Artificial Neural Network Modeling for the Management of Oil Slick Transport in the Marine Environments

Due to an increase in demand of petroleum products which are transported by vessels or exported by pipelines, oil spill management becomes a controversial issue in coastal environment safety as well as making serious financial problems. After spilling oil in the water body, oil spreads as a thin layer on the water surface. Currents, waves and wind are the main causes of oil slick transport. These phenomena depend on the overall interaction among gravity, viscosity, surface tension and interfacial tension of oil in water bodies. In the current study, Artificial Neural Network (ANN) models have been designed and trained for the prediction of oil spreading and advection under different hydrodynamic conditions. In this regard, results obtained from a multiphase Lagrangian numerical model are deployed to train ANN model. The mentioned numerical model which is based on the moving particle semi-implicit (MPS) method is developed in the earlier stage of the study. In this research study, the MPS numerical model is first validated and verified against the analytical formulas which are based on experimental data cited in the literature. Then, various hydrodynamic conditions and oil spill scenarios were chosen to obtain different numerical model results. Finally, numerical model results are then deployed for training ANN model to provide a useful tool for urgent prediction of oil slick trajectory in order to manage the oil slick transport in the coastal environments.