Real-time and autonomous rock physical stratigraphy assessment and machine learning deployment

Abstract

A computer-implemented method is described. The method includes streaming data comprising real-time acquired rock physical data associated with at least one subsurface stratum. The method includes analyzing the data stream to determine at least one model configured to evaluate at least one subsurface stratum. The method includes using the data stream as input to execute at least one model to evaluate at least one subsurface stratum. Additionally, the method includes outputting a representation of the stratum characteristics in real time.

Description

[0001] Priority Statement

[0002] This application claims priority to U.S. Patent Application No. 18 / 497,557, filed October 30, 2023, the entire contents of which are incorporated herein by reference. Technical Field

[0003] This disclosure generally relates to autonomous rock physical stratigraphy assessment and machine learning deployment. Background Technology

[0004] During drilling, data associated with subsurface formations is acquired. The format of data captured during drilling can vary based on a variety of factors. Data is often unavailable until it is manually interpreted. Partially informed or uninformed decisions are made when necessary, based on partial or no data analysis. Attached Figure Description

[0005] Figure 1 The workflow for achieving real-time and autonomous rock physical stratigraphy assessment and machine learning deployment is shown.

[0006] Figure 2 The workflow between the state storage system and the data consultation and retrieval system is shown.

[0007] Figure 3 This is a process flow diagram illustrating the process of achieving real-time and autonomous rock physical stratigraphy assessment and machine learning deployment.

[0008] Figure 4 This illustrates an oil and gas production operation that includes both one or more field operations and one or more computational operations, which exchange information and control exploration for oil and gas production.

[0009] Figure 5 This is a schematic diagram of an example controller (or control system) for enabling real-time and autonomous rock physical stratigraphy assessment and machine learning deployment. Detailed Implementation

[0010] This describes real-time and autonomous rock-physical formation assessment and machine learning deployment. The system and technology include an autonomous rock physics (AP) platform. The AP platform operates according to a workflow that analyzes rock-physical and drilling data as it is acquired. The system and technology simultaneously enable real-time and autonomous rock-physical formation assessment associated with multiple wellbores. In the example, visualizations of the wellbores are generated in real time, and users can provide feedback on any number of wells simultaneously. The system and technology autonomously detect data from new, unseen wells and automatically add these new, unseen wells to the user's available asset stack. In the example, the technology stores the acquired rock-physical and drilling data and can also provide remote access to the rock-physical and drilling data over a network. The data is transformed to obtain a representation of formation characteristics. In some examples, visualizations are automatically generated whenever updated rock-physical and drilling data are stored. Additionally, the updated data is deployed to users. Therefore, this technology enables users to share information in real time to derive formation characteristics, regardless of the format of the acquired data. This technology eliminates the costs, risks, and lost opportunities associated with waiting to perform manual analysis on the captured data.

[0011] Figure 1 A workflow 100 for achieving real-time and autonomous rock physics stratigraphy assessment and machine learning deployment is illustrated. The workflow is described based on the communication and functionality of various systems within the autonomous rock physics platform. The software, hardware, or any combination thereof that implements the autonomous rock physics platform includes, for example, information regarding... Figure 1 The described workflow and system. Table 1 lists the relevant information. Figure 1 The system is described, with exemplary software implementations of each of these components given in parentheses.

[0012]

[0013] Although Table 1 lists specific software packages, this technique is not limited to the listed software. This technique is based on... Figure 1 The workflow described is implemented using 100.

[0014] exist Figure 1 The example illustrates a Data Consultation and Extraction System (DC&ES) 102. DC&ES 102 is used to acquire rock physical data and drilling data. In some embodiments, the rock physical data and drilling data are evaluated in real time during acquisition. Rock physical data and drilling data are continuously acquired while drilling the wellbore. This system and technology are connected to a database that collects rock physical data and drilling data in real time.

[0015] In the example, the DC&ES 102 extracts data from multiple databases. The DC&ES 102 obtains data from at least one database that hosts data pushed to it in real-time, periodically, or at very short intervals. The pushed data is streaming data continuously generated by one or more data sources. In the example, rock physics data is streaming data, including continuous measurements of formation properties using electric instruments to infer properties and make decisions about drilling and production operations. Drilling data is streaming data associated with the electric instruments, tools, and other infrastructure associated with drilling operations at the corresponding wellbore.

[0016] exist Figure 1 In the example, DC&ES 102 obtains data from state storage system 120. Rock physics data and drilling data are stored in state storage system 120. Arrow 132 illustrates the data flow from periodic data consultation and extraction module 102 to state storage system 120. State storage system 120 also stores data output by data analysis system (DAS) 106. Arrow 134 illustrates the data flow from DAS 106 to state storage system 120.

[0017] In the example, the state storage system 120 is a database with a high update frequency and contains most of the data to be analyzed. This data includes actual rock physics and drilling measurements that are continuously updated as drilling operations proceed. Other databases with lower update frequencies are referenced to obtain contextual information used by the expert system (ES) 104, which provides accurate analysis parameters to the DAS 106.

[0018] In the example, the implementation of DC&ES 102 is achieved by using Application 102A, which hosts data based on the Well Site Information Transmission Standard Markup Language (WITSML) standard (e.g., WITSML version 2.0 released by Energistics in February 2017). For example, Application 102A is a Python client for the server that hosts the data.

[0019] In the example, the implementation of DC&ES 102 is achieved by using Application 102B, a relational database management system (RDBMS) built using Structured Query Language, to host the data. For example, Application 102B is a Python client for an Oracle server used to host the data. Application 102B extracts well-related metadata as contextual input to ES 104. ES 104 uses the contextual data to derive parameters for evaluating data originating from any given well.

[0020] exist Figure 1In the example, during drilling operations, as data is continuously collected, DAS 106 is used to continuously analyze rock physical data and drilling data. The analysis of this data is autonomous because the parameters used to perform the analysis are derived by ES 104. Arrow 122 illustrates the data flow from the periodic data consultation and extraction module 102 to ES 104. ES 104 constructs these parameters based on the data itself and metadata collected from several databases. In the example, the parameters derived by ES 104 appear in the form of specific predefined models corresponding to each of the DAS subsystems 126A, 126B, and 126C. DAS 106 receives these specific predefined models and executes them to autonomously assess the rock physical formation.

[0021] This workflow analyzes data at DAS 106 based on parameters exported from ES 104 and data from the periodic data consultation and extraction module 102. Arrow 124 illustrates the data flow from ES 104 to DAS 106.

