The Prediction of Permeability in a Carbonate Reservoir Using Adaptive Neurofuzzy Inference System

In this paper a new approach for the prediction of permeability from well logs is presented. It is based on adaptive neurofuzzy inference system (ANFIS) which is fuzzy inference system implemented in the framework of adaptive networks. A combination of back propagation and least-squares method referred to as hybrid learning method is applied to estimate membership function parameters of fuzzy inference system and learning purpose of ANFIS. The constructed model is optimized for the number of epochs to avoid overfitting and to provide maximum generalization by considering the error index of validation sets during training. To verify the effectiveness of the methodology, a case study in one of the carbonate reservoirs of Iran is carried out. Core and well-log data from two wells that are located in the center of the field provide the data for the learning task. Model validation of proposed ANFIS model is implemented by using core permeability and well-log data from a third well that is located on the corner of the field. Numerical simulation results show that the adaptive neurofuzzy inference system is capable for the prediction of permeability.