شماره ركورد كنفرانس :
5041
عنوان مقاله :
Comparison of artificial neural network and response surface methodology for kinetics modeling of propylene production via dehydrogenation process
Author/Authors :
H. Khorrami Chemical Engineering Group - Faculty of Engineering - University of Mohaghegh Ardabili, Ardabil, Iran , A. Heydari Chemical Engineering Group - Faculty of Engineering - University of Mohaghegh Ardabili, Ardabil, Iran , M. Fattahi Chemical Engineering Department - Abadan Faculty of Petroleum Engineering - Petroleum University of Technology, Abadan, Ira
كليدواژه :
Propane dehydrogenation , Artificial neural network , Response surface methodology , Propylene
عنوان كنفرانس :
The 10th International Chemical Engineering Congress & Exhibition (IChEC 2018)
چكيده فارسي :
فاقد چكيده فارسي
چكيده لاتين :
The effects of operational conditions on the dehydrogenation of propane in a fixed-bed reactor was studied through artificial neural network (ANN) and response surface methodology (RSM). RSM with three independent variables and ANN-MLP network (LM training algorithm) with four neurons in hidden layer used to develop model for propane conversion to propylene by dehydrogenation process. The ANN and RSM were compared with experimental data as well as the obtained data from thermodynamic model. Temperature, water amount and time were selected as the input data and output data was propane conversion. Results revealed that the correlation coefficient for the ANN, thermodynamic model and RSM were 0.9985, 0.9921 and 0.9678, respectively.
