| dc.description.abstract | This study discusses the prediction of petrophysical rock typing in reservoir in a field using core and log data with an Artificial Neural Network (ANN) model. The objective of this study is to determine the effect of core-log data in rock type prediction; create test models, validation, and prediction with ANN; and apply the ANN model to wells with limited input data. The method used in this study is to build the ANN model architecture which is divided into three stages: training-test, validation, and prediction. The result of the study is that the neural network model of core data (porosity and permeability) and log data (gamma-ray, neutron, density, photoelectric, and sonic) affected forming rock type predictions. The result of the test and validation of the ANN model obtained an accuracy of 84.19% with core-log input data. Based on the analysis, the ANN model can be used for limited data input (log data) with the result that the more wells used, the better the accuracy value. | en_US |
