Ai-assisted hydrocarbon anomaly identification

Abstract

A method for generating a seismic profile includes receiving input data. The method also includes generating seismic features based upon the input data. The method also includes extracting seismic tiles and / or sub-volumes from the seismic features. The method also includes training a 2D or 3D model based upon the seismic features and the seismic tiles and / or sub-volumes to produce a trained 2D or 3D model. The method also includes generating the seismic profile using the trained 2D or 3D model.

Description

AI-ASSISTED HYDROCARBON ANOMALY IDENTIFICATIONCross-Reference to Related Applications

[0001] This application claims priority to U.S. Provisional Patent Application No. 63 / 568,247, filed on March 21, 2024, which is incorporated by reference in its entirety.Background

[0002] The challenge of identifying seismic anomalies for exploration purposes is caused by the diversity of geological settings for hydrocarbon accumulation and the limitations and ambiguity of the imaging capabilities of seismic data. For this reason, geoscientists perform extensive tests and validations to determine whether the identified seismic anomaly is a drillable prospect. In addition, the ability to learn from previous successes or failures is limited by how many examples were available to the geoscientist. Moreover, applying the same standards from one case to another is an interpretive task that is dependent on the geoscientist’s experience and skills.

[0003] On the other hand, the ability of machine-learning (ML) models to generalize over a large set of successes and failures and apply the same standards when assessing each case makes it possible to detect the potential exploration successes or failures in a more consistent manner. Therefore, what is needed is a system and method to leverage the artificial intelligence (Al) technologies to automate this process and increase the certainty of the results.Summary

[0004] A method for generating a seismic profile is disclosed. The method includes receiving input data. The method also includes generating seismic features based upon the input data. The method also includes extracting seismic tiles and / or sub-volumes from the seismic features. The method also includes training a 2D or 3D model based upon the seismic features and the seismic tiles and / or sub-volumes to produce a trained 2D or 3D model. The method also includes generating the seismic profile using the trained 2D or 3D model.

[0005] A computing system is also disclosed. The computing system includes one or more processors and a memory system. The memory system includes one or more non-transitory computer-readable media storing instructions that, when executed by at least one of the one or more processors, cause the computing system to perform operations. The operations includereceiving input data. The input data includes well log data, binary or multi-class success labels, and / or seismic data. The operations also include generating seismic features based upon the input data. The operations also include training a one-dimensional (ID) model based upon the input data and the seismic features to produce a trained ID model. The operations also include generating classification sub-cubes using the trained ID model. The operations also include extracting seismic tiles and / or sub-volumes from the input data and the seismic features. The operations also include extracting masks and / or labels from the classification sub-cubes. The masks and / or labels correspond to the seismic tiles and / or sub-volumes. The operations also include training a 2D or 3D model based upon the seismic features, the seismic tiles and / or subvolumes, and the corresponding masks and / or labels to produce a trained 2D or 3D model. The seismic features are used to guide the trained 2D or 3D model to generate geologically consistent anomalies. The operations also include generating a seismic profile using the trained 2D or 3D model.

[0006] A non-transitory computer-readable medium is also disclosed. The medium stores instructions that, when executed by one or more processors of a computing system, cause the computing system to perform operations. The operations include receiving input data. The input data includes well log data, binary or multi-class success labels, and / or seismic data. The multiclass success labels include a reservoir type, a fluid content, a gas pay, a stacked oil pay, a thin oil pay, a tight oil pay, a fluid contact, wet sand, a channel margin, mud facies, wet reservoir, and / or non-reservoir. The operations also include generating seismic features based upon the input data. The seismic features include derivatives of the seismic data. The seismic features include poststack seismic reflection data, machine learning (ML)-based seismic attributes, amplitude versus offset (AVO) attributes, energy envelopes of seismic traces, and / or geological context attributes in relative geologic time (RGT). The operations also include training a one-dimensional (ID) model based upon the input data and the seismic features to produce a trained ID model. The trained ID model is trained to predict the binary or multi-class success labels. The trained ID model includes a deep learning model. The seismic features are used to guide the trained ID model to generate geologically consistent anomalies. The geologically consistent anomalies include a class, the gas pay, the stacked oil pay, the thin oil pay, the fluid contact, the wet sand, the channel margin, and / or the mud facies. Being geologically consistent indicates that multi-class labelling is performed in a context of geomorphic and lithostratigraphic interpretation. The operations alsoinclude generating classification sub-cubes using the trained ID model. The classification subcubes include the seismic traces around well trajectories. The operations also include extracting seismic tiles and / or sub-volumes from the input data and the seismic features. The seismic tiles are two-dimensional (2D) portions of a seismic section. The seismic tiles are extracted in different directions and depths. The sub-volumes are three-dimensional (3D) cuboidal chunks of a larger 3D seismic volume. The operations also include extracting masks and / or labels from the input data, the seismic features, and the classification sub-cubes. The masks and / or labels correspond to the seismic tiles and / or sub-volumes. The operations also include training a 2D or 3D model based upon the seismic features, the seismic tiles and / or sub-volumes, and the corresponding masks and / or labels to produce a trained 2D or 3D model. The trained 2D or 3D model includes a deep learning model. The seismic features are used to guide the trained 2D or 3D model to generate the geologically consistent anomalies. The operations also include generating a seismic profile using the trained 2D or 3D model. The seismic profile includes a 3D success cube. The trained 2D or 3D model is applied on the seismic tiles in inline and crossline directions or applied on the sub-volumes to generate the 3D success cube. The 3D success cube includes a binary or multi-class cube. The 3D success cube illustrates probabilities that a plurality of seismic pixels or voxels belong to one or more predefined classes that are derived from the ML-based seismic attributes and / or model features. The model features include output from intermediate layers of the trained ID, 2D, and / or 3D model.

