Prediction of absorption and stripping factors in natural gas processing industries using feedforward artificial neural network

In dynamic simulators, mathematical models are applied in order to study the time-dependent behavior of a system, meaning the system process units and the corresponding control units. Absorption and stripping are the unit operations that are widely used in the natural gas processing industries. Many attempts have been made to define an average absorption factor method to short-cut the time consuming rigorous calculation procedures. One of the options for this complex engineering modeling problem is artificial intelligence approach. Artificial neural networks have been shown to be able to approximate any continuous nonlinear functions and have been used to build data base empirical models for nonlinear processes. In this study, feedforward neural networks (FANN) models were used to model the absorption efficiency. The mean square error (MSE), residue analysis and coefficient of determination based on the observed and prediction output is chosen as the performance criteria of model. It was found that the developed FANN models provided satisfactory model with the MSE and coefficient of determination of 0.0003 and 0.9998 for new unseen data from literature respectively.