Modeling of Shrimp Chitosan Polymer Adsorption Using Artificial Neural Network

One phenomenon that can occur when a polymer solution is injected into an oil reservoir is adsorption. Adsorption occurs due to interactions between polymer molecules and the reservoir pore surface. Adsorption causes some polymer molecules to be removed from solution. So, this process results in a reduction in the polymer concentration in the solution. In this study, an artificial neural network (ANN) model is used to estimate the adsorption of shrimp chitosan polymer on the surface of 40 mesh and 60 mesh sand grains. The ANN model can estimate adsorption more accurately than previous models. This is because previous models only predicted certain adsorption patterns, while the ANN model is able to predict adsorption with complex relationships. The comparison of the mean absolute relative errors (MAREs) of the ANN, Langmuir, Freundlich, Henry, and Harkins-Jura models is 5.7%, 15.9%, 14.6%, 15.2%, and 14.5%, respectively.