2025 5th International Conference on Innovative Research in Applied Science, Engineering and Technology (IRASET) | 979-8-3315-3297-0/25/$31.

00 ©2025 IEEE | DOI: 10.1109/IRASET64571.2025.11008203

Intelligent Control Systems for Power and Energy

Management in Smart Buildings

Benkhada Aya Obayd Merouane Farajy Nouhaila OUABIDA Elhoussaine

Advanced Systems Engineering Advanced Systems Engineering Engineering Sciences Advanced Systems Engineering
Lab Lab Lab Lab
National School of Applied National School of Applied National School of Applied National School of Applied
Sciences, Sciences, Sciences, Sciences,
Ibn Tofail University Ibn Tofail University Ibn Tofail University Ibn Tofail University
Kenitra, Morocco Kenitra, Morocco Kenitra, Morocco Kenitra, Morocco
[email protected] [email protected] [email protected] [email protected]

Abstract— In this paper, we present a systematic literature Component Analysis (PCA), have demonstrated their
review on the role of intelligent control systems in optimizing capability to intelligently manage power consumption, ensuring
power and energy management for smart buildings. Using the accurate and efficient energy allocation [2].
PRISMA methodology to ensure the quality of the review, we Systems for managing home energy consumption (HEMS) play
analyzed 214 relevant articles retrieved from databases such as
a vital role in integrating renewable energy sources and storage
ScienceDirect, Scopus, Google Scholar, and IEEE Xplore. Our
review synthesizes recent findings and identifies trends in the solutions into smart buildings. These systems effectively
integration of intelligent systems for improving energy schedule energy usage based on electricity pricing models,
efficiency, balancing energy consumption, and enabling smart leading to reduced peak average ratios (PAR), lower electricity
automation in buildings. This work highlights the latest costs, and enhanced energy system stability [3].
advances, key research areas, and future opportunities for Multi-agent reinforcement learning platforms, such as
deploying intelligent control systems to address energy GridLearn, align building-level energy goals with grid-level
challenges in smart buildings. objectives. By managing behind-the-meter resources and
regulating voltage, these platforms ensure energy efficiency
Keywords—Intelligent Control Systems, Energy Management,
while preserving occupant privacy and comfort [4].
Smart Buildings, Energy Efficiency, Renewable Energy
Integration, SLR. The built environment accounts for over one-third of global
carbon emissions, making energy use in buildings a key focus
I. INTRODUCTION for sustainability efforts. However, renewable energy sources
are often intermittent, meaning they cannot always be accessed
This study offers a comprehensive overview of the current on demand. To address this challenge, complementary
state of intelligent control systems utilized for power and energy solutions are essential to maximize the efficiency of renewable
management in smart buildings. These systems incorporate energy utilization. Among these solutions, hydrogen-based
advanced technologies and methodologies to enhance energy systems have demonstrated significant potential. Hydrogen can
efficiency, lower operational costs, and support sustainable store excess energy or be directly converted into thermal
practices within the built environment. energy, providing a sustainable method for balancing energy
Decentralized automation systems have been developed to supply and demand. Moreover, incorporating machine learning
manage energy demand more efficiently by employing load- models, optimization algorithms, and smart control systems in
shifting techniques. These systems not only minimize energy building energy management can reduce energy consumption,
costs during peak periods but also ensure occupant comfort by lower carbon emissions, and enhance thermal comfort.
considering thermal and visual conditions in energy Dynamic load-priority-based scheduling approaches further
management strategies [1]. contribute to optimizing energy usage in smart buildings. These
Applying machine learning and deep learning models in smart adaptive algorithms address the uncertainties of renewable
buildings has revolutionized energy optimization. Techniques energy sources, prioritize energy loads according to user
including Random Forest (RF) and Deep Neural Networks preferences, and reduce peak loads, ensuring cost-effectiveness
(DNN) combined with feature reduction methods like Principal and operational efficiency [6].

979-8-3315-3297-0/25/$31.00 ©2025 IEEE

Authorized licensed use limited to: NATIONAL INSTITUTE OF TECHNOLOGY SILCHAR. Downloaded on July 15,2025 at 15:31:46 UTC from IEEE Xplore. Restrictions apply.
 Wireless communication technologies also enhance energy
management in smart home systems by optimizing power 4. Screening and selection
consumption in both active and standby modes. These We began with 214 articles, removing duplicates and
technologies offer scalable and cost-effective solutions for ineligible items, leaving 208 for further review. After
integrating energy management systems into modern building assessing titles and abstracts, we excluded 44 reports,
infrastructures [7]. ultimately including 164 relevant articles in the study.
In conclusion, intelligent control systems represent a
transformative approach for managing energy in smart 5. Inclusion and Reporting
buildings. By leveraging cutting-edge technologies, such as
machine learning, decentralized automation, and renewable
energy integration, these systems address key challenges in
energy efficiency, cost reduction, and sustainability.
Therefore, while the integration of renewable energy into smart
buildings and the use of advanced data management techniques
offer numerous benefits, addressing cybersecurity concerns is
essential to ensure the safe and secure implementation of these
technologies. Our Systematic Literature Review (SLR) on the
impact of renewable energy on energy efficiency and the
integration of advanced data management techniques in smart
buildings is positioned as an in-depth survey, analyzing 164
relevant articles from Google Scholar, Direct Science, Scopus,
and IEEE databases. By focusing on the results in a targeted
manner, this study aims to provide an in-depth look at recent
advances, key research areas, and future opportunities in the
field of renewable energy integration and advanced data
management for intelligent energy systems in smart buildings.