[0022] DAS 106 comprises three subsystems: a stratigraphic analysis subsystem 106A, a machine learning model deployment subsystem 106B, and a dedicated computational subsystem 106C. These three subsystems are communicatively coupled, as indicated by arrows 126A, 126B, and 126C. The stratigraphic assessment subsystem 106A takes measurements (density, gamma rays, resistivity, etc.) and at least one petrophysical model as input to generate output. This system and technology automatically specify these petrophysical models (and real-time input measurements) required for the analysis. Workflow 100 includes a visualization and supervision system (VSS) 108 for continuously displaying the collected data and its associated analyses. Arrow 128 illustrates the data flow from DAS 106 to the VSS system 108. The VSS system 108 is used to collect feedback from users so that they can refine the form of parameters derived by the autonomous system. Parameters input in this manner are obtained in ES 104 and enforced in the remainder of the real-time autonomous assessment. Arrow 130 illustrates the data flow from VSS 108 to ES 104.

[0023] ES 104 provides autonomy to autonomous rock physics platforms. ES 104 enables autonomous rock physics platforms to provide rock physics solutions without human interpretation. ES 104 maps any wellbore's real-time rock physics data stream to the set of instructions required for a comprehensive assessment of the corresponding wellbore. These instructions are: (i) formation analysis model parameters (corresponding to...) Figure 1 (ii) The stratigraphic assessment subsystem 106A); (ii) The machine learning model to be deployed and its parameters (corresponding to Figure 1 (iii) machine learning deployment subsystem 106B); and dedicated computation and its parameters (corresponding to Figure 1The dedicated subject-specific computing subsystem 106C).

[0024] In some embodiments, the stratigraphic analysis at stratigraphic assessment subsystem 106A is performed by a rock physics solver (e.g., the rock physics solver shown in Table 1). In this example, the rock physics solver is an application that generates robust subsurface properties. The rock physics solver obtains raw rock physics measurements extracted by application 102A (e.g., the WITSML client (Table 1)) along with a set of physical parameters. The physical parameters provide context to the rock physics solver. The fully defined set of rock physics parameters is referred to as a rock physics model. In this example, the rock physics model is predefined using an industry-standard workflow. The rock physics model is stored (e.g., stored on disk or cloud storage). In this example, metadata associated with the physical model is stored, and this metadata includes context for each corresponding rock physics model. For example, the context describes the conditions for designing and deploying the rock physics model.

[0025] In the example, the rock physics model includes metadata describing the set of measurements it accepts as input. These measurements include, but are not limited to, density, neutron porosity, and gamma rays. The measurements depend on the specific tools used to capture downhole data at the wellbore.

[0026] In the example, the rock physics model includes metadata describing which physical method was chosen to describe each measurement. This depends not only on the technology of the specific tool used to perform the particular measurement, but also on the decision of which physical method best describes the particular strata. Examples of available physical methods include the Archie equation and its parameters, the two-water equation and its parameters, and specific neutron tool physics methods.

[0027] In the example, the rock physics model includes metadata describing the uncertainties associated with each measurement at each depth. This depends on the specific measurement and borehole conditions.

[0028] In this example, the rock physics model includes metadata describing the set of volumes it is intended to solve. Examples of volume sets include calcite, dolomite, anhydrite, water, and petroleum. This depends on the geology of the formation through which the borehole passes. Additionally, in this example, the rock physics model includes metadata describing the properties of the relevant minerals and fluids. Examples of properties include petroleum API, formation temperature, and pressure. In this example, the rock physics model also includes metadata describing the properties of the borehole environment, such as drilling mud parameters.

[0029] In the example, the rock physics model is based on data that can be analyzed by stratigraphic assessment software (e.g., The decoding format specifies a structured object for all the aforementioned parameters.

[0030] In the example, context is a structured object that summarizes metadata about the well being analyzed. Many context parameters are those directly used by the rock physics model; for example, the context includes information to identify which tools are used in the well and therefore includes which measurements the model should accept and what uncertainties should be specified for those measurements. Another example of context that directly affects the rock physics model is the mud type / parameters. This context includes what type of mud (oil-based, water-based) is used in the system and what its parameters are (density, salinity).

[0031] In the example, the context also includes parameters that must undergo indirect processes to affect the actual values ​​of the underlying rock physics model. For example, the context includes the name of the formation the borehole is currently traversing, but this information must be processed by ES 104 to associate the specific volume to be solved with that formation name. For example, the context includes information indicating the borehole location, such as drilling the Arab-D formation, but the expert subsystem converts this information into actual volumes to solve for calcite, water, oil, and porosity in this example.

[0032] In the example, the autonomous platform defines the context as the value of the parameter in Table 2.

[0033]

[0034] In the example, the context is formed by a combination of subsurface units, oil and gas types, mud types, drill bit sizes, and service companies. ES 104 derives the context from available data and ultimately determines which rock physics model to deploy based on the provided associated context.

[0035] The instruction set generated by ES 104 for a comprehensive evaluation of the corresponding wellbore also includes the machine learning model to be deployed. In the example, the autonomous platform is capable of deploying the machine learning model. In the example, the machine learning model is trained and stored on disk or in a cloud storage location before deployment. The machine learning model is associated with the following metadata: specifying the inputs executed by the machine learning model, the outputs generated by the machine learning model, and the context used to deploy the machine learning model.

[0036] In the example, the input is used to determine whether to deploy a machine learning model. For instance, when the rock physics data and drilling data include data types that correspond to the inputs of a particular machine learning model, that specific model is deployed. ES104 derives context from the available data and ultimately decides which (or which) machine learning models to deploy based on the provided context.

[0037] The instruction set generated by ES 104 for a comprehensive evaluation of the corresponding wellbore also includes dedicated calculations. The autonomous platform is capable of deploying any type of calculation or analysis model in real time. These calculations or models are predefined and stored along with the context in which they should be deployed, as described in Table 2. Specific parameters input into the dedicated calculations are also stored along with the corresponding dedicated calculation. In the example, the metadata associated with the dedicated calculation specifies how specific parameters are constructed from the raw rock physics / drilling parameters. The way these parameters are constructed varies depending on the context described in Table 2.

[0038] In some embodiments, ES 104 is a mapping tool that specifies a context based on data extracted from several databases. Based on this context, ES 104 then determines which analysis tools to deploy. This is done by selecting these analysis tools whose predefined context information is as close as possible to the currently specified context. In the example, ES 104 derives the context by comparing predefined context information with current context information extracted from rock physics data and drilling data.

[0039] In the first example, the current subsurface unit indicated by the context is extracted from a database (e.g., an enterprise database including wellbore metadata interpreted by geologists). ES 104 utilizes an aliasing system to reconcile different terms used by different disciplines or stakeholders to refer to the same area. For example, rock physics data and drilling data (e.g., flow transport data) are converted to a standardized format in real time. If ES 104 cannot extract or reconcile information, it predicts the subsurface unit currently being drilled by using a pre-trained machine learning model. This ML model uses rock physics data and well location information as input to predict the subsurface unit being drilled.

