Exxonmobil Upstream Research Company

Antonio R C Paiva

Matthias Imhof

Pavel Dimitrov

Leslie A Wahrmund

Damian N Burch

Sara E Hanson-Hedgecock

Rainer Van Den Bosch

Amit Kushwaha

The systems and methods described herein include training a well performance predictor based on field data corresponding to a hydrocarbon field in which a well is to be drilled; generating a number of candidate well parameter combinations for the well and predicting a performance of the well for each candidate well parameter combination using the trained well performance predictor; and determining an optimized well parameter combination for the well such that the predicted performance of the well is maximized.

Determining well parameters for optimization of well performance

A method to classify one or more seismic surfaces or surface patches based on measurements from seismic data, including: obtaining, by a computer, a training set including a plurality of previously obtained and labeled seismic surfaces or surface patches and one or more training seismic attributes measured or calculated at, above, and/or below the seismic surfaces; obtaining, by the computer, one or more unclassified seismic surfaces or surface patches and one or more seismic attributes measured or calculated at, above, and/or below the unclassified seismic surfaces; learning, by the computer, a classification model from the previously obtained and labeled seismic surfaces or surface patches and the one or more training seismic attributes; and classifying, by the computer, the unclassified seismic surfaces or surface patches based on a comparison between the classification model and the unclassified seismic surfaces or surface patches.

Seismic stratigraphic surface classification

A method to select a representative subset of a plurality of horizon surfaces or surface patches from geophysical subsurface imaging data, including: defining a score function on one or more horizon surfaces or surface patches; calculating, by a computer, the score for each of the plurality of horizon surfaces or surface patches with regard to other horizon surfaces or surface patches and whether the other horizon surfaces or surface patches have been selected or not for inclusion or exclusion in the subset of the plurality of horizon surfaces; selecting, by a computer, one or more of the plurality of horizon surfaces or surface patches to be included in the subset of the plurality of horizon surfaces or surface patches or excluded from the subset of the plurality of horizon surfaces or surface patches based on their respective scores; iteratively repeating the selecting and calculating steps until a stopping condition is reached and the subset of the plurality of horizon surfaces or surface patches is determined; and performing interpretation on the subset of the plurality of horizon surfaces or surfaces patches.

Method for selecting horizon surfaces

Systems and methods for determining well parameters for optimization of well performance. The method includes training, via a computing system, a well performance predictor based on field data corresponding to a hydrocarbon field in which a well is to be drilled. The method also includes generating, via the computing system, a number of candidate well parameter combinations for the well and predicting, via the computing system, a performance of the well for each candidate well parameter combination using the trained well performance predictor. The method further includes determining, via the computing system, an optimized well parameter combination for the well such that the predicted performance of the well is maximized.

Method for decomposing a complexly shaped object in a data set, such as a geobody () in a seismic data volume, into component objects more representative of the true connectivity state of the system represented by the data set. The geobody is decomposed using a basis set of eigenvectors () of a connectivity matrix () describing the state of connectivity between voxels in the geobody. Lineal subspaces of the geobody in eigenvector space are associated with likely component objects (), either by a human interpreter () cross plotting () two or more eigenvectors, or in an automated manner in which a computer algorithm () detects the lineal sub-spaces and the clusters within them.

Growing community of inventors

Branchburg, NJ, United States of America

Antonio R C Paiva