Optimization of automotive HVAC performance and passenger comfort through intelligent algorithms

Abstract

Optimizing passenger thermal comfort while maintaining energy efficiency remains a critical challenge in automotive climate control. This study investigates the complex thermal interaction between a passenger and a car seat, introducing a novel technique for direct thermal management through the vehicle's heating, ventilation, and air-conditioning system. An experimental setup was developed using evaporator coils integrated beneath seat surfaces, addressing a significant gap in the literature regarding two-way (heating and cooling) seat thermal management. Artificial Neural Networks were employed to model the ambiguous system parameters and to predict thermal performance. The results show an excellent predictive accuracy with Mean Squared Error of 1.29E-02 and correlation coefficient of 0.99659. The average deviations were under -0.21% for cooling capacity, -0.02% for heating capacity and 0.38% for coefficient of performance. These results show that the integration of intelligent algorithms into modified heating, ventilation, and air conditioning architectures significantly improves occupant comfort and system efficiency, thereby providing a powerful data-driven framework for next-generation automotive climate control solutions.