Systems and methods for refining estimated parameter values in seismic imaging
AssigneeChevron U.S.A. Inc.
InventorsAdam Dean Halpert, Laura L. Bandura, Shuxing Cheng, Konstantin Osypov
Life of this patent
Estimating the term…
Abstract
Systems and methods for training a model that refines estimated parameter values include computer processors and non-transitory electronic storage that stores subsurface map data sets that correspond to different subsurface volumes of interest, the system configured to obtain training data including unrefined subsurface map data sets specifying estimated parameter values of a first parameter as a function of position within corresponding subsurface volumes of interest, obtain an initial seismic mapping model, generate a conditioned seismic mapping model, and store the conditioned seismic mapping model in the electronic storage.
Claims20
A computer-implemented method for training a model that refines estimated parameter values within subsurface map data sets, the method being implemented in a computer system, the computer system including one or more physical computer processors and non-transitory electronic storage that stores subsurface map data sets that correspond to different subsurface volumes of interest, wherein a given subsurface map data set corresponding to a given subsurface volume of interest specifies estimated parameter values of different parameters as a function of position within the given subsurface volume of interest, the method comprising:
- obtaining, from the electronic storage, training data, the training data including (i) unrefined subsurface map data sets specifying estimated parameter values of a first parameter as a function of position within corresponding subsurface volumes of interest, and (ii) refined subsurface map data sets specifying refined estimations of parameter values of the first parameter as a function of position within the corresponding subsurface volumes of interest, wherein the estimated parameter values and the refined estimations of parameter values corresponding to the first parameter are seismic amplitude values, porosity values, acoustic impedance values, density values, or lithology values, and wherein the estimated parameter values include pre-processed noise derived from collecting the parameter values and processed noise derived from processing the estimated parameter values;
- obtaining, from the electronic storage, an initial seismic mapping model;
- generating a conditioned seismic mapping model by training, on the one or more physical computer processors using the training data, the initial seismic mapping model;
- and
- storing the conditioned seismic mapping model in the electronic storage.
A computer-implemented method for refining estimated parameter values within subsurface map data sets, the method being implemented in a computer system, the computer system including one or more physical computer processors and non-transitory electronic storage that stores subsurface map data sets that correspond to different subsurface volumes of interest, wherein a given subsurface map data set corresponding to a given subsurface volume of interest specifies estimated parameter values of different parameters as a function of position within the given subsurface volume of interest, the method comprising:
- obtaining an initial target subsurface map data set;
- obtaining, from the electronic storage, a conditioned seismic mapping model, the conditioned seismic mapping model having been conditioned by training an initial seismic mapping model, wherein the training data includes (i) unrefined subsurface map data sets specifying estimated parameter values of a first parameter as a function of position within corresponding subsurface volumes of interest, and (ii) refined subsurface map data sets specifying refined estimations of parameter values of the first parameter as a function of position within the corresponding subsurface volumes of interest, wherein the estimated parameter values and the refined estimations of parameter values corresponding to the first parameter are seismic amplitude values, porosity values, acoustic impedance values, density values, or lithology values, and wherein the estimated parameter values include pre-processed noise derived from collecting the parameter values and processed noise derived from processing the estimated parameter values;
- applying the conditioned seismic mapping model to the initial target subsurface map data set to generate a refined target subsurface map data set;
- generating an image that represents the refined target subsurface map data set using visual effects to depict at least a portion of the parameter values in the refined target subsurface map data set as a function of position within the subsurface volume of interest;
- and
- displaying, on a graphical user interface, the image.
A computer-implemented method for refining estimated parameter values within subsurface map data sets, the method being implemented in a computer system, the computer system including one or more physical computer processors and non-transitory electronic storage that stores subsurface map data sets that correspond to different subsurface volumes of interest, wherein a given subsurface map data set corresponding to a given subsurface volume of interest specifies estimated parameter values of different parameters as a function of position within the given subsurface volume of interest, the method comprising:
- obtaining multiple initial target subsurface map data sets that have been derived from different seismic pre-stack data sets;
- obtaining, from the electronic storage, a conditioned seismic mapping model, the conditioned seismic mapping model having been conditioned by training an initial seismic mapping model, wherein the training data includes (i) unrefined subsurface map data sets specifying estimated parameter values of a first parameter as a function of position within corresponding subsurface volumes of interest, and (ii) refined subsurface map data sets specifying refined estimations of parameter values of the first parameter as a function of position within the corresponding subsurface volumes of interest, wherein the estimated parameter values and the refined estimations of parameter values corresponding to the first parameter are seismic amplitude values, porosity values, acoustic impedance values, density values, or lithology values, and wherein the estimated parameter values include pre-processed noise derived from collecting the parameter values and processed noise derived from processing the estimated parameter values;
- applying the conditioned seismic mapping model individually to the initial target subsurface map data sets to generate metric values of an estimated quality metric for the individual initial target subsurface map data sets;
- selecting from among the multiple sets of seismic pre-stack data based on the metric values of the estimated quality metric determined for the initial target subsurface map data sets derived from the seismic pre-stack data sets;
- generating a stacked seismic image using selected ones of the seismic pre-stack data sets;
- and
- displaying, on a graphical user interface, the stacked seismic image.
Dependencies are inferred from the claim text.