Speeding up PVT flash calculations in basin modeling with machine learning

Abstract

Disclosed are methods and systems directed to: determining a flash model including a phase state sub-model, a K-values sub-model, and a critical point sub-model; generating, based on the phase state sub-model, the K-values sub-model, and the critical point sub-model: a first training dataset including first feature data and first label data for the phase state sub-model; a second training dataset including second feature data and second label data for the phase state sub-model; and a third training dataset including third feature data and third label data for the phase state sub-model. Furthermore, the method comprises: configuring, using the training data, the phase state sub-model, the K-values sub-model, and the critical point sub-model to generate a configured flash model; and executing, using the configured flash model, a flash computation, wherein the flash computation quantitatively and / or qualitatively specifies fluid component amount data associated with the subsurface of the resource site.

Description

Attorney Docket No.: IS24.0565-WO-PCT SPEEDING UP PVT FLASH CALCULATIONS IN BASIN MODELING WITH MACHINE LEARNING CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims priority to U.S. Provisional Patent App. No. 63 / 654,518, filed on May 31, 2024, and titled "Speeding Up PVT Flash Calculations In Basin Modeling With Machine Learning,'' which is incorporated herein by reference in its entirety for all purposes. INTRODUCTION

[0002] The disclosed methods and systems relate to optimizing flash computations during geological modeling. BACKGROUND

[0003] In computer modeling, flash computations (e.g., calculations specifying hydrocarbon component amount data) can take up a significant fraction of time, especially in geological modeling such as basin modeling or simulations.

[0004] Furthermore, modeling complex subsurface structures can be rather challenging due to hardware and / or logic constraints associated with such modeling. This is because such modeling can include: superimposing or coloring of single and / or multiphase subsurface indicators associated with flash computations in geological or basin modeling reports; critical point computations associated with such reports; and optimally using acceleration hardware for such modeling.

[0005] There is therefore a need to accelerate and / or optimize flash computations during geological modeling to quickly generate data outputs needed for energy development at a resource site. SUMMARY

[0006] Disclosed are methods, systems, and computer programs that optimize flash computations during geological modeling. According to an embodiment, a method for optimizing flash computations during geological modeling comprises determining a flash model comprising: a phase state sub-model associated with a subsurface of a resource site; a K-Attorney Docket No.: IS24.0565-WO-PCT values sub-model associated with the subsurface of the resource site; and a critical point sub- model associated with the subsurface of the resource site.

[0007] The method also comprises generating, based on the phase state sub-model, the K- values sub-model, and the critical point sub-model, training data associated with the subsurface of the resource site by applying a pressure-volume-temperature (PVT) function to the flash model thereby generating: a first training dataset including first feature data and first label data for the phase state sub-model; a second training dataset including second feature data and second label data for the phase state sub-model; and a third training dataset including third feature data and third label data for the phase state sub-model.

[0008] Furthermore, the method comprises: configuring, using the training data, the phase state sub-model, the K-values sub-model, and the critical point sub-model to generate a configured flash model; determining, at a first timepoint during the geological modeling, that a first flash computation is required for the geological modeling; and executing, using the configured flash model, the first flash computation, wherein the first flash computation quantitatively specifies fluid component amount data associated with the subsurface of the resource site.

[0009] In other embodiments, a system and a computer program can include or execute the method described above. These and other implementations may each optionally include one or more of the following features.

[0010] The PVT function may be configured to model a linear heating of a source rock associated with the subsurface of the resource site with an expulsion of hydrocarbons for different geological scenarios and subsequent generation of the training data.

[0011] According to one embodiment, the geological modeling comprises simulating or executing a plurality of flash computations including the first flash computation and a second flash computation such that the plurality of flash computations indicate hydrocarbon amount data associated with a basin in the subsurface of the resource site due to migration of hydrocarbons in the subsurface of the resource site.

[0012] Furthermore, the phase state sub-model may be configured to indicate phase-state data associated with hydrocarbons in the subsurface of the resource site.

[0013] In some cases, the phase-state data comprises one of a single-phase liquid state data, a single-phase vapor state data, or a vapor-liquid equilibrium state data.Attorney Docket No.: IS24.0565-WO-PCT

[0014] Moreover, the K-values sub-model is configured to approximately indicate equilibrium values of a vapor-liquid equilibrium state for initializing a subsequent update of the flash model for subsequent flash calculations according to some embodiments.

[0015] In addition, the critical point sub-model may be configured to indicate critical point data of one or more phase states of hydrocarbons in the subsurface of the resource site.

[0016] According to one embodiment, the first flash computation is comprised in a geological report for the resource site, the geological report being adapted for executing one or more of: well placement operations at the resource site; equipment placement operations at the resource site; locating a resource including hydrocarbons at the resource site; and configuring equipment associated with energy development at the resource site. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The disclosure is illustrated by way of example, and not by way of limitation in the figures of the accompanying drawings in which like reference numerals are used to refer to similar elements. It is emphasized that various features may not be drawn to scale and the dimensions of various features may be arbitrarily increased or reduced for clarity of discussion.

[0018] FIG. 1 shows an exemplary high-level flowchart for determining equilibrium data values for a flash model according to some embodiments of this disclosure.

[0019] FIG.2 depicts a cross-sectional view of a resource site for which the process of FIG.4 may be executed.

[0020] FIG. 3 depicts a network system comprising a communicative coupling of devices or systems associated with the resource site of FIG.2.

[0021] FIG.4 shows an exemplary detailed workflow for optimizing flash computations during geological modeling.

[0022] FIG. 5 provides an exemplary plot of hydrocarbon components data indicating fluid kinetic information following rising temperatures in a source rock for multiple scenarios with or without secondary cracking of hydrocarbon data associated with the resource site of FIG.2.

[0023] FIG.6 shows a pressure-temperature (PT) diagram indicating liquid molar fraction data for specific compositional subsurface structures at specific temperatures associated with a source rock at the resource site of FIG.2.Attorney Docket No.: IS24.0565-WO-PCT

[0024] FIGS. 7A and 7B show exemplary visualizations indicating iteration data of PT flash computations with or without vapor liquid inference.

[0025] FIG.8A shows actual quality control data used as benchmark information for assessing the predictive ability of the disclosed flash models.

[0026] FIG. 8B shows predicted results data generated by using the disclosed flash models based on three predictions and corrective iterations. DETAILED DESCRIPTION

[0027] 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 disclosed subject-matter. However, it will be apparent to one of ordinary skill in the art that the solutions disclosed 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.

