Method for estimating formation permeability from wireline logs using neural networks

摘要

A method for determining rock formation permeability from wireline well logs utilizes neural networks. The neural networks provide consistency, accuracy and overall quality without bias to the calculations. A network is provided having input, intermediate and output layers (12,14,16). Wireline log data and corresponding core permeability data are acquired and the network is trained by iterative weight adjustment to recognize known permeability from known input data, the trained network is tested on known data and permeabilities and then applied to new well log data to determine permeability.