[0040] In the second example, the oil and gas associated data included in the context is extracted from a database (e.g., a corporate database) containing wellbore metadata transmitted by the asset owner before drilling operations commence at the well site. In this example, oil and gas are expected to be found in the data. In another example, more than one target oil and gas exists; in this case, the asset owner associates each expected subsurface unit with its expected oil and gas type. If ES 104 cannot extract target oil and gas information, ES predicts the target oil and gas in the wellbore using a pre-trained machine learning model. An ML model is a model that uses well location and target subsurface unit information to predict which target oil and gas is present at a particular subsurface unit.

[0041] In the third example, the data associated with mud type / mud parameters included in the context is extracted from a database (e.g., an enterprise database) that includes wellbore metadata transmitted by field mud engineers and populated into the database.

[0042] In the fourth example, ES 104 extracts drill bit size information included in the context from the WITSML database. This is reliable information that is automatically populated into the WITSML database, which provides real-time rock physics data.

[0043] In the fifth example, ES 104 extracts service company information included in the context from the WITSML database. This is reliable information that is automatically populated into the WITSML database, which provides real-time rock physics data.

[0044] ES 104 specifies the context for each drilling depth. Since any context parameter (e.g., the five context parameters listed in Table 2) can change with drilling operations, the context can vary with drilling depth. Using this context information, ES 104 specifies the specific model to be used by all available analysis tools (Table 1). In the example, ES 104 specifies a model that is as close as possible to the current context using predefined context information. If the match is not perfect, ES generates a warning to the user indicating a mismatch between the current context and the predefined context information being used. For example, the generated warning might specify a difference between the currently specified context and the context in which the model is designed for deployment.

[0045] The context and other ES specifications are displayed in system 108, which executes a web application accessible to the end user. Users with access can modify the context and specifications made by ES 104 via the web application, thereby overriding the autonomous decision-making of the ES, as indicated by arrow 130 from system 108 to expert system 104. After modification, the web application of system 108 stores these new specifications in a state storage module at state storage system 120. Users with access can change the context (e.g., the ES estimation context) (e.g., modifying the target subsurface unit) or the result of the estimated context (e.g., selecting a different rock physics model from a list of all rock physics models available to the platform). The autonomous platform will then follow the modifications made by the user to complete the remaining operations in the wellbore where these changes were submitted. Nevertheless, the ES system will continue to operate to ensure the autonomy of any parameters not included in the modifications made by the user.

[0046] The DAS 106 computes the rock physics output characterizing the wellbore. The DAS 106 receives raw rock physics data and drilling data, as well as instructions from the ES 104. The rock physics data and drilling data execute these instructions. In the example, the instructions generated by the ES 104 are specific predefined models received as input by the corresponding DAS subsystems (106A, 106B, and 106C). The DAS is integrated by the following independent but interoperable subsystems: the formation analysis subsystem 106A, the machine learning model deployment subsystem 106B, and the dedicated computation subsystem 106C.

[0047] In this example, the formation analysis subsystem is a rock physics solver. A rock physics solver is an error-minimizing algorithm capable of converting raw rock physics measurements into a representation of formation properties near the wellbore. Examples of formation properties include total porosity, water saturation, and matrix lithology. In this example, the rock physics solver is an application written in Python. Because the rock physics solver can convert any available rock physics measurements into a representation of formation properties and solve nonlinear rock physics models without making simplifying assumptions, it is commercial-grade. In some embodiments, the formation analysis subsystem 106A manages the rock physics solver within the context of an autonomous platform.

[0048] In the example, the formation analysis subsystem 106A communicates with the machine learning deployment subsystem 106B. For instance, if the rock physics model specified by the expert system (ES) requires currently unacquired rock physics measurements, the formation analysis subsystem 106A requests the machine learning deployment subsystem 106B to deploy a pre-trained, context-sensitive machine learning model capable of outputting the currently unacquired rock physics measurements (e.g., curves associated with well logging measurements, including but not limited to density, gamma ray, and resistivity logging). If available, the machine learning deployment subsystem 106B deploys the pre-trained, context-sensitive machine learning model before the formation analysis subsystem 106A performs the formation analysis, enabling the formation analysis to be completed using the relevant information.

[0049] In another example, if the machine learning model specified by the expert system (ES) 104 uses rock physical measurements that cannot be obtained from the raw measurements but are calculated by the stratigraphic analysis subsystem 106A, then the input is obtained from the stratigraphic analysis subsystem 106A. In this way, any existing machine learning model can be deployed after performing stratigraphic analysis.

[0050] In the examples, the machine learning model takes rock physics measurements (e.g., rock physics curves) as input and outputs a new curve with as many elements as the input in a depth-by-depth computation. For example, the inputs to the machine learning model are density, gamma rays, and resistivity (measurements); an example output of the machine learning model is sonic slowness (another measurement, but one not measured in this well, so it is predicted based on data from nearby wells where such measurements have been performed). In another example, the inputs to the machine learning model are density, porosity, and sonic slowness (measurements); an example output of the machine learning model is rock strength (a geomechanical parameter). In this example, the model is used to predict parameters measured in a laboratory from actual rock plug samples. The model is trained using laboratory data and then deployed downhole at scale as continuous curves.

[0051] The machine learning deployment subsystem 106B uses raw rock physical measurements, the output of specialized calculations, and / or the output of the stratigraphic analysis subsystem 106A to deploy a pre-trained machine learning model. In the example, a custom deployer tool is used to deploy the machine learning model. A custom deployer is a deployment tool for a software package used to train and deploy machine learning models that receive deep indexed data (e.g., specifically, rock physical data). The custom deployer is configured to deploy in environments where data is acquired in different formats. For example, different measurement service providers may use different names, sampling rates, and corrections for the same physical quantity being measured. The custom deployer identifies the different names, sampling rates, and corrections for the same physical quantity being measured and transforms and standardizes the data using an aliasing system and quality control steps before deployment. In the example, the data is converted to a standardized format before being input into the trained machine learning model. The trained model can be deployed based on commands.

[0052] In some embodiments, the machine learning deployment subsystem 106B deploys a set of pre-trained machine learning models according to instructions from the expert system (ES) 104. The outputs (predictions) form part of a set of stratigraphic characteristics provided by the autonomous platform system. In the example, the output of the rock physics model can be used for other machine learning models to be deployed, stratigraphic analysis, and specialized calculations. In the example, the order in which subsystems 126A, 126B, and 126C execute instructions is determined by ES 104, ensuring that all inputs / outputs are utilized efficiently.