[0028] The disclosed systems and methods may be accomplished using interconnected devices and systems that obtain a plurality of data associated with various parameters of interest at a resource site. The workflows / flowcharts described in this disclosure, according to some embodiments, implicate a new processing approach (e.g., hardware, special purpose processors, and specially programmed general-purpose processors) because such analyses are too complex and cannot be done by a person in the time available or at all. Thus, the described systems and methods are directed to tangible implementations or solutions to specific technological problems in developing natural resources such as oil, gas, water well industries, and other mineral exploration operations. More specifically, the systems and methods presently disclosed may be applicable to operations associated with stratigraphic analysis associated with a resource site.

[0029] Attention is now directed to methods, techniques, infrastructure, and workflows for operations that may be carried out at a resource site. Some operations in the processing procedures, methods, techniques, and workflows disclosed herein may be combined while the order of some operations may be changed. Some embodiments include an iterative refinement of one or more data models associated with the resource site via feedback loops executed byAttorney Docket No.: IS24.0565-WO-PCT one or more computing device processors and / or through other control devices / mechanisms that make determinations regarding whether a given action, template, or resource data, etc., is sufficiently accurate. Overview

[0030] FIG.1 shows an exemplary high-level flowchart 100 for determining equilibrium data values for a flash model (e.g., a flash computing model) according to some embodiments of this disclosure. The following variable definitions are provided for the various variables shown in FIG.1: • p: pressure • T: temperature • zi: first subsurface property • pci:critical pressure • Tci:critical temperature of a component, e.g., methane • ωci: acentric factor • kij: binary interaction parameters • Ki: equilibrium value of a specific component (e.g., fluid and / or gas) • yi: molar fraction of component i in vapor (e.g., gas) • xi: molar fraction of component i in liquid (e.g., oil) • nv: second subsurface property • ϕ: fugacity coefficients

[0031] At block 102, data values associated with pci, Tci, ωci, and kijare applied to estimate, at block 104, fluid equilibrium value data. It is appreciated that the data values, according to one embodiment comprise data values derived from synthetic data (e.g., synthetic data obtained or generated using a geological modeling tool) associated with a resource site and / or actual real- time or non-real-time data associated with the resource site.

[0032] At block 106, the equilibrium value data from block 104 may be analyzed to determine subsurface fluid properties or parameters including molar fraction data associated with fluid (e.g., liquid and / or gas hydrocarbon phases) in a subsurface of the resource site.

[0033] The molar fraction data may be applied, at block 108, to determine fugacity coefficients for a flash model comprising at least 3 sub-models (e.g., sub-models described herein).Attorney Docket No.: IS24.0565-WO-PCT

[0034] At block 110, the fugacity coefficients may be used to generate or otherwise create binary or non-binary interaction parameters for the flash model. The binary or non-binary interaction parameters may be tested, at block 112, to determine their ability to drive the flash model to generate accurate predictions (e.g., subsurface data predictions). If the binary model or non-binary interaction parameters are determined to enhance or otherwise make the flash model performant, then they are stored at block 114 for usage as training data for the 3 sub- models disclosed. If the binary or non-binary interaction parameters are deemed non- performant, then the workflow is fed back to block 106 in subsequent iterations for redetermination of the molar fraction data. Resource Site

[0035] FIG. 2 shows a cross-sectional view of a resource site 200 for which the process of FIG. 1 may be executed. While the illustrated resource site 200 represents a subterranean formation, the resource site, according to some embodiments, may be below water bodies such as oceans, seas, lakes, ponds, wetlands, rivers, or other marine environments.

[0036] According to one embodiment, various measurement tools capable of sensing one or more resource site data such as seismic two-way travel time, density, resistivity, production rate, etc., of a subterranean formation and / or geological formations may be provided at the resource site. As an example, wireline tools may be used to obtain measurement information related to geological attributes (e.g., geological attributes of a wellbore and / or reservoir) including geophysical and / or chemical information. For example, the chemical information may include chemical information associated with the subsurface and / or chemical information associated with the surface / above ground areas of the resource site 200.

[0037] In some embodiments, various sensors may be located at various locations around the resource site 200 to monitor and / or collect data for executing the process of FIGS.1 and 4. In other embodiments, the techniques disclosed herein may be applied to surface and / or subsurface seismic monitoring applications, surface and / or subsurface gravity applications, surface and / or subsurface electromagnetic applications, subsurface or surface ground heave applications, and surface or subsurface measurement of induced seismicity applications. According to some implementations, the disclosed techniques may be applied to remote sensing applications, subsea applications associated with permanent sensors, temporary sensor applications,Attorney Docket No.: IS24.0565-WO-PCT applications associated with remotely operated vehicles, and applications associated with aerial-based measurements.

[0038] Part, or all, of the resource site 200 may be on land, on water, or below water. In addition, while a resource site 200 is depicted, the technology described herein may be used with any combination of one or more resource sites (e.g., multiple oil fields or multiple wellsites, one or more saline aquifers, one or more depleted oil / gas fields, etc.), one or more processing facilities, etc. As can be seen in FIG. 2, the resource site 200 may have data acquisition tools 202a, 202b, 202c, and 202d positioned at various locations within the resource site 200. The subterranean structure 204 may have a plurality of geological formations 206a- 206d. As shown, this structure may have several formations or layers, including a shale layer 206a, a carbonate layer 206b, a shale layer 206c, and a sand layer 206d. A fault 207 may extend through the shale layer 206a and the carbonate layer 206b. The data acquisition tools, for example, may be adapted to take measurements and detect geophysical and / or chemical characteristics of the various formations shown.

[0039] While a specific subterranean formation with specific geological structures is depicted, it is appreciated that the resource site 200 may contain a variety of geological structures and / or formations, sometimes having extreme complexity. In some locations of a given geological structure, for example, below a water line (e.g., aquifer) relative to the given geological structure, fluid may occupy pore spaces of the formations. Each of the measurement devices may be used to measure properties of the formations and / or other geological features. While each data acquisition tool is shown as being in specific locations in FIG.2, it is appreciated that one or more types of measurement may be taken at one or more locations across one or more sources of the resource site 200 or other locations for comparison and / or analysis.

