An ANFIS-based Model to Predict the Oil Spill Consequences on the Ground

Abstract

Considering the randomness and complexity of oil spill accidents on the ground, the oil spill volume, the spilled oil density, the spreading coefficient of oil product on the surface layer and ground thickness, were taken as the initial influencing attributes for the prediction of oil contamination into the ground. Based on the study of the Adaptive Neural Fuzzy Inference System (ANFIS), a nonlinear fuzzy model to evaluate oil spill damage to the ground was established. Combined with the oil spill on the ground data obtained from the linear oil spill model and opinions of experts, the ANFIS-based prediction model for oil spill contamination to the ground has been proposed in this paper. Study results show that the proposed model is able to predict the oil spill contamination into the ground with reasonable accuracy. Its performance was assessed through the correlation coefficient (R), the coefficient of determination (R2) and the root-mean-square error (RMSE).