[0053] In the example, the dedicated calculation subsystem 126C is an index of pre-established analytical calculations. This subsystem stores context triggers (Table 2) and maps them to the stored analytical calculations. For example, a context trigger could be defined as follows: Routines "A" & "B" must always be executed whenever drilling through any wellbore passing through the target subsurface unit "FOO1," regardless of the other four context parameters. Routines A & B are stored functions accessible to the subsystem. These analytical calculations can accept inputs of any form and complexity. The range of calculations can be from simple arithmetic to complex software calculations. Furthermore, a calculation can be any function that accepts input and returns an output. In the autonomous platform, dedicated calculations include, but are not limited to, geomechanics, layered shale analysis, mud logging analysis, NMR porosity analysis, NMR heavy oil analysis, automatic interpretation of density images, automatic interpretation of resistivity images, automatic bedding boundary detection, or any combination thereof. Although a specific calculation is described as a dedicated calculation, any predetermined calculation can be used. For example, a dedicated calculation can be selected based on data from a specific wellbore that allows for further analysis and also based on the specific objectives of the relevant wellbore. ES 104 uses available data and specific targets, and transforms the context (wellbore location, target geological formations, etc.) into a dedicated computational set. The system determines in real time whether the required input can be provided based on the data available during drilling.

[0054] The output of the dedicated computing subsystem 106C forms part of the formation characteristics transmitted in real time by the autonomous platform. In this example, the dedicated computing subsystem 106C is implemented using a physical model.

[0055] exist Figure 1 In the example, the State Storage System (SSS) 120 is a relational database that stores the latest snapshot of the state of the platform at any given time. Specifically, it stores: (i) data acquired by DC&ES 102; (ii) data output by DAS subsystems 106A-106C; (iii) context estimates made by ES 104 and parameters passed to the DAS subsystem; and (iv) user-defined contexts and parameters specified through a web application interface (such as...). Figure 1 (As shown by arrow 128).

[0056] Figure 2 The workflow 200 between State Storage System (SSS) 220 and Data Consultation and Extraction System (DC&ES) 202 is illustrated. In this example, DC&ES 202, Application 202A, Application 202B, Expert System 204, and State Storage System 220 are connected to… Figure 1 The DC&ES 102, Application 102A, Application 102B, Expert System 104, and State Storage System 120 are the same as or similar.

[0057] In some embodiments, DC&ES 202 frequently requests data headers present in the WITSML database. At reference numeral 242, DC&ES 202 consults the headers for the active wellbore. The headers store a description of the data present in the corresponding database. The headers are small, allowing for low latency in requests to determine the presence of data in the corresponding database, thus enabling frequent real-time queries. The header information is compared with the system status in SSS 220. At reference numeral 244, it is determined whether the headers describe data not yet stored in SSS 220. If the queried headers indicate the presence of data in the corresponding database not yet stored in SSS 220, the status system specifies the wellbore, channel, and depth range associated with the missing data, as described by the headers at reference numeral 246. Missing data may be due to data being retrieved only after the last request for header data. In response to the presence of data not yet stored in SSS 220 in the corresponding database, at reference numeral 248, a query is prepared for the actual missing data. In this example, the query includes the channel and depth range for a specific wellbore. This gives the system the characteristics of high memory efficiency, streamlined design, and rapid response. At reference numeral 250, the extracted data is appended to SSS 220 and also sent to ES 204. Once ES 204 receives the new data, the Data Analysis System (DAS) acquires it, as shown at reference numeral 252. The newly extracted data, typically with a short depth range, is analyzed, providing rapid responsiveness and formation characteristics upon receipt.

[0058] In the example, the output of DAS for this newly acquired interval is appended to SSS 220, so that SSS 220 includes the latest state of the system. For example, as Figure 1 As shown, arrow 134 indicates that the DAS output for the newly acquired interval is sent to SSS120. (See again...) Figure 1System 108 is a Visualization and Monitoring System (VSS). The fundamental objectives of VSS 108 are twofold. First, VSS 108 continuously displays the data available in SSS 120 to multiple end users via a network. VSS 108 includes dashboards and layouts that present rock physics and drilling data in a clear and easily understandable manner. These visualization tools are interactive and enable users to take quick action based on the data presented on the displays associated with VSS 108. Additionally, VSS 108 allows users to export data, enabling further analysis via external tools if necessary. Second, VSS 108 allows end users to override Expert System 104 and customize how the data is analyzed. These instructions are passed to Expert System 104 and enforced for subsequent data acquisition operations for the relevant specific wellbore. In this example, VSS 108 achieves this by creating a web application called WebAPP, which connects to the SSS relational database and displays its data. WebAPP frequently and periodically queries the SSS database to ensure that a quasi-instantaneous snapshot of the system's state is always displayed.

[0059] In the example, the real-time formation assessment results from DAS 106 are a set of curves describing the volume of the formation, and VSS visualizes these volumes. The volumes are characterized by (i) matrix composition (e.g., solid rock, including water bound to clay minerals); (ii) pore space (e.g., how much of the volume is voids between solid rock and water bound to clay minerals); and (iii) fluid composition (e.g., which fluids fill the aforementioned pore spaces). Examples of each corresponding volume category are: (i) the volume of quartz, illite, or calcite; (ii) PHIT (total porosity); and (iii) the volume of oil, natural gas, or water. These curves describing the volumes have the same length as the input curves of DAS 106 (e.g., rock physics measurements). Therefore, if measurements such as density, resistivity, and gamma rays have been acquired from depth 5,000' to depth 6,000' at any given time, and samples are taken every 1 foot, then each of these curves has 1,001 data points. When analyzed using the stratigraphic assessment subsystem 106A, it outputs curves describing the volume, with each curve also having 1,001 points. In other words, the analysis is performed depth-by-depth.

[0060] The specific volumetric curves output by the formation assessment subsystem 106A depend on the rock physics model specified for analyzing that particular formation. For example, one rock physics model might need to solve for the volumes of calcite, dolomite, porosity, water, and natural gas, while another might need to solve for the volumes of quartz, illite, porosity, water, and petroleum. In this example, porosity and water are determined by the rock physics solver.

[0061] Figure 3 This is a process flow diagram 300 illustrating the process of achieving real-time and autonomous rock physical stratigraphy assessment and machine learning deployment.

[0062] At box 302, the streaming includes real-time acquired rock physical data associated with at least one subsurface formation. In the example, real-time acquisition includes both rock physical data and drilling data associated with at least one subsurface formation.

[0063] At box 304, the flow transport data is continuously analyzed to derive models / parameters, which are then executed according to the context of the assessment to evaluate the corresponding subsurface strata.

[0064] At box 306, the model / parameters are executed in the order they appear in the context, based on the available streaming data (e.g., the machine learning deployment subsystem deploys the ML model before performing the stratigraphic analysis, so that the stratigraphic analysis can be completed).