[0007] It will be appreciated that this summary is intended merely to introduce some aspects of the present methods, systems, and media, which are more fully described and / or claimed below. Accordingly, this summary is not intended to be limiting.Brief Description of the Drawings

[0008] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the present teachings and together with the description, serve to explain the principles of the present teachings. In the figures:

[0009] Figure 1 illustrates an example of a system that includes various management components to manage various aspects of a geologic environment, according to an embodiment.

[0010] Figure 2 illustrates well log data (six logs on the left) versus binary success (on the right) where black represents success (1) and white represents failure (0), according to an embodiment.

[0011] Figure 3 illustrates seismic and its attributes (eight on the left) along the well trajectory versus a binary success log (on the right) where black represents a success (1) and white represents a failure (0), according to an embodiment.

[0012] Figure 4 illustrates three examples of seismic 2D tiles (left) and corresponding binary success tiles (right), according to an embodiment.

[0013] Figure 5 illustrates seismic inputs displayed on a 2D section (left) versus a corresponding binary success section (right), according to an embodiment.

[0014] Figure 6 illustrates a flowchart of a method for generating a seismic profde, according to an embodiment.

[0015] Figure 7 illustrates a schematic view of an Al model-building framework, according to an embodiment.

[0016] Figure 8 illustrates a schematic view of a computing system for performing at least a portion of the method(s) described herein, according to an embodiment.Detailed Description

[0017] Reference will now be made in detail to embodiments, examples of which are illustrated in the accompanying drawings and figures. In the following detailed description, numerous specific details are set forth in order to provide a thorough understanding of the invention. However, it will be apparent to one of ordinary skill in the art that the invention may be practiced without these specific details. In other instances, well-known methods, procedures, components, circuits, and networks have not been described in detail so as not to unnecessarily obscure aspects of the embodiments.

[0018] It will also be understood that, although the terms first, second, etc. may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. For example, a first object or step could be termed a second object or step, and, similarly, a second object or step could be termed a first object or step, without departing from the scope of the present disclosure. The first object or step, and the second object or step, are both, objects or steps, respectively, but they are not to be considered the same object or step.

[0019] The terminology used in the description herein is for the purpose of describing particular embodiments and is not intended to be limiting. As used in this description and the appendedclaims, the singular forms “a,” “an” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will also be understood that the term “and / or” as used herein refers to and encompasses any possible combinations of one or more of the associated listed items. It will be further understood that the terms “includes,” “including,” “comprises” and / or “comprising,” when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. Further, as used herein, the term “if’ may be construed to mean “when” or “upon” or “in response to determining” or “in response to detecting,” depending on the context.

[0020] Attention is now directed to processing procedures, methods, techniques, and workflows that are in accordance with some embodiments. Some operations in the processing procedures, methods, techniques, and workflows disclosed herein may be combined and / or the order of some operations may be changed.

[0021] The present disclosure may help to identify any possible hydrocarbon (HC) anomalies even if the same response was not encountered in the available well penetration. This raises a question about how sufficient the current labels are. Amplitude versus offset (AVO) background and any anomalous behavior is already depicted in pre-stack seismic data, and AVO anomalies are any deviation from the background trend. The method described herein may employ the appropriate unsupervised technique to identify the background trend in the data and its change with depth and compaction along with any deviations from it. These deviations may be the anomalies that will be validated in the next steps.

[0022] The method makes use of the available well data and seismic interpretation to validate the identified anomalies by the unsupervised model. It is expected that some identified anomalies may not be covered by the current interpretation. Other validation methods may be adopted. The output anomaly cube may be used as an input (along with other attributes) to the supervised learning model.

[0023] The solution starts from the ground truth information (e.g., well data) and extends this information around the well location, and then extends from around well locations to full 3D cube. This may be achieved by unsupervised learning to create initial anomaly cube. It may also or instead be achieved by ID machine learning (ML) so that an ML model can predict the exploration success (e g., classes) given ID seismic traces and their attributes. This model may be deployedaround the well locations so that the information in the wells is extended in 3D sub-cubes around the well trajectory. It may also or instead be achieved by reconciliation and / or validation of the predictions around the well traj ectories with available seismic interpretation. It may also or instead be achieved by multi-attribute 2D learning using 2D seismic tiles around the well trajectories.System Overview

[0024] Figure 1 illustrates an example of a system 100 that includes various management components 110 to manage various aspects of a geologic environment 150 (e.g., an environment that includes a sedimentary basin, a reservoir 151, one or more faults 153-1, one or more geobodies 153-2, etc.). For example, the management components 110 may allow for direct or indirect management of sensing, drilling, injecting, extracting, etc., with respect to the geologic environment 150. In turn, further information about the geologic environment 150 may become available as feedback 160 (e.g., optionally as input to one or more of the management components 110).

[0025] In the example of Figure 1, the management components 110 include a seismic data component 112, an additional information component 114 (e.g., well / logging data), a processing component 116, a simulation component 120, an attribute component 130, an analysis / visualization component 142 and a workflow component 144. In operation, seismic data and other information provided per the components 112 and 114 may be input to the simulation component 120.

[0026] In an example embodiment, the simulation component 120 may rely on entities 122. Entities 122 may include earth entities or geological objects such as wells, surfaces, bodies, reservoirs, etc. In the system 100, the entities 122 can include virtual representations of actual physical entities that are reconstructed for purposes of simulation. The entities 122 may include entities based on data acquired via sensing, observation, etc. (e.g., the seismic data 112 and other information 114). An entity may be characterized by one or more properties (e.g., a geometrical pillar grid entity of an earth model may be characterized by a porosity property). Such properties may represent one or more measurements (e.g., acquired data), calculations, etc.

