Reverse Osmosis Desalination Performance Using Artificial Neural Network Approach with Optimization

Abstract

Reverse osmosis (RO) has found extensive usage in the fields of desalination and pollution control. The intention of the proposed work is to predict the thermal efficiency and average flux in RO process using Artificial Neural Network (ANN) with optimization process. These prediction processes initially optimize the network structure hidden layer and hidden neuron using different training algorithms and get the better network structure. For improving the prediction accuracy of RO in ANN process different optimization techniques are used. The optimal hidden layer and neuron attained in hybridization of GA and PSO technique based predict the parameters. From the results the ANN training algorithm predicts the error accuracy in LM and also in HA technique 75.2% and 89.25% respectively in this work compared to the GA and PSO techniques.