Prediction of Flow Units and Permeability Using Artificial Neural Networks

Abstract

Abstract An accurate description of the reservoir is necessary for predicting the reservoir performance. The flow unit model is most practical approach for reservoir simulation purposes. The flow units are defined according to geological and petrophysical properties that influence the flow of fluids in the reservoir. Identification of flow units requires permeability data. Predicting permeability distribution is a difficult problem in heterogeneous reservoirs due to insufficient permeability measurements. The geophysical well logs are the most abundant source of data in a reservoir. Therefore, a well log-based methodology for predicting permeability can enhance prediction of the flow unit distribution. In this study, in a complex reservoir was successfully characterized using flow unit modeling. Statistical techniques were employed to identify flow units based on limited data obtained from core analysis supplemented by mini-permeameter measurements and geological interpretations. Permeability was then predicted from geophysical well log and preliminary flow units using artificial neural network (ANN). An innovative approach for training and testing of the ANN was developed which provided consistent and reliable predictions. The neural network predictions were then verified for the wells with core data. Well log data, available on substantial number of wells in the reservoir, were then utilized to predict the distribution of flow units, permeability, and porosity in the reservoir. This approach led to development of a reliable reservoir model. The accuracy of model was verified by successful simulation of the production performance. The methodology presented in this paper can serve as a new guideline for the characterization of reservoirs with limited core data.