Evaluation of the Reduction in Uncertainty Obtained by Conditioning a 3D Stochastic Channel to Multiwell Pressure Data

Feng Jun Zhang, Doctoral Student, Petroleum Engineering, Albert C. Reynolds, Dean S. Oliver

 

A stochastic channel embedded in a background facies is conditioned to data observed at wells, including well-test pressure data, well observations of the channel thickness and the depth of the top of the channel top. The background facies is a fixed rectangular box. The model parameters consist of geometric parameters that describe the shape, size and location of the channel, and permeability and porosity in the channel and nonchannel facies. The main objective of this work is to characterize the reduction in uncertainty in channel model parameters and predicted reservoir performance that can be achieved by conditioning to well-test pressure data at one or more wells. Multiple conditional realizations of the geometric parameters and rock properties are generated by using maximum randomized likelihood method. Some statistical information is computed and histograms are plotted to evaluate the uncertainty in model parameters. Generation of a realization of the model that is conditional to pressure data requires the minimization of an appropriate objective function that consists of data mismatch part and regularization term. The Levenberg-Marquardt algorithm was used as our minimization algorithm. The ensemble of predictions of reservoir performance generated from the suite of realizations provides a Monte Carlo estimate of the uncertainty in future performance predictions.

 

Pressure data from an active well and from an observation well are used to condition the model. The examples of this work show that conditioning a channel model to active and observation well pressure data leads to a significant reduction in uncertainty in the model and in future performance predictions. The integration of observation well pressure data is most useful for reducing the uncertainty in porosity.