[0040] The data collected from various sources at the resource site 200 may be processed and / or evaluated and / or used as training data, and or used to generate high resolution result sets for characterizing a resource at the resource site, and / or used for generating resource models, etc. In one embodiment, the data collected by a set of sensors at the resource site may include data associated with the number of wells of a first reservoir or second reservoir at the resource site, data associated with the number of grid cells of the first or second reservoir, data associated with the average permeability of the first or second reservoir, data associated with theAttorney Docket No.: IS24.0565-WO-PCT production duration history (e.g., number of years of production) of the first reservoir or second, etc.

[0041] Data acquisition tool 202a is illustrated as a measurement truck, which may comprise devices or sensors that take measurements of the subsurface through sound vibrations such as, but not limited to, seismic measurements. Drilling tool 202b may include a downhole sensor adapted to perform logging while drilling (LWD) data collection. The wireline tool 202c may include a downhole sensor deployed in a wellbore or borehole. Production tool 202d may be deployed from a production unit or Christmas tree into a completed wellbore. Examples of resource site data that may be measured include weight on bit, torque on bit, subterranean pressures (e.g., underground fluid pressure), temperatures, flow rates, compositions, rotary speed, particle count, voltages, currents, and / or other parameters of operations as further discussed below.

[0042] Sensors may be positioned about the resource site 200 to collect data relating to various resource site operations, such as sensors deployed by the data acquisition tools 202. The sensors may include any type of sensor such as a metrology sensor (e.g., temperature, humidity), an automation enabling sensor, an operational sensor (e.g., pressure sensor, H2S sensor, thermometer, depth, tension), evaluation sensors, that can be used for acquiring data regarding the formation, wellbore, formation fluid / gas, wellbore fluid, gas / oil / water comprised in the formation / wellbore fluid, or any other suitable sensor. For example, the sensors may include accelerometers, flow rate sensors, pressure transducers, electromagnetic sensors, acoustic sensors, temperature sensors, chemical agent detection sensors, nuclear sensor, and / or any additional suitable sensors.

[0043] In one embodiment, the data captured by the one or sensors may be used to characterize, or otherwise generate one or more parameter values for a high-resolution result set used to, for example, label or configure a machine learning (ML) engine, a resource model as the case may require. In other embodiments, test data or synthetic data may also be used in developing the ML engine or resource model via one or more parameterization / labeling operations such as those discussed in association with the workflows presented herein.

[0044] Evaluation sensors may be featured in downhole tools such as tools 202b-202d and may include for instance electromagnetic, acoustic, nuclear, and optic sensors. Examples of tools including evaluation sensors that can be used in the framework of the disclosed methods andAttorney Docket No.: IS24.0565-WO-PCT systems include electromagnetic tools comprising imaging sensors such as FMITMor QuantaGeoTM(mark of SLB, Houston, TX); induction sensors such as Rt ScannerTM(mark of SLB, Houston, TX), multifrequency dielectric dispersion sensor such as Dielectric ScannerTM(mark of SLB, Houston, TX); acoustic tools including sonic sensors, such as Sonic ScannerTM(mark of SLB, Houston, TX) or ultrasonic sensors, such as pulse-echo sensor as in UBITMor PowerEchoTM(marks of SLB, Houston, TX) or flexural sensors PowerFlexTM(mark of SLB, Houston, TX); nuclear sensors such as Litho ScannerTM(mark of SLB, Houston, TX) or nuclear magnetic resonance sensors; fluid sampling tools including fluid analysis sensors such as InSitu Fluid AnalyzerTM(mark of SLB, Houston, TX); distributed sensors including fiber optic. Such evaluation sensors may be used in particular for evaluating the formation in which the well is formed (e.g., determining petrophysical or geological properties of the formation), for verifying the integrity of the well (such as casing or cement properties) and / or analyzing the produced fluid (e.g., fluid flow analysis, fluid type analysis, etc.).

[0045] As shown, data acquisition tools 202a-202d may generate data plots or measurements 208a-208d, respectively. These data plots are depicted within the resource site 200 to demonstrate that data can be generated based on some of the operations executed at the resource site 200.

[0046] Data plots 208a-208c are examples of static data plots that may be generated by data acquisition tools 202a-202c, respectively. However, it is herein contemplated that data plots 208a-208c may also be data plots that may be generated and / or updated in real time. These measurements may be analyzed to better define properties of subsurface formation(s) and / or determine the accuracy of the surface or subsurface measurements and / or check for and compensate for measurement errors associated with the surface or subsurface measurements. The plots of each of the respective measurements may be aligned and / or scaled for comparison and verification purposes. In some embodiments, base data associated with the plots may be incorporated into site planning, and / or modeling a test at the resource site 200. The respective measurements that can be taken may be any of the measurements referenced above.

[0047] Other data may also be collected, such as historical data of the resource site 200 and / or data from resource sites similar to the resource site 200, user inputs, information (e.g., economic information) associated with the resource site 200 and / or resource sites similar to the resource site 200, and / or other measurement data and other parameters of interest. Similar measurementsAttorney Docket No.: IS24.0565-WO-PCT may also be used to determine over time changes in formation information associated with the resource site.

[0048] Computer facilities such as those discussed in association with FIG. 3 may be positioned at various locations about the resource site 200 (e.g., a surface unit) and / or at remote locations. A surface unit (e.g., one or more terminals 320) may be used to communicate with the onsite tools and / or offsite operations, as well as with other surface or downhole sensors. The surface unit may be capable of sending commands to the oil field equipment / systems, and receiving data therefrom. The surface unit may also collect data generated during production operations and can produce output data, which may be stored or transmitted for further processing.

