, AI systems could potentially escalate
operational expenses;
• AI algorithms have the potential to display bias linked to the training data, potentially
resulting in outcomes that are unfair or inequitable;
• Depending excessively on AI could potentially diminish human oversight and decision-
making, introducing possible risks;
• The gathering and analyzing of extensive data could potentially cause concerns about
the privacy of consumer data.

5.3. Scenarios
Overcoming these challenges will be essential for the successful integration of smart
homes and smart cities and efficient usage of energy in the future. It will require a collabo-
rative effort from manufacturers, policymakers, and consumers.
Summing up the challenges, the following scenarios for future usage of AI within
energy can be considered:
• AI-driven smart home adaptability: creating adaptive AI systems that learn user
behaviors and preferences to optimize energy usage in smart homes. These systems
should harmonize various devices, predict consumption patterns, and adjust settings
in response to changing conditions, ensuring energy efficiency without compromising
user comfort.
• Secure and privacy-enhanced charging infrastructure: development of AI-powered
charging stations equipped with robust cybersecurity measures and privacy protocols.
These systems ensure seamless operation, user data protection, and effective energy
management, overcoming privacy concerns and fostering greater EV adoption.
• Standardized integration of AI in urban grids: establishing standardized protocols for
integrating AI into urban grids and infrastructure. This involves collaborative efforts to
ensure the compatibility, cybersecurity, and seamless integration of AI solutions across
various urban energy systems, ensuring reliable and efficient energy distribution.
• Emission reduction through AI-optimized driving: implementing AI-based systems
that actively encourage eco-friendly driving behaviors. These systems utilize real-time
data analysis, offering personalized feedback and incentives to drivers, promoting
fuel-efficient driving habits and reducing vehicle emissions.
• Fairness and bias mitigation in AI algorithms: addressing biases in AI algorithms used
for energy management by implementing fairness-aware and transparent AI models.
Efforts should focus on developing tools that detect and mitigate biases, ensuring
equitable outcomes and fair decision-making in energy-related AI applications.
• Collaborative international AI standards: facilitating international collaboration to
establish unified AI standards for energy solutions. This involves harmonizing regula-
tions, sharing best practices, and fostering a global framework that promotes ethical
use of AI in managing energy systems.
These scenarios represent potential directions for the use of AI in energy-related do-
mains, offering solutions to overcome existing challenges while emphasizing collaboration
Energies 2023, 16, 7988 26 of 34

among stakeholders, policymakers, manufacturers, and consumers to ensure sustainable

and efficient energy usage in the future.

6. Conclusions
The paper presents a comprehensive review of AI-driven solutions in the urban energy
sector, shedding light on the significant impacts and promising applications of artificial
intelligence. Through a literature review including papers from 2019 to 2023, the study
categorizes these solutions into two main areas: residential and individual user applications,
and urban infrastructure integration for both individual users and communities. The review
has achieved its objectives, as follows:
• O1: to identify trends, emerging technologies, and applications using artificial intelli-
gence in the energy field:
- The examination has shown key emerging technologies in AI-driven energy
solutions for residential users and society at large. They include solutions for
individual users in homes, such as AI-driven heating and cooling, lighting, win-
dows and blinds, home devices—refrigerators, and energy management systems.
When it comes to society, the following are most popular: electric vehicle charging
infrastructure, vehicle emission reduction, smart grid, and energy storages.
• O2: to provide up-to-date insights into the use of artificial intelligence in energy-
related applications:
- Focusing on recent research, the paper has provided valuable insights into the
current state of AI-driven urban energy solutions. It highlights the rapid evolution
of technology and its growing role in shaping urban energy systems.
• O3: to gain a comprehensive understanding of the current state of AI-driven urban
energy solutions:
- The review has deepened our understanding of the dynamic field of AI-driven ur-
ban energy solutions. It elucidates how AI is being integrated into various aspects
of urban living, from individual homes to a broader community infrastructure.
• O4: to explore future directions, emerging trends, and challenges in the field of AI-
driven energy solutions:
- The paper acknowledges the transformative potential of AI in urban energy
management while recognizing the challenges ahead. It shows the way for future
research activities by offering a view of AI-driven solutions in homes and cities.
The research questions have been addressed with meaningful insights:
• R1: What are the key emerging technologies in AI-driven energy solutions for residen-
tial users and society?
- The paper identifies emerging technologies that are set to transform the energy
landscape, including smart home devices, electric vehicle infrastructure, smart
grids, and more. These technologies promise to improve energy efficiency, reduce
carbon emissions, and enhance the quality of life of urban residents.
• R2: How is artificial intelligence integrated into urban infrastructure to enhance
energy-related solutions?
- Artificial intelligence is integrating into urban infrastructure to optimize energy-
related solutions. This includes the enhancement of electric vehicle charging
infrastructure, reduction in vehicle emissions, development of smart grids, and
efficient energy storage.
• R3: What challenges are associated with the implementation of AI-driven solutions in
urban energy management?
- The challenges include the need to balance resident comfort with energy effi-
ciency in smart homes, ensuring compatibility and cooperation among various
devices, and preventing unintended energy consumption increases due to con-
Energies 2023, 16, 7988 27 of 34

stant connectivity. The challenges also extend to managing renewable energy

sources, coordinating energy consumption, and optimizing HVAC systems in
smart homes. In the field of urban infrastructure, challenges involve high upfront
costs, privacy concerns about user data, potential job displacement, cybersecurity
risks, technology adoption, and others.
In conclusion, this paper serves as a valuable resource for researchers, policymakers,
and practitioners in the field of urban energy management. Researchers should focus
on interdisciplinary collaborations, exploring AI’s depth in urban energy management.
Policymakers could help by incentivizing AI integration and establishing standards, while
practitioners should focus on real-world trials and user-centric AI implementations. This
paper not only highlights the transformative potential of AI but also underscores the need
for a collaborative effort to overcome challenges and harness the full benefits of artificial
intelligence in creating smarter, more sustainable, and energy-efficient urban environments.
The biggest limitation of this study is the lack of a detailed analysis of individual
solutions based on artificial intelligence, although this is due to the huge number of
these solutions. While the article reviews these solutions broadly, it refrains from delving
into specific details and intricacies due to the overwhelming volume and scope of the
discussed solutions.

Author Contributions: Conceptualization, K.S. and R.W.; literature review: K.S. and R.W.; method-
ology, K.S., R.W. and W.W.G.; validation, K.S., R.W. and W.W.G.; formal analysis, K.S. and R.W.;
investigation, K.S. and R.W.; writing—original draft preparation, K.S.; writing—review and editing,
K.S.; funding acquisition, K.S. and R.W. All authors have read and agreed to the published version of
the manuscript.
Funding: This research received no external funding.