II. METHODS
A. Research Design
Preferred Reporting Items for Systematic Reviews and
Meta-Analyses in English. PRISMA is a set of guidelines and
recommendations designed to improve the quality of publication Fig. 1. Research structure according to the PRISMA Protocol.
of systematic literature reviews and meta-analyses in medical
research and other fields. These guidelines aim to ensure the For this study, we gathered resources from several key
transparency and methodological quality of these studies, platforms, including Direct Science, Scopus, IEEE, and Google
providing a framework for the preparation and presentation of Scholar. These diverse sources offered a wide array of articles,
results clearly and rigorously. which allowed us to cover all aspects of smart buildings, energy
management, and intelligent control systems. By leveraging
B. Identification: Inclusion and Exclusion these databases, we ensured that the selected studies were
1. Data Base selection relevant and of high-quality, contributing to the depth and
We selected databases from 2014 to 2025, ensuring recent breadth of the review.
and relevant studies on smart buildings, energy III. RESULTS
management, and intelligent control systems. We used
Google Scholar, Direct Science, Scopus, and IEEE for
comprehensive coverage. This section synthesizes key insights from the analyzed articles,
highlighting advancements in AI-driven energy management,
2. Search Strings smart grids, and smart buildings.
We used targeted search strings: "smart buildings",
"energy management", and "intelligent control systems" to A. VOSviewer Analysis of authors' citations
focus on key topics and retrieve relevant articles related to Among the authors analyzed in this dataset, several
our research area. individuals stand out for their significant collaborations in co-
authoring articles, each having made substantial contributions to
3. Inclusion and Exclusion criteria multiple documents. Siano Pierluigi, Alhelou Hassan Haes, and
We included open-access articles published in English Khalid Muhammad emerge as central figures within the co-
from 2014 to 2025. We excluded non-English articles, authorship network. These authors have collectively co-
those with inaccessible content, and duplicates, ensuring authored numerous papers, focusing on energy management
data cleanliness. systems, renewable energy integration, and optimization

Authorized licensed use limited to: NATIONAL INSTITUTE OF TECHNOLOGY SILCHAR. Downloaded on July 15,2025 at 15:31:46 UTC from IEEE Xplore. Restrictions apply.
 techniques in smart buildings and factories. Their research is In addition to Siano, Alhelou, and Khalid, the co-authorship
pivotal in advancing methodologies for optimizing energy network reveals several clusters of authors who have
efficiency through intelligent control systems and advanced data significantly contributed to research in smart energy
management. management systems. These clusters represent groups of
researchers collaborating closely, with some exhibiting stronger
The force metrics further emphasize their considerable intra-cluster connections.
influence within the co-authorship network, with Siano,
Alhelou, and Khalid exhibiting substantial total privilege For instance, one prominent cluster includes authors such as
strengths. Their collaborations provide valuable insights into the Jean-Louis Scartezzini and Haris Doukas, who have extensively
intersection of renewable energy, smart technologies, and worked together on integrating renewable energy into smart
energy management in industrial and residential settings. The buildings and developing intelligent energy management
combined contributions of these researchers underline the strategies. Their collaborative efforts have led to advancements
importance of interdisciplinary approaches and collaborative in optimizing energy efficiency through innovative technologies
efforts in driving innovation within the energy sector, and methodologies.
particularly in the context of creating efficient and sustainable
energy systems for smart buildings and factories. Another notable cluster features Anuar Nor Badrul and
Bolandnazar Elham, whose joint research focuses on the
application of artificial intelligence in industrial processes,
particularly in enhancing energy efficiency and system
optimization. Their work complements the broader research
landscape by providing specialized insights into the role of AI
in modern energy systems.
These clusters, while distinct, collectively enrich the overall
discourse on smart energy management. Their contributions
underscore the importance of collaborative research in
advancing knowledge and innovation within the energy sector,
particularly in the context of developing efficient and
sustainable systems for smart buildings and industrial
applications.
Visualizing these clusters highlights the interconnected nature
of research efforts and the pivotal role of teamwork in driving
progress in energy efficiency and management.
Fig. 2. Co-authorship.

B. VOSviewer Analysis of keywords occurrence

Fig. 4. Co-occurrence.

Of the 10 keywords examined, several keywords exhibit

noteworthy strengths in terms of ’total link strength,’ signifying
their robust connections to other keywords within the context
of the collected articles on the adoption of renewable energy
Fig. 3. Author Clusters in Co-Authorship Network. and AI for enhancing energy efficiency in intelligent buildings.