[0049] The data collected by sensors may be used alone or in combination with other data associated with the resource site 200. The data may be collected in one or more databases and / or transmitted onsite or offsite. The data may be historical data, real time data, or combinations thereof. The real time data may be used in real time, or stored for later use. The data may also be combined with historical data or other inputs for further analysis or for modeling purposes to optimize production processes at the resource site 200. In one embodiment, the data is stored in separate databases, or combined into a single database. Network System

[0050] FIG.3 shows a high-level network system 300 comprising a communicative coupling of devices or systems associated with the resource site 200 of FIG.2. The system shown in the figure may include a set of processors 302a, 302b, and 302c for executing one or more processes discussed herein. The set of processors 302 may be electrically coupled to one or more servers (e.g., computing systems) including memory 306a, 306b, and 306c that may store for example, program data, databases, and other forms of data. Each server of the one or more servers may also include one or more communication devices 308a, 308b, and 308c. The set of servers may provide a cloud-computing platform 310. In one embodiment, the set of servers includes different computing devices that are situated in different locations and may be scalable based on the needs and workflows associated with the resource site 200. The communication devices of each server may enable the servers to communicate with each other through a local or global network such as an Internet network. In some embodiments, the servers may be arranged as aAttorney Docket No.: IS24.0565-WO-PCT town 312, which may provide a private or local cloud service for users. A town may be advantageous in remote locations with poor connectivity. Additionally, a town may be beneficial in scenarios with large networks where security may be of concern. A town in such large network embodiments can facilitate implementation of a private network within such large networks. The town may interface with other towns or a larger cloud network, which may also communicate over public communication links. Note that cloud-computing platform 310 may include a private network and / or portions of public networks. In some cases, cloud-computing platform 310 may include remote storage and / or other application processing capabilities.

[0051] The network system 300 of FIG.3 may also include one or more user terminals 314a and 314b each including at least a processor to execute programs, a memory (e.g., 316a and 316b) for storing data, a communication device and one or more user interfaces and devices that enable the user to receive, view, and transmit information. In one embodiment, the user terminals 314a and 314b is a computing system having interfaces and devices including keyboards, touchscreens, display screens, speakers, microphones, a mouse, styluses, etc. The user terminals 314 may be communicatively coupled to the one or more servers of the cloud- computing platform 310. The user terminals 314 may be client terminals or expert terminals, enabling collaboration between clients and experts through the network system 300 of FIG.3.

[0052] The network system 300 of FIG.3 may also include at least one or more resource sites 200 having, for example, a set of terminals 320, each including at least a processor, a memory, and a communication device for communicating with other devices communicatively coupled to the cloud-computing platform 310. The resource site 200 may also have a set of sensors (e.g., one or more sensors described in association with FIG.2) or sensor interfaces 322a and 322b communicatively coupled to the set of terminals 320 and / or directly coupled to the cloud- computing platform 310. In some embodiments, data collected by the set of sensors / sensor interfaces 322a and 322b may be processed to generate a one or more resource models (e.g., reservoir models) or one or more resolved data sets used to generate the resource model which may be displayed on a user interface associated with the set of terminals 320, and / or displayed on user interfaces associated with the set of servers of the cloud computing platform 310, and / or displayed on user interfaces of the user terminals 314. Furthermore, various equipment / devices discussed in association with the resource site 200 may also be communicatively coupled to the set of terminals 320 and or communicatively coupled directly to the cloud-computing platformAttorney Docket No.: IS24.0565-WO-PCT 310. The equipment and sensors may also include one or more communication device(s) that may communicate with the set of terminals 320 to receive orders / instructions locally and / or remotely from the resource site 200 and also send statuses / updates to other terminals such as the user terminals 314.

[0053] The network system 300 of FIG. 3 may also include one or more client servers 324 including a processor, memory, and communication device. For communication purposes, the client servers 324 may be communicatively coupled to the cloud-computing platform 310, and / or to the user terminals 314a and 314b, and / or to the set of terminals 320 at the resource site 200 and / or to sensors at the oil field, and / or to other equipment at the resource site 200.

[0054] A processor, as discussed with reference to the system of FIG. 3, may include a microprocessor, a graphical processing unit (GPU), a microcontroller, a processor module or subsystem, a programmable integrated circuit, a programmable gate array, or another control or computing device.

[0055] The memory / storage media discussed above in association with FIG. 3 can be implemented as one or more computer-readable or machine-readable storage media that are non-transitory. In some embodiments, storage media may be distributed within and / or across multiple internal and / or external enclosures of a computing system and / or additional computing systems. Storage media 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), BluRays or any other type of optical media; or other types of storage devices. “Non-transitory” computer readable medium refers to the medium itself (i.e., tangible, not a signal) and not data storage persistency (e.g., RAM vs. ROM).

[0056] Note that instructions can be provided on one computer-readable or machine-readable storage medium, or alternatively, can be provided on multiple computer-readable or machine- readable storage media distributed in a large system having possibly plural nodes and / or non- transitory storage means. Such computer-readable or machine-readable storage medium or media is (are) considered to be part of an article (or article of manufacture). The storageAttorney Docket No.: IS24.0565-WO-PCT medium or media can be located either in a computer system running the machine-readable instructions, or located at a remote site from which machine-readable instructions can be downloaded over a network for execution.

[0057] It is appreciated that the described network system 300 of FIG. 3 is an example that may have more or fewer components than shown, may combine additional components, and / or may have a different configuration or arrangement of the components. The various components shown may be implemented in hardware, software, or a combination of both, hardware, and software, including one or more data processing and / or application-specific integrated circuits.

[0058] Further, the steps in the flowcharts described herein may be implemented by running one or more functional modules in an information processing apparatus such as general-purpose processors or application specific chips, such as ASICs, FPGAs, PLDs, GPUs or other appropriate devices associated with the network system 300 of FIG. 3. For example, the flowchart of FIG.1 as well as the flowchart of FIG. 4 below may be executed using a data engine / a data processing module (e.g., computing module) stored in memory 306a, 306b, or 306c such that the data engine / data processing module includes instructions that are executed by the one or more processors such as processors 302a, 302b, or 302c as the case may be. The various modules of FIG. 3, combinations of these modules, and / or their combination with general hardware, are included within the scope of protection of the disclosure. While one or more computing processors (e.g., processors 302a, 302b, or 302c) may be described as executing steps associated with one or more of the flowcharts described in this disclosure, the one or more computing device processors may be associated with the cloud-based computing platform 310 and may be located at one location or distributed across multiple locations. In one embodiment, the one or more computing device processors may also be associated with other systems of FIG.3 other than the cloud-computing platform 310.

[0059] In some embodiments, a computing system is provided that includes at least one processor, at least one memory, and one or more programs stored in the at least one memory, such that the programs comprise instructions, which when executed by the at least one processor, are configured to perform any method disclosed herein.

[0060] In some embodiments, a computer readable storage medium is provided, which has stored therein one or more programs, the one or more programs including instructions, which when executed by a processor, cause the processor to perform any method disclosed herein. InAttorney Docket No.: IS24.0565-WO-PCT some embodiments, a computing system is provided that includes at least one processor, at least one memory, and one or more programs stored in the at least one memory for performing any method disclosed herein. In some embodiments, an information processing apparatus for use in a computing system is provided for performing any method disclosed herein. Embodiments

[0061] The disclosed methods and systems can comprise an optimized setup of machine learning (ML) models or sub-models which help achieve significant speedups of geological modeling simulations (e.g., basin simulations) including flash calculations with negligible to no loss of accuracy when applied as disclosed herein.

