International Business Machines Corporation

Shell Usa, Inc.

Matthias Kormaksson

Theodore Gerard Van Kessel

Lior Horesh

Andrew R Conn

Moshood Omolade Saliu

Paul Borrel

Bianca Zadrozny

Carmen Nilda Mena Paz

Paulo Rodrigo Cavalin

Marcos Rodrigues Vieira

Luciano De Andrade Barbosa

Ligang Lu

Vagner Figueredo De Santana

Peter Bak

Gilad Mordechai Saadoun

Alvaro Bueno Buoro

Cicero Nogueira Dos Santos

Harold Jeffrey Ship

Marcelo Nery Dos Santos

Yuval Nardi

Jorge Luis Guevara Diaz

Alecio Pedro Delazari Binotto

Michael Raghib

Daria Bogdanova

Michael Raghib Moreno

Breno William Santos Rezende De Carvalho

Mingqi Wu

Detlef Hohl

John Tolle

Carlos Raoni De Alencar Mendes

Jan Limbeck

A system, method and program product for implementing a machine learning platform that processes a data map having feature and operational information. A system is disclosed that includes an interpretable machine learning model that generates a function in response to an inputted data map, wherein the data map includes feature data and operational data over a region of interest, and wherein the function relates a set of predictive variables to one or more response variables; an integration/interpolation system that generates the data map from a set of disparate data sources; and an analysis system that evaluates the function to predict outcomes at unique points in the region of interest.

Machine learning platform for processing data maps

An embodiment includes a method for use by at least one machine learning classifier. The method comprises the machine learning classifier obtaining one or more recent results from at least one geomechanical simulation; the machine learning classifier comparing the recent results to stored historical data; and, based on the comparing, the machine learning classifier deciding at least one reservoir model for use by at least one reservoir simulation.

Machine learning assisted reservoir simulation

A system, method and program product for stratigraphic layer identification using a stratigraphic knowledge base for machine learning. Reservoir data includes seismic data and well log data for a reservoir area. The well log data is processed to identify well stratigraphic layer features and the seismic data is processed to identify seismic stratigraphic layer features. A feature matching algorithm based on a stratigraphic knowledge base is selected to match the well stratigraphic layer features to the seismic stratigraphic layer features. The matched features are used to define a stratigraphic layer interpretation for the reservoir area and the interpretation is presented to a user.

Stratigraphic layer identification from seismic and well data with stratigraphic knowledge base

A system, method and program product for processing borehole images to delineate between natural fractures and induced fractures. A system is disclosed that includes: an image analysis platform that inputs a noisy image from a borehole, processes the image using a set of filtering strategies, and renders a set of suggested filtered images via a user interface, the user interface including a mechanism for allowing a user to choose a selected filtered image from the set of suggested filtered images that best delineates between natural fractures and induced fractures, and wherein the image analysis platform further includes a feedback system for packaging and outputting the noisy image and selected filtered image as feedback; and a learning platform having a knowledge registration system that collects and stores training data and the feedback and in a knowledgebase, and a machine learning system that generates filtering strategies.

Identification of natural fractures in wellbore images using machine learning

A method for identifying seismic horizons in digitized seismic images includes deploying a swarm of agents at an initial position in a seismic image to be analyzed, where the swarm of agents includes picking agents that define a direction for the swarm and averaging agents that smooth the direction of the swarm, identifying a direction to follow in the seismic image for each picking agent, and advancing each picking agent in the identified direction, and averaging, by the averaging agents, the directions identified by the picking agents, wherein if an information concentration measured by a picking agent at a current time step is greater than a previous time step, the picking agent keeps a previous direction, otherwise the picking agent changes direction according to the average current state of the set of averaging agents within its neighborhood.

Unsupervised identification of seismic horizons using swarms of cooperating agents

A method for identifying a resource in a field using historic well data including vertical well logs for the resource and historic horizontal well production data for the resource, the method including extracting a plurality of features from the vertical well logs, performing a spatial interpolation of the plurality of features extracted from the vertical well logs onto coordinates of the horizontal well production data to determine a plurality of interpolated features, and building a model predicting production of the resource in the field by regressing the horizontal well production data onto the interpolated features, wherein the model is displayed as a visualization of the resource production predicted in the field.

Resource identification using historic well data

A fracture pattern characterization method, system, and computer program product, include classifying fracture patterns of reservoir layers using a correlation of reservoir information associated with target images and the target images extracted from a database and determining a fracture pattern characterization of a new reservoir layer input based on the classified fracture patterns.