Data Availability Statement: Not applicable.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Ranosz, R.; Bluszcz, A.; Kowal, D. Conditions for the innovation activities of energy sector enterprises shown on the example of
mining companies. Inżynieria Miner. 2020, 1, 249–256. [CrossRef]
2. Bluszcz, A.; Manowska, A. Differentiation of the level of sustainable development of energy markets in the European Union
countries. Energies 2020, 13, 4882. [CrossRef]
3. Jelonek, D.; Chomiak-Orsa, I. The Application of ICT in the Area of Value Co-Creation Mechanisms Support as a Determinant of
Innovation Activities. Int. J. Ambient Comput. Intell. 2018, 9, 32–42. [CrossRef]
4. Sadeeq, M.M.; Abdulkareem, N.M.; Zeebaree, S.R.M.; Ahmed, D.M.; Sami, A.S.; Zebari, R.R. IoT and Cloud computing issues,
challenges and opportunities: A review. Qubahan Acad. J. 2021, 1, 1–7. [CrossRef]
5. Awan, F.M.; Minerva, R.; Crespi, N. Using Noise Pollution Data for Traffic Prediction in Smart Cities: Experiments Based on
LSTM Recurrent Neural Networks. IEEE Sens. J. 2021, 21, 20722–20729. [CrossRef]
6. Koval, V.; Borodina, O.; Lomachynska, I.; Olczak, P.; Mumladze, A.; Matuszewska, D. Model Analysis of Eco-Innovation for
National Decarbonisation Transition in Integrated European Energy System. Energies 2022, 15, 3306. [CrossRef]
7. Kulpa, J.; Kamiński, P.; Stecuła, K.; Prostański, D.; Matusiak, P.; Kowol, D.; Kopacz, M.; Olczak, P. Technical and Economic Aspects
of Electric Energy Storage in a Mine Shaft—Budryk Case Study. Energies 2021, 14, 7337. [CrossRef]
8. Stecuła, K.; Olczak, P.; Kamiński, P.; Matuszewska, D.; Duong Duc, H. Towards Sustainable Transport: Techno-Economic Analysis
of Investing in Hydrogen Buses in Public Transport in the Selected City of Poland. Energies 2022, 15, 9456. [CrossRef]
9. Kinelski, G. Smart-city trends in the environment of sustainability as support for decarbonization processes. Polityka Energetyczna
2022, 25, 109–136. [CrossRef]
10. Sribna, Y.; Koval, V.; Olczak, P.; Bizonych, D.; Matuszewska, D.; Shtyrov, O. Forecasting solar generation in energy systems to
accelerate the implementation of sustainable economic development. Polityka Energetyczna 2021, 24, 5–28. [CrossRef]
11. Jia, L.; Cheng, P.; Yu, Y.; Chen, S.; Wang, C.; He, L.; Nie, H.; Wang, J.; Zhang, J.; Fan, B.; et al. Regeneration mechanism of a novel
high-performance biochar mercury adsorbent directionally modified by multimetal multilayer loading. J. Environ. Manag. 2023,
326, 116790. [CrossRef]
12. Salehi-Amiri, A.; Akbapour, N.; Hajiaghaei-Keshteli, M.; Gajpal, Y.; Jabbarzadeh, A. Designing an effective two-stage, sustainable,
and IoT based waste management system. Renew. Sustain. Energy Rev. 2022, 157, 112031. [CrossRef]
Energies 2023, 16, 7988 28 of 34

13. Anastasi, G.; Bartoli, C.; Conti, P.; Crisostomi, E.; Franco, A.; Saponara, S.; Testi, D.; Thomopulos, D.; Vallati, C. Optimized energy
and air quality management of shared smart buildings in the COVID-19 scenario. Energies 2021, 14, 2124. [CrossRef]
14. Loska, A.; Paszkowski, W. Smartmaintenance—The concept of supporting the exploitation decision-making process in the
selected technical network system. Adv. Intell. Syst. Comput. 2018, 637, 64–73. [CrossRef]
15. Chomiak-Orsa, I.; Domagała, P.; Greńczuk, A.; Grzelak, W.; Hauke, K.; Kotwica, A.; Perechuda, K.; Pondel, M. Open Data for
simulation to determine the efficient management of parking spaces in Smart City. Procedia Comput. Sci. 2022, 207, 3625–3634.
[CrossRef]
16. Paszkowski, W.; Loska, A. The use of neural network model in the assessment of annoyance of the industrial noise sources. In
International Conference on Intelligent Systems in Production Engineering and Maintenance; Springer: Berlin/Heidelberg, Germany,
2017; pp. 428–439.
17. Stecuła, K. Virtual Reality Applications Market Analysis—On the Example of Steam Digital Platform. Informatics 2022, 9, 100.
[CrossRef]
18. Zhang, C.; Lu, Y. Study on artificial intelligence: The state of the art and future prospects. J. Ind. Inf. Integr. 2021, 23, 100224.
[CrossRef]
19. Bluszcz, A.; Tobór-Osadnik, K.; Tomiczek, K.; Mansora, N.S.; Awang, H. The Use of Geomatics Tools in Critical Infrastructure
Management. Inżynieria Miner. 2023, 1, 169–174. [CrossRef]
20. Bluszcz, A.; Manowska, A. The use of hierarchical agglomeration methods in assessing the Polish energy market. Energies 2021,
14, 3958. [CrossRef]
21. Morse, H.F.; AlGhamdi, M.S.; Almusaied, H.M.; Alsubaie, S.A.; Almarri, D.M.; Alqahtani, M.A.; Alqahtani, R.A.; Tabeidi, R.A.;
Balharith, T. The Smart Heater Based on Internet of Things. In The International Conference on Artificial Intelligence and Computer
Vision; Springer: Berlin/Heidelberg, Germany, 2021; Volume 1377, pp. 145–153.
22. Sadeghian Broujeny, R.; Madani, K.; Chebira, A.; Amarger, V.; Hurtard, L. A Heating Controller Designing Based on Living Space
Heating Dynamic’s Model Approach in a Smart Building. Energies 2021, 14, 998. [CrossRef]
23. Png, E.; Srinivasan, S.; Bekiroglu, K.; Chaoyang, J.; Su, R.; Poolla, K. An internet of things upgrade for smart and scalable heating,
ventilation and air-conditioning control in commercial buildings. Appl. Energy 2019, 239, 408–424. [CrossRef]
24. Al Essa, M.J.M. Energy management of space-heating systems and grid-connected batteries in smart homes. Energy Ecol. Environ.
2022, 7, 1–14. [CrossRef]
25. Huchuk, B.; Sanner, S.; O’Brien, W. Development and evaluation of data-driven controls for residential smart thermostats. Energy
Build. 2021, 249, 111201. [CrossRef]
26. Afram, A.; Janabi-Sharifi, F.; Fung, A.S.; Raahemifar, K. Artificial neural network (ANN) based model predictive control (MPC)
and optimization of HVAC systems: A state of the art review and case study of a residential HVAC system. Energy Build. 2017,
141, 96–113. [CrossRef]
27. Vázquez-Canteli, J.R.; Nagy, Z. Reinforcement learning for demand response: A review of algorithms and modeling techniques.