[0062] According to some embodiments, flash calculations can take up to 50% or at least 60% of the simulation time of fluid flow simulators such as compositional flow simulators. In addition, flash calculations can comprise calculations that specify various amounts of components or composition data in liquid and / or vapor (e.g., sometimes called oil / gas) of a given fluid with an overall total specific composition. In particular, flash calculations can be used thermodynamically to determine liquid-vapor component data associated with fluid in a subsurface of a resource site.

[0063] On modern computers, a simple flash calculation for basin modeling without stability analysis with, for example, 4 – 10 components and no binary interaction parameters, can take up to 20 μs. When run in batches, an ML prediction (e.g., an ML inference prediction) can take less than 1 μs. Thus, in optimal cases when a flash calculation can be replaced (e.g., replaced completely) by inference predictions, a speedup of about 10 times, or 20 times, or 30 times, or 40 times can be achieved. Since ML inference predictions can be regarded as a regression-type statistic with accuracy given by availability of training data and complexity of the learning process, sufficient accuracy is often difficult to achieve in some implementations thereby making achieving significant simulation speed-ups difficult.

[0064] According to one embodiment, the disclosed methods and systems apply training of three different flash sub-models for basin modeling to implement inference or model prediction computing operations in a manner that solves problems of: phase state vapor liquid equilibrium (VLE) ratio determinations with stability analysis; and critical point calculations. It is appreciated that the three different sub-models can be independently or separately trained andAttorney Docket No.: IS24.0565-WO-PCT subsequently applied in the aforementioned prediction computing operations. These flash sub- models can be specifically optimized in accuracy and performance for geological modeling operations including basin modeling purposes thereby yielding a significant overall speed up according to some embodiments.

[0065] Furthermore, training data for the three sub-models can be created or otherwise received by generating PVT training data using, for example, kinetic functions (e.g., Petromod kinetic functions). In one embodiment, PVT indicates pressure data, volume data, and temperature data, all of which can comprise parameters that specify a thermodynamic setup. PVT can be used in a sense of PVT properties of fluids. These fluid properties can include, for example, critical temperature or critical temperatures of components, critical pressures, acentric factors of molecules, etc. For example, the kinetic functions referenced above can be used to model linear heating of a source rock with expulsion of hydrocarbons (e.g., petroleum) for different geological scenarios such that subsequent generation of PVT training labels can be attained by applying similar flash calculations which are used in a given basin simulator under consideration. According to one embodiment, the training data can be substantially created to enforce or impose accurate statistics on the three sub-models when training said sub-models.

[0066] Additionally, a prediction of VLE fractions (ln K) may also be leveraged in the disclosed methods which can be used as an improved starting point for final VLE calculation iterations. As a result, there is a trade-off between maximum possible acceleration of computing code or logic (e.g., code or logic associated with model predictions) and full accuracy of VLE calculation results. Furthermore, stability analysis and critical point calculations associated with the foregoing computer modeling can also be incorporated in some embodiments to enhance model predictions of the foregoing sub-models. It is appreciated that flash functionality in the disclosed modeling (e.g., basin modeling) simulator can be replaced by fast ML predictions yielding a speedups of almost a factor of two while including expensive stability analysis and critical point calculations, which, in some cases, do not degrade model performance in any way.

[0067] According to one embodiment, variables / parameters / settings of each of the three sub- models can be beneficially configured to perform flash calculations by defining component properties associated with a basin in a subsurface of a resource site. These variables canAttorney Docket No.: IS24.0565-WO-PCT comprise quantities that are based on a predefined or real-time kinetic fluid data associated with the subsurface at the resource site. Exemplary Flowchart

[0068] FIG. 4 shows an exemplary detailed workflow / flowchart 400 for optimizing flash computations during geological modeling. It is appreciated that a data engine stored in a memory device may cause a computer processor to execute the various processing stages of the workflow 400. For example, the disclosed techniques may be implemented as a data engine within a geological software tool such that the data engine enables optimally executing flash computations based on the processes outlined herein.

[0069] At block 402, the data engine may determine a flash model comprising: a phase state sub-model associated with a subsurface of a resource site; a K-values sub-model associated with the subsurface of the resource site; and a critical point sub-model associated with the subsurface of the resource site.

[0070] The data engine may further generate, based on the phase state sub-model, and based on the K-values sub-model, and based on the critical point sub-model, training data associated the subsurface of the resource site, as indicated at block 404, by applying a pressure-volume- temperature (PVT) function to the flash model thereby generating: a first training dataset including first feature data and first label data for the phase state sub-model; a second training dataset including second feature data and second label data for the phase state sub-model; and a third training dataset including third feature data and third label data for the phase state sub- model.

[0071] At block 406, the data engine configures, using the training data, the phase state sub- model, the K-values sub-model, and the critical point sub-model to generate a configured flash model.

[0072] Turning to block 408, the data engine may determine, at a first timepoint during the geological modeling, that a first flash computation is required for the geological modeling following which the data engine executes, at block 410, using the configured flash model, the first flash computation. The first flash computation, according to one embodiment, quantitatively and / or qualitatively specifies and / or indicates and / or characterizes and / orAttorney Docket No.: IS24.0565-WO-PCT quantifies and / or describes and / or represents fluid component amount data associated with the subsurface of the resource site.

[0073] In other embodiments, a system and a computer program can include or execute the method described above in association with FIG.4. These and other implementations may each optionally include one or more of the following features.

[0074] According to one embodiment, the PVT function may be configured to model a linear heating of a source rock associated with the subsurface of the resource site with an expulsion of hydrocarbons for different geological scenarios and subsequent generation of the training data.

[0075] According to one embodiment, the geological modeling comprises simulating or executing a plurality of flash computations including the first flash computation and a second flash computation such that the plurality of flash computations indicate hydrocarbon amount data associated with a basin in the subsurface of the resource site due to migration of hydrocarbons in the subsurface of the resource site.

[0076] Furthermore, the phase state sub-model may be configured to indicate phase-state data associated with hydrocarbons in the subsurface of the resource site.