[0027] In an example embodiment, the simulation component 120 may operate in conjunction with a software framework such as an object-based framework. In such a framework, entities may include entities based on pre-defined classes to facilitate modeling and simulation. Acommercially available example of an object-based framework is the MICROSOFT® .NET® framework (Redmond, Washington), which provides a set of extensible object classes. In the .NET® framework, an object class encapsulates a module of reusable code and associated data structures. Object classes can be used to instantiate object instances for use in by a program, script, etc. For example, borehole classes may define objects for representing boreholes based on well data.

[0028] In the example of Figure 1, the simulation component 120 may process information to conform to one or more attributes specified by the attribute component 130, which may include a library of attributes. Such processing may occur prior to input to the simulation component 120 (e.g., consider the processing component 116). As an example, the simulation component 120 may perform operations on input information based on one or more attributes specified by the attribute component 130. In an example embodiment, the simulation component 120 may construct one or more models of the geologic environment 150, which may be relied on to simulate behavior of the geologic environment 150 (e.g., responsive to one or more acts, whether natural or artificial). In the example of Figure 1, the analysis / visualization component 142 may allow for interaction with a model or model-based results (e.g., simulation results, etc.). As an example, output from the simulation component 120 may be input to one or more other workflows, as indicated by a workflow component 144.

[0029] As an example, the simulation component 120 may include one or more features of a simulator such as the ECLIPSE™ reservoir simulator (SLB, Houston Texas), the INTERSECT™ reservoir simulator (SLB, Houston Texas), etc. As an example, a simulation component, a simulator, etc. may include features to implement one or more meshless techniques (e.g., to solve one or more equations, etc.). As an example, a reservoir or reservoirs may be simulated with respect to one or more enhanced recovery techniques (e.g., consider a thermal process such as SAGD, etc ).

[0030] As an example, the simulation component 120 may include one or more features of a simulator such as SYMMETRY software (SLB, Houston, Texas). More particularly, SYMMETRY may process workflows in a single integrated environment with accurate thermodynamic fluid representation and consistent modeling across multiple disciplines including process, production, and HSE. The simulator integrates steady-state and transient (e.g., dynamic) analyses that can be tailored for each domain. This approach enables users to optimize processesin upstream, midstream, and downstream sectors while maximizing profits and minimizing capital expenditures. It may also help reduce emissions, energy consumption, and waste.

[0031] As an example, the simulation component 120 may include one or more features of a simulator such as PIPESIM (SLB, Houston, Texas). More particularly, PIPESIM is steady-state multiphase flow simulator that incorporates the three areas of flow modeling: multiphase flow, heat transfer and fluid behavior.

[0032] As an example, the simulation component 120 may include one or more features of a simulator such as OLGA™ (SLB, Houston, Texas). More particularly, OLGA™ is a dynamic multiphase flow simulator that models transient flow (e.g., time-dependent behaviors) to maximize production potential. Transient modeling is a component for feasibility studies and field development design. Dynamic simulation is useful in deep water and is used in both offshore and onshore developments to investigate transient behavior in pipelines and wellbores. Transient simulation with the OLGA™ simulator provides an added dimension to steady-state analysis by predicting system dynamics, such as time-varying changes in flow rates, fluid compositions, temperature, solids deposition, and operational changes.

[0033] In an example embodiment, the management components 110 may include features of a commercially available framework such as the PETREL® seismic to simulation software framework (SLB, Houston, Texas). The PETREL® framework provides components that allow for optimization of exploration and development operations. The PETREL® framework includes seismic to simulation software components that can output information for use in increasing reservoir performance, for example, by improving asset team productivity. Through use of such a framework, various professionals (e.g., geophysicists, geologists, and reservoir engineers) can develop collaborative workflows and integrate operations to streamline processes. Such a framework may be considered an application and may be considered a data-driven application (e.g., where data is input for purposes of modeling, simulating, etc.).

[0034] In an example embodiment, various aspects of the management components 110 may include add-ons or plug-ins that operate according to specifications of a framework environment. For example, a commercially available framework environment marketed as the OCEAN® framework environment (SLB, Houston, Texas) allows for integration of add-ons (or plug-ins) into a PETREL® framework workflow. The OCEAN® framework environment leverages .NET® tools (Microsoft Corporation, Redmond, Washington) and offers stable, user-friendly interfacesfor efficient development. In an example embodiment, various components may be implemented as add-ons (or plug-ins) that conform to and operate according to specifications of a framework environment (e.g., according to application programming interface (API) specifications, etc.).

[0035] Figure 1 also shows an example of a framework 170 that includes a model simulation layer 180 along with a framework services layer 190, a framework core layer 195 and a modules layer 175. The framework 170 may include the commercially available OCEAN® framework where the model simulation layer 180 is the commercially available PETREL® model-centric software package that hosts OCEAN® framework applications. In an example embodiment, the PETREL® software may be considered a data-driven application. The PETREL® software can include a framework for model building and visualization.

[0036] As an example, a framework may include features for implementing one or more mesh generation techniques. For example, a framework may include an input component for receipt of information from interpretation of seismic data, one or more attributes based at least in part on seismic data, log data, image data, etc. Such a framework may include a mesh generation component that processes input information, optionally in conjunction with other information, to generate a mesh.

[0037] In the example of Figure 1, the model simulation layer 180 may provide domain objects 182, act as a data source 184, provide for rendering 186 and provide for various user interfaces 188. Rendering 186 may provide a graphical environment in which applications can display their data while the user interfaces 188 may provide a common look and feel for application user interface components.

[0038] As an example, the domain objects 182 can include entity objects, property objects and optionally other objects. Entity objects may be used to geometrically represent wells, surfaces, bodies, reservoirs, etc., while property objects may be used to provide property values as well as data versions and display parameters. For example, an entity object may represent a well where a property object provides log information as well as version information and display information (e g., to display the well as part of a model).

