Reinforcement Learning: A Smarter Control Approach for Balancing Occupant Welfare and Energy Efficiency in Commercial HVACs

This research explores how artificial intelligence can make commercial heating and cooling systems smarter and more energy efficient. Traditional building control systems rely on fixed rules and often waste energy or reduce indoor comfort. In this study, a learning-based control approach was developed using real building data from Monash University. A virtual environment was created to safely train the intelligent system before testing it. The results show that the proposed approach can significantly reduce energy consumption while improving indoor comfort and air quality. This work supports the development of more sustainable and intelligent commercial buildings.