Authors : Newsha Valadbeygi

Volume/Issue : Volume 8 - 2023, Issue 9 - September

Google Scholar : https://tinyurl.com/nt7yk4v4

Scribd : https://tinyurl.com/mujvhpt3

DOI : https://doi.org/10.5281/zenodo.8420643

Abstract : Calculating and predicting the performance of cooling towers has posed a significant challenge for researchers in this field. Over time, various methods, including the utilization of artificial intelligence and algorithms, have been proposed to address this issue. In this study, experimental data pertaining to cooling tower performance has been employed to develop a novel model based on neural networks. The objective is to predict the performance of cooling towers and analyze performance trends in this particular type of structure. To achieve this, a multi-layer perceptron neural network is utilized due to its high capacity, with real data serving as input. Subsequently, the efficiency of the neural network model is assessed by comparing the results with real-world samples. The validation process involves predicting cooling tower performance, examining performance trends, and analyzing tower behavior under windy conditions.

Keywords : Cooling Tower Performance, Functional Prediction, Multilayer Perceptron Neural Network.