[0039] In the example of Figure 1, data may be stored in one or more data sources (or data stores, generally physical data storage devices), which may be at the same or different physical sites and accessible via one or more networks. The model simulation layer 180 may be configured to model projects. As such, a particular project may be stored where stored project informationmay include inputs, models, results and cases. Thus, upon completion of a modeling session, a user may store a project. At a later time, the project can be accessed and restored using the model simulation layer 180, which can recreate instances of the relevant domain objects.

[0040] In the example of Figure 1, the geologic environment 150 may include layers (e.g., stratification) that include a reservoir 151 and one or more other features such as the fault 153-1, the geobody 153-2, etc. As an example, the geologic environment 150 may be outfitted with any of a variety of sensors, detectors, actuators, etc. For example, equipment 152 may include communication circuitry to receive and to transmit information with respect to one or more networks 155. Such information may include information associated with downhole equipment 154, which may be equipment to acquire information, to assist with resource recovery, etc. Other equipment 156 may be located remote from a well site and include sensing, detecting, emitting or other circuitry. Such equipment may include storage and communication circuitry to store and to communicate data, instructions, etc. As an example, one or more satellites may be provided for purposes of communications, data acquisition, etc. For example, Figure 1 shows a satellite in communication with the network 155 that may be configured for communications, noting that the satellite may additionally or instead include circuitry for imagery (e.g., spatial, spectral, temporal, radiometric, etc.).

[0041] Figure 1 also shows the geologic environment 150 as optionally including equipment 157 and 158 associated with a well that includes a substantially horizontal portion that may intersect with one or more fractures 159. For example, consider a well in a shale formation that may include natural fractures, artificial fractures (e.g., hydraulic fractures) or a combination of natural and artificial fractures. As an example, a well may be drilled for a reservoir that is laterally extensive. In such an example, lateral variations in properties, stresses, etc. may exist where an assessment of such variations may assist with planning, operations, etc. to develop a laterally extensive reservoir (e.g., via fracturing, injecting, extracting, etc.). As an example, the equipment 157 and / or 158 may include components, a system, systems, etc. for fracturing, seismic sensing, analysis of seismic data, assessment of one or more fractures, etc.

[0042] As mentioned, the system 100 may be used to perform one or more workflows. A workflow may be a process that includes a number of worksteps. A workstep may operate on data, for example, to create new data, to update existing data, etc. As an example, a may operate on one or more inputs and create one or more results, for example, based on one or more algorithms. Asan example, a system may include a workflow editor for creation, editing, executing, etc. of a workflow. In such an example, the workflow editor may provide for selection of one or more predefined worksteps, one or more customized worksteps, etc. As an example, a workflow may be a workflow implementable in the PETREL® software, for example, that operates on seismic data, seismic attribute(s), etc. As an example, a workflow may be a process implementable in the OCEAN® framework. As an example, a workflow may include one or more worksteps that access a module such as a plug-in (e.g., external executable code, etc.).AI-Assisted Hydrocarbon Anomaly Identification Using ID and 2D Trace and Seismic Labelling and Binary Prediction

[0043] Figure 2 illustrates well log data 210A-210f versus binary success 220 where black represents success (1) and white represents failure (0), according to an embodiment. Broadly, the method may include ID learning and 2D label generation and validation. The method may also identify exploration success and / or failure logs. More particularly, on a well level, depending on the number of available wells with the assumption that time-depth relations are available, well log data and binary success log labels (e.g., success and / or failure) may be manually interpreted for the wells available. The success log may be further detailed into multi-class labels that include information on the reservoir type and the fluid content. An example of the multiclass labels is shown in Table 1 below. The classes may be defined based on the available data and the feasibility of distinguishing them.Seismic Feature Generation

[0044] Seismic inputs including pre-stack volumes may be enhanced by generating other seismic attributes, leveraging already-available ML models such as machine-learning (ML) relativegeologic time (RGT) and ML seismic conditioning (Table 1). One aim of this task is to generate appropriate seismic features that are commonly used to assess direct hydrocarbon indicators such as downdip and lateral conformance of amplitude to structure, phase change, and consistency of amplitude above background as inputs to the ML model. The exact set attributes and features may be confirmed by testing the ML model behavior.Labelling at Well Control

[0045] In an example, the assumption is that a well tie has already been performed, the input seismic data has already been processed, and the AVO has already been conditioned so that the theoretical AVO response matches the observed response of the pre-stack seismic data. These assumptions may be confirmed during the data ingestion / quality control (QC).Well Trajectory versus Binary Success

[0046] Figure 3 illustrates seismic data and its attributes 310A-310H along the well trajectory versus a binary success log 320 where black represents a success (1) and white represents a failure (0), according to an embodiment. Said another way, Figure 3 shows a training dataset of seismic input(s) 310A-310H along well trajectories and the corresponding labels 320. Depending on the availability of data types, a multi-modal ML formulation may be adopted, and a suitable deep learning model may be trained to predict binary (or multiclass) success.Reconciliation / validation of the sub-cubes predictions

[0047] The ID trained model may be deployed on seismic traces around well trajectories to generate classification sub-cubes around each well. The predictions in these sub-cubes may be validated against the 3D interpretation. When validation is completed, the method may extend the 3D interpretation into the 3D label space.2D Learning - Seismic 2D tiles and Corresponding Binary Success Tiles