[0065] At box 308, formation characteristics are output in real time by executing the model / parameters. In the example, real-time output of formation characteristics includes rendering formation characteristics for multiple instances of the visualization system. Additionally, in the example, formation characteristics are used for geological steering operations, formation assessment, well completion operations, drilling optimization, core / casing point selection, formation testing / sampling design, or any combination thereof. In the example, formation characteristics are used for geological steering operations, where, during drilling, the wellbore location (inclination and azimuth) is dynamically adjusted to reach one or more geological targets. These changes are based on formation characteristics determined in real time during drilling. In the example, formation characteristics can be used to guide oil and gas production operations, such as reference... Figure 4 As shown.

[0066] Figure 4 An oil and gas production operation 400 is illustrated, comprising one or more field operations 410 and one or more computational operations 412, which exchange information and control exploration for oil and gas production. In some embodiments, the outputs of the technology of this disclosure may be performed before, during, or in combination with the oil and gas production operation 400, specifically, for example, as field operation 410 or computational operation 412 or both.

[0067] Examples of field operations 410 include forming / drilling boreholes, hydraulic fracturing, production through boreholes, and injecting fluids (e.g., water) through boreholes. In some embodiments, the methods of this disclosure can trigger or control field operations 410. For example, the methods of this disclosure can generate data from hardware / software including sensors and physical data collection devices (e.g., seismic sensors, logging tools, flow meters, and temperature and pressure sensors). The methods of this disclosure can include sending data from hardware / software to field operations 410 and responsively triggering field operations 410, including, for example, generating plans and signals that provide feedback to and control the physical components of field operations 410. Alternatively or additionally, field operations 410 can trigger the methods of this disclosure. For example, implementing physical components deployed in field operations 410 (including hardware such as sensors) can generate plans and signals that can be provided as input or feedback (or both) to the methods of this disclosure.

[0068] Examples of computational operation 412 include one or more computer systems 420, each including one or more processors and a computer-readable medium (e.g., a non-transitory computer-readable medium) operatively coupled to the processors to perform computer operations to perform the methods of this disclosure. Computational operation 412 may be implemented using one or more databases 418 that store data received from field operation 410 and / or data generated within computational operation 412 (e.g., by implementing the methods of this disclosure) or both. For example, one or more computer systems 420 process inputs from field operation 410 to assess conditions in the physical world, and their outputs are stored in database 418. For example, seismic sensors of field operation 410 may be used to perform seismic surveys to map subsurface features, such as topography and faults. During a seismic survey, a source (e.g., a seismic vibrator or explosion) generates seismic waves that propagate through the Earth, and a seismic receiver (e.g., a seismic detector) measures reflections that occur when the seismic waves interact with the boundaries between layers of subsurface strata. The source signal and the received signal are provided to the computing operation 412, where they are stored in the database 418 and analyzed by one or more computer systems 420.

[0069] In some implementations, one or more outputs 422 generated by one or more computer systems 420 may be provided as feedback / input to field operation 410 (or as direct input or stored in database 418). Field operation 410 may use the feedback / input to control the physical components used to perform field operation 410 in the real world.

[0070] For example, computational operation 412 can process seismic data to generate three-dimensional (3D) maps of the subsurface formation. Computational operation 412 can use these 3D maps to provide plans for locating and drilling exploratory wells. In some operations, logging-while-drilling (LWD) technology is used to drill these wells, integrating logging tools into the drill string. LWD technology allows computational operation 412 to process new information about the formation and control drilling to adjust in real time to observed conditions.

[0071] Upon receiving information from an exploration well, one or more computer systems 420 can update a 3D map of the subsurface formation, and computational operation 412 can adjust the location of the next exploration well based on the updated 3D map. Similarly, computational operation 412 can use data received from production operations to control components of the production operation. For example, production well and pipeline data can be analyzed to predict slugs in pipelines leading to the refinery, and computational operation 412 can control upstream machinery operating valves at the refinery to reduce the likelihood of plant interruptions, which pose a risk of bringing the plant offline.

[0072] In some implementations of computational operation 412, a customized user interface can present intermediate or final results of the above process to the user. Information can be presented in one or more text, tabular, or graphical formats (e.g., via a dashboard). This information can be presented at one or more field locations (e.g., at an oil well or other facility), on the Internet (e.g., a webpage), on a mobile application (or app), or at a central processing facility.

[0073] For example, in the exploration, production, and / or testing of petrochemical processes or facilities, the presented information may include user-selectable feedback (e.g., changes in parameters or processing inputs) to improve the production environment. For instance, this feedback may include parameters that, when selected by the user, could cause changes or improvements to drilling parameters (including bit speed and direction) or the overall production of a gas or oil well. When implemented by the user, this feedback can improve the speed and accuracy of calculations, streamline processes, improve models, and address issues related to efficiency, performance, safety, reliability, cost, downtime, and the need for manual interpretation.

[0074] In some implementations, feedback can be implemented in real time, for example, to provide immediate or near-immediate changes in operation or model. The term "real-time" (or a similar term as understood by one of ordinary skill in the art) means that the action and response are close in time, such that an individual perceives the action and response as occurring substantially simultaneously. For example, the time difference between the response to the data display (or the time required to initiate the display) after an individual has performed an action to access the data can be less than 1 millisecond (ms), less than 1 second (s), or less than 5 seconds. Although the requested data does not need to be displayed (or initiated to be displayed) immediately, the data is displayed (or initiated to be displayed) without any intentional delay, taking into account the processing limitations of the described computing system and the time required for, for example, collection, precise measurement, analysis, processing, storage, or transmission.

[0075] Events may include readings or measurements captured by downhole equipment (e.g., sensors, pumps, bottomhole components, or other equipment). These readings or measurements can be analyzed at the surface, for example, using applications that may include modeling applications and machine learning. This analysis can be used to generate changes to the settings of downhole equipment (e.g., drilling rigs). In some implementations, changes in oil or gas well exploration, production / drilling, or testing can be implemented automatically (e.g., by using rules) using determined values ​​of parameters or other variables. For example, the outputs of this disclosure can be used as inputs to other equipment and / or systems at a facility. This is particularly useful for systems or various devices located meters or miles apart, or in different countries or other jurisdictions.

[0076] Figure 5 This is a schematic diagram of an example controller 500 (or control system) for enabling real-time and autonomous rock physical stratigraphy assessment and machine learning deployment. For example, controller 500 can, based on... Figure 1 Workflow 100 or Figure 2 The workflow 200 is operated. In some embodiments, the controller 500 and... Figure 4 The computer system 420 is identical or similar. The controller 500 is intended to include various forms of digital computers, such as printed circuit boards (PCBs), processors, digital circuitry, or other components for supplying interlocking alarm management systems. Additionally, the system may include portable storage media, such as a Universal Serial Bus (USB) flash drive. For example, a USB flash drive may store an operating system and other applications. The USB flash drive may include input / output components, such as a wireless transmitter or USB connector that can be plugged into a USB port of another computing device.

