MPC-Based HVAC Control System for Energy Efficiency and User Comfort

Rule-based control methods in HVAC (Heating, Ventilation, and Air Conditioning) systems fail to adapt to unexpected situations and environmental changes, resulting in energy inefficiency. This paper proposes an energy-efficient control method for HVAC systems by leveraging Model Predictive Control (MPC). To design the MPC, we develop AI-based prediction models and introduce a cost function that optimizes both energy consumption and user comfort simultaneously. Performance evaluation of the prediction model shows that the stacking ensemble model significantly improves prediction accuracy compared to individual models.