[0048] Figure 4 illustrates three examples 41 OA-410C of seismic 2D tiles (left) and corresponding binary success tiles 420A-420C (right), according to an embodiment. In the binary success tiles, black represents a success (1), and white represents a failure (0). Seismic tiles around the wells available may be extracted in different directions and depths within the sub-cube predictions. Thisextraction may generate 2D training data including seismic inputs and corresponding validated classes. Simplified examples showing seismic amplitude and single class masks are shown in Figure 4. This may allow for 2D training. One advantage of 2D is that the model can benefit from the contextual information of seismic pixels which may allow the model to achieve good generalization over various geological settings.Seismic Inputs Versus Corresponding Binary Success Section

[0049] Figure 5 illustrates seismic inputs (displayed on a 2D section) 510A-510E versus a corresponding binary success section 520, according to an embodiment. Black represents a success (1), and white represents a failure (0). After preparing the input and corresponding 2D masks, a 2D deep learning model may be applied on the seismic (and its attributes) tiles with the corresponding 2D masks as labels. At this stage, the selected seismic features may be revised and optimized. The trained model may be deployed on 2D seismic sections in both inline and crossline directions to generate the binary (or multiclass) 3D success cube, as shown in Figure 5. Input features may be used to guide the refined model to select geologically consistent attributes. Structural conformance, access to charge, absence of oil and / or gas escape features, cap rock thickness, and / or presence of other DHI may be validation concepts that the model may be optimized for which to account.Exemplary Method

[0050] Figure 6 illustrates a flowchart of a method 600 for generating a seismic profile, according to an embodiment. The method 600 may also or instead identify hydrocarbon anomalies using ID and 2D traces, seismic labelling, and binary predictions. The method 600 may also or instead classify pixels or voxels in seismic data based upon a set of classes related to exploration success and failure. An illustrative order of the method 600 is provided below; however, one or more portions of the method 600 may be performed in a different order, simultaneously, repeated, or omitted. At least a portion of the method 600 may be performed by a computing system (described below).

[0051] The method 600 may include receiving input data, as at 605. An example of input data 210A-210F is shown in Figure 2. The input data may be or include well log data, binary or multiclass success labels, seismic data, or a combination thereof. The multi-class success labels maybe or include a reservoir type, a fluid content, a gas pay, a stacked oil pay, a thin oil pay, a tight oil pay, a fluid contact, wet sand, a channel margin, mud facies, wet reservoir, non-reservoir, or a combination thereof.

[0052] The method 600 may also include generating seismic features based upon the input data, as at 610. An example of seismic features 310A-310H is shown in Figure 3. The seismic features may be or include derivatives of the seismic data. The seismic features may be or include poststack seismic reflection data, traditional and / or machine learning (ML)-based seismic attributes, amplitude versus offset (AVO) attributes, energy envelopes of seismic traces, geological context attributes in relative geologic time (RGT), or a combination thereof.

[0053] The method 600 may also include training a one-dimensional (ID) model based upon the input data and / or the seismic features to produce a trained ID model, as at 615. The ID model may be trained to predict the binary or multi-class success labels. The trained ID model may be or include a deep learning model. The seismic features may be used to guide the trained ID model to generate geologically consistent anomalies. The geologically consistent anomalies may be or include a class, the gas pay, the stacked oil pay, the thin oil pay, the fluid contact, the wet sand, the channel margin, the mud facies, or a combination thereof. Being geologically consistent indicates that multi-class labelling is done in the context of geomorphic and lithostratigraphic interpretation.

[0054] The method 600 may also include generating classification sub-cubes using the trained ID model, as at 620. The classification sub-cubes may be or include the seismic traces around well trajectories.

[0055] The method 600 may also include extracting seismic tiles and / or sub-volumes from the input data and / or the seismic features, as at 625. An example of the seismic tiles and / or subvolumes 410A-410C is shown in Figure 4. The seismic tiles are two-dimensional (2D) portions of a seismic section. The seismic tiles may be extracted in different directions and depths. The sub-volumes are three-dimensional (3D) cuboidal chunks of a larger 3D seismic volume.

[0056] The method 600 may also include extracting masks and / or labels from the input data, the seismic features, the classification sub-cubes, or a combination thereof, as at 630. An example of the masks and / or labels 420A-420C is shown in Figure 4. The masks and / or labels correspond to the seismic tiles and / or sub-volumes.

[0057] The method 600 may also include training a 2D or 3D model based upon the seismic features, the seismic tiles and / or sub-volumes, the corresponding masks and / or labels, or a combination thereof to produce a trained 2D or 3D model, as at 635. An example of this is shown in Figure 5. The trained 2D or 3D model may be or include a deep learning model. The seismic features may be used to guide the trained 2D or 3D model to generate the geologically consistent anomalies.

[0058] The method 600 may also include generating a seismic profile using the trained ID, 2D, and / or 3D model, as at 640. The seismic profile may be or include a 3D success cube. The trained ID, 2D, and / or 3D model may be applied on the seismic tiles in inline and crossline directions or applied on the sub-volumes to generate the 3D success cube. The 3D success cube may be or include a binary or multi-class cube. The 3D success cube illustrates probabilities that a plurality of seismic pixels or voxels belong to one or more predefined classes that are derived from the traditional and machine learning (ML)-based seismic attributes and / or model features. The model features may be or include output(s) from intermediate layers of the trained ID, 2D, and / or 3D model.

[0059] The method 600 may also include identifying the geologically consistent anomalies in or based upon the 3D success cube, as at 645. These may be or include one or more of the geologically consistent anomalies described above.

[0060] The method 600 may also include displaying the 3D success cube and / or the geologically consistent anomalies in or based upon the 3D success cube, as at 650.