[0077] The controller 500 includes a processor 510, a memory 520, a storage device 530, and an input / output interface 540 communicatively coupled to an input / output device 560 (e.g., a display, keyboard, measuring device, sensor, valve, pump). Each of the components 510, 520, 530, and 540 is interconnected using a system bus 550. The processor 510 is capable of processing instructions for execution within the controller 500. The processor can be designed using any of a variety of architectures. For example, the processor 510 can be a CISC (Complex Instruction Set Computer) processor, a RISC (Reduced Instruction Set Computer) processor, or a MISC (Minimum Instruction Set Computer) processor.

[0078] In one embodiment, processor 510 is a single-threaded processor. In another embodiment, processor 510 is a multi-threaded processor. Processor 510 is capable of processing instructions stored in memory 520 or storage device 530 to display graphical information of a user interface on input / output interface 540.

[0079] Memory 520 stores information within controller 500. In one embodiment, memory 520 is a computer-readable medium. In one embodiment, memory 520 is a volatile memory cell. In another implementation, memory 520 is a non-volatile memory cell.

[0080] Storage device 530 provides mass storage for controller 500. In one embodiment, storage device 530 is a computer-readable medium. In various other embodiments, storage device 530 may be a floppy disk device, hard disk device, optical disk device, or magnetic tape device.

[0081] Input / output interface 540 provides input / output operations for controller 500. In one implementation, input / output device 560 includes a keyboard and / or pointing device. In another embodiment, input / output device 560 includes a display unit for displaying a graphical user interface.

[0082] Any number of controllers 500 may exist, associated with or outside the computer system containing controller 500, wherein each controller 500 communicates via a network. Furthermore, the terms "client," "user," and other suitable terms may be used interchangeably without departing from the scope of this disclosure. Additionally, this disclosure includes the possibility that a plurality of users may use one controller 500 and that a user may use multiple controllers 500.

[0083] According to some non-limiting embodiments or examples, a computer implementation method is provided for real-time and autonomous rock physical stratigraphy assessment and machine learning deployment, comprising: using at least one hardware processor to stream data including real-time acquired rock physical data associated with at least one subsurface stratigraphy; using at least one hardware processor to analyze the data stream to determine at least one model configured to assess at least one subsurface stratigraphy; using at least one hardware processor to execute at least one model using the data stream as input to assess at least one subsurface stratigraphy; and using at least one hardware processor to output a representation of stratigraphic characteristics in real time.

[0084] According to some non-limiting embodiments or examples, a system is provided, comprising: at least one processor; and at least one non-transitory storage medium storing instructions that, when executed by the at least one processor, cause the at least one processor to: stream data including real-time acquired rock physical data associated with at least one subsurface stratum; analyze the data stream to determine at least one model configured to evaluate at least one subsurface stratum; use the data stream as input to execute the at least one model to evaluate at least one subsurface stratum; and output a representation of stratum characteristics in real time.

[0085] According to some non-limiting embodiments or examples, a non-transitory storage medium is provided for at least one storage instruction that, when executed by at least one processor, causes at least one processor to: stream data including real-time acquired rock physical data associated with at least one subsurface stratum; analyze the data stream to determine at least one model configured to evaluate at least one subsurface stratum; use the data stream as input to execute at least one model to evaluate at least one subsurface stratum; and output a representation of stratum characteristics in real time.

[0086] The following numbered examples illustrate other non-limiting aspects or embodiments:

[0087] Example 1: A computer-implemented method for real-time and autonomous rock-physical stratigraphy assessment and machine learning deployment, comprising: using at least one hardware processor to stream data including real-time acquired rock-physical data associated with at least one subsurface stratigraphy; using at least one hardware processor to analyze the data stream to determine at least one model configured to assess at least one subsurface stratigraphy; using at least one hardware processor to execute at least one model using the data stream as input to assess at least one subsurface stratigraphy; and using at least one hardware processor to output a representation of stratigraphic characteristics in real time.

[0088] Example 2: A computer-implemented method according to Example 1, comprising: analyzing a data stream to determine a model configured to simultaneously evaluate data acquired in real time associated with multiple subsurface strata.

[0089] Example 3: The computer-implemented method according to Example 1 or 2, wherein real-time output of formation characteristics includes: rendering formation characteristics for multiple instances of the visualization system.

[0090] Example 4: The computer-implemented method according to Examples 1 to 3, wherein the data stream is converted into a standardized format in real time.

[0091] Example 5: A computer-implemented method according to Examples 1 to 4, comprising: analyzing a data stream to determine at least one model configured to evaluate at least one subsurface formation based on context extracted from rock physics data and drilling data.

[0092] Example 6: A computer-implemented method according to Examples 1 to 5, wherein at least one model is a trained machine learning model deployed based on the type of input to the trained machine learning model found in rock physics data.

[0093] Example 7: A computer-implemented method according to Examples 1 to 6, wherein at least one model is deployed using a custom deployer configured to deploy in an environment where data is obtained in different formats.

[0094] Example 8: A system comprising: at least one processor; and at least one non-transitory storage medium storing instructions that, when executed by the at least one processor, cause the at least one processor to: stream data including real-time acquired rock physical data associated with at least one subsurface stratum; analyze the data stream to determine at least one model configured to evaluate at least one subsurface stratum; use the data stream as input to execute the at least one model to evaluate at least one subsurface stratum; and output a representation of strata characteristics in real time.

[0095] Example 9: The system according to Example 8 includes: analyzing data streams to determine a model configured to simultaneously evaluate data acquired in real time associated with multiple subsurface strata.

[0096] Example 10: According to the system described in Example 8 or 9, the real-time output of formation characteristics includes: rendering formation characteristics for multiple instances of the visualization system.

[0097] Example 11: The system according to Examples 8 to 10, wherein the data stream is converted into a standardized format in real time.

[0098] Example 12: The system according to Examples 8 to 11 includes: analyzing data streams to determine at least one model configured to evaluate at least one subsurface formation based on context extracted from rock physics data and drilling data.

[0099] Example 13: The system according to Examples 8 to 12, wherein at least one model is a trained machine learning model that is deployed based on the type of input to the trained machine learning model found in rock physics data.

[0100] Example 14: The system according to Examples 8 to 13, wherein at least one model is deployed using a custom deployer configured to deploy in an environment where data is obtained in different formats.

[0101] Example 15: A non-transitory storage medium for at least one storage instruction, which, when executed by at least one processor, causes at least one processor to: stream data including real-time acquired rock physical data associated with at least one subsurface stratum; analyze the data stream to determine at least one model configured to evaluate at least one subsurface stratum; use the data stream as input to execute at least one model to evaluate at least one subsurface stratum; and output a representation of stratum characteristics in real time.

[0102] Example 16: At least one non-transitory storage medium according to Example 15, comprising: analyzing data streams to determine a model configured to simultaneously evaluate data acquired in real time associated with multiple subsurface strata.