[0077] In some cases, the phase-state data comprises one of a single-phase liquid state data, a single-phase vapor state data, or a vapor-liquid equilibrium state data.

[0078] Moreover, the K-values sub-model is configured to approximately indicate equilibrium values of a vapor-liquid equilibrium state for initializing a subsequent update of the flash model for subsequent flash calculations according to some embodiments.

[0079] In addition, the critical point sub-model may be configured to indicate critical point data of one or more phase states of hydrocarbons in the subsurface of the resource site.

[0080] According to one embodiment, the first flash computation is comprised in a geological report for, or associated with the resource site, such that the geological report can be adapted or used for executing one or more of: well placement operations at the resource site; equipment placement operations at the resource site; locating (e.g., surgically locating) a resource including hydrocarbons or nonhydrocarbons (e.g., Salar brines) at the resource site; and configuring equipment associated with energy development at the resource site.

[0081] According to one embodiment, data used to train one or more of the various sub-models referenced herein can be generated or otherwise derived synthetically using, for example, aAttorney Docket No.: IS24.0565-WO-PCT geological application. For example, the geological application may be configured to model heating of petroleum bearing source rocks in a subsurface according to a specific geological setup. For these specific petroleum results, flash calculations separating the petroleum into liquid and vapor for different or assumed pressures and temperatures can be collected or otherwise determined. In some embodiments, the aforementioned synthetic data used for training the various sub-models may be supplemented by, or substituted with actual subsurface data captured at a resource site as the case may require.

[0082] According to one embodiment: a first model (e.g., first computing model) comprised in the three sub-models is a phase-state model; a second model (e.g., second computing model) comprised in the three sub-models is a K values model or an equilibrium ln K values model; and a third model (e.g., third computing model) comprised in the three sub-models is a critical point model. It is appreciated that the three ML models can be regarded as fast and performant replacement models or surrogate models for flash calculations in during geological modeling including basin modeling. It is further appreciated that the disclosed sub-models can also replace flash calculations in subsequent basin analysis tools including geological analysis tools such as PetroFlash.

[0083] According to one embodiment of this disclosure, compositional data indicating kinetic functionality of a subsurface are provided for generating of petrochemical (e.g., petroleum) component data at a constant heating rate. FIG. 5 provides exemplary plots of hydrocarbon components data 502-508 indicating fluid kinetic information following rising temperatures in a source rock for multiple scenarios with or without secondary cracking of hydrocarbon data associated with a resource site. This can be done at multiple temperatures where three different training data sets for the three flash sub-models with features and labels are generated for the three different flash sub-models. In addition, this figure illustrates an overall fluid composition data generated during, for example, basin simulations including flash computations or calculations.

[0084] For each compositional data at a specific petroleum generation temperature, flash computations with pressure data and temperature data ranging between -20 degrees Celsius up to 580 degrees Celsius and pressure between 0 to 70 MPa can be performed to generate training data. In addition, simulations or computational tests may be executed using a model (e.g., a first flash computing model comprised in the three flash models) to determine the occurrenceAttorney Docket No.: IS24.0565-WO-PCT of specific petroleum data at a plurality of specific pressures and temperatures which might occur in a basin due to migration of fluid (e.g., petroleum) to a point in the subsurface having ambient conditions. Temperature and pressure data may be sampled at about 2, or 3, or 4-, or 5-degrees Celsius steps and at about 1, or 2, or 3 MPa steps. In some cases, an integer phase state label defining if the petroleum is in a single-phase liquid, a single-phase vapor, or in a vapor liquid equilibrium (VLE) can be determined or calculated as a label for the ML classifications indicated in FIG. 6. In particular, FIG. 6 shows a pressure-temperature (PT) diagram 600 indicating liquid molar fraction data for specific compositional subsurface structures at specific temperatures associated with a source rock. According to one embodiment, phase labels could be designated as: 0 thereby defining a liquid-like single phase state 602 of petroleum; 1 for a single-phase vapor-like state 604 of the petroleum; or 2 for the petroleum in VLE state 606 which finally is transformed into a one hot encoding for classification purposes. According to one embodiment, the one hot encoding can comprise a specific encoding of data during machine learning where a variable can have integer values of 0, 1, or 2 such that these values describe a label associated with one or more of the sub-models. For example, the integer values can be designated as: 0 for oil only characterizations; 1 for gas only characterizations; and 2 for two-phase characterizations. In some cases, the one hot encoding defines three variables where: the first variable defines whether an oil only characterization is true or false; the second variable defines whether a gas only characterization is true or false; and the third variable defines whether a two-phase characterization (e.g., gas or oil / fluid characterization) is true or false. If any of these variables are true, said variables may be interpreted as having a value of 1 and if false, said variables may be interpreted as having a value of 0. This beneficially enables the application of statistical computations including logistic regression computing operations to one or more of the sub-models disclosed.

[0085] It is appreciated that the “unphysical” separation of single-phase petroleum into liquid and vapor can constitute a technical requirement of the basin simulator and / or be associated with a mandatory user request. In addition, FIG.6 also shows a vertical line 608 indicating a type of separation of the undersaturated liquid from the under-saturated vapor under consideration. It is further appreciated that a critical point 610 can represents transition data that enable assessing or interpreting flash calculations.Attorney Docket No.: IS24.0565-WO-PCT

[0086] According to one embodiment, each generated compositional data following the same PT sampling can have equilibrium K values specified as labels for an ML regression model (e.g., a second flash model comprised in the three flash models). It is appreciated that a logarithm of K (ln K) label may be applied to one or more of the modeling processing instead of using just plain K. According to some embodiments, K or Kirepresents the equilibrium value of a specific component in a two-phase vapor / liquid (e.g., gas / fluid (petroleum)) equilibrium in a basin such that K or Ki= yi / xiwhere yiis the molar fraction of component i in vapor (e.g., gas) and xiis the molar fraction of component i in liquid (e.g., oil). This technical detail beneficially enhances accuracy as a regression of K could generate negative K prediction values which could rise during the modeling process. It is appreciated that data samples in the dataset which are not in the vapor liquid phase state may be excluded for use in training the second flash model. This is one of the distinctions of the disclosed process over other methods.

[0087] Furthermore, predicted equilibrium values generated may be used as starting values for flash iterations. In a stable subsequent iteration computing method using one or more of the three flash models, an ML inference method can be advantageous for Newton-Raphson gradient-based iteration as first problematic iterations in Newton-Raphson could be skipped and thus the performance gain of the disclosed methods could even be higher if this method is applied.