Authorized licensed use limited to: NATIONAL INSTITUTE OF TECHNOLOGY SILCHAR. Downloaded on July 15,2025 at 15:31:46 UTC from IEEE Xplore. Restrictions apply.
 Here are some of these keywords and their respective total link
strengths:
Intelligent Buildings (92): The keyword” intelligent
buildings” occurs frequently and demonstrates a substantial
total link strength, indicating its central role in discussions on
enhancing energy efficiency. This underscores its importance
as a foundational concept in the exploration of the subject
matter.
Energy Management (75): ”Energy management” emerges as
a key term, showing strong associations with other keywords.
This highlights its pivotal role in discussions surrounding
advanced technologies for energy optimization, suggesting its
prominence as a focal point in the studies.
Smart Power Grids (63): With significant occurrences and a
notable total link strength, ”smart power grids” stand out as a Fig. 5. Word cloud.
crucial aspect of the research, showcasing their significance and
interconnection with other keywords, particularly in the context In our examination of the keyword cloud, the most
of optimizing energy distribution. prominent terms reflect major themes and central discussions
Energy Efficiency (55): The term ”energy efficiency” within the context of energy efficiency and smart systems. The
possesses a substantial influence and relevance within the largest dimensions, being the most frequent, provide significant
context of the articles, indicating its role in driving insights into the priorities and perspectives within this field. A
advancements in energy-saving technologies and smart thorough analysis of the collected articles revealed that
building systems. keywords such as "Energy," "Management," "Smart Systems,"
Automation (49): Occurring frequently, ”automation” exhibits "Optimization," "Control," and "Renewable" recurrently emerge
in the discourse. These terms underscore the importance of
a strong total link strength, suggesting its association with AI-
smart energy management, the integration of renewable sources,
driven control systems and its role in the discourse on and the optimization of energy systems as crucial themes in the
intelligent energy management. current strategies for improving energy efficiency.
Smart Grid (42): With considerable occurrences, ”smart grid”
is highlighted as a significant keyword, emphasizing its
interconnection with other renewable energy and energy
transmission concepts in the research.
Renewable Energy Source (38): With a strong presence in the
dataset, ”renewable energy source” showcases its relevance and
strong connections within the scope of the collected articles,
particularly in the context of sustainable energy solutions.
Demand Side Management (36): ”Demand side
management,” demonstrating a notable total link strength,
underscores its importance in the development and
optimization of energy consumption strategies, reflecting its
key role in energy-efficient solutions.
Electric Power Transmission (33): The keyword ”electric
power transmission” signifies its significance in discussions,
particularly related to energy distribution and control in smart
grids, showcasing its critical role in overall energy efficiency
discourse.
Solar Energy (31): The term ”solar energy” plays a key role in Fig. 6. Descriptive statistics by year.
discussions surrounding renewable energy integration,
highlighting its importance in the transition towards more 2. Processing: The descriptive analysis by year highlights
sustainable and self-sufficient energy systems. a clear shift in the distribution of research items related
to energy efficiency and the use of renewable energy
and AI. The year 2024 stands out with a significant
C. Contextual analysis using NVIVO peak, marking the highest number of matching items,
1. Pre-processing: Word cloud, after eliminating words to which indicates an intensifying focus on the topic. This
ignore increase likely reflects recent advancements and
growing interest in smart systems, energy management,
and sustainable technologies. The following years, 2025
and 2023, show a noticeable decrease in comparison,
although they still maintain a meaningful number of

Authorized licensed use limited to: NATIONAL INSTITUTE OF TECHNOLOGY SILCHAR. Downloaded on July 15,2025 at 15:31:46 UTC from IEEE Xplore. Restrictions apply.
 studies, suggesting a continuous interest in these areas. reference type. The analysis categorizes items into journal
The sharp spike in 2024 may represent an influx of articles, conference papers, and book sections. Journal articles
research stemming from specific initiatives or funding dominate the dataset, comprising 81.88% of the total references,
opportunities, driving attention to energy efficiency emphasizing the preference for peer-reviewed research in this
solutions during that time. This analysis underscores the field. Conference papers make up 8.70%, indicating ongoing
importance of considering the temporal dynamics when academic discussions and the significance of recent
evaluating research trends in the field of energy developments. Book sections account for 9.42%, reflecting a
management and AI. smaller contribution, while full books are virtually absent. This
distribution highlights the central role of journal articles in
advancing research on the integration of AI and renewable
energy within sugar factories.

Fig. 7. Descriptive statistics by database.

Contextual analysis using NVIVO resulted in descriptive

statistical processing by the database. The graph illustrates the
distribution of articles across different database providers. A Fig. 9. Descriptive statistics by journal.
total of 72 articles were found in ScienceDirect, 56 in Google
Scholar, 12 in Scopus, and 3 in IEEE Xplore. The distribution is The descriptive analysis by title reveals a varied distribution
quite notable, with ScienceDirect contributing the largest share of references in the field of renewable energy and AI integration.
at 31.16%, followed by Scopus at 28.99%, and Google Scholar The analysis shows that there is no single title dominating the
at 24.64%. IEEE Xplore accounted for 15.22% of the total corpus. Instead, several titles contribute equally, with "Applied
articles. There are no entries for 'Non applicable' or 'Non Energy" and "Power Systems and IT" leading the distribution,
assignée' categories. Additionally, the initial discrepancy with each representing a significant portion of the total references.
the Web of Science database, which showed zero articles, was However, the category labeled "Other" stands out, accounting
attributed to a change in the database provider after exporting for 36.2% of the references. This category includes various
the RIS file from Zotero. The sources initially attributed to Web lesser-known works or niche titles. The dispersion across
of Science were, in fact, from IEEE Xplore. multiple titles reflects the interdisciplinary nature of the field,
where a range of studies addresses diverse aspects of renewable
energy and AI integration. The prominence of the "Other"
category underscores the broad scope of research that may not
align with the most commonly cited works but still significantly
contributes to advancing the field.

