Design of neural network for manipulating gas refinery sweetening regenerator column outputs

In this study, a new approach for the prediction collection outputs of regenerator column in gas sweetening plant is suggested. The experimental input data, including inlet temperatures of reflux, difference between inlet and outlet condenser temperatures, amount of H2O and inlet amine temperatures and outlet down temperature of tower and amount of reflux as outputs have been used to create artificial neural network (ANN) models. The testing results from the model are in good agreement with the experimental data. The new proposed method was evaluated by a case study in HASHEMI NEJAD gas refinery in KHORASAN of Iran. Design of topology and parameters of the neural networks as decision variables was done by trial and error, high performance efficiency networks was obtained to predict the output parameters of regenerator column.

Graphical abstractFigure optionsDownload full-size imageDownload as PowerPoint slideHighlights► A new ANN models for the prediction of outputs of regenerator column in gas plant is suggested. ► The calculation process that takes place in regenerator column is simplified by ANN. ► Using ANN leads to the acceptable prediction of the output of regenerator column.