Abstract:The 'one-size-fits-all' setting of heating, ventilation, and air conditioning (HVAC) systems in university classrooms not only compromises occupant comfort but also leads to potential energy wastage. This study proposes a user-centered approach to optimize the thermal environment in university classrooms by integrating computer vision and Q-XGBoost learning. Computer vision facilitates non-intrusive and real-time sensing of thermal comfort in dynamic university classroom environments. Simultaneously, the Q-XGBoost model identifies optimal operations for HVAC systems by learning from the interactions between occupants and their environment, ensuring a balance between comfort and energy efficiency. These innovative methodologies are harmoniously integrated into a smart air conditioning control box (SACC-Box), achieving intelligent operation of HVAC systems in the real world. Subsequently, a controlled study was conducted to test the user-centered approach's performance in improving thermal comfort and energy efficiency. The results reveal that classrooms equipped with the SACC-Box significantly improve occupant satisfaction with thermal comfort by approximately 13.5 % while concurrently achieving up to a 5 % reduction in energy consumption compared to classrooms operating with traditional built-in smart control systems. This investigation offers a viable solution to balance energy efficiency and user comfort in university classrooms, underlining the transformative potential of integrating intelligent technologies that inspire a greener, more adaptable, and user-centric futures in higher education establishments.

A multi-occupants’ comfort-driven and energy-efficient control strategy of VAV system based on learned thermal comfort profiles

A Multiple Model Approach for Predictive Control of Indoor Thermal Environment with High Resolution

Energy-efficient control of thermal comfort in multi-zone residential HVAC via reinforcement learning

Automatic Control System for Thermal Comfort Based on Predicted Mean Vote and Energy Saving

User-centric Approach to Optimizing Thermal Comfort in University Classrooms: Utilizing Computer Vision and Q-XGBoost Reinforcement Learning

Experimental Assessment of a Satisfaction Based Thermal Comfort Control for a Group of Occupants

Group Comfort Models: Predicting Indoor Group Thermal Comfort by Learning Preferences of Multiple Occupants

DeepComfort: Energy-Efficient Thermal Comfort Control in Buildings Via Reinforcement Learning

Application of energy combined thermal comfort in intelligent building management in complex environments

A Unified Control Framework of HVAC System for Thermal and Acoustic Comforts in Office Building.

Heating Control Strategy Based on Dynamic Programming for Building Energy Saving and Emission Reduction

Computer vision-based smart HVAC control system for university classroom in a subtropical climate

Investigating Occupancy-Driven Air-Conditioning Control Based on Thermal Comfort Level

Double Deep Q-learning Based on Personalized Thermal Comfort Model for HVAC Optimization

A Self-learning intelligent passenger vehicle comfort cooling system control strategy

Research on intelligent regulation of air conditioning energy saving based on human thermal comfort

Enhancing personalised thermal comfort models with Active Learning for improved HVAC controls

Comfort-Aware Building Climate Control Using Distributed-Parameter Models

Occupant-centric HVAC and window control: A reinforcement learning model for enhancing indoor thermal comfort and energy efficiency

Reinforcement learning-based thermal comfort control for vehicle cabins

A Model-Driven Learning Approach for Predicting the Personalized Dynamic Thermal Comfort in Ordinary Office Environment