فهرست مطالب mahnaz yasemi

The aim of this study was to model the removal efficiency of cefixime by the Fenton method using a neural network. This model predicts experimental results well. In this model, the amount of hydrogen peroxide, iron catalyst, cefixime removal time, initial concentration of cefixime, and pH are the input parameters. The output variable is the removal percentage of cefixime. Total error squares (SSE), mean the square root of error (RMSE), adjusted coefficient of determination (), and coefficient of determination   in determining the number of optimal neurons in the middle of the performance index. According to the obtained results, the neural network model was able to predict the absorption efficiency with the sigmoid tangent transfer function in the hidden layer and the linear stimulus transfer function in the output layer. Also, the results of modeling the neural network with org-art showed that the grid with a 1-13-5 arrangement (5 neurons in the input layer, 13 neurons in the hidden layer, and 1 neuron in the output layer) had the best result in predicting the output. The correlation coefficients of all the levels of training, validation, and test 0.3 were 0.99436, 0.9993, and 0.96901, respectively. To predict the trend of changes, neural network tools have been used in MATLAB software.

In this work, the artificial neural networks (ANN) technology was applied to the simulation of oleuropein extraction process. For this technology, a 3-layer network structure is applied, and the operation factors such as  amount  of  flow  intensity  ratio,  temperature,  residence  time,  and  pH  are  used  as  input  variables  of  the network,  whereas  the  extraction  yield  is  considered  as  response  value.  Performance  indicators
RMSE, SSE, R2adj, R2

have  been  used  to  determine  the  number  of  optimal  midway neurons. Neural  network trained  with  an  error  back-propagation  algorithm,  was  used  to  evaluate  the  effects  of  parameters  on extraction yield.The obtained optimal architecture of artificial neural network model involved a feed-forward neural network with 4 input neurons, 1 hidden layer with 6 neurons and one output layer including single neuron.The trained network gave the minimum value in the RMSE of 1.6423 and the maximum value in the
=  0.9641,  which  implied  a  good  agreement  between  the  predicted  value  and  the  actual  value,  and confirmed  a  good  generalization  of  the  network.Functional  structure  of  modeling,  related  to  education,validation and test were obtained 0.99229,0.95591and 0.99224 respectively. The overall agreement between the  experimental  data  and  ANN  predictions  was  satisfactory  showing  a  determination  coefficient  of 0.9838.The  neural  network  tools  implemented  in  MATLAB  software  were  used  to  predict  the  change process.

In this paper, solubility of oxygen in organic solvents has been estimated using Artificial Neural Networks (ANN). Solubility data were studied for wide ranges of temperature (298.2-348.29 K) and pressure (0.0535 to 9.2338 MPa). Solvents are included of methanol, n-propanol, octane, toluene, dibutyl ether and 2-methyltetrahydrofuran. Network model consists of four inputs in input layer for acentric factor, molecular weight, TR and PR of the system and one neuron in output layer corresponding to solubility of oxygen. The best structure for feed-forward back propagation neural network is logarithmic sigmoid transfer function for hidden layer, 13 neurons in this layer and linear transfer function for output layer. Results show that optimum neural network architecture is able to predict the solubility of oxygen in organic solvents with an acceptable level of accuracy, R2 of 0.999997, ARD % of 0.8103 and AAD% of 0.0042. Sensitivity analysis shows that TR has the greatest effect on the solubility of oxygen.