[0061] The method 600 may also include performing an action in response to the seismic profile (e.g., the 3D success cube), and / or the geologically consistent anomalies in or based upon the 3D success cube, as at 655. The action may be or include generating and / or transmitting a signal that recommends, instructs, or causes a physical action to occur at a wellsite. The physical action may be or include extracting geobodies, selecting where to drill a wellbore, drilling the wellbore, varying a weight and / or torque on a drill bit that is drilling the wellbore, varying a drilling trajectory of the wellbore, or varying a concentration and / or flow rate of a fluid pumped into the wellbore.Al Model-Building Framework

[0062] Figure 7 illustrates a schematic view of an Al model-building framework, according to an embodiment. The proposed solution generates a seismic cube illustrating the probability for each seismic voxel to belong to one of the predefined classes. High-probability voxels of one class may provide insights into the 3D geometry of the reservoir and indications of fluid type and contact. Given the diverse training data set, the model may be general enough to classify any given seismic data set without any prior knowledge of the geological settings.

[0063] Moreover, ML automates the process of identification and evaluation of the seismic anomaly. The ML models can be refined and generalized by adding more data and learnings to it. This allows for the construction of a large foundation model that may be trained or many more examples than any expert can comprehend. In contrast, existing technology usually uses limited seismic inputs (full stack amplitude data in most cases) and outputs a limited number of classes. The training data is usually limited to a single seismic cube or a few others.

[0064] The proposed solution may provide the ability to identify and de-risk the AVO anomaly at the same time. In contrast, conventional methods are focused on the prediction while the validation and de-risking is performed using conventional methods. The proposed solution may be packaged in a graphical user interface (GUI) that consumes the user’s data and provides the de-risked AVO anomalies as output.Unsupervised Learning

[0065] To account for potential hydrocarbon (HC) indicators that may not be defined in the interpretation dataset, the method uses an unsupervised learning scheme where latent geophysical variables are projected into a seismic attribute space in the form of a self-organized map (SOM). SOM clusters may be compared to known interpretation labels and can be validated through fluid substitution modeling and comparison to relevant analogs.Seismic Imaging and Processing Challenges - Mitigation

[0066] Whether the targets are above or under salt, shallow or deep, and seismic survey and processing parameters, the seismic imaging and resolution capabilities vary, and similar geological settings may exhibit different seismic signatures for different imaging capabilities. The quality of AVO conditioning applied to the data has a direct impact on the seismic signature observed. If various datasets are to be used in training, the differences in processing can impact the quality ofthe ML model. Moreover, the offset coverage of each dataset may be sufficient for detecting some AVO anomalies. Insufficient offset coverage may affect the quality of the results.

[0067] Limited seismic resolution, in comparison with well data, is another challenge. The use of other seismic attributes as inputs may enhance the predictability of the ML model in this case. Large and diverse dataset of seismic data along with corresponding ground truth labels may indicate the presence or absence of hydrocarbon anomalies. This dataset may cover a variety of geological formations, depths, noise characteristics, and seismic processing parameters to ensure robustness and generalization of the Al model. Quality control of input data plays a role to ensure the best possible performance of the ML model(s).Depositional Systems and Exploration Play Diversity + Mitigation

[0068] HC accumulations may occur in many different geological settings and pressure regimes. Different settings show different seismic responses. Like the previous point, training data may cover the possibilities so that a general model is achieved, and velocity data may provide good guidance to the ML model to distinguish the different pressure regimes.ML performance metrics

[0069] Performance metrics may be defined to evaluate the effectiveness and accuracy of the AL assisted system. These metrics may include precision, recall, Fl -score, accuracy, and area under the receiver operating characteristic curve (AUC-ROC).Useful Data

[0070] The data may be or include seismic data processed to emphasize the relevant play-type, well data (e.g., petrophysical, geological interpretation), relevant interpretations (e.g., ID, 3D labels, geological context), an adequate number of training samples (e.g., pay, no pay, etc. notionally 100+), or a combination thereof.Hardware

[0071] Deep learning models may be employed to conduct the above analysis. The hardware may include or possess access to networked standalone machines or cloud based virtual machines with large disks, sufficient RAM, powerful CPUs, and recent nVidia GPUs (architecturally Volta series,VI 00, or higher). The machines / VMs may be Linux-based. The specific choice of GPUs may be made so that the underlying drivers are compatible with recent releases of common ML frameworks such as PyTorch, TensorFlow, Keras, etc. Machines / VMs with multiple GPUs may be used to permit faster ML model training. These machines / VMs may be equipped with sufficiently large (standard) hard disk drives, with a capacity to hold several terabytes of data. The machines / VMs may also be equipped with SSD disks to allow data staging to permit faster access via memory maps. The SSD disks may be large but may not have the same capacity as the standard disk drives.Exemplary Computing System

[0072] In some embodiments, the methods of the present disclosure may be executed by a computing system. Figure 8 illustrates an example of such a computing system 800, in accordance with some embodiments. The computing system 800 may include a computer or computer system 801 A, which may be an individual computer system 801 A or an arrangement of distributed computer systems. The computer system 801 A includes one or more analysis modules 802 that are configured to perform various tasks according to some embodiments, such as one or more methods disclosed herein. To perform these various tasks, the analysis module 802 executes independently, or in coordination with, one or more processors 804, which is (or are) connected to one or more storage media 806. The processor(s) 804 is (or are) also connected to a network interface 807 to allow the computer system 801 A to communicate over a data network 809 with one or more additional computer systems and / or computing systems, such as 80 IB, 801C, and / or 80 ID (note that computer systems 80 IB, 801C and / or 80 ID may or may not share the same architecture as computer system 801 A, and may be located in different physical locations, e.g., computer systems 801 A and 801B may be located in a processing facility, while in communication with one or more computer systems such as 801 C and / or 80 ID that are located in one or more data centers, and / or located in varying countries on different continents).

[0073] A processor may include a microprocessor, microcontroller, processor module or subsystem, programmable integrated circuit, programmable gate array, or another control or computing device.

