Abstract:
ABSTRACT
Advancements in machine learning have revolutionized various industries, including building
energy management and thermal comfort optimization. The integration of these technologies
offers transformative potential for developing intelligent, adaptive systems in the built
environment. This paper provides a comprehensive review of machine learning-based
approaches in dynamic thermal comfort control systems, focusing on their potential for
energy optimization in various building typologies. As HVAC systems evolve to balance
thermal comfort with energy efficiency, machine learning algorithms such as artificial neural
networks, fuzzy logic, and reinforcement learning are increasingly being applied to predict
and adjust environmental settings dynamically. By analyzing key studies in the field, this
review identifies the advantages and limitations of different machine learning models in
terms of energy savings and occupant comfort. The paper also highlights the gaps in current
research, particularly the need for more real-time, adaptive models that can integrate both
occupant behavior and external environmental factors. The findings suggest that machine
learning offers significant potential for reducing energy consumption in buildings while
maintaining or improving thermal comfort, but further development is necessary to refine
these systems for broader and more reliable applications. Ultimately, this review aims to
serve as a foundation for future research, fostering advancements in smart building
technologies that prioritize both sustainability and human well-being.
Keywords: Machine Learning, Thermal Comfort, Energy Optimization, Smart Buildings