[0088] FIGS. 7A and 7B show exemplary visualizations 700a and 700b indicating iteration data of PT flash computations with or without vapor liquid inference. In particular, FIG.7A shows iteration data for PT flash computations without vapor liquid inference while FIG. 7B shows one with vapor liquid inference. As can be seen in these two figures, there is an overall speedup in iterations of almost two in average. According to one embodiment, iteration numbers at most pressure-temperature points are significantly lower in FIG.7B (e.g., improved or optimized workflow) than in FIG. 7A (e.g., an implementation without the disclosed methods and systems). This is because the overhead in tensor flow is substantially minimal (e.g., very small) in FIG.7B thereby resulting in less computation iterations that translate to a speedup in flash calculations.

[0089] In some embodiments, a third flash model comprised in the three flash models may be determined by calculating or determining a critical point for each composition and assigned as a label for the third flash model. It is appreciated that the disclosed method enhances samplingAttorney Docket No.: IS24.0565-WO-PCT of source rock temperature data by a factor of 10. The first flash phase state model (e.g., a first sub-model of the three models) may be configured to predict a non-strictly straight-line separating oil-only and gas-only areas in a PT diagram. However, with results of a critical point model, it is easy to correct this discrepancy to account for straight line scenarios.

[0090] FIG.8A shows actual quality control data or results 800a used as benchmark data that facilitates assessing the predictive ability of the disclosed flash models. FIG. 8B shows predicted data or results 800b generated using the disclosed flash models based on three predictions and corrective iterations. As seen in FIG.8B, there is a critical point 802 (e.g., a critical point indicator) which represents transition data needed to assess or otherwise interpret flash calculations that have been performed. Furthermore, the shape of the two phase “bubble” (e.g., indicators 804 and 806 representing single and / or multiphase aspects) shown in FIG.8B can sometimes be slightly deformed, and the location of the vertical black-white separation line 808 can be slightly deviated by a few degrees Celsius. These are acceptable deviations on the level of accuracy in basin modeling. It is however relevant to note that VLE results in the area of indicator 806 of FIG.8B can represent full accuracy data values without any deviations. It is appreciated that FIG.8B can be beneficially included in a geological report generated from implementing the disclosed approach.

[0091] It is appreciated that all three flash ML models can be trained with flash results including stability analysis results which can be expensive in flash calculations but comes for free or at lower computational costs using the disclosed methods and systems (e.g., the disclosed methods and systems directed to model inference computing operations). Stability analysis can be expensive in some embodiments when generating the training data. The benefit of an overall iteration speedup with a factor of almost two including expensive stability analysis and expensive critical point calculations which come almost for free based on the disclosed methods and systems and which are currently not affordable at all in some basin modeling thus making the disclosed approach a significant leap in subsurface modeling.

[0092] As previously noted, the disclosed methods and systems may be adapted for Basin modeling but could be used in a modeling tool which performs flash calculations computations for PVT analysis including extreme and non-extreme cases of reservoir simulations. While some workflows might speed up certain modeling calculations, such workflows may be very specialized, computationally expensive, restrictive in applicability, and complex to implement.Attorney Docket No.: IS24.0565-WO-PCT Not only is the disclosed approach fast, but it is also easier to implement from a hardware- software perspective.

[0093] It is appreciated that the geological report referenced above can be used for at least one of: well placement operations at the resource site; equipment placement operations at the resource site; locating (e.g., surgically locating) fluid (e.g., hydrocarbons) at the resource site; and configuring equipment associated with energy development at the resource site.

[0094] It is further appreciated that because of the incorporation of the disclosed approach in addressing issues surrounding flash computations during geological modeling, modeling complexities including superimposing or coloring of single and / or multiphase indicators associated with flash computations in geological or basin modeling reports can be resolved due to the speed up in the flash computations resulting from the use of the disclosed approach. Furthermore, critical point computations associated with the flash computations, as well as optimally using acceleration hardware for geological or basin modeling can be achieved using the disclosed methods and systems due to the use of the disclosed three sub-models comprised in the flash model.

[0095] While any discussion of, or citation to related art in this disclosure may or may not include some prior art references, such discussions are neither concessions nor acquiescence to the position that any given reference is prior art or analogous prior art.

[0096] The foregoing description, for purpose of explanation, has been described with reference to specific embodiments. However, the illustrative discussions above are not intended to be exhaustive or limited to the precise forms disclosed. Many modifications and variations are possible in view of the above teachings. The embodiments were chosen and described in order to explain the disclosed principles and their practical applications, to enable others skilled in the art to use various embodiments with various modifications as are suited to the particular use contemplated.

[0097] It is appreciated that the term optimize / optimal and its variants (e.g., efficient or optimally) may simply indicate improving, rather than the ultimate form of 'perfection' or the like.

[0098] It is further appreciated that, although the terms first, second, etc., may be used herein to describe various elements. These elements should not be limited by these terms. Rather, these terms are used to distinguish one element from another. For example, a first object orAttorney Docket No.: IS24.0565-WO-PCT 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. The first object or step, and the second object or step, are both objects or steps, respectively, but are not to be considered the same object or step.

[0099] 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 the description and the appended claims, the singular forms “a,” “an” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It is appreciated that the term “and / or” as used herein refers to, and encompasses any possible combination of one or more of the associated listed items. It is further appreciated 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.

[0100] 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.

[0101] Those with skill in the art will appreciate that while some terms in this disclosure may refer to absolutes (e.g., all source receiver traces, each of a plurality of objects, etc.), the methods and techniques disclosed herein may also be performed on fewer than all of a given thing (e.g., performed on one or more components and / or performed on one or more source receiver traces, etc.). Accordingly, in instances in the disclosure where an absolute is used, the disclosure may also be interpreted to be referring to a subset.

