Occupant-Feedback-Integrated Model Predictive Control for Air Conditioning: An Experimental Study in a Commercial Building

This paper presents the implementation and analysis of a ML-based MPC system integrated with a real-time occupant feedback mechanism through field tests in a commercial building.We demonstrate the efficacy of the occupant-centric control strategy that dynamically adjusts thermal comfort setpoints based on direct user input.Furthermore, we propose a comprehensive framework for analyzing the resulting data, offering insights into system performance, energy consumption, and occupant satisfaction.The findings illustrate that incorporating a human-in-the-loop approach can enhance building energy efficiency without compromising occupant comfort.