System, method and computer program product for fracture pattern characterization in naturally fractured reservoirs

One aspect includes a method for use with a plurality of steam assisted gravity drainage wells. The method includes designing a first modeling framework which provides a model of total output for the plurality of wells from one or more input variables for respective ones of the plurality of wells. The method also includes designing a second modeling framework which provides a model of output for a given one of the plurality of wells based at least in part on one or more input variables for the given one of the plurality of wells and one or more input variables for at least another one of the plurality of wells. The method further includes utilizing the first and second modeling frameworks to determine an optimal configuration for the wells and configuring the well to operate with the determined optimal configuration.

Cross-well allocation optimization in steam assisted gravity drainage wells

One aspect includes a method for use with a steam assisted gravity drainage well. The method includes generating an analytical framework for the well at least in part by modeling one or more observable variables as a function at least of one or more control variables; and modeling one or more output variables as a function at least of at least one of the one or more control variables; and at least one of the one or more observable variables. The method further includes utilizing the analytical framework to determine an optimal configuration for the well, the optimal configuration including a set of one or more values for the one or more control variables which produce one or more desired values for the one or more output variables; and configuring the well to operate with the determined optimal configuration.

Multi-level modeling of steam assisted gravity drainage wells

A method for controlling operation of a steam-assisted gravity drainage (SAGD) oil well system to increase efficiency in emulsion production includes obtaining a predictive model of the SAGD oil well system. The predictive model has control parameters including one or more continuous control parameters and one or more discrete control parameters. The method also includes utilizing the predictive model to determine values for the control parameters that meet one or more objectives for the SAGD oil well system subject to one or more constraints based on a sequence of iterations of solving a mixed integer non-linear problem. The method further includes adjusting controls of the SAGD oil well system based on the determined values of the control parameters.

Controlling operation of a steam-assisted gravity drainage oil well system utilizing continuous and discrete control parameters

Visualizations of a serial process are provided. A process model of a serial process is constructed using an underlying statististical model. The process model represents each instance of the serial process as a series of interconnected nodes, wherein (i) the serial process includes a plurality of events; (ii) each node is associated with an event and a time at which the associated event occurred in an instance of the serial process; and (iii) connectors link nodes together to form the series of interconnected nodes, the connectors representing stages of serial process. The stages are scored and categorized into various categories based, at least in part, on the process model. One or more of the categories is associated with outlier stages. A visualization of the process model is presented such that the visualization associates the stages of the instances of serial process with respective categories.

Visualization of serial processes

A method for increasing efficiency in emulsion production for a steam-assisted gravity drainage (SAGD) oil well system includes generating a causal model of the SAGD oil well system and training the causal model of the SAGD oil well system utilizing historical time series data relating to one or more SAGD oil wells at one or more SAGD production sites of the SAGD oil well system. The historical time series data is obtained from a plurality of sensors in the SAGD oil well system. The method also includes utilizing the causal model to determine a forecast emulsion production and a forecast set of control parameters associated with one or more of the SAGD production sites of the SAGD oil well system. The method further includes adjusting a set of controls of the SAGD oil well system based on the forecast emulsion production and the forecast set of control parameters and subject to one or more constraints associated with the SAGD oil well system.

Controlling operation of a steam-assisted gravity drainage oil well system by adjusting controls based on forecast emulsion production

A method for increasing efficiency in emulsion production for a steam-assisted gravity drainage (SAGD) oil well system includes generating a multiple time step model of the SAGD oil well system and training the multiple time step model of the SAGD oil well system utilizing historical time series data relating to one or more SAGD oil wells at one or more SAGD production sites of the SAGD oil well system. The historical time series data is obtained from a plurality of sensors in the SAGD oil well system. The method also includes utilizing the multiple time step model to determine, based on a set of objectives and subject to one or more constraints, two or more sets of values for control parameters associated with one or more of the SAGD production sites of the SAGD oil well system for respective ones of two or more time steps. The method further includes adjusting, in each of the two or more time steps, a set of controls of the SAGD oil well system based on the determined values of the control parameters for the corresponding time step.

Controlling operation of a stem-assisted gravity drainage oil well system by adjusting multiple time step controls

A method for increasing efficiency in emulsion production for a steam-assisted gravity drainage (SAGD) oil well system includes generating a model of the SAGD oil well system and training the model of the SAGD oil well system utilizing historical time series data relating to one or more SAGD oil wells at one or more SAGD production sites of the SAGD oil well system. The historical time series data is obtained from a plurality of sensors in the SAGD oil well system. The method also includes determining an uncertainty of the model as a function of a control space of the model and utilizing the model to determine values for control variables associated with one or more of the SAGD production sites of the SAGD oil well system which reduce the model uncertainty while meeting one or more objectives subject to one or more constraints. The method further includes adjusting a set of controls of the SAGD oil well system based on the determined values for the control variables.

Growing community of inventors

Rio de Janeiro, Brazil

Matthias Kormaksson