Claims

Attorney Docket No.: IS24.0565-WO-PCT CLAIMS What is claimed is:

1. A method for optimizing flash computations during geological modeling, the method comprising: determining a flash model comprising: a phase state sub-model associated with a subsurface of a resource site, a K-values sub-model associated with the subsurface of the resource site, and a critical point sub-model associated with the subsurface of the resource site; generating, based on the phase state sub-model, the K-values sub-model, and the critical point sub-model, training data associated with the subsurface of the resource site by applying a pressure-volume-temperature (PVT) function to the flash model thereby generating: a first training dataset including first feature data and first label data for the phase state sub-model, a second training dataset including second feature data and second label data for the phase state sub-model, and a third training dataset including third feature data and third label data for the phase state sub-model; configuring, using the training data, the phase state sub-model, the K-values sub-model, and the critical point sub-model to generate a configured flash model; determining, at a first timepoint during the geological modeling, that a first flash computation is required for the geological modeling; and executing, using the configured flash model, the first flash computation, wherein the first flash computation quantitatively specifies fluid component amount data associated with the subsurface of the resource site.

2. The method of Claim 1, wherein the PVT function is configured to model a linear heating of a source rock associated with the subsurface of the resource site with an expulsion of hydrocarbons for different geological scenarios and subsequent generation of the training data.

3. The method of Claim 1, wherein:Attorney Docket No.: IS24.0565-WO-PCT the geological modeling comprises simulating or executing a plurality of flash computations including the first flash computation and a second flash computation; and the plurality of flash computations indicate hydrocarbon amount data associated with a basin in the subsurface of the resource site due to migration of hydrocarbons in the subsurface of the resource site.

4. The method of Claim 1, wherein the phase state sub-model is configured to indicate phase-state data associated with hydrocarbons in the subsurface of the resource site.

5. The method of Claim 4, wherein the phase-state data comprises one of a single-phase liquid state data, a single-phase vapor state data, or a vapor-liquid equilibrium state data.

6. The method of Claim 1, wherein the K-values sub-model is configured to approximately indicate equilibrium values of a vapor-liquid equilibrium state for initializing a subsequent update of the flash model for subsequent flash calculations.

7. The method of Claim 1, wherein the critical point sub-model is configured to indicate critical point data of one or more phase states of hydrocarbons in the subsurface of the resource site.

8. The method of Claim 1, wherein the first flash computation is comprised in a geological report for the resource site, the geological report being adapted for executing one or more of: well placement operations at the resource site; equipment placement operations at the resource site; locating a resource including hydrocarbons at the resource site; and configuring equipment associated with energy development at the resource site.

9. A system for optimizing flash computations during geological modeling, the system comprising: a computer processor, and a memory storing a data processing engine that comprises instructions which are executable by the computer processor to: determine a flash model comprising:Attorney Docket No.: IS24.0565-WO-PCT a phase state sub-model associated with a subsurface of a resource site, a K-values sub-model associated with the subsurface of the resource site, and a critical point sub-model associated with the subsurface of the resource site; generate, based on the phase state sub-model, the K-values sub-model, and the critical sub-model, training data associated the subsurface of the resource site by applying a pressure-volume-temperature (PVT) function to the flash model thereby generating: a first training dataset including first feature data and first label data for the phase state sub-model, a second training dataset including second feature data and second label data for the phase state sub-model, a third training dataset including third feature data and third label data for the phase state sub-model; configure, using the training data, the phase state sub-model, the K-values sub- model, and the critical point sub-model to generate a configured flash model; determine, at a first timepoint during the geological modeling, that a first flash computation is required for the geological modeling; and execute, using the configured flash model, the first flash computation, wherein the first flash computation quantitatively specifies fluid component amount data associated with the subsurface of the resource site.

10. The system of Claim 9, wherein the PVT function is configured to model a linear heating of a source rock associated with the subsurface of the resource site with an expulsion of hydrocarbons for different geological scenarios and subsequent generation of the training data.

11. The system of Claim 9, wherein: the geological modeling comprises simulating or executing a plurality of flash computations including the first flash computation and a second flash computation; andAttorney Docket No.: IS24.0565-WO-PCT the plurality of flash computations indicate hydrocarbon amount data associated with a basin in the subsurface of the resource site due to migration of hydrocarbons in the subsurface of the resource site.

12. The system of Claim 9, wherein the phase state sub-model is configured to indicate phase-state data associated with hydrocarbons in the subsurface of the resource site.

13. The system of Claim 9, wherein the K-values sub-model is configured to approximately indicate equilibrium values of a vapor-liquid equilibrium state for initializing a subsequent update of the flash model for subsequent flash calculations.

14. The system of Claim 9, wherein the critical point sub-model is configured to indicate critical point data of one or more phase states of hydrocarbons in the subsurface of the resource site.

15. The method of Claim 1, wherein the first flash computation is comprised in a geological report for the resource site, the geological report being adapted for executing one or more of: well placement operations at the resource site; equipment placement operations at the resource site; locating a resource including hydrocarbons at the resource site; and configuring equipment associated with energy development at the resource site.

16. A computer program for optimizing flash computations during geological modeling, the computer program including a non-transitory computer-readable medium comprising code configured to: determine a flash model comprising: a phase state sub-model associated with a subsurface of a resource site, a K-values sub-model associated with the subsurface of the resource site, and a critical point sub-model associated with the subsurface of the resource site; generate, based on the phase state sub-model, the K-values sub-model, and the critical sub-model, training data associated the subsurface of the resource site by applying a pressure- volume-temperature (PVT) function to the flash model thereby generating:Attorney Docket No.: IS24.0565-WO-PCT a first training dataset including first feature data and first label data for the phase state sub-model, a second training dataset including second feature data and second label data for the phase state sub-model, a third training dataset including third feature data and third label data for the phase state sub-model; configure, using the training data, the phase state sub-model, the K-values sub-model, and the critical point sub-model to generate a configured flash model; determine, at a first timepoint during the geological modeling, that a first flash computation is required for the geological modeling; and execute, using the configured flash model, the first flash computation, wherein the first flash computation quantitatively specifies fluid component amount data associated with the subsurface of the resource site.

17. The computer program of Claim 16, wherein the PVT function is configured to model a linear heating of a source rock associated with the subsurface of the resource site with an expulsion of hydrocarbons for different geological scenarios and subsequent generation of the training data.

18. The computer program of Claim 16, wherein the phase state sub-model is configured to indicate phase-state data associated with hydrocarbons in the subsurface of the resource site.

19. The computer program of Claim 16, wherein the K-values sub-model is configured to approximately indicate equilibrium values of a vapor-liquid equilibrium state for initializing a subsequent update of the flash model for subsequent flash calculations.

20. The computer program of Claim 16, wherein the critical point sub-model is configured to indicate critical point data of one or more phase states of hydrocarbons in the subsurface of the resource site.

Images