Energy and Comfort Efficient Adaptive Learning Model for Service Areas in an Outdoor Sports Facility

Purpose: This research investigates the development and performance of advanced supply air control strategies aimed at simultaneously enhancing energy efficiency and sustaining acceptable indoor thermal comfort within the service areas of a sports facility. The primary objective is to derive control methodologies that strategically balance these two critical factors while promoting the maximum utilization of the spaces. Method: The research employed computational simulations to evaluate two novel control strategies, such as an adaptive model and a learning model by an artificial neural network algorithm, against a conventional thermostat over a week-long period. Results: The simulation results demonstrated significant energy savings: the adaptive model achieved an approximate 12% decrease in average energy consumption, and the learning model, which utilizes an artificial neural network algorithm, delivered a more substantial 22% reduction, respectively. These savings stem from the models’ capability to minimize system overshooting and mitigate signal fluctuations during operation. Furthermore, both proposed models showed a very high level of effectiveness in maintaining thermal comfort homeostasis compared to the thermostat. The network-based learning model proved particularly advantageous, confirming its improved ability to reduce unnecessary energy consumption via adaptive control and produce robust, effective regression results under varying environmental and operational conditions.