Your browser is not recommended for EDX, we suggest using the latest version of Google Chrome.

Development, Testing and Validation of a Neural Model to Predict Porosity and Permeability from Well Logs, Grayburg Formation, McElroy Field, West Texas

License(s):

Other (Open)

Accurate, high-resolution, three-dimensional (3D) reservoir characterization can provide substantial benefits for effective oilfield management. By doing so, the predictive reliability of reservoir flow models, which are routinely used as the basis for investment decisions involving hundreds of millions of dollars and designed to recover millions of barrels of oil, can be significantly improved. Even a small improvement in incremental recovery for high-value assets can result in important contributions to bottom-line profitability. Today's standard practice for developing a 3D reservoir description is to use seismic inversion techniques. These techniques make use of geostatistics and other stochastic methods to solve the inverse problem, i.e., to iteratively construct a likely geologic model and then upscale and compare its acoustic response to that actually observed in the field.

Followers: 0

Citation (Click to Copy)

Data and Resources

    Gathering Resources...

Keywords

Additional Info

Field Value
Last Updated September 29, 2016, 15:03 (LMT)
Created September 29, 2016, 15:03 (LMT)
Citation Jack W. Steen Scott R. Reeves ---- Roy Long, Development, Testing and Validation of a Neural Model to Predict Porosity and Permeability from Well Logs, Grayburg Formation, McElroy Field, West Texas, 2016-09-29, https://edx.netl.doe.gov/dataset/development-testing-and-validation-of-a-neural-model-to-predict-porosity-and-permeability-from-wel
Netl Product yes
Poc Email Roy.long@netl.doe.gov
Point Of Contact Roy Long
Program Or Project KMD
Publication Date 2003-7-1