Appl. Energy 2019, 235, 1072–1089. [CrossRef]
28. Claessens, B.J.; Vanhoudt, D.; Desmedt, J.; Ruelens, F. Model-free control of thermostatically controlled loads connected to a
district heating network. Energy Build. 2018, 159, 1–10. [CrossRef]
29. Marantos, C.; Lamprakos, C.; Siozios, K.; Soudris, D. Towards Plug&Play smart thermostats for building’s heating/cooling
control. In IoT for Smart Grids Design Challenges and Paradigms; Springer: Berlin/Heidelberg, Germany, 2019; pp. 183–207.
30. Goman, V.V.; Korolev, S.I. Smart thermostats in building automation systems and smart homes. KnE Eng. 2020, 5, 67–74.
[CrossRef]
31. Huang, K.; Hallinan, K.P.; Lou, R.; Alanezi, A.; Alshatshati, S.; Sun, Q. Self-learning algorithm to predict indoor temperature and
cooling demand from smart WiFi thermostat in a residential building. Sustainability 2020, 12, 7110. [CrossRef]
32. Duman, A.C.; Erden, H.S.; Gönül, Ö.; Güler, Ö. A home energy management system with an integrated smart thermostat for
demand response in smart grids. Sustain. Cities Soc. 2021, 65, 102639. [CrossRef]
33. An, Y.; Fu, Y.; Dai, J.-G.; Yin, X.; Lei, D. Switchable radiative cooling technologies for smart thermal management. Cell Rep. Phys.
Sci. 2022, 3, 101098. [CrossRef]
34. Daneshvar, Y.; Sabzehparvar, M.; Hashemi, S.A.H. Energy efficiency of small buildings with smart cooling system in the summer.
Front. Energy 2020, 16, 651–660. [CrossRef]
35. Nezhad, A.E.; Rahimnejad, A.; Gadsden, S.A. Home energy management system for smart buildings with inverter-based air
conditioning system. Int. J. Electr. Power Energy Syst. 2021, 133, 107230. [CrossRef]
36. Hou, X.; Wang, J.; Huang, T.; Wang, T.; Wang, P. Smart home energy management optimization method considering energy
storage and electric vehicle. IEEE Access 2019, 7, 144010–144020. [CrossRef]
37. Dong, B.; Shi, Q.; Yang, Y.; Wen, F.; Zhang, Z.; Lee, C. Technology evolution from self-powered sensors to AIoT enabled smart
homes. Nano Energy 2021, 79, 105414. [CrossRef]
38. Alipio, M.; Bures, M. Intelligent Network Maintenance Modeling for Fixed Broadband Networks in Sustainable Smart Homes.
IEEE Internet Things J. 2023, 10, 18067–18081. [CrossRef]
39. Garcia-Magarino, I.; Muttukrishnan, R.; Lloret, J. Human-centric AI for trustworthy IoT systems with explainable multilayer
perceptrons. IEEE Access 2019, 7, 125562–125574. [CrossRef]
Energies 2023, 16, 7988 29 of 34

40. Ayan, O.; Turkay, B. IoT-based energy efficiency in smart homes by smart lighting solutions. In Proceedings of the 2020 21st
International Symposium on Electrical Apparatus & Technologies (SIELA), Bourgas, Bulgaria, 3–6 June 2020; IEEE: Toulouse,
France, 2020; pp. 1–5.
41. Li, Z.; Li, J.; Li, X.; Yang, Y.; Xiao, J.; Xu, B. Design of office intelligent lighting system based on Arduino. Procedia Comput. Sci.
2020, 166, 134–138. [CrossRef]
42. Chiesa, G.; Di Vita, D.; Ghadirzadeh, A.; Herrera, A.H.M.; Rodriguez, J.C.L. A fuzzy-logic IoT lighting and shading control
system for smart buildings. Autom. Constr. 2020, 120, 103397. [CrossRef]
43. Andramuño, J.; Mendoza, E.; Núñez, J.; Liger, E. Intelligent distributed module for local control of lighting and electrical outlets
in a home. J. Phys. Conf. Ser. 2021, 1730, 12001. [CrossRef]
44. Cho, Y.; Seo, J.; Lee, H.; Choi, S.; Choi, A.; Sung, M.; Hur, Y. Platform design for lifelog-based smart lighting control. Build.
Environ. 2020, 185, 107267. [CrossRef]
45. Ozeer, U.; Letondeur, L.; Ottogalli, F.-G.; Salaün, G.; Vincent, J.-M. Designing and implementing resilient IoT applications in the
fog: A smart home use case. In Proceedings of the 2019 22nd Conference on Innovation in Clouds, Internet and Networks and
Workshops (ICIN), Paris, France, 19–21 February 2019; IEEE: Toulouse, France, 2019; pp. 230–232.
46. Khoa, T.A.; Nhu, L.M.B.; Son, H.H.; Trong, N.M.; Phuc, C.H.; Phuong, N.T.H.; Van Dung, N.; Nam, N.H.; Chau, D.S.T.; Duc,
D.N.M. Designing efficient smart home management with IoT smart lighting: A case study. Wirel. Commun. Mob. Comput. 2020,
2020, 8896637. [CrossRef]
47. Touqeer, H.; Zaman, S.; Amin, R.; Hussain, M.; Al-Turjman, F.; Bilal, M. Smart home security: Challenges, issues and solutions at
different IoT layers. J. Supercomput. 2021, 77, 14053–14089. [CrossRef]
48. Zhou, Y.; Fan, F.; Liu, Y.; Zhao, S.; Xu, Q.; Wang, S.; Luo, D.; Long, Y. Unconventional smart windows: Materials, structures and
designs. Nano Energy 2021, 90, 106613. [CrossRef]
49. Cannavale, A.; Ayr, U.; Fiorito, F.; Martellotta, F. Smart electrochromic windows to enhance building energy efficiency and visual
comfort. Energies 2020, 13, 1449. [CrossRef]
50. Ke, Y.; Chen, J.; Lin, G.; Wang, S.; Zhou, Y.; Yin, J.; Lee, P.S.; Long, Y. Smart windows: Electro-, thermo-, mechano-, photochromics,
and beyond. Adv. Energy Mater. 2019, 9, 1902066. [CrossRef]
51. Wang, S.; Zhou, Y.; Jiang, T.; Yang, R.; Tan, G.; Long, Y. Thermochromic smart windows with highly regulated radiative cooling
and solar transmission. Nano Energy 2021, 89, 106440. [CrossRef]
52. Zhang, S.; Cao, S.; Zhang, T.; Lee, J.Y. Plasmonic Oxygen-Deficient TiO2-x Nanocrystals for Dual-Band Electrochromic Smart
Windows with Efficient Energy Recycling. Adv. Mater. 2020, 32, 2004686. [CrossRef]
53. Li, G.; Chen, J.; Yan, Z.; Wang, S.; Ke, Y.; Luo, W.; Ma, H.; Guan, J.; Long, Y. Physical crosslinked hydrogel-derived smart windows:
Anti-freezing and fast thermal responsive performance. Mater. Horiz. 2023, 10, 2004–2012. [CrossRef] [PubMed]
54. Zhou, Y.; Dong, X.; Mi, Y.; Fan, F.; Xu, Q.; Zhao, H.; Wang, S.; Long, Y. Hydrogel smart windows. J. Mater. Chem. A 2020, 8,
10007–10025. [CrossRef]
55. Zhou, Y.; Wang, S.; Peng, J.; Tan, Y.; Li, C.; Boey, F.Y.C.; Long, Y. Liquid thermo-responsive smart window derived from hydrogel.
