Halliburton Energy Services, Inc.

Jacky Muri Wiener

Muthukumarappan Ramurthy

Harry D Smith, Jr

Dingding Chen

Syed Hamid

John Andrew Quirein

Jeffery L Grable

Donald Pat Kundert

John Bosco Le Tran

Meagan Renee Stephens

A procedure that integrates petrophysics and rock typing taken from vertical well measurements, 3D seismic elastic properties and seismic attributes, and geostatistical modeling to build a 3D reservoir model is provided. The 3D reservoir model may be directly incorporated into horizontal fracture model designs. The developed 3D reservoir model for a subsurface volume may be used in a fracture model to optimize fracturing designs and maximize well performance.

Rock physics based method of integrated subsurface reservoir characterization for use in optimized stimulation design of horizontal wells

Permeability prediction systems and methods using quadratic discriminant analysis are presented. At least one disclosed method embodiment includes: acquiring formation property measurements at a plurality of positions along at least one borehole in a study area; identifying clusters in a plurality of points representing the formation property measurements at the plurality of postions; and determining a system permeability value for each cluster. Quadratic Discriminant Analysis ('QDA') is used to associate one the clusters with each position along the one or more boreholes, thereby determining a system permeability prediction for each position. The total system permeability can then be predicted by aggregating the system permeability predictions.

Permeability prediction systems and methods using quadratic discriminant analysis

A system and method for selecting a training data set from a set of multidimensional geophysical input data samples for training a model to predict target data. The input data may be data sets produced by a pulsed neutron logging tool at multiple depth points in a cases well. Target data may be responses of an open hole logging tool. The input data is divided into clusters. Actual target data from the training well is linked to the clusters. The linked clusters are analyzed for variance, etc. and fuzzy inference is used to select a portion of each cluster to include in a training set. The reduced set is used to train a model, such as an artificial neural network. The trained model may then be used to produce synthetic open hole logs in response to inputs of cased hole log data.

Neural network training data selection using memory reduced cluster analysis for field model development

Growing community of inventors

Aurora, CO, United States of America

Jacky Muri Wiener