[0074] The storage media 806 may be implemented as one or more computer-readable or machine-readable storage media. Note that while in the example embodiment of Figure 8 storagemedia 806 is depicted as within computer system 801 A, in some embodiments, storage media 806 may be distributed within and / or across multiple internal and / or external enclosures of computing system 801A and / or additional computing systems. Storage media 806 may include one or more different forms of memory including semiconductor memory devices such as dynamic or static random access memories (DRAMs or SRAMs), erasable and programmable read-only memories (EPROMs), electrically erasable and programmable read-only memories (EEPROMs) and flash memories, magnetic disks such as fixed, floppy and removable disks, other magnetic media including tape, optical media such as compact disks (CDs) or digital video disks (DVDs), BLURAY® disks, or other types of optical storage, or other types of storage devices. Note that the instructions discussed above may be provided on one computer-readable or machine-readable storage medium, or may be provided on multiple computer-readable or machine-readable storage media distributed in a large system having possibly plural nodes. Such computer-readable or machine-readable storage medium or media is (are) considered to be part of an article (or article of manufacture). An article or article of manufacture may refer to any manufactured single component or multiple components. The storage medium or media may be located either in the machine running the machine-readable instructions, or located at a remote site from which machine-readable instructions may be downloaded over a network for execution.

[0075] In some embodiments, computing system 800 contains one or more hydrocarbon anomaly identification module(s) 808. In the example of computing system 800, computer system 801A includes the hydrocarbon anomaly identification module 808. In some embodiments, a single hydrocarbon anomaly identification module may be used to perform some aspects of one or more embodiments of the methods disclosed herein. In other embodiments, a plurality of hydrocarbon anomaly identification modules may be used to perform some aspects of methods herein.

[0076] It should be appreciated that computing system 800 is merely one example of a computing system, and that computing system 800 may have more or fewer components than shown, may combine additional components not depicted in the example embodiment of Figure 8, and / or computing system 800 may have a different configuration or arrangement of the components depicted in Figure 8. The various components shown in Figure 8 may be implemented in hardware, software, or a combination of both hardware and software, including one or more signal processing and / or application specific integrated circuits.

[0077] Further, the steps in the processing methods described herein may be implemented by running one or more functional modules in information processing apparatus such as general purpose processors or application specific chips, such as ASICs, FPGAs, PLDs, or other appropriate devices. These modules, combinations of these modules, and / or their combination with general hardware are included within the scope of the present disclosure.

[0078] Computational interpretations, models, and / or other interpretation aids may be refined in an iterative fashion; this concept is applicable to the methods discussed herein. This may include use of feedback loops executed on an algorithmic basis, such as at a computing device (e.g., computing system 800, Figure 8), and / or through manual control by a user who may make determinations regarding whether a given step, action, template, model, or set of curves has become sufficiently accurate for the evaluation of the subsurface three-dimensional geologic formation under consideration.

[0079] The foregoing description, for purposes of explanation, has been described with reference to specific embodiments. However, the illustrative discussions above are not intended to be exhaustive or limiting to the precise forms disclosed. Many modifications and variations are possible in view of the above teachings. Moreover, the order in which the elements of the methods described herein are illustrated and described may be re-arranged, and / or two or more elements may occur simultaneously. The embodiments were chosen and described in order to best explain the principles of the disclosure and its practical applications, to thereby enable others skilled in the art to best utilize the disclosed embodiments and various embodiments with various modifications as are suited to the particular use contemplated.

Claims

CLAIMSWhat is claimed is:

1. A method for generating a seismic profile, the method comprising: receiving input data; generating seismic features based upon the input data; extracting seismic tiles and / or sub-volumes from the seismic features; training a 2D or 3D model based upon the seismic features and the seismic tiles and / or subvolumes to produce a trained 2D or 3D model; and generating the seismic profile using the trained 2D or 3D model.

2. The method of Claim 1, wherein the input data comprises well log data, binary or multi - class success labels, and / or seismic data.

3. The method of Claim 1, further comprising training a one-dimensional (ID) model based upon the input data and the seismic features to produce a trained ID model.

4. The method of Claim 3, further comprising generating classification sub-cubes using the trained ID model, wherein classification sub-cubes comprise seismic traces around well trajectories.

5. The method of Claim 4, further comprising extracting masks and / or labels from the input data, the seismic features, and / or the classification sub-cubes, wherein the masks and / or labels correspond to the seismic tiles and / or sub-volumes.

6. The method of Claim 5, wherein the 2D or 3D model is also trained based upon the corresponding masks and / or labels.

7. The method of Claim 1, wherein the seismic tiles are two-dimensional (2D) portions of a seismic section, wherein the seismic tiles are extracted in different directions and depths, andwherein the sub-volumes are three-dimensional (3D) cuboidal chunks of a larger 3D seismic volume.

8. The method of Claim 1, wherein the trained 2D or 3D model comprises a deep learning model, and wherein the seismic features are used to guide the trained 2D or 3D model to generate geologically consistent anomalies.

9. The method of Claim 1, further comprising displaying the seismic profile.

10. The method of Claim 1, further comprising performing an action in response to the seismic profile.

11. A computing system, comprising: one or more processors; and a memory system comprising one or more non-transitory computer-readable media storing instructions that, when executed by at least one of the one or more processors, cause the computing system to perform operations, the operations comprising: receiving input data, wherein the input data comprises well log data, binary or multi-class success labels, and / or seismic data; generating seismic features based upon the input data; training a one-dimensional (ID) model based upon the input data and the seismic features to produce a trained ID model; generating classification sub-cubes using the trained ID model; extracting seismic tiles and / or sub-volumes from the input data and the seismic features; extracting masks and / or labels from the classification sub-cubes, wherein the masks and / or labels correspond to the seismic tiles and / or sub-volumes; training a 2D or 3D model based upon the seismic features, the seismic tiles and / or sub-volumes, and the corresponding masks and / or labels to produce a trained 2D or 3D model, and wherein the seismic features are used to guide the trained 2D or 3D model to generate geologically consistent anomalies; andgenerating a seismic profile using the trained 2D or 3D model.

