Modeling the removal of cefixime by the Fenton method with an artificial neural network.

Document Type : Research Article


1 Department of Chemical Engineering, Eyvan-e-Gharb Branch, Islamic Azad University, Eyvan-e-Gharb, Iran

2 Department of chemical engineering , Ilam branch,, Islamic Azad University, Ilam, Iran.


The purpose of this investigation, modeling is the efficiency of the removal of cefixime by the Fenton method using a neural network. Artificial Neural Network was used in prediction of results of experimental. In this modeling, the amount of hydrogen peroxide, iron catalyst, the removal time of cefixime, the concentration of cefixime and the input pH and the elimination rate are output. It includes the sum of error squares (SSE), the mean square of error squares (RMSE), the adjusted coefficient of determination (R_adj^2), and the coefficient of determination of R^2 in determining the number of optimal middle neurons. According to the results, the neural network model was able to predict the absorption efficiency with the tansigmoid transfer function in the hidden layer and the purelin stimulus transfer function in the output layer. Also, the results of neural network modeling with Levenberg – Marquardt showed that a network with a layout of 1-13-5 (5 neurons in the input layer, 13 neurons in the hidden layer and 1 neuron in the output layer) has the best result in predicting the output. The correlation coefficient was obtained for values of training, validation and test are 0.99436 and 0.99993 and 0.96901, respectively. To predict the trend of changes, neural network tools have been used in MATLAB software.


Main Subjects