D. Analysis Grid
We analyzed 164 articles and selected 16 Research Papers
(RPs) to extract key elements related to methodology and
results. The selected studies focus on AI-driven energy
management, renewable energy integration, smart
buildings, and smart grid optimization, covering diverse
approaches such as swarm intelligence, deep learning, IoT,
and hybrid renewable systems. This classification provided a
structured understanding of advancements in smart energy
Fig. 8. Descriptive statistics by reference type. systems, refining research questions and highlighting impactful
technological innovations.
A descriptive analysis of the corpus on renewable energy and AI
adoption in sugar factories reveals the breakdown of items by

Authorized licensed use limited to: NATIONAL INSTITUTE OF TECHNOLOGY SILCHAR. Downloaded on July 15,2025 at 15:31:46 UTC from IEEE Xplore. Restrictions apply.
 TABLE I. SYNTHESIS GRID OF REPS

Problematic Methodology Results

The study proposes SWAPP, a decision-making algorithm

How can swarm using swarm intelligence (SI) agents for urban energy SWAPP surpasses contemporary reinforcement
intelligence optimize management. Key components include a K-Means learning (RL) techniques. and rule-based
energy storage classifier for tailored decision-making, a reward controllers (RBCs) by optimizing electricity
[8] management in smart cities mechanism for strategic performance assessment, and a costs, reducing carbon emissions, and enhancing
to improve cost efficiency, two-phase policy learning approach that combines grid stability, proving its effectiveness in
grid stability, and carbon cognitive and social learning. A sequential batch CityLearn's approach to building an energy
reduction? sampling method ensures efficient swarm training over management environment.
long decision horizons.

How can an intelligent

system be designed to The proposed system operates in two phases: data collection
The system successfully maintained an optimized
manage and monitor and monitoring of lighting conditions, followed by analysis
[9] brightness index of 79, ensuring efficient lighting
lighting in a smart city to and optimization of intelligent control using a numerical
management and improved energy performance.
optimize energy efficiency model.
and user comfort?

The proposed system integrates multiple components,

How can a hybrid solar- including a parabolic trough collector, hydrogen storage,
The system achieved 80.6% primary energy
wind energy system with wind turbine, and fuel cells, with a rule-based control
savings, a significant carbon dioxide reduction
hydrogen storage be strategy for intelligent energy management. A techno-
[10] rate, and lower overall energy costs, demonstrating
optimized to enhance grid economic-environmental analysis is conducted using
its effectiveness in optimizing energy efficiency
stability and reduce peak TRNSYS, an engineering equation solver, and MATLAB
and grid stability.
load? with a grey wolf optimization algorithm coupled with an
artificial neural network.

How can Deep

Reinforcement Learning The study proposes a DRL-based energy management
(DRL) be integrated into framework that dynamically adapts to environmental Experimental findings confirm that the deep
Building Energy conditions in real time to optimize electricity usage, HVAC reinforcement learning (DRL) method considerably
[11] Management Systems control, and energy savings. Compared to conventional reduces electricity costs while maintaining stable
(BEMS) to optimize rule-based and predictive control systems, the model's comfort levels, outperforming traditional energy
energy consumption and effectiveness is assessed using energy cost and room management models.
occupant satisfaction in temperature stability metrics.
business centers?

The study develops a network of wireless sensors (WSN)

How can ZigBee and LoRa that utilizes ZigBee and LoRa technologies afin de suivre
The system effectively monitors and controls
communication protocols et réguler temperature, humidity, and lighting in smart
smart building environments, leveraging the low-
be integrated into wireless environments. A multi-protocol gateway is designed to
[12] power, long-range capabilities of ZigBee and
sensor networks to enhance facilitate data transfer between the LoRa end device and a
LoRa technologies for scalable IoT applications
smart building energy central data collection (CDC) unit through ZigBee-based
in energy management.
management systems? XBee S2 transceivers. Performance metrics such as
communication range and data throughput. are analyzed.

How can AI-based real-

The study employs GRU-BSO and Real-Time Electricity
time energy management The system saved 17 kW and reduced costs by 47
Scheduling (RTES) with Price Forecasting (PF),
[13] optimize residential energy cents, proving its efficiency in home energy
integrated into an Internet of Energy framework using a
consumption and reduce management.
two-level communication network.
costs?
How can smart water
The study integrates photovoltaic panels, an
management and The study integrates photovoltaic panels, an ESP32
ESP32 microcontroller, and real-time sensor
photovoltaic pumping microcontroller, and real-time sensor monitoring to
monitoring to optimize energy-efficient water
[14] systems ensure a optimize energy-efficient water distribution. A battery
distribution. A charge controller and battery
sustainable water supply in and charge controller.system regulate the power supply
system regulate the power supply for solar-
rural areas? for solar-powered water pumps.
powered water pumps.

