In this research, a new hybrid artificial intelligence (AI)-meshless approach was presented for modeling contaminant transport in porous media. The key innovation of the proposed hybrid model is that both black box and physical-based models were used for simulating contaminant transport in porous media. An experimental model was also used to test the effectiveness of the proposed approach. In this method, for each test point (TP), artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS) models were calibrated to predict temporal contaminant concentrations (CCs). Then, considering the predicted CCs of TPs as interior conditions, the multiquadric radial basis function (MQ-RBF) as a meshless method which solves partial differential equation (PDE) of contaminant transport modeling in porous media, was used to estimate CC value at any point within the study area (in the experiment, sand tank) where there is not any TP. In this stage, optimal values of dispersion coefficient in advection-dispersion PDE and shape coefficient of MQ-RBF were determined using imperialist competitive algorithm. Optimizing these parameters could handle some uncertainties of the phenomenon. Results showed that the efficiency of ANFIS-meshless model is almost the same as ANN-meshless model due to less uncertainties involved in the obtained data under controlled experiments.

You do not currently have access to this content.