--> Artificial Neural Network (ANN) Prediction of porosity and Water Saturation of Shaly Sandstone Reservoirs
[First Hit]

AAPG Asia Pacific Region, The 4th AAPG/EAGE/MGS Myanmar Oil and Gas Conference:
Myanmar: A Global Oil and Gas Hotspot: Unleashing the Petroleum Systems Potential

Datapages, Inc.Print this page

Artificial Neural Network (ANN) Prediction of Previous HitporosityNext Hit and Water Saturation of Shaly Previous HitSandstoneNext Hit Reservoirs

Abstract

This paper presents a successful application of neural networks (ANN) in Previous HitpredictingNext Hit Previous HitporosityNext Hit, water saturation and identifying lithofacies of shalysand Previous HitreservoirNext Hit using well logging data. ANN technique utilizes the prevailing unknown nonlinear relationship in data between well logs and the Previous HitreservoirNext Hit rock petrophysical properties. In heterogeneous reservoirs classical methods face problems in determining accurately the relevant petrophysical parameters due to assumptions and uncertainties of input parameters. Applications of artificial intelligence have recently made this challenge a possible practice and in this study neural network has been proposed to supplement or replace the existing conventional techniques to determine water saturation using shaly water saturation models (total shale, Simandoux and medium effective) and effective Previous HitporosityNext Hit in shalysand reservoirs. Two neural networks were presented to determine Previous HitporosityNext Hit and water saturation using GR, resistivity and density logging data and the cut off values for Previous HitporosityNext Hit and water saturation. Water saturation and Previous HitporosityNext Hit have been determined using conventional techniques and neural network approach for two wells drilled in shalysand Previous HitreservoirNext Hit. ANN outputs have shown good matching with core data and the reference calculated petrophysical parameters; Previous HitporosityNext Hit, water saturation and defined pay zones in a new well that projects its application for new wells. Neural network approached have trained for Previous HitporosityNext Hit and water saturation using the available well logging data. The predicted Previous HitporosityNext Hit and water saturation values have shown excellent matching with core data in the two wells comparing to the Previous HitporosityNext Hit and water saturation of the conventional techniques. Consequently, the developed network (ANN) can successfully deduce Previous HitporosityTop, water saturation and defined pay zones of for new wells in shalysand.