How can real-time Scenario 4 demonstrated a balanced competitive

The study introduces the High Street Electricity Market
metering and Peer-to-Peer environment, showcasing the model’s
model, enabling P2P energy trading while maintaining
(P2P) transactions enhance effectiveness in ensuring fair competition,
[15] traditional retail relationships. It analyzes regulatory
Smart Renewable Energy regulatory alignment, and enhanced energy self-
frameworks, economic models, and grid impact, with a
Communities while consumption within Smart Renewable Energy
real-world implementation in a Dublin residential
ensuring regulatory Communities.
community.

Authorized licensed use limited to: NATIONAL INSTITUTE OF TECHNOLOGY SILCHAR. Downloaded on July 15,2025 at 15:31:46 UTC from IEEE Xplore. Restrictions apply.
 Problematic Methodology Results
compliance and grid
stability?

How can deep learning and The study combines Deep Learning Neural Networks
Simulations on a 33-bus radial distribution
optimization techniques (DLNN) with the Artificial Bee Colony (ABC)
network show that the DLNN-ABC framework
improve voltage regulation optimization algorithm to optimize Voltage Source
[16] significantly improves voltage regulation,
in smart microgrids with Converter (VSC) controllers. EVs are leveraged as
reducing Root Mean Squared Error (RMSE) by
electric vehicle (EV) reactive power compensators to enhance voltage stability
50%, and outperforming traditional methods.
integration? while maintaining optimal state-of-charge (SoC) levels.

How can AI-driven

technologies enhance The study introduces the Integrated AI-based Smart
operational efficiency, Buildings Framework (IAI-DSBF) which analyzes, Simulation results show that IAI-DSBF improves
occupant comfort, and optimizes, and autonomously manages building building performance, enhances regulatory
[17]
sustainability in smart operations. It integrates AI-based decision-making, data compliance, and fosters community engagement,
buildings while addressing analytics, and automated systems to improve efficiency, achieving a building performance ratio of 99.1%.
regulatory and compliance, and social inclusivity.
socioeconomic challenges?
How can an intelligent
The study develops a web-based platform that monitors
system optimize energy The system improves energy efficiency, adapts to
sensor data and power consumption using AI-driven
usage in smart buildings evolving power usage patterns, and ensures
[18] energy conservation strategies. A Zigbee network in a
while ensuring adaptability continuous operation even during a bulk grid
star formation enables seamless communication between
and resilience to grid blackout.
client devices and a central server.
conditions?

How can deep learning The study applies Long Short-Term Memory (LSTM) The LSTM model achieves high accuracy and
improve energy neural networks to forecast hourly electricity usage using robustness in electricity consumption prediction,
[19] consumption prediction real-world data from a building with fan coils and gas optimizing energy management strategies and
and management in high- boilers. It also evaluates temperature control algorithms enabling fault diagnosis and renewable energy
occupancy buildings? for rooms with fan coil systems. integration in smart buildings.

The study utilizes Advanced Control and Monitoring

How can Intelligent with Digital Signal Processor (DSP) programming to
Distribution Systems (SDS) actively manage distributed energy sources. A Controller The proposed system enhances grid stability,
ensure stable control of Area Network (CANbus) enables grid synchronization, improves renewable energy integration, and
[20]
power flow and seamless while Ethernet and GSM communication support remote effectively reduces harmonic distortion,
integration of renewable monitoring. A harmonic mitigation strategy reduces promoting future smart energy management.
energy sources? Total Harmonic Distortion (THD) during DC-to-AC
conversion.

The study proposes a Software-Defined Networking

How can an efficient and
(SDN)-enabled Energy Management System (EMS) to Simulations demonstrate that the system generates
intelligent energy
optimize renewable energy integration, distribution, and 1000 W of solar energy and achieves a
management system be
[21] efficiency. A robust energy management algorithm geothermal heat exchange rate of 45.12 W per
implemented for a hybrid
dynamically controls the hybrid system using real-time day, ensuring sustainable and efficient energy
solar-geothermal integrated
data, addressing integration complexity, resource management.
system?
allocation, and storage challenges.

How can demand response

(DR) be effectively
The study reviews state-of-the-art research on DR in AI- The survey highlights key advancements in DR
integrated into AI-based
BEMS, examining DR techniques, AI and machine optimization, and the role of AI-driven adaptive
Building Energy
[22] learning applications, and adaptive control strategies for strategies, and identifies future challenges in
Management Systems (AI-
optimizing energy usage. It also explores future challenges achieving efficient and intelligent energy
BEMS) to optimize energy
and opportunities in DR implementation within AI-BEMS. management in smart buildings.
consumption and enhance
grid reliability?

The study develops a multimodal optimization system for

How can an intelligent Simulations using the METASYS building
energy resource management in data centers, integrating
building system optimize control system demonstrate the system’s
ZigBee-based sensor networks, tilt sensors, and
energy management and effectiveness, with successful Modbus
[23] temperature sensors for security monitoring. A semantic
ensure seamless integration communication for power conversion and
data model enhances data representation, while a
of various subsystems in distribution, validating its practical application
hierarchical platform with communication and control
smart buildings? in smart buildings.
layers ensures system efficiency.

