Year: 2026
Type: article
Abstract: An Artificial Neural Network (ANN) model is developed for simultaneous prediction of six performance parameters of 42CT and 42DT fan coil units. Twelve geometric, aerodynamic, and hydraulic inputs are used to predict airflow rate, cooling capacity, heating capacity, fan power, sound power level, and unit price. A three-hidden-layer network with [36, 18, 9] neurons trained via Bayesian Regularization on 282 experimental samples achieves R² = 0.9997, 0.9999, and 0.9992 for airflow, cooling, and heating capacity. The universal model generalizes across both product series without model-specific calibration
Source: Scientific works/Elmi eserler
Authors: Buğra Avcı, Ahmet Selim Dalkılıç, Burak Uğuz
Institution: Yıldız Technical University
Language: English
Subfield: Building and Construction
Field: Engineering
Domain: Physical Sciences
Open Access status: bronze
OpenAlex ID: W7169834524
