Artificial Neural Network Approach with Back Propagations for Modelling SSremoval in GMF

摘要

In this paper, removal efficiency of suspended solid SS from water was investigated in glasses media filter called GMF. Removal efficiency of SS is obtained by using laboratory glasses media filter "GMF" where this efficiency is used as target function in Artificial neural networks .The remain characteristics are used as input parameters for ANNs where these parameters include raw water quality, operation conditions and glasses media characteristics. The model result showed that optimal number of neurons is nine neurons. As a final observations, the study shows that Artificial neural networks with back propagation algorithm is a good tool that can be used in Prediction Removal efficiency of GMF whereas the results was indicated that the BP model has good convergence performance during training, and the predictions of outflow suspended solid removal efficiency coincided well with the measured values.