Joule 2020, 4, 2458–2474. [CrossRef]
56. Chen, L.; Duan, G.; Zhang, C.; Cheng, P.; Wang, Z. 3D printed hydrogel for soft thermo-responsive smart window. Int. J. Extrem.
Manuf. 2022, 4, 25302. [CrossRef]
57. Aburas, M.; Soebarto, V.; Williamson, T.; Liang, R.; Ebendorff-Heidepriem, H.; Wu, Y. Thermochromic smart window technologies
for building application: A review. Appl. Energy 2019, 255, 113522. [CrossRef]
58. Nundy, S.; Mesloub, A.; Alsolami, B.M.; Ghosh, A. Electrically actuated visible and near-infrared regulating switchable smart
window for energy positive building: A review. J. Clean. Prod. 2021, 301, 126854. [CrossRef]
59. Tällberg, R.; Jelle, B.P.; Loonen, R.; Gao, T.; Hamdy, M. Comparison of the energy saving potential of adaptive and control-
lable smart windows: A state-of-the-art review and simulation studies of thermochromic, photochromic and electrochromic
technologies. Sol. Energy Mater. Sol. Cells 2019, 200, 109828. [CrossRef]
60. Jung, W.; Hong, T.; Oh, J.; Kang, H.; Lee, M. Development of a prototype for multi-function smart window by integrating
photovoltaic blinds and ventilation system. Build. Environ. 2019, 149, 366–378. [CrossRef]
61. Kang, H.; Hong, T.; Jung, S.; Lee, M. Techno-economic performance analysis of the smart solar photovoltaic blinds considering
the photovoltaic panel type and the solar tracking method. Energy Build. 2019, 193, 1–14. [CrossRef]
62. Kang, H.; Hong, T.; Lee, M. Technical performance analysis of the smart solar photovoltaic blinds based on the solar tracking
methods considering the climate factors. Energy Build. 2019, 190, 34–48. [CrossRef]
63. Wang, K.; Ti, Y.; Liu, D.; Chen, S. A Smart Refrigerator Architecture that Reduces Food Ingredients Waste Materials and Energy
Consumption. Ekoloji Derg. 2019, 107, 4873–4878.
64. Phuchamniphatthananun, A.; Pora, W. A smart domestic refrigerator with energy efficiency improvement. In Proceedings of
the 2019 IEEE Asia Pacific Conference on Circuits and Systems (APCCAS), Bangkok, Thailand, 11–14 November 2019; IEEE:
Toulouse, France, 2019; pp. 345–348.
65. Sabegh, M.R.Z.; Bingham, C.M. Impact of hysteresis control and internal thermal mass on the energy efficiency of IoT-controlled
domestic refrigerators. In Proceedings of the 2019 IEEE 7th International Conference on Smart Energy Grid Engineering (SEGE),
Oshawa, ON, Canada, 12–14 August 2019; IEEE: Toulouse, France, 2019; pp. 103–107.
Energies 2023, 16, 7988 30 of 34

66. Griva, A.I.; Rekkas, V.P.; Koritsoglou, K.; Sotiroudis, S.P.; Boursianis, A.D.; Papadopoulou, M.S.; Goudos, S.K. Energy Consump-
tion Assessment in Refrigeration Equipment: The SmartFridge Project. In Proceedings of the 2023 12th International Conference
on Modern Circuits and Systems Technologies (MOCAST), Athens, Greece, 28–30 June 2023; IEEE: Toulouse, France, 2023;
pp. 1–4.
67. Pérez-Gomariz, M.; López-Gómez, A.; Cerdán-Cartagena, F. Artificial neural networks as artificial intelligence technique for
energy saving in refrigeration systems—A review. Clean Technol. 2023, 5, 116–136. [CrossRef]
68. Cai, S. Research on Intelligent Refrigerator Control based on Artificial Intelligence Algorithm. Highlights Sci. Eng. Technol. 2023,
35, 12–16. [CrossRef]
69. Eltawil, M.A.; Mohammed, M.; Alqahtani, N.M. Developing Machine Learning-Based Intelligent Control System for Performance
Optimization of Solar PV-Powered Refrigerators. Sustainability 2023, 15, 6911. [CrossRef]
70. Lu, R.; Hong, S.H.; Yu, M. Demand response for home energy management using reinforcement learning and artificial neural
network. IEEE Trans. Smart Grid 2019, 10, 6629–6639. [CrossRef]
71. Majchrzak, K.; Olczak, P.; Matuszewska, D.; Wdowin, M. Economic and environmental assessment of the use of electric cars in
Poland. Polityka Energ.-Energy Policy J. 2021, 24, 153–168. [CrossRef]
72. Markowska, K.; S˛ekala, A.; Stecuła, K.; Kawka, T.; Sirovitskiy, K.; Pankova, O.; Vnukova, N.; Shulyak, M.; Kharchenko, S.; Shchur,
T. Comparison of the Sustainability and Economic Efficiency of an Electric Car and an Aircraft—A Case Study. Sustainability 2023,
15, 1238. [CrossRef]
73. Bazhinov, O.; Kravtson, M.; Bazhynova, T.; Haiek, Y.; Kharchenko, S.; Shchur, T.; Markowska, K.; S˛ekala, A.; Stecuła, K.; Kawka, T.
