Prediction of H2S Concentration in Gas Sweetening Processing Units by Using Artificial Neural Networks

Gas sweetening is a fundamental step in gas treatment processes. In gas sweetening units, acid gases (H2 S and CO2 ) are chemically absorbed from a gas using aqueous alkanolamine solutions, to product a sweet gas. The solvent is regenerated in a desorption column and the purified solvent is recycled to the absorption column. Gas sweetening units can be controlled if all of operation data for example sweet gas, lean amine and rich amine flow rates, concentration and temperatures existed. In this paper a new method based an Artificial Neural Network (ANN) for prediction of H S concentration in lean amine stream. Sour gas and sweet gas have been input variables of the network and have been set as network output. These predictions can be prevented operation problems. The results according to R value and mean squared error shows good accuracy of this type of modeling.

Keywords:

Lean Amine , H2 S Concentration , Artificial Neural Network , Prediction , Gas sweetening