Authorized licensed use limited to: NATIONAL INSTITUTE OF TECHNOLOGY SILCHAR. Downloaded on July 15,2025 at 15:31:46 UTC from IEEE Xplore. Restrictions apply.
 Demand-side management strategies help optimize energy
consumption by balancing grid loads and improving system
IV. DISCUSSION resilience. Battery storage technologies play a key role in
storing excess renewable energy and supplying it during
peak demand, reducing energy volatility and enhancing
The energy sector is rapidly evolving with the integration of overall grid stability [25].
artificial intelligence, machine learning, and renewable energy.
This transformation is particularly noticeable in smart buildings, 7. Community-Driven and Decentralized Energy
where energy management systems are improving efficiency, Models
sustainability, and reliability. Beyond individual buildings, decentralized energy models
1. Intelligent Energy Management and Optimization are emerging as efficient solutions for energy management.
Community-based smart grids allow local energy
The use of advanced technologies such as multi-agent deep production, such as solar microgrids, to be managed
reinforcement learning and AI-based control mechanisms collectively. Blockchain technology is also being explored
has significantly improved real-time energy optimization. to improve transparency in energy transactions and promote
These systems rely in the Internet of Things (IoT) networks sustainable energy sharing [29].
and big data analytics to monitor energy consumption,
manage demand, and predict energy needs, reducing waste 8. The Future of AI in intelligent buildings and
and improving overall efficiency [24]. Sustainability
2. Renewable Energy Integration and Smart Grid The combination of AI, IoT, and renewable energy is
Advancements shaping the next generation of smart buildings. Future
developments will focus on self-adaptive AI systems, real-
Hybrid energy solutions, including solar and geothermal time optimization, and enhanced grid interaction. To fully
systems, are enhancing the performance of smart buildings. realize these benefits, continued investment in
Smart power grids play a crucial role in integrating infrastructure, regulatory policies, and interdisciplinary
renewable energy, allowing better demand-side collaboration will be essential [30].
management and grid stability. These systems ensure a more
sustainable and resilient power supply by optimizing energy The growing integration of artificial intelligence and
distribution[25]. advanced computational methods with renewable energy is
no longer just a concept of the future it is actively shaping
3. Predictive Maintenance for Sustainable Operations the way the energy sector evolves. Research in this field
Machine learning-based predictive maintenance is helping highlights how AI-driven optimization is transforming
extend the lifespan of critical infrastructure and reduce energy management, predictive maintenance, and the
operational costs. By analyzing data from energy systems, seamless incorporation of renewable sources. These
AI-driven diagnostics can anticipate failures, enabling innovations are not only improving efficiency in industries
proactive maintenance and minimizing energy losses. This and smart buildings but are also fostering the development
approach ensures long-term sustainability and enhances of decentralized energy networks and intelligent grid
overall building efficiency [26]. systems. As these technologies continue to advance, there
is an urgent need for ongoing adaptation, whether in
4. Automation and Smart Decision-Making in Energy regulatory policies, infrastructure upgrades, or new
Management frameworks that support a smarter and greener sustainable
Automated control systems optimize heating, ventilation, energy landscape.
air conditioning, and lighting in buildings. These systems
dynamically adjust energy use based on environmental
conditions and user behavior, reducing unnecessary V. CONCLUSION
consumption. Cloud-based energy management systems This study provides a clear overview of how intelligent
further enhance automation, providing scalable and control systems are improving energy management in smart
efficient solutions for modern buildings[27]. buildings. By analyzing 164 research articles from Google
5. Cybersecurity Issues in Smart Energy Systems Scholar, ScienceDirect, Scopus, and IEEE Xplore, we identified
key advancements that are shaping the energy landscape of the
As smart buildings become more interconnected through future efficiency in buildings.
cloud computing and AI-driven systems, cybersecurity
risks increase. The integration of next-generation Artificial intelligence and machine learning are playing a major
communication networks exposes energy infrastructure to role in optimizing energy consumption, enabling real-time
potential threats. To ensure secure energy management, it is monitoring, and improving predictive maintenance. The
essential to implement encryption protocols, decentralized integration of renewable energy sources, such as solar and wind
authentication, and AI-based intrusion detection power, along with smart power grids and automation, is making
systems[28]. buildings more efficient and environmentally friendly.
6. Demand-side management and Energy Storage However, despite these advancements, challenges remain,
Solutions particularly in cybersecurity. As smart buildings become more