12. The computing system of Claim 11, wherein the multi-class success labels comprise a reservoir type, a fluid content, a gas pay, a stacked oil pay, a thin oil pay, a tight oil pay, a fluid contact, wet sand, a channel margin, mud facies, wet reservoir, and / or non-reservoir.

13. The computing system of Claim 12, wherein the trained ID model is trained to predict the binary or multi-class success labels, wherein the trained ID model comprises a deep learning model, wherein the seismic features are used to guide the trained ID model to generate the geologically consistent anomalies, wherein the geologically consistent anomalies comprise a class, the gas pay, the stacked oil pay, the thin oil pay, the fluid contact, wet sand, the channel margin, and / or the mud facies, and wherein being geologically consistent indicates that multi-class labelling is performed in a context of geomorphic and lithostratigraphic interpretation.

14. The computing system of Claim 11, wherein the seismic features comprise post-stack seismic reflection data, machine learning (ML)-based seismic attributes, amplitude versus offset (AVO) attributes, energy envelopes of the seismic traces, and / or geological context attributes in relative geologic time (RGT).

15. The computing system of Claim 14, wherein the seismic profile comprises a 3D success cube, wherein the trained 2D or 3D model is applied on the seismic tiles in inline and crossline directions or applied on the sub-volumes to generate the 3D success cube, wherein the 3D success cube comprises a binary or multi-class cube, wherein the 3D success cube illustrates probabilities that a plurality of seismic pixels or voxels belong to one or more predefined classes that are derived from the machine learning (ML)-based seismic attributes and / or model features, and wherein the model features comprise output from intermediate layers of the trained 2D or 3D model.

16. A non-transitory computer-readable medium storing instructions that, when executed by one or more processors of a computing system, cause the computing system to perform operations, the operations comprising:receiving input data, wherein the input data comprises well log data, binary or multi-class success labels, and / or seismic data, and wherein the multi-class success labels comprise a reservoir type, a fluid content, a gas pay, a stacked oil pay, a thin oil pay, a tight oil pay, a fluid contact, wet sand, a channel margin, mud facies, wet reservoir, and / or non-reservoir; generating seismic features based upon the input data, wherein the seismic features comprise derivatives of the seismic data, and wherein the seismic features comprise post-stack seismic reflection data, machine learning (ML)-based seismic attributes, amplitude versus offset (AVO) attributes, energy envelopes of seismic traces, and / or geological context attributes in relative geologic time (RGT); training a one-dimensional (ID) model based upon the input data and the seismic features to produce a trained ID model, wherein the trained ID model is trained to predict the binary or multi-class success labels, wherein the trained ID model comprises a deep learning model, wherein the seismic features are used to guide the trained ID model to generate geologically consistent anomalies, wherein the geologically consistent anomalies comprise a class, the gas pay, the stacked oil pay, the thin oil pay, the fluid contact, the wet sand, the channel margin, and / or the mud facies, and wherein being geologically consistent indicates that multi-class labelling is performed in a context of geomorphic and lithostrati graphic interpretation; generating classification sub-cubes using the trained ID model, wherein the classification sub-cubes comprise the seismic traces around well trajectories; extracting seismic tiles and / or sub-volumes from the input data and the seismic features, wherein the seismic tiles are two-dimensional (2D) portions of a seismic section, wherein the seismic tiles are extracted in different directions and depths, and wherein the sub-volumes are three-dimensional (3D) cuboidal chunks of a larger 3D seismic volume; extracting masks and / or labels from the input data, the seismic features, and the classification sub-cubes, wherein the masks and / or labels correspond to the seismic tiles and / or sub-volumes; training a 2D or 3D model based upon the seismic features, the seismic tiles and / or subvolumes, and the corresponding masks and / or labels to produce a trained 2D or 3D model, wherein the trained 2D or 3D model comprises a deep learning model, and wherein the seismic features are used to guide the trained 2D or 3D model to generate the geologically consistent anomalies; andgenerating a seismic profile using the trained 2D or 3D model, wherein the seismic profile comprises a 3D success cube, wherein the trained 2D or 3D model is applied on the seismic tiles in inline and crossline directions or applied on the sub-volumes to generate the 3D success cube, wherein the 3D success cube comprises a binary or multi-class cube, wherein the 3D success cube illustrates probabilities that a plurality of seismic pixels or voxels belong to one or more predefined classes that are derived from the ML-based seismic attributes and / or model features, and wherein the model features comprise output from intermediate layers of the trained ID, 2D, and / or 3D model.

17. The non-transitory computer-readable medium of Claim 16, wherein the operations further comprise identifying the geologically consistent anomalies in or based upon the 3D success cube.

18. The non-transitory computer-readable medium of Claim 17, wherein the operations further comprise displaying the 3D success cube and the geologically consistent anomalies in or based upon the 3D success cube.

19. The non-transitory computer-readable medium of Claim 17, wherein the operations further comprise performing an action in response to the geologically consistent anomalies in or based upon the 3D success cube, and wherein the action comprises generating and / or transmitting a signal that recommends, instructs, or causes a physical action to occur at a wellsite.

20. The non-transitory computer-readable medium of Claim 19, wherein the physical action comprises selecting where to drill a wellbore, drilling the wellbore, varying a weight and / or torque on a drill bit that is drilling the wellbore, varying a drilling trajectory of the wellbore, or varying a concentration and / or flow rate of a fluid pumped into the wellbore.

Images