شماره ركورد كنفرانس :
5041
عنوان مقاله :
Prediction of H2S adsorption modeling on Fe2O3 by artificial neural network technique and Thomas kinetic model
Author/Authors :
M. Ghajari Department of Chemical Engineering - Abadan Faculty of Petroleum Engineering - Petroleum University of Technology, Abadan, Iran , M. R. Shishesaz Department of Inspection and Safety Engineering - Abadan Faculty of Petroleum Engineering - Petroleum University of Technology, Abadan, Iran , M. Fattahi Department of Chemical Engineering - Abadan Faculty of Petroleum Engineering - Petroleum University of Technology, Abadan, Iran
كليدواژه :
Artificial Neural Network , H2S removal , Kinetic model , Adsorption
عنوان كنفرانس :
The 10th International Chemical Engineering Congress & Exhibition (IChEC 2018)
چكيده فارسي :
فاقد چكيده فارسي
چكيده لاتين :
In this investigation, Artificial Neural Network (ANN) and Thomas Kinetic model for H2S removal modeling on Fe2O3 was studied. ANN simulation and modeling of H2S adsorption on Fe2O3 was performed through the data are provided in the open literature. ANN is a way that can solve the problem with any complexity in which their results compared with kinetic modeling. Modeling results predictions through Thomas model are lower in compare to ANN for H2S removal. Various ANN learning rule and properties of H2S adsorption were investigated through this research. Results showed that ANN has the best accuracy and more reliability. Thomas model is also good for prediction behavior of adsorption. By utilizing the Traincgf algorithm the lowest RMSEslope equal to 0.0763 obtained as well as Trainlm has lowest RMSEintersect, RMSEData and number of epochs that are equal to 0.0013, 0.0017 and 4.
