Development of an AI-Based Adaptive Control System for Real-Time HVAC Performance Enhancement

Abstract

This paper presents a Model Predictive Control (MPC)-based HVAC system for a small vehicle equipped with a 2.2 kW compressor, designed to regulate cabin temperature efficiently. The proposed system dynamically adjusts the compressor's operation to maintain the desired temperature while optimizing energy consumption. The MPC controller minimizes the temperature error and smoothes control actions by optimizing a cost function that considers future temperature predictions and control signal variations. The system operates through a closed-loop process involving four main components: the compressor, condenser, expansion valve, and evaporator. The compressor increases the refrigerant’s pressure and temperature, which is then cooled in the condenser by releasing heat to the environment. The refrigerant is expanded through the expansion valve to lower its temperature and pressure before entering the evaporator, where it absorbs heat from the cabin, thus cooling the vehicle’s interior. Mathematical models for each component are integrated into the MPC framework to improve accuracy and system response. This advanced control mechanism provides superior temperature regulation compared to conventional methods by anticipating future disturbances and optimizing real-time control actions. The proposed system enhances passenger comfort, reduces energy consumption, and improves the overall performance of the vehicle’s HVAC system.