Authorized licensed use limited to: NATIONAL INSTITUTE OF TECHNOLOGY SILCHAR. Downloaded on July 15,2025 at 15:31:46 UTC from IEEE Xplore. Restrictions apply.
 interconnected, the the threat of cyber risks increases, making it [5] A. Kaabinejadian, A. Pozarlik, and C. Acar, “A
necessary to develop stronger security frameworks to protect systematic review of predictive, optimization, and smart control
data and energy management systems. strategies for hydrogen-based building heating systems,” Appl.
Another important development is the emergence of Energy, vol. 379, p. 124994, 2025, doi:
community-driven energy models. Decentralized smart grids https://doi.org/10.1016/j.apenergy.2024.124994.
and blockchain-based energy transactions are helping to [6] S. Panda et al., “Priority-based scheduling in
improve energy distribution, increase resilience, and reduce residential energy management systems integrated with
dependence on traditional power networks. These solutions are renewable sources using adaptive Salp swarm algorithm,”
supporting the transition to smarter and more sustainable energy Results in Engineering, vol. 23. 2024. doi:
systems. 10.1016/j.rineng.2024.102643.
[7] A. Ożadowicz, “Generic IoT for smart buildings and
Although valid in the analyzed context, this study's results
field-level automation—challenges, threats, approaches, and
present limitations of generalization due to specific climatic,
economic, or technical factors. For example, a solution effective solutions,” Computers, vol. 13, no. 2, p. 45, 2024.
in a cold climate may prove unsuitable in a hot zone. To [8] C. E. Tungom, B. Niu, and H. Wang, “SWAPP:
strengthen the proposals' applicability, additional research could Swarm precision policy optimization with dynamic action
explore regional adaptations or modular models adjustable to bound adjustment for energy management in smart cities,”
different environments. Appl. Energy, vol. 377, p. 124410, 2025, doi:
https://doi.org/10.1016/j.apenergy.2024.124410.
The analysis of recent studies shows a growing interest in [9] T.-C. Chen, N. K. A. Dwijendra, S. Singhal, R.
renewable energy integration and AI-based energy management Sivaraman, and A. Mamdouh, “Intelligent system application
between 2014 and 2025, reflecting the global focus on to monitor the smart city building lighting,” Computers,
sustainability and energy efficiency. Continued research and
Materials and Continua, vol. 75, no. 2. pp. 3159–3169, 2023.
innovation will be essential to address emerging challenges and
doi: 10.32604/cmc.2023.035418.
maximize the potential of these technologies.
[10] A. Behzadi, S. M. Alirahmi, H. Yu, and S. Sadrizadeh,
In summary, this review highlights the latest advancements, “An efficient renewable hybridization based on hydrogen
key challenges, and future opportunities in energy management storage for peak demand reduction: A rule-based energy control
for smart buildings. The findings from this study will be crucial and optimization using machine learning techniques,” Journal
for researchers, industry professionals, and policymakers in of Energy Storage, vol. 57. 2023. doi:
shaping the next generation of sustainable and intelligent energy 10.1016/j.est.2022.106168.
solutions. [11] N. S. Alsharafa, R. Suguna, R. J. Krishna, V. K.
ACKNOWLEDGMENT Sonthi, S. M. Padmaja, and P. Mariaraja, “Optimizing building
energy management with deep reinforcement learning for smart
This research was supported by the Doctoral Scholarship and sustainable infrastructure,” Journal of Machine and
"PhD-Associate Scholarship – PASS" provided by the Computing, vol. 4, no. 2. pp. 381–391, 2024. doi:
CNRST (Centre National de la Recherche Scientifique et 10.53759/7669/jmc202404036.
Technique). The author sincerely appreciates this financial
[12] A. I. Ali and S. Zorlu Partal, “Development and
support, which contributed significantly to the progress of this
work. performance analysis of a ZigBee and LoRa-based smart
building sensor network,” Frontiers in Energy Research, vol.
REFERENCES 10. 2022. doi: 10.3389/fenrg.2022.933743.
[1] C. Spagkakas, D. Stimoniaris, and D. Tsiamitros, [13] S. Zhao, S. Chen, T. R. Alsenani, B. Alotaibi, and M.
“Efficient demand side management using a novel Abuhussain, “Residential energy consumption and price
decentralized building automation algorithm,” Energies, vol. forecasting in smart homes based on the internet of energy,”
16, no. 19. 2023. doi: 10.3390/en16196852. Sustain. Energy Technol. Assess., vol. 73, p. 104081, 2025, doi:
[2] M. M. Talib and M. S. Croock, “Intelligent power https://doi.org/10.1016/j.seta.2024.104081.
management models for buildings: a comparative analysis,” [14] M. Rumbayan, I. Pundoko, S. R. Sompie, and D. G.
Journal Europeen des Systemes Automatises, vol. 57, no. 1. pp. Ruindungan, “Integration of smart water management and
95–103, 2024. doi: 10.18280/jesa.570110. photovoltaic pumping system to supply domestic water for rural
[3] M. Razghandi, H. Zhou, M. Erol-Kantarci, and D. communities,” Results Eng., vol. 25, p. 103966, 2025, doi:
Turgut, “Smart home energy management: VAE-GAN https://doi.org/10.1016/j.rineng.2025.103966.
synthetic dataset generator and q-learning,” IEEE Transactions [15] R. Peeren, D. Dabhi, and J. Dalton, “Levelling the
on Smart Grid, vol. 15, no. 2. pp. 1562–1573, 2024. doi: playing field for smart renewable energy community in the
10.1109/TSG.2023.3288824. electricity market through the high street electricity market
[4] A. Pigott, C. Crozier, K. Baker, and Z. Nagy, model,” Appl. Energy, vol. 377, p. 124660, 2025, doi:
“GridLearn: Multiagent reinforcement learning for grid-aware https://doi.org/10.1016/j.apenergy.2024.124660.
building energy management,” Electric Power Systems [16] K. M., M. D., and K. Rajagopal, “Enhancing voltage
Research, vol. 213. 2022. doi: 10.1016/j.epsr.2022.108521. control and regulation in smart micro-grids through deep
learning - optimized EV reactive power management,” Energy