Determination of the quality index of cars. Sci. J. Silesian Univ. Technol. Ser. Transp. 2023, 118, 17–28. [CrossRef]
74. Ma, Y.; Chen, X.; Wang, L.; Yang, J. Study on smart home energy management system based on artificial intelligence. J. Sens. 2021,
2021, 9101453. [CrossRef]
75. Chauhan, R.K.; Chauhan, K.; Badar, A.Q.H. Optimization of electrical energy waste in house using smart appliances management
System-A case study. J. Build. Eng. 2022, 46, 103595. [CrossRef]
76. Olczak, P.; Żelazna, A.; Stecuła, K.; Matuszewska, D.; Lelek, Ł. Environmental and economic analyses of different size photovoltaic
installation in Poland. Energy Sustain. Dev. 2022, 70, 160–169. [CrossRef]
77. Kulpa, J.; Olczak, P.; Stecuła, K.; Sołtysik, M. The Impact of RES Development in Poland on the Change of the Energy Generation
Profile and Reduction of CO2 Emissions. Appl. Sci. 2022, 12, 11064. [CrossRef]
78. Lissa, P.; Deane, C.; Schukat, M.; Seri, F.; Keane, M.; Barrett, E. Deep reinforcement learning for home energy management system
control. Energy AI 2021, 3, 100043. [CrossRef]
79. Rocha, H.R.O.; Honorato, I.H.; Fiorotti, R.; Celeste, W.C.; Silvestre, L.J.; Silva, J.A.L. An Artificial Intelligence based scheduling
algorithm for demand-side energy management in Smart Homes. Appl. Energy 2021, 282, 116145. [CrossRef]
80. Nutakki, M.; Mandava, S. Review on optimization techniques and role of Artificial Intelligence in home energy management
systems. Eng. Appl. Artif. Intell. 2023, 119, 105721. [CrossRef]
81. Han, B.; Zahraoui, Y.; Mubin, M.; Mekhilef, S.; Seyedmahmoudian, M.; Stojcevski, A. Home Energy Management Systems: A
Review of the Concept, Architecture, and Scheduling Strategies. IEEE Access 2023, 11, 19999–20025. [CrossRef]
82. Ajitha, A.; Radhika, S. A comprehensive review of demand response strategies and the role of emergent technologies for
sustainable home energy management systems. Int. J. Ambient Energy 2023, 44, 2262–2282. [CrossRef]
83. Chen, Z.; Xiao, F.; Guo, F.; Yan, J. Interpretable machine learning for building energy management: A state-of-the-art review. Adv.
Appl. Energy 2023, 9, 100123. [CrossRef]
84. Ahmad, T.; Zhang, D.; Huang, C.; Zhang, H.; Dai, N.; Song, Y.; Chen, H. Artificial intelligence in sustainable energy industry:
Status Quo, challenges and opportunities. J. Clean. Prod. 2021, 289, 125834. [CrossRef]
85. Tushar, W.; Yuen, C.; Saha, T.K.; Morstyn, T.; Chapman, A.C.; Alam, M.J.E.; Hanif, S.; Poor, H.V. Peer-to-peer energy systems for
connected communities: A review of recent advances and emerging challenges. Appl. Energy 2021, 282, 116131. [CrossRef]
86. Li, J.; Herdem, M.S.; Nathwani, J.; Wen, J.Z. Methods and applications for Artificial Intelligence, Big Data, Internet of Things, and
Blockchain in smart energy management. Energy AI 2023, 11, 100208. [CrossRef]
87. Sun, D.; Ou, Q.; Yao, X.; Gao, S.; Wang, Z.; Ma, W.; Li, W. Integrated human-machine intelligence for EV charging prediction in
5G smart grid. EURASIP J. Wirel. Commun. Netw. 2020, 2020, 139. [CrossRef]
88. Ahmed, M.; Zheng, Y.; Amine, A.; Fathiannasab, H.; Chen, Z. The role of artificial intelligence in the mass adoption of electric
vehicles. Joule 2021, 5, 2296–2322. [CrossRef]
89. Raza, A.; Baloch, M.H.; Ali, I.; Ali, W.; Hassan, M.; Karim, A. Artificial Intelligence and IoT-Based Autonomous Hybrid Electric
Vehicle with Self-Charging Infrastructure. In Proceedings of the 2022 International Conference on Emerging Technologies in
Electronics, Computing and Communication (ICETECC), Jamshoro, Pakistan, 7–9 December 2022; IEEE: Toulouse, France, 2022;
pp. 1–6.
90. Bajaj, D.K.; Siddharth, P. Artificial Intelligence (AI) in Electrical Vehicles. In Recent Advances in Energy Harvesting Technologies;
River Publishers: Roma, Italy, 2023; pp. 57–76.
91. ElHusseini, H.; Assi, C.; Moussa, B.; Attallah, R.; Ghrayeb, A. Blockchain, AI and smart grids: The three musketeers to a
decentralized EV charging infrastructure. IEEE Internet Things Mag. 2020, 3, 24–29. [CrossRef]
92. Paret, P.; Finegan, D.; Narumanchi, S. Artificial Intelligence for Power Electronics in Electric Vehicles: Challenges and Opportuni-
ties. J. Electron. Packag. 2023, 145, 34501. [CrossRef]
Energies 2023, 16, 7988 31 of 34

93. Swarnkar, R.; Harikrishnan, R.; Thakur, P.; Singh, G. Electric Vehicle Lithium-ion Battery Ageing Analysis Under Dynamic
Condition: A Machine Learning Approach. SAIEE Afr. Res. J. 2022, 114, 4–13. [CrossRef]
94. Li, C.; Dong, Z.; Chen, G.; Zhou, B.; Zhang, J.; Yu, X. Data-driven planning of electric vehicle charging infrastructure: A case
study of Sydney, Australia. IEEE Trans. Smart Grid 2021, 12, 3289–3304. [CrossRef]
95. Chen, Q.; Folly, K.A. Application of Artificial Intelligence for EV Charging and Discharging Scheduling and Dynamic Pricing: A
Review. Energies 2022, 16, 146. [CrossRef]
96. Dharavat, N.; Golla, N.K.; Sudabattula, S.K.; Velamuri, S.; Kantipudi, M.V.V.; Kotb, H.; AboRas, K.M. Impact of plug-in electric
vehicles on grid integration with distributed energy resources: A review. Front. Energy Res. 2023, 10, 1099890. [CrossRef]
97. Cao, Y.; Cui, J.; Liu, S.; Li, X.; Zhou, Q.; Hu, C.; Zhuang, Y.; Liu, Z. A Holistic Review on E-Mobility Service Optimization:
Challenges, Recent Progress and Future Directions. IEEE Trans. Transp. Electrif. 2023. [CrossRef]
98. Mosayebi, M.; Gheisarnejad, M.; Farsizadeh, H.; Andresen, B.; Khooban, M.H. Smart extreme fast portable charger for electric
vehicles-based artificial intelligence. IEEE Trans. Circuits Syst. II Express Briefs 2022, 70, 586–590. [CrossRef]
99. Lazari, V.; Chassiakos, A. Multi-Objective Optimization of Electric Vehicle Charging Station Deployment Using Genetic Algo-
rithms. Appl. Sci. 2023, 13, 4867. [CrossRef]
100. Rimal, B.P.; Kong, C.; Poudel, B.; Wang, Y.; Shahi, P. Smart electric vehicle charging in the era of internet of vehicles, emerging
trends, and open issues. Energies 2022, 15, 1908. [CrossRef]
101. Herrlich, E.; Schaich, E.; Wagner, S.; Uckelmann, D. Parking and charging: New concepts for the use of intelligent charging
infrastructure in car parks. In iCity. Transformative Research for the Livable, Intelligent, and Sustainable City; Springer: Cham,
Switzerland, 2022; pp. 183–195.
