Journal of the Korean Society of Mineral and Energy Resources Engineers ISSN:2288-0291(Print) 2288-2790(Online) 한국자원공학회지

Reservoir permeability is one of the most important parameters of oil and gas fields for reservoir characterization. Although formation permeability is often measured in a laboratory from cores or evaluated from well test data, core analysis and well test data are only available from a few wells in a field, while most wells are logged. However, permeability determination from geophysical well logs in heterogeneous formation has a difficult and complex problem. In this study, an artificial neural network(ANN) model has been developed using core analysis data and well logs to determine reservoir permeability. The ANN is trained to predict formation permeability with well log data as input. The technique is demonstrated with an application to the well data in offshore Korea. The results of this study show that permeability distribution derived from the ANN model is consistent with core analysis results. The ANN is capable of providing more accurate and reliable results compared with conventional methods and statistical regression. The ANN is a practical and powerful tool for predicting reservoir permeability of a heterogeneous formation utilizing well logs.