Experimental Investigation on Improvement of Wet Cooling Tower Efficiency with Diverse Packing Compaction Using ANN-PSO Algorithm

In this study, a numerical and empirical scheme for increasing cooling tower performance is developed by combining the particle swarm optimization (PSO) algorithm with a neural network and considering the packing’s compaction as an effective factor for higher accuracies. An experimental setup is used to analyze the effects of packing compaction on the performance. The neural network is optimized by the PSO algorithm in order to predict the precise temperature difference, efficiency, and outlet temperature, which are functions of air flow rate, water flow rate, inlet water temperature, inlet air temperature, inlet air relative humidity, and packing compaction. The effects of water flow rate, air flow rate, inlet water temperature, and packing compaction on the performance are examined. A new empirical model for the cooling tower performance and efficiency is also developed. Finally, the optimized performance conditions of the cooling tower are obtained by the presented correlations. The results reveal that cooling tower efficiency is increased by increasing the air flow rate, water flow rate, and packing compaction.