Prediction of improved cyclone system efficiency: Multi objective optimization by hybrid approach based on the genetic algorithm and artificial neural network

An integrated process which is proposed for improving dust removal efficiency of cyclones in another paper is considered here for simulation by Artifitial Neural Networks (ANNs) and hybrid ANN and Genetic Algorithm (GA). The process incorporates of two cyclones coupled with a specially designed cylindrical chamber, which includes a rotating tube inside it with air-impinging nozzles drilled on the peripheral surface of the tube. The chamber includes a tube with nozzles on its peripheral surface from which jet-impingement flow throws the particles nearer to wall of the chamber. Efficiency of the jet-impingement chamber, as a function of the feed flow rate, recycle flow rate, jet-impingement flow rate as well as the jet-impingement tube rotational speed has been tested on a pilot scale apparatus of the process for fitting and simulating by ANN and hybrid ANN and GA. ANN and hybrid ANN and GA were able to accurately capture the non-linear characteristics of the chamber even for a new condition that has not been used in the training process (tested data).