Landmark Graphics Corporation

Landmark Graphics Corporation, a Halliburton Company

Genbao Shi

Jeffrey Marc Yarus

Richard L Chambers

Zitao Xu

Rae Mohan Srivastava

Veronica Liceras

ChiKang David Chien

Yogendra Narayan Pandey

Maurice Christopher Gehin

Andres Cesar Callegari

Mehran Hassanpour

Zhaoyang Wang

Chikang David Chien

Randy Guetter

Marko Maucec

Gustavo Carvajal

Steven Ward

Yevgeniy Zagayevskiy

Jin Fei

Joe Lynch

Gaetan Bardy

Gaetan Pierre Louis Bardy

Michael Dunn

Jeffrey M Yarus

Richard Chambers

Andres C Callegari

Mayrice Gehin

Hassanpour Mehran

Gaetan Pierre Bardy

Maurice Gehin

Jeffrey Yarus

A method of predicting rock properties at a selectable scale is provided, including receiving coordinates of locations of respective sample points, receiving measurement data associated with measurements or measurement interpretations for each sample point, receiving for each sample point a scale that indicates the scale used to obtain the measurements and/or measurement interpretations, wherein different scales are received for different sample points. A deep neural network (DNN) is trained by applying the received coordinates, measurement data, and scale associated with each sample point and associating the sample point with a rock property as a function of the coordinates, measurement data, and scale applied for the sample point. The DNN is configured to generate rock property data for a received request point having coordinates and a selectable scale, wherein the rock property data is determined for the request point as a function of the coordinates and the selectable scale.

Building scalable geological property models using machine learning algorithms

A neural network trainer trains neural networks to estimate secondary data at locations throughout a geological formation where secondary data is unknown. The neural networks are trained to estimate secondary data using locations in the geological formation as input. Subsequently, the secondary data is deleted from memory using the trained neural network as a proxy representation to reduce memory footprint and allow for estimation of secondary data at locations where it is unknown.

Geological property modeling with neural network representations

A method for processing a well data log may comprise adding one or more boundary areas to the well data log, dividing the well data log into one or more segments using the one or more boundary areas, processing each of the one or more segments on one or more information handling systems, and reforming each of the one or more segments into a final simulation. A system for processing a well data log may comprise one or more information handling systems in a cluster. The one or more information handling systems may be configured to perform the method for processing the well data log.

Distributed Sequential Gaussian Simulation

A method for creating a seamless scalable geological model may comprise identifying one or more geological scales, establishing a geological tied system, identifying one or more graphical resolution levels for each of the one or more geological scales, constructing the seamless scalable geological model, and producing a post-process model. A system for creating a seamless scalable geological model may comprise an information handling system, which may comprise a random access memory, a graphics module, a main memory, a secondary memory, and one or more processors configured to run a seamless scalable geological model software.

Seamless scaling geomodeling

Target objects are simulated using different triangle mesh sizes to improve processing performance. To perform the simulation, a seed point for the target object within a constraint volume is determined, the seed point representing a vertex of a first triangle forming part of the target object. One or more hexagonal orbits of triangles adjacent the first triangle are propagated, whereby the hexagonal orbits of triangles form the target object. The size of each triangle is determined based upon dimensions of the target object, and the target object is generated.

Target object simulation using orbit propagation

Systems and methods for mapping vertices from one coordinate system in an earth model to another coordinate system in a two-dimensional (2D) array without disrupting the topology of the vertices.

Mapping vertices from an earth model to a 2D array

A method includes generating a faulted point cloud representing a faulted geological formation including a first fault block having a first surface, a second fault block having a second surface, and a fault formed therebetween and having a fault surface, generating a first fault trace from an intersection of the first surface and the fault surface and a second fault trace from an intersection of the second surface and the fault surface, generating a first fault polygon using the first and second fault traces, generating a center polyline, generating third and fourth fault traces separated from the first and second fault traces, respectively, generating a second fault polygon using the third and fourth fault traces, expanding a first area including the first and third fault traces, expanding a second area including the second and fourth fault traces, and generating an unfaulted point cloud representing an unfaulted geological formation.