102. Hu, S.; Mao, F.; Chen, J.; Li, Z. A review of charging infrastructure layout planning and electric vehicle traffic simulation
application. IOP Conf. Ser. Earth Environ. Sci. 2021, 821, 12028. [CrossRef]
103. Rani, K.S.; Suguna, R.; Kumar, R.S.; Ram, P.S.; Haran, V.H. A smart AIS based portable wireless electric charging vehicles. J. Phys.
Conf. Ser. 2021, 1916, 12145. [CrossRef]
104. Sumanasena, V.; Gunasekara, L.; Kahawala, S.; Mills, N.; De Silva, D.; Jalili, M.; Sierla, S.; Jennings, A. Artificial Intelligence for
Electric Vehicle Infrastructure: Demand Profiling, Data Augmentation, Demand Forecasting, Demand Explainability and Charge
Optimisation. Energies 2023, 16, 2245. [CrossRef]
105. Tabassum, T.; Khalghani, M.R. Cybersecurity Challenges in Microgrids: Inverter-Based Resources and Electric Vehicles. In Power
Systems Cybersecurity: Methods, Concepts, and Best Practices; Springer: Berlin/Heidelberg, Germany, 2023; pp. 91–114.
106. Elma, O.; Cali, U.; Kuzlu, M. An overview of bidirectional electric vehicles charging system as a Vehicle to Anything (V2X) under
Cyber–Physical Power System (CPPS). Energy Reports 2022, 8, 25–32. [CrossRef]
107. Ahmad, F.; Ashraf, I.; Iqbal, A.; Marzband, M.; Khan, I. A novel AI approach for optimal deployment of EV fast charging station
and reliability analysis with solar based DGs in distribution network. Energy Rep. 2022, 8, 11646–11660. [CrossRef]
108. Shurrab, M.; Singh, S.; Otrok, H.; Mizouni, R.; Khadkikar, V.; Zeineldin, H. An efficient vehicle-to-vehicle (v2v) energy sharing
framework. IEEE Internet Things J. 2021, 9, 5315–5328. [CrossRef]
109. Kakkar, R.; Kumari, A.; Gupta, R.; Agrawal, S.; Tanwar, S. Artificial Neural Network and Game Theory for Secure Optimal
Charging Station Selection for EVs. In Proceedings of the IEEE INFOCOM 2023-IEEE Conference on Computer Communications
Workshops (INFOCOM WKSHPS), Hoboken, NJ, USA, 20 May 2023; IEEE: Toulouse, France, 2023; pp. 1–6.
110. Vershinin, Y.A.; Pashchenko, F.; Olaverri-Monreal, C. Technologies for Smart Cities; Springer Nature: Berlin, Germany, 2022;
ISBN 3031055160.
111. Silva, C.; Martins, F. Traffic flow prediction using public transport and weather data: A medium sized city case study. In Trends
and Innovations in Information Systems and Technologies; Springer: Berlin/Heidelberg, Germany, 2020; Volume 28, pp. 381–390.
112. Zhao, J.; Xi, X.I.; Na, Q.I.; Wang, S.; Kadry, S.N.; Kumar, P.M. The technological innovation of hybrid and plug-in electric vehicles
for environment carbon pollution control. Environ. Impact Assess. Rev. 2021, 86, 106506. [CrossRef]
113. Piccialli, F.; Giampaolo, F.; Prezioso, E.; Crisci, D.; Cuomo, S. Predictive analytics for smart parking: A deep learning approach in
forecasting of iot data. ACM Trans. Internet Technol. 2021, 21, 1–21. [CrossRef]
114. Shatnawi, N.; Al-Omari, A.A.; Al-Qudah, H. Optimization of bus stops locations using GIS techniques and artificial intelligence.
Procedia Manuf. 2020, 44, 52–59. [CrossRef]
115. Abduljabbar, R.; Dia, H.; Liyanage, S.; Bagloee, S.A. Applications of artificial intelligence in transport: An overview. Sustainability
2019, 11, 189. [CrossRef]
116. Kumar, P.; Hariharan, K.; Manikandan, M.S.K. Hybrid long short-term memory deep learning model and Dijkstra’s Algorithm
for fastest travel route recommendation considering eco-routing factors. Transp. Lett. 2023, 15, 926–940.
117. Nousias, S.; Tselios, C.; Bitzas, D.; Amaxilatis, D.; Montesa, J.; Lalos, A.S.; Moustakas, K.; Chatzigiannakis, I. Exploiting
gamification to improve eco-driving behaviour: The GamECAR approach. Electron. Notes Theor. Comput. Sci. 2019, 343, 103–116.
[CrossRef]
118. Ghaffarpasand, O.; Jahromi, A.M.; Maleki, R.; Karbassiyazdi, E.; Blake, R. Intelligent geo-sensing for moving toward smart,
resilient, low emission, and less carbon transport. In Artificial Intelligence and Data Science in Environmental Sensing; Elsevier:
Amsterdam, The Netherlands, 2022; pp. 39–55.
Energies 2023, 16, 7988 32 of 34

119. Tselios, C.; Nousias, S.; Bitzas, D.; Amaxilatis, D.; Akrivopoulos, O.; Lalos, A.S.; Moustakas, K.; Chatzigiannakis, I. Enhancing an
eco-driving gamification platform through wearable and vehicle sensor data integration. In Proceedings of the Ambient Intelli-
gence: 15th European Conference, AmI 2019, Rome, Italy, 13–15 November 2019; Proceedings 15. Springer: Berlin/Heidelberg,
Germany, 2019; pp. 344–349.
120. Delnevo, G.; Di Lena, P.; Mirri, S.; Prandi, C.; Salomoni, P. On combining Big Data and machine learning to support eco-driving
behaviours. J. Big Data 2019, 6, 1–15. [CrossRef]
121. Boru İpek, A. Multi-Objective Simulation Optimization Integrated with Analytic Hierarchy Process and Technique for Order
Preference by Similarity to Ideal Solution for Pollution Routing Problem. Transp. Res. Rec. 2023, 2677, 1658–1674. [CrossRef]
122. Mandžuka, S. Cooperative systems in traffic technology and transport. In New Technologies, Development and Application 4;
Springer: Berlin/Heidelberg, Germany, 2019; pp. 299–308.