Authorized licensed use limited to: NATIONAL INSTITUTE OF TECHNOLOGY SILCHAR. Downloaded on July 15,2025 at 15:31:46 UTC from IEEE Xplore. Restrictions apply.
 Rep., vol. 13, pp. 1095–1107, 2025, doi: [24] V. Vishnoi, I. Budhiraja, D. Garg, S. Garg, B. J. Choi,
https://doi.org/10.1016/j.egyr.2024.12.072. and M. S. Hossain, “Multiagent deep reinforcement learning
[17] M. Arun, D. Barik, S. S. R. Chandran, S. based Energy efficient resource management scheme for RIS
Praveenkumar, and K. Tudu, “Economic, policy, social, and assisted D2D users in 6G-aided smart cities environment,”
regulatory aspects of AI-driven smart buildings,” J. Build. Eng., Alex. Eng. J., vol. 116, pp. 609–620, 2025, doi:
vol. 99, p. 111666, 2025, doi: https://doi.org/10.1016/j.aej.2024.12.004.
https://doi.org/10.1016/j.jobe.2024.111666. [25] H. Liu, Z. Du, T. Xue, and T. Jiang, “Enhancing smart
[18] S. Saluja et al., “Smart multi-building energy building performance with waste heat recovery: Supply-side
monitoring system,” in 2024 IEEE international conference on management, demand reduction, and peak shaving via
electro information technology (eIT), May 2024, pp. 342–347. advanced control systems,” Energy Build., vol. 327, p. 115070,
doi: 10.1109/eIT60633.2024.10609877. 2025.
[19] R. A. Pour, M. Brenna, A. Saldarini, D. Zaninelli, M. [26] H. Wang, Z. Huang, X. Zhang, X. Huang, X. wei
Longo, and S. Miraftabzadeh, “Development of intelligent Zhang, and B. Liu, “Intelligent power grid monitoring and
models for building energy management,” in 2024 IEEE management strategy using 3D model visual computation with
International Conference on Environment and Electrical deep learning,” Energy Reports, vol. 8. pp. 3636–3648, 2022.
Engineering and 2024 IEEE Industrial and commercial power doi: 10.1016/j.egyr.2022.02.123.
systems Europe (EIC / I&CPS europe), Jun. 2024, pp. 1–6. doi: [27] P. H. Shaikh, N. B. M. Nor, P. Nallagownden, I.
10.1109/EIC/ICPSEurope61470.2024.10751373. Elamvazuthi, and T. Ibrahim, “A review on optimized control
[20] R. S. K. Naik and T. Devaraju, “Intelligent distribution systems for building energy and comfort management of smart
systems and grid integration of renewable energy resources,” in sustainable buildings,” Renew. Sustain. Energy Rev., vol. 34,
2024 second international conference on smart technologies pp. 409–429, 2014.
for power and renewable energy (second), Apr. 2024, pp. 1–5. [28] H. Xie et al., “Eco-Energetical analysis of the circular
doi: 10.1109/SPECon61254.2024.10537538. economy and community-based virtual power plants (CE-
[21] N. Jahan and Md. A. B. S. Abir, “SDN-enabled smart cVPP): A systems engineering-engaged life cycle assessment
building energy management for a hybrid solar-geothermal (SE-LCA) method for sustainable renewable energy
integrated system,” in 2024 IEEE International Conference on development,” Appl. Energy, vol. 365, p. 123191, Jul. 2024,
Interdisciplinary approaches in Technology and Management doi: 10.1016/j.apenergy.2024.123191.
for Social Innovation (IATMSI), Mar. 2024, pp. 1–6. doi: [29] H. Xie et al., “Eco-Energetical analysis of the circular
10.1109/IATMSI60426.2024.10502637. economy and community-based virtual power plants (CE-
[22] M. Ibrar et al., “Survey on demand response in the cVPP): A systems engineering-engaged life cycle assessment
landscape of adaptive and intelligent building energy (SE-LCA) method for sustainable renewable energy
management systems,” in 2024 International Wireless development,” Appl. Energy, vol. 365, p. 123191, Jul. 2024,
Communications and Mobile Computing (IWCMC), May 2024, doi: 10.1016/j.apenergy.2024.123191.
pp. 1203–1209. doi: 10.1109/IWCMC61514.2024.10592593. [30] I. Ahmed, M. Asif, H. H. Alhelou, M. Khalid, and
[23] D. Du, “Intelligent building system based on multi- others, “A review on enhancing energy efficiency and
modal optimization of the energy data center,” in 2024 3rd adaptability through system integration for smart buildings,” J.
International Conference on sentiment analysis and deep Build. Eng., p. 109354, 2024.
learning (ICSADL), Mar. 2024, pp. 406–412. doi:
10.1109/ICSADL61749.2024.00072.

Authorized licensed use limited to: NATIONAL INSTITUTE OF TECHNOLOGY SILCHAR. Downloaded on July 15,2025 at 15:31:46 UTC from IEEE Xplore. Restrictions apply.