Modeling of HVAC system for balancing indoor thermal comfort level and energy efficiency

Abstract

Modeling the heating, ventilation and air-conditioning (HVAC) system plays an important role in modern society as it provides an effective solution for the controlled environment. However, HVAC can consume a large amount of electricity, therefore, there is a need to make HVAC systems more energy efficient. Reducing the HVAC energy consumption can be rather challenging due to several constraints. First, the energy consumption depends on a number of factors and components which are not conveniently modeled. The thermal comfort is another constraint whereby reducing HVAC energy consumption may compromise the cooling performance and affect the occupants’ comfortability. This work therefore proposes a machine learning approach to simulate a specific HVAC system based on the experimental data of HVAC system in Nanyang Technology University and then integrated with three different population-based meta-heuristic optimization algorithms: Seagull optimization algorithm, Whale optimization algorithm and Butterfly optimization algorithm, to optimize the control strategy of the system. The optimization results of three algorithms are compared and the Whale optimization algorithm is chosen as the final algorithm for HVAC system control strategy optimization. The model not only has a good optimization performance but also a low computational complexity. The system is then simulated in various conditions under different occupants’ preferences and the feasibility of the system is analyzed and shows that this system is suitable for tuning the thermal comfort level of occupant in high temperature conditions. For low temperature, the system can hardly change the thermal sensation of the occupants even if the occupants have higher preference of keeping a neutral thermal comfort level than saving energy.