PERMESBILITY PREDICTION FROM WELL LOG DATA USING NON PARAMETRIC TECHNIQUES FOR INPROVED RESEROIR CHARACTERIZATION OF ONSHORE FIELDS

Abstract

Reservoir characterization plays an important role in modern reservoir management. It helps in maximizing multidisciplinary data and increases the chances and reliability of reservoir predictions. Permeability is one of the most important parameters affecting the productivity of hydrocarbon bearing reservoir. Thus, understanding the heterogeneity of reservoir and characterizing it with consistent input of permeability is very crucial. Formation permeability is measured directly from the core sample studies performed in laboratory. However, it is very costly and is not feasible for the whole reservoir. Also, Permeability cannot be inferred directly from any well log measurements. Earlier, methods various methods have been used for the permeability prediction using empirical relationship, statistical regression, etc., These methods are not applicable for all the reservoirs, since permeability varies largely due to different depositional environment. In recent past, a new approach has been introduced for predicting permeability by non-parametric methods, such as Artificial Neural Network (ANN) and Alternating Conditional Expectations (ACE). These methods prove to be more robust since they require no priori assumption regarding functional form. Using core log data of a given well, both ANN and ACE has been applied for Cauveiy basin. It has been found that ANN proves to be a better technique for permeability prediction in the area of study. Using ANN permeability of wells without core log has been predicted.