Prediction and Simulation the Breakthrough of Residual Chlorine Removal by Granular Activated Carbon Adsorbent Using Artificial Neural Networks

Rusul Naseer ； Ala'a Abdulrazaq Jassim ； Lü Xi-Wu ； Saad Abualhail

《Research Journal of Applied Sciences, Engineering and Technology》 4卷13期 (2012/07) Pp. 1883-1887

並列摘要

This study has included two parts. The first part has dealt with carbon production whereas the date Palm was used to produce Granular Activated Carbon (GAC) with specific physical characteristics. The new produced of GAC is used to adsorbate the Residual chlorine from water by deep bed filter column. In the second part, the experimental results of the breakthrough of residual chlorine curves is predicted and simulated using artificial neural network with back propagation algorithm whereas the optimum number of neuron was investigated based on RMSE. The removal of residual chlorine has been used as target function in ANN while the other properties of adsorption process such as operation conditions, chlorine concentration in raw water and GAC characteristics has been used as input parameters. The results showed that ANN with back propagation algorithm is a good tool that can be used to predict the best operating parameter for designing GAC layer in multimedia filter whereas 35 neuron gave the best fitting with experimental data. In addition to that, the simulation result was showed that the predictions of breakthrough curve model has been coincided well with the measured values which explained that the depth 25cm with grain size 1.5mm of GAC filter bed will be give the optimum removal of residual chlorine from chlorinated water.

並列關鍵字

Activated carbon ； Artificial Neural Network (ANN) ； breakthrough curve ； chlorine adsorption