123. Zhou, J. Research on Multi-objective Green Vehicle Path Optimization Based on Whale Algorithm. In Proceedings of the 2023
International Conference on Networking, Informatics and Computing (ICNETIC), Palermo, Italy, 29–31 May 2023; pp. 385–389.
124. Fernández, E.I.; Kühne, N.G.; Mulero, N.B.; Jara, A.J. Advancing Sustainability Impact Assessment: A Comprehensive Tool
for Low Emissions Zone Management. In Proceedings of the 2023 8th International Conference on Smart and Sustainable
Technologies (SpliTech), Bol, Hotel Elaphusa, 20–23 June 2023; pp. 1–6.
125. Gong, K. Research and analysis on technical problems of new energy vehicles in China based on big data and artificial intelligence
algorithm. J. Phys. Conf. Ser. 2021, 2138, 12020. [CrossRef]
126. Manoharan, D.S. An improved safety algorithm for artificial intelligence enabled processors in self driving cars. J. Artif. Intell.
Capsul. Networks 2019, 1, 95–104. [CrossRef]
127. Miao, Y.; Qiu, Z.; Zhang, M.; He, B.; Zhang, X.; Li, M. Integrated artificial intelligence and life cycle assessment to predict
environmental and economic impacts for natural gas hybrid electric vehicle with multi-dimension. J. Phys. Conf. Ser. 2022,
2198, 12016. [CrossRef]
128. Ali, S.S.; Choi, B.J. State-of-the-art artificial intelligence techniques for distributed smart grids: A review. Electronics 2020, 9, 1030.
[CrossRef]
129. Lytras, M.D.; Chui, K.T. The recent development of artificial intelligence for smart and sustainable energy systems and applications.
Energies 2019, 12, 3108. [CrossRef]
130. You, S.; Zhao, Y.; Mandich, M.; Cui, Y.; Li, H.; Xiao, H.; Fabus, S.; Su, Y.; Liu, Y.; Yuan, H. A review on artificial intelligence for grid
stability assessment. In Proceedings of the 2020 IEEE International Conference on Communications, Control, and Computing
Technologies for Smart Grids (SmartGridComm), Virtual, 11–13 November 2020; pp. 1–6.
131. Hussain, M.; Dhimish, M.; Titarenko, S.; Mather, P. Artificial neural network based photovoltaic fault detection algorithm
integrating two bi-directional input parameters. Renew. Energy 2020, 155, 1272–1292. [CrossRef]
132. Ahmad, T.; Chen, H.; Shah, W.A. Effective bulk energy consumption control and management for power utilities using artificial
intelligence techniques under conventional and renewable energy resources. Int. J. Electr. Power Energy Syst. 2019, 109, 242–258.
[CrossRef]
133. Seh, Z.W.; Jiao, K.; Castelli, I.E. Artificial intelligence and machine learning in energy storage and conversion. Energy Adv. 2023, 2,
1237–1238. [CrossRef]
134. Zhou, S.; Hu, Z.; Gu, W.; Jiang, M.; Zhang, X.-P. Artificial intelligence based smart energy community management: A
reinforcement learning approach. CSEE J. Power Energy Syst. 2019, 5, 1–10. [CrossRef]
135. Darab, C.; Tarnovan, R.; Turcu, A.; Martineac, C. Artificial intelligence techniques for fault location and detection in distributed
generation power systems. In Proceedings of the 2019 8th International Conference on Modern Power Systems (MPS), Cluj
Napoca, Romania, 21–23 May 2019; pp. 1–4.
136. Arrieta, A.B.; Díaz-Rodríguez, N.; Del Ser, J.; Bennetot, A.; Tabik, S.; Barbado, A.; García, S.; Gil-López, S.; Molina, D.; Benjamins,
R. Explainable Artificial Intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI. Inf. Fusion
2020, 58, 82–115. [CrossRef]
137. Xu, Y.; Ahokangas, P.; Louis, J.-N.; Pongrácz, E. Electricity market empowered by artificial intelligence: A platform approach.
Energies 2019, 12, 4128. [CrossRef]
138. Sulaiman, A.; Nagu, B.; Kaur, G.; Karuppaiah, P.; Alshahrani, H.; Reshan, M.S.A.; AlYami, S.; Shaikh, A. Artificial Intelligence-
Based Secured Power Grid Protocol for Smart City. Sensors 2023, 23, 8016. [CrossRef]
139. Omitaomu, O.A.; Niu, H. Artificial intelligence techniques in smart grid: A survey. Smart Cities 2021, 4, 548–568. [CrossRef]
140. Wang, H.; Li, Z.; Meng, Q. Design and Research of Smart Grid Based on Artificial Intelligence. In Proceedings of the 2023 IEEE
International Conference on Image Processing and Computer Applications (ICIPCA), Changchun, China, 28–30 June 2023; pp.
1467–1471.
141. Zhao, X.; Guo, Y.; Guo, X.; Li, H. Artificial Intelligence Applications and Prospects for The Smart Grid. In Proceedings of the 2023
Panda Forum on Power and Energy (PandaFPE), Chengdu, China, 27–30 April 2023; pp. 1844–1848.
142. Davlyatov, S. Artificial Intelligence Techniques: Smart Way to Smart Grid. In Proceedings of the 2023 International Conference on
Artificial Intelligence and Smart Communication (AISC), Greater Noida, India, 27–29 January 2023; IEEE: Toulouse, France, 2023;
pp. 838–842.
143. Onen, A. Role of artificial intelligence in smart grids. Electr. Eng. 2022, 104, 231. [CrossRef]
Energies 2023, 16, 7988 33 of 34

144. Yadeo, D.; Sen, S.; Saxena, V. Smart grid: Solid-state transformer and load forecasting techniques using artificial intelligence. In
Artificial Intelligence and Machine Learning in Smart City Planning; Elsevier: Amsterdam, The Netherlands, 2023; pp. 181–197.
145. Fang, X.; Lai, Y.; Fan, L.; Sun, Z.; Wang, Y.; Xuan, Y. Power grid user asset value evaluation method and application under the
background of artificial intelligence and smart grid. J. Comput. Methods Sci. Eng. 2023, 23, 351–360. [CrossRef]
146. Khan, A.A.; Laghari, A.A.; Rashid, M.; Li, H.; Javed, A.R.; Gadekallu, T.R. Artificial intelligence and blockchain technology for
secure smart grid and power distribution Automation: A State-of-the-Art Review. Sustain. Energy Technol. Assess. 2023, 57, 103282.