Systems and methods for unfaulting point clouds

The disclosed embodiments include a method, apparatus, and computer program product for modifying a three-dimensional geocellular model. For example, one disclosed embodiment includes a system that includes at least one processor and at least one memory coupled to the at least one processor. The memory stores instructions that when executed by the at least one processor performs operations that includes loading into memory a three-dimensional geocellular model that corresponds to a two-dimensional geological model. The operations include determining a portion of the three-dimensional geocellular model affected by a change to the two-dimensional geological model and performing a local update to the portion of the three-dimensional geocellular model affected by the change.

Local updating of 3D geocellular model

Fracture networks are simulated using a large triangle mesh size for large fractures and a smaller triangle mesh size for small fractures. Input data defining parameters of one or more fractures are input, the fractures being comprised of a triangle mesh. A first triangle mesh size for the fractures is determined based upon the input data. A second smaller triangle mesh size is then determined based upon the input data. The fracture network is then simulated using the large and small triangle mesh sizes.

Fracture network triangle mesh adjustment

In various examples, a method includes storing one or more data structures on a storage device, the one or more data structures identifying a plurality of faults in a geographical formation and a plurality of fault blocks on either side of the plurality of faults in the geographic formation; for each pair of faults blocks on opposite sides of a fault identified in the one or more data structures: determining, using at least one processor, a fault polygon of a respective pair of fault blocks with respect to a fault of the plurality of faults; and calculating a matching factor between the respective pair of fault blocks based on the fault polygon; selecting a pair of fault blocks to merge based on the calculated matching factor; and updating the one or more data structures to indicate the selected pair of fault blocks has been merged.

Global grid building unfaulting sequence for complex fault-network topologies

A consecutive set of data points, P, P, P, and P, is selected. A first line is created through Pand P. A first tangent vector originating at P, parallel to the first line, is created. A second line through Pand Pis created. A second tangent vector through Pis created. The second tangent vector is parallel to the second line. A baseline through Pand Pis created. A Bezier curve between Pand Pis created, wherein the Bezier curve has a degree. The degree of the Bezier curve is determined based on a comparison of the first tangent vector, the second tangent vector, and the baseline.

Tri-cubic and hybrid interpolation in a 3D texture shader

Target objects having undulating surfaces are simulated using different triangle mesh sizes to improve processing performance. To perform the simulation, a target object is generated using a triangle mesh formed by a plurality of triangles. The target object has an X, Y, and Z direction, wherein the Z direction is perpendicular to an X-Y plane of the target object. The undulating surface on the target object is generated using a Z value in the Z direction.

Target object simulation using undulating surfaces

Systems and methods for optimizing hardware resource utilization when processing variable-precision data are provided. Application data objects are processed using either a central processing unit (CPU) or the relatively lower precision data processing requirements of a dedicated math processing unit, e.g., a graphics processing unit (GPU), based on a level of precision determined for each application data object. The level of precision is used to calculate at least one bounding value for each application data object. The bounding value is compared to a selected precision threshold in order to determine whether the application data object can be processed by the GPU at a relatively lower level of precision without an undesirable loss of computational precision.

Optimizing computer hardware resource utilization when processing variable precision data

A computing device facilitates the organization of a plurality of three-dimensional geological data realizations into respective one-dimensional arrays of geological property values, with each geological property value corresponding to a three-dimensional grid location of a respective three-dimensional geological data realization. The computing device then facilitates the grouping of the one-dimensional arrays into two or more array clusters based on a comparison of geometric locations of the geological property values within the respective arrays, and selects at least one of the plurality of three-dimensional geological data realizations for each of the two or more array clusters. The selected data realizations are then provided at a user interface.

Cluster analysis for selecting reservoir models from multiple geological realizations

Systems and methods for creating a surface in a faulted space, which includes using interpolation techniques.

Growing community of inventors

Sugar Land, TX, United States of America

Genbao Shi