Porosity and Permeability Prediction Using Artificial Neural Network based on Flow Zone Index in an Iranian Heterogeneous Carbonate Reservoir

In this study , an artificial neural network ANN model for porosity and permeability prediction is presented . This is an improvement for ANN to use data of well logs in order to predict targets in un-cored wells/ intervals. The well logs and ogher data are gathered form and iranian heterogeneous carbonate reservoir. core porodity is then predicted using ANN approach in which log-derived porosity NPHI, RHOB,DT and PEF log values were selected as input values in to ANN modeling procedure. having done so, core permeability is predicted indirectly using flow zone indicator FZI as target and ANN approach in which predicted -porosity from pervious step RHOB log , CGR log, DT log,PEFlog and depth point values were selected as inputs with FZI as output of the ANN procedure. having predicted FZI , the predicted -porosity values were used to calculate permeability.

کلیدواژه ها:

artificial neural network (ANN) ، neutron log ، density log ، sonic log ، photo electric factor log ، corrected gamma ray log ، flow zone index ، hydraulic flow unit.