147. Rodgers, W.; Cardenas, J.A.; Gemoets, L.A.; Sarfi, R.J. A smart grids knowledge transfer paradigm supported by experts’
throughput modeling artificial intelligence algorithmic processes. Technol. Forecast. Soc. Chang. 2023, 190, 122373. [CrossRef]
148. Khan, M.A.; Saleh, A.M.; Waseem, M.; Sajjad, I.A. Artificial Intelligence Enabled Demand Response: Prospects and Challenges in
Smart Grid Environment. IEEE Access 2022, 11, 1477–1505. [CrossRef]
149. Behara, R.K.; Saha, A.K. artificial intelligence control system applied in smart grid integrated doubly fed induction generator-
based wind turbine: A review. Energies 2022, 15, 6488. [CrossRef]
150. Khosrojerdi, F.; Akhigbe, O.; Gagnon, S.; Ramirez, A.; Richards, G. Integrating artificial intelligence and analytics in smart grids:
A systematic literature review. Int. J. Energy Sect. Manag. 2021, 16, 318–338.
151. Kumar, R.S.; Saravanan, S.; Pandiyan, P.; Tiwari, R. Impact of artificial intelligence techniques in distributed smart grid monitoring
system. In Smart Energy and Electric Power Systems; Elsevier: Amsterdam, The Netherlands, 2023; pp. 79–103.
152. Chaurasia, K.; Kamath, H.R. Artificial Intelligence and Machine Learning Based: Advances in Demand-Side Response of
Renewable Energy-Integrated Smart Grid. In Proceedings of the Smart Systems: Innovations in Computing: Proceedings of SSIC
2021; Springer: Berlin/Heidelberg, Germany, 2022; pp. 195–207.
153. Slama, S. Ben Prosumer in smart grids based on intelligent edge computing: A review on Artificial Intelligence Scheduling
Techniques. Ain Shams Eng. J. 2022, 13, 101504. [CrossRef]
154. Zambrano, A.F.; Giraldo, L.F. Solar irradiance forecasting models without on-site training measurements. Renew. Energy 2020,
152, 557–566. [CrossRef]
155. Ruhnau, O.; Hennig, P.; Madlener, R. Economic implications of forecasting electricity generation from variable renewable energy
sources. Renew. Energy 2020, 161, 1318–1327. [CrossRef]
156. Dutta, S.; Sadhu, P.K.; Cherikuri, M.; Mohanta, D.K. Application of artificial intelligence and machine learning techniques in
island detection in a smart grid. Intell. Renew. Energy Syst. 2022, 79–109. [CrossRef]
157. Xin, C.; Yu, Z.; Yu, J.; Bian, J.; Cao, Q. Application of Smart Grid Energy Storage Technology Based on Artificial Intelligence
Technology. In Proceedings of the 2021 3rd International Conference on Artificial Intelligence and Advanced Manufacture,
Manchester, UK, 23–25 October 2021; pp. 1483–1487.
158. Liu, Z.; Gao, Y.; Liu, B. An artificial intelligence-based electric multiple units using a smart power grid system. Energy Rep. 2022,
8, 13376–13388. [CrossRef]
159. Yun, J.J.; Zhao, X.; Ma, L.; Liu, Z.; Yang, B. Open innovation dynamics and evolution in the mobile payment industry–comparative
analysis among Daegu, Cardiff, and Nanjing. Eur. Plan. Stud. 2023, 31, 862–884. [CrossRef]
160. Gönenç, A.; Acar, E.; Demir, İ.; Yılmaz, M. Artificial Intelligence Based Regression Models for Prediction of Smart Grid Stability.
In Proceedings of the 2022 Global Energy Conference (GEC), Batman, Turkey, 26–29 October 2022; pp. 374–378.
161. Ucar, F. A Comprehensive Analysis of Smart Grid Stability Prediction along with Explainable Artificial Intelligence. Symmetry
2023, 15, 289. [CrossRef]
162. Zheng, L.; Zhang, S.; Huang, H.; Liu, R.; Cai, M.; Bian, Y.; Chang, L.; Du, H. Artificial intelligence-driven rechargeable batteries in
multiple fields of development and application towards energy storage. J. Energy Storage 2023, 73, 108926. [CrossRef]
163. He, Z.; Guo, W.; Zhang, P. Performance prediction, optimal design and operational control of thermal energy storage using
artificial intelligence methods. Renew. Sustain. Energy Rev. 2022, 156, 111977. [CrossRef]
164. Rojek, I.; Mikołajewski, D.; Mroziński, A.; Macko, M. Machine Learning-and Artificial Intelligence-Derived Prediction for Home
Smart Energy Systems with PV Installation and Battery Energy Storage. Energies 2023, 16, 6613. [CrossRef]
165. Chakrapani, K.; Kavitha, T.; Safa, M.I.; Kempanna, M.; Chakrapani, B. Applications of Artificial Intelligence in Intelligent
Combustion and Energy Storage Technologies. In Applications of Big Data and Artificial Intelligence in Smart Energy Systems; River
Publishers: Roma, Italy, 2023; pp. 27–45.
166. Long, H.U.O.; Zhang, Y.; Xin, C. Artificial Intelligence Applications in Distributed Energy Storage Technologies. Power Gener.
Technol. 2022, 43, 707.
167. Zehra, S.S.; Rahman, A.U.; Armghan, H.; Ahmad, I.; Ammara, U. Artificial intelligence-based nonlinear control of renewable
energies and storage system in a DC microgrid. ISA Trans. 2022, 121, 217–231. [CrossRef] [PubMed]
168. Abdalla, A.N.; Nazir, M.S.; Tao, H.; Cao, S.; Ji, R.; Jiang, M.; Yao, L. Integration of energy storage system and renewable energy
sources based on artificial intelligence: An overview. J. Energy Storage 2021, 40, 102811. [CrossRef]
169. Athari, M.H.; Ardehali, M.M. Operational performance of energy storage as function of electricity prices for on-grid hybrid
renewable energy system by optimized fuzzy logic controller. Renew. Energy 2016, 85, 890–902. [CrossRef]
170. Zahedi, R.; Ardehali, M.M. Power management for storage mechanisms including battery, supercapacitor, and hydrogen of
autonomous hybrid green power system utilizing multiple optimally-designed fuzzy logic controllers. Energy 2020, 204, 117935.
[CrossRef]
Energies 2023, 16, 7988 34 of 34

171. Yunhao, H.; Xin, C.; Erde, W.; Jianfeng, C.; Tingting, H.; Yang, M. Hierarchical control strategy for distributed energy storage units
in isolated DC microgrid. In Proceedings of the 2019 Chinese Control Conference (CCC), Guangzhou, China, 27–30 July 2019;
pp. 7410–7415.
172. Franco, A.A.; Barnard, A.S. Artificial Intelligence in Electrochemical Energy Storage. Batter. Supercaps 2022, 5, e202200149.
[CrossRef]
173. Lee, D.; Ooka, R.; Matsuda, Y.; Ikeda, S.; Choi, W. Experimental analysis of artificial intelligence-based model predictive control
for thermal energy storage under different cooling load conditions. Sustain. Cities Soc. 2022, 79, 103700. [CrossRef]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.