[0103] Example 17: At least one non-transitory storage medium according to Example 15 or 16, wherein real-time output of formation characteristics includes: rendering formation characteristics for multiple instances of the visualization system.

[0104] Example 18: At least one non-transitory storage medium according to Examples 15 to 17, wherein the data stream is converted into a standardized format in real time.

[0105] Example 19: At least one non-transitory storage medium according to Examples 15 to 18, comprising: analyzing data streams to determine at least one model configured to evaluate at least one subsurface formation based on context extracted from rock physical data and drilling data.

[0106] Example 20: At least one non-transitory storage medium according to Examples 15 to 19, wherein at least one model is a trained machine learning model deployed based on the type of input to the trained machine learning model found in rock physics data.

[0107] The embodiments of the subject matter and functional operation described in this specification can be implemented in digital electronic circuits, in tangibly implemented computer software or firmware, in computer hardware, including the structures disclosed in this specification and their equivalents, or combinations thereof. The software implementation of the described subject matter can be implemented as one or more computer programs. Each computer program may include one or more computer program instruction modules encoded on a tangible, non-transitory, computer-readable computer storage medium for execution by or control of the operation of a data processing device. Alternatively or additionally, program instructions may be encoded in / on an artificially generated propagated signal. For example, this signal may be a machine-generated electrical, optical, or electromagnetic signal generated to encode information for transmission to a suitable receiver for execution by the data processing device. The computer storage medium may be a machine-readable storage device, a machine-readable storage substrate, a random or serial access storage device, or a combination of computer storage media.

[0108] The terms “data processing apparatus,” “computer,” and “electronic computer equipment” (or their equivalents as understood by one of ordinary skill in the art) refer to data processing hardware. For example, a data processing apparatus may include all kinds of means, devices, and machines for processing data, exemplarily including programmable processors, computers, or multiple processors or computers. The apparatus may also include special-purpose logic circuitry, including, for example, a central processing unit (CPU), a field-programmable gate array (FPGA), or an application-specific integrated circuit (ASIC). In some embodiments, the data processing apparatus or special-purpose logic circuitry (or a combination of data processing apparatus or special-purpose logic circuitry) may be hardware-based or software-based (or a combination of hardware-based and software-based). Optionally, the apparatus may include code that creates an execution environment for a computer program, such as code constituting a combination of processor firmware, a protocol stack, a database management system, an operating system, or an execution environment. This disclosure contemplates the use of data processing apparatuses with or without conventional operating systems (e.g., LINUX, UNIX, WINDOWS, MACOS, ANDROID, or IOS).

[0109] Computer programs (which may also be referred to or described as programs, software, software applications, modules, software modules, scripts, or code) can be written in any form of programming language. Programming languages ​​can include, for example, compiled languages, interpreted languages, declarative languages, or procedural languages. Programs can be deployed in any form, including as standalone programs, modules, components, subroutines, or units used in a computing environment. Computer programs can (but are not required to) correspond to files in a file system. Programs can be stored as part of a file that holds other programs or data (e.g., one or more scripts stored in a markup language document), as a single file dedicated to the program in question, or as multiple collaborating files that store one or more modules, subroutines, or code portions. Computer programs can be deployed to execute on a single computer or on multiple computers located at, for example, a site or distributed across multiple sites interconnected by a communication network. Although portions of the programs shown in the figures can be depicted as individual modules implementing various features and functions through various objects, methods, or processes, programs may alternatively include multiple submodules, third-party services, components, libraries, etc., where appropriate. Conversely, the features and functions of various components can be combined into a single component where appropriate. The threshold used for calculation can be determined statistically, dynamically, or both statistically and dynamically.

[0110] The methods, processes, or logic flows described in this specification can be executed by one or more programmable computers that execute one or more computer programs to perform functions by manipulating input data and generating output. These methods, processes, or logic flows can also be executed by special-purpose logic circuitry (e.g., a CPU, FPGA, or ASIC), and the apparatus can also be implemented as special-purpose logic circuitry (e.g., a CPU, FPGA, or ASIC).

[0111] A computer suitable for executing computer programs can be based on one or more general-purpose and special-purpose microprocessors, as well as other types of CPUs. The components of a computer are a CPU for executing instructions and one or more storage devices for storing instructions and data. Typically, the CPU can receive instructions and data from memory and write them to memory. A computer may also include one or more mass storage devices for storing data, or be operatively coupled to one or more mass storage devices for storing data; in some embodiments, the computer can receive data from and transfer data to mass storage devices, including, for example, magnetic disks, magneto-optical disks, or optical disks. Furthermore, the computer may be embedded in another device, such as a mobile phone, a personal digital assistant (PDA), a mobile audio or video player, a game console, a global positioning system (GPS) receiver, or a portable storage device (e.g., a universal serial bus (USB) flash drive).

[0112] Computer-readable media (appropriately temporary or non-temporary) suitable for storing computer program instructions and data can include all forms of permanent / non-permanent and volatile / non-volatile memory, media, and storage devices. Computer-readable media can include, for example, semiconductor storage devices such as random access memory (RAM), read-only memory (ROM), phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), and flash memory devices. Computer-readable media can also include, for example, magnetic devices such as magnetic tape, cassette tape, tape cartridges, and internal / removable discs. Computer-readable media can also include magneto-optical disks and optical storage devices and technologies, such as digital video discs (DVDs), CD-ROMs, DVD+ / -Rs, DVD-RAMs, DVD-ROMs, HD-DVDs, and BLURAYs. Memory can store various objects or data, including caches, classes, frames, applications, modules, backup data, jobs, web pages, web page templates, data structures, database tables, repositories, and dynamic information. The types of objects and data stored in memory can include parameters, variables, algorithms, instructions, rules, constraints, and references. Additionally, memory can include logs, policies, security or access data, and report files. Processors and memory can be supplemented by dedicated logic circuitry or integrated into dedicated logic circuitry.

[0113] Embodiments of the subject matter described in this disclosure can be implemented on a computer having a display device for providing interaction with a user, including displaying information to the user (and receiving input from the user). Types of display devices may include, for example, cathode ray tube (CRT), liquid crystal display (LCD), light-emitting diode (LED), and plasma monitors. The display device may include a keyboard and pointing devices including, for example, a mouse, a trackball, or a touchpad. User input can also be provided to the computer using a touchscreen, such as a pressure-sensitive tablet computer surface or a multi-touch screen using capacitive or inductive touchscreens. Other types of devices can be used to provide interaction with the user, including receiving user feedback, such as sensory feedback, including visual, auditory, or tactile feedback. Input from the user can be received in the form of sound, speech, or tactile input. Additionally, the computer can interact with the user by sending or receiving documents to or from the device used by the user. For example, the computer may send a webpage to a webpage in response to a request received from a webpage on a user's client device.

[0114] The term "graphical user interface" or GUI can be used in the singular or plural to describe one or more graphical user interfaces and each display of a particular graphical user interface. Therefore, a GUI can represent any graphical user interface, including but not limited to a web browser, a touchscreen, or a command-line interface (CLI) that processes information and effectively presents the results to a user. Typically, a GUI may include multiple UI elements (some or all of which are associated with a web browser), such as interactive fields, dropdown lists, and buttons. These and other UI elements may be related to or represent the functionality of a web browser.

[0115] Implementations of the subject matter described in this specification can be implemented in computing systems that include back-end components (e.g., as a data server) or middleware components (e.g., an application server). Furthermore, the computing system may include front-end components, such as a client computer having one or both a graphical user interface or a web browser, through which a user can interact with the computer. Components of the system can be interconnected via a medium of wired or wireless digital data communication (or a combination of data communication) in a communication network or any form of communication network. Examples of communication networks include local area networks (LANs), radio access networks (RANs), metropolitan area networks (MANs), wide area networks (WANs), Global Microwave Access Interoperability (WIMAX), wireless local area network (WLANs) (e.g., using 802.11a / b / g / n or 802.20 or a combination of protocols), all or part of the Internet, or any other communication system (or combination of communication networks) at one or more locations. The network can transmit, for example, Internet Protocol (IP) packets, Frame Relay frames, Asynchronous Transfer Mode (ATM) cells, voice, video, data, or combinations of communication types between network addresses.

[0116] Computer systems can include clients and servers. Clients and servers are generally geographically isolated and typically interact via a communication network. The client-server relationship can be established by computer programs running on the respective computers and having a client-server relationship with each other. A clustered file system can be any file system type that can be accessed from multiple servers for reading and updating. Because locking in a swap file system can be done at the application layer, locking or consistency tracking may not be necessary. Furthermore, Unicode data files can differ from non-Unicode data files.

[0117] While this specification contains numerous specific implementation details, these details should not be construed as limiting the scope of the claims, but rather as descriptions of features specific to particular examples. In a single implementation, certain features described in the context of a single implementation may also be combined. Conversely, various features described in the context of a single implementation may also be implemented separately or in any suitable sub-combination in multiple implementations. Furthermore, although previously described features may be described as functioning in certain combinations and even initially claimed in this way, in some cases, one or more features from the claimed combination may be removed from the combination, and the claimed combination may refer to a sub-combination or a variation of a sub-combination.

[0118] Specific embodiments of this subject matter have been described. It will be apparent to those skilled in the art that other embodiments, modifications, and substitutions of the described embodiments are within the scope of the appended claims. Although operations are described in a specific order in the drawings or claims, this should not be construed as requiring that these operations be performed in the specific order shown or in a successive order to achieve the desired result, or requiring that all of the shown operations be performed (some operations may be considered optional). In some cases, multitasking or parallel processing (or a combination of multitasking and parallel processing) may be advantageous and performed as appropriate.

Claims

1. A computer-implemented method for achieving real-time and autonomous petrophysical stratigraphy assessment and machine learning deployment, comprising: Use at least one hardware processor to stream data including real-time rock physical data associated with at least one subsurface stratum; The at least one hardware processor is used to analyze the data stream to determine at least one model configured to evaluate the at least one underground stratum; Using the at least one hardware processor, the at least one model is executed to evaluate the at least one underground stratum using the data stream as input; as well as The at least one hardware processor is used to output a representation of the formation characteristics in real time.

2. The computer-implemented method according to claim 1, comprising: The data stream is analyzed to determine the model configured to simultaneously evaluate real-time data associated with multiple subsurface strata.

3. The computer-implemented method according to claim 1, wherein, Real-time output of the formation characteristics includes rendering the formation characteristics for multiple instances of the visualization system.

4. The computer-implemented method according to claim 1, wherein, The data stream is converted into a standardized format in real time.

5. The computer-implemented method according to claim 1, comprising: The data stream is analyzed to determine at least one model configured to evaluate the at least one subsurface formation based on context extracted from the rock physics data and drilling data.

6. The computer-implemented method according to claim 1, wherein, The at least one model is a trained machine learning model, which is deployed based on the type of input to the trained machine learning model found in the rock physics data.

7. The method for implementing a computer according to claim 1, wherein, The at least one model is deployed using a custom deployer configured to deploy in environments where data is obtained in different formats.

8. A system comprising: At least one processor; as well as At least one non-transitory storage medium for storing instructions, which, when executed by the at least one processor, cause the at least one processor to: Streaming includes real-time acquisition of rock physical data associated with at least one subsurface stratum; Analyze the data stream to determine at least one model configured to evaluate the at least one underground stratum; Using the data stream as input, execute the at least one model to evaluate the at least one underground stratum; as well as It outputs a representation of formation characteristics in real time.

9. The system according to claim 8, comprising: The data stream is analyzed to determine the model configured to simultaneously evaluate real-time data associated with multiple subsurface strata.

10. The system according to claim 8, wherein, Real-time output of the formation characteristics includes rendering the formation characteristics for multiple instances of the visualization system.

11. The system according to claim 8, wherein, The data stream is converted into a standardized format in real time.

12. The system according to claim 8, comprising: The data stream is analyzed to determine at least one model configured to evaluate the at least one subsurface formation based on context extracted from the rock physics data and drilling data.

13. The system according to claim 8, wherein, The at least one model is a trained machine learning model, which is deployed based on the type of input to the trained machine learning model found in the rock physics data.

14. The system according to claim 8, wherein, The at least one model is deployed using a custom deployer configured to deploy in environments where data is obtained in different formats.

15. A non-transitory storage medium for storing at least one instruction, said instruction, when executed by at least one processor, causing said at least one processor to: Streaming includes real-time acquisition of rock physical data associated with at least one subsurface stratum; Analyze the data stream to determine at least one model configured to evaluate the at least one underground stratum; Using the data stream as input, execute the at least one model to evaluate the at least one underground stratum; as well as It outputs a representation of formation characteristics in real time.

16. The at least one non-transitory storage medium according to claim 15, comprising: The data stream is analyzed to determine the model configured to simultaneously evaluate real-time data associated with multiple subsurface strata.

17. The at least one non-transitory storage medium according to claim 15, wherein, Real-time output of the formation characteristics includes rendering the formation characteristics for multiple instances of the visualization system.

18. The at least one non-transitory storage medium according to claim 15, wherein, The data stream is converted into a standardized format in real time.

19. The at least one non-transitory storage medium according to claim 15, comprising: The data stream is analyzed to determine at least one model configured to evaluate the at least one subsurface formation based on context extracted from the rock physics data and drilling data.

20. The at least one non-transitory storage medium according to claim 15, wherein, The at least one model is a trained machine learning model, which is deployed based on the type of input to the trained machine learning model found in the rock physics data.

Images