Reservoir property estimation

MYPI2026000095A0Undetermined Publication Date: 2025-01-07SCHLUMBERGER TECHNOLOGY BV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
MYPI2026000095
Authority / Receiving Office
MY · MY
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-07-07
Filing Date
2024-07-05
Publication Date
2025-01-07

Smart Images

  • Figure PI2026000095_DRAWING_001
    Figure PI2026000095_DRAWING_001
  • Figure PI2026000095_DRAWING_002
    Figure PI2026000095_DRAWING_002
  • Figure PI2026000095_DRAWING_003
    Figure PI2026000095_DRAWING_003
Patent Text Reader

Abstract

Techniques and systems for reservoir characterization. One embodiment includes receiving first raw data corresponding to at least one attribute of a well, applying a pre processing operation to correct at least one error in the first raw data to generate prepared data, transmitting the prepared data to a machine learning system; training the machine learning system into a trained machine learning system using the prepared data, and generating a final model via the trained machine learning system, wherein the final model operates to generate a characterization of a reservoir in a subsurface region of Earth when second raw data is input to the final model. (Fig. 2)
Need to check novelty before this filing date? Find Prior Art

Description

RESERVOIR PROPERTY ESTIMATIONCROSS REFERENCE PARAGRAPH

[0001] This application claims the benefit of U.S. Provisional Application No. 63 / 512,351, entitled "RESERVOIR PROPERTY ESTIMATION" filed July 07, 2023, the disclosure of which is hereby incorporated herein by reference.BACKGROUND

[0002] The subject matter disclosed herein relates to systems and methods of fluid saturation predictions and reservoir properties predictions.

[0003] Well logging tools, such as downhole tools, are utilized to measure well properties for well evaluation. However, some harsh environments are ill suited to the use of well logging tools. For example, Ultra-high temperature (uHT) wells expose well logging tools to temperatures that can exceed 175°C. When exposed to certain harsh environments, including uHT wells, fail rates of well logging tool increase. However, as there is a need for determining and / or predicting well properties, reservoir properties, alternative solutions to direct measurement via downhole tools, such as surface logging tools, would be useful, especially in situations in which typical well logging tools are not available and / or have limited usage.SUMMARY

[0004] A summary of certain embodiments disclosed herein is set forth below. It should be understood that these aspects are presented merely to provide the reader with a brief summary of these certain embodiments and that these aspects are not intended to limit the scope of this disclosure. Indeed, this disclosure may encompass a variety of aspects that may not be set forth below.

[0005] There are limited tools and techniques to log wells exposed to certain environmental factors, such as Ultra-high temperature wells (uHT), where, for example, tool failure rates are high. Present embodiments include systems and methods which utilize machine learning in conjunction with surface logging and drilled formation gas data to estimate fluid saturation which is utilized to provide reserve estimation (e g., reserve volume, productivity). In some embodiments, the techniques derive porosity within acceptable accuracy from surface drill cutting, formation gas and drilling mechanics data and combines derived fluid saturation data with derived porosity data to estimate reserve and reservoir properties. The solution proposed can be utilized in all drilling conditions to add another dimension to determining reservoir properties.BRIEF DESCRIPTION OF THE DRAWINGS

[0006] Various aspects of this disclosure may be better understood upon reading the following detailed description and upon reference to the drawings in which:

[0007] FIG. 1 depicts an example wellsite system for measuring borehole data using various downhole tools and surface tools, in accordance with embodiments of the present disclosure;

[0008] FIG. 2 depicts a well control system configured to control the wellsite system of FIG. 1, in accordance with embodiments of the present disclosure;

[0009] FIG. 3 depicts a flow chart illustrating use of a machine learning model to generate a final model used in reservoir prediction, in accordance with embodiments of the present disclosure; and

[0010] FIG. 4 depicts a second flow chart illustrating application of the final model of FIG. 3 in conjunction with reservoir prediction, in accordance with embodiments of the present disclosure.DETAILED DESCRIPTION OF SPECIFIC EMBODIMENTS

[0011] One or more specific embodiments will be described below. In an effort to provide a concise description of these embodiments, not all features of an actual implementation are described in the specification. It should be appreciated that in the development of any such actual implementation, as in any engineering or design project, numerous implementation-specific decisions must be made to achieve the developers’ specific goals, such as compliance with system -related and business-related constraints, which may vary from one implementation to another. Moreover, it should be appreciated that such a development effort might be complex and time consuming, but would nevertheless be a routine undertaking of design, fabrication, and manufacture for those of ordinary skill having the benefit of this disclosure.

[0012] In some well environments (e.g., Ultra-high temperature (uHT) wells), logging while drilling data acquisition is difficult because traditional logging devices will not function under harsh conditions (e.g., in harsh environments). Alternative logging techniques and devices, for example, wireline logging, may be utilized in these circumstances. However, the data collected may not be as reliable, borehole condition deteriorates over time and drilling fluid invasion into formation adulterates the logging responses, at the same time fail rates of getting logging tool all the way to the bottom of the borehole can be high, and greater cost is imposed when these alternative logging techniques and devices are utilized. Accordingly, in some embodiments, present systems and techniques operate to provide logging of a well utilizing a model that can be applied to a well of interest. This model may initially be generated using, for example, previously collected data samples from wells similar to the well of interest (i.e., the target well), whereby the data samples may be processed utilizing, for example, machine learning to generate the model. Particular inputs (e.g., surface data, such as drill cutting, and drill formation gas together with, for example, available surface or downhole gamma ray and / or surface or downhole drilling mechanics data) can be provided to generate the model to predict reservoir properties such as porosity, saturation, fraction of shale, and permeability using machine learning.

[0013] With the foregoing in mind, FIG. 1 illustrates an example wellsite system 10 for measuring borehole data using various downhole tools (e.g., logging tools) and surface tools according to one or more aspects of the present disclosure. Surface acquisition systems 12 are located on a wellsite surface 16 above a geological formation 14 (e.g., a subsurface formation) into which multiple wellbores 18A-18C extend from the wellsite surface 16. The wellbores 18A and 18B are located within offset wells 20A and 20B, respectively. The wellbore 18C is located within a subject well 20C. Each of the offset wells 20A and 20B are located at a site having an offset distance with respect to a site of the subject well 20C.

[0014] An offset well (e.g., offset well 20A or 20B) may include an existing wellbore (e.g., 18A or 18B) that may be used as a guide for planning a well (e.g., subject well 20C) and / or well performance benchmarking. In some cases, offset well data may be combined with seismic data and other relevant information (e.g., local geological surveys, prior experience). In some cases, offset well data may be limited (e.g., due to competition between different oil and gas operators).

[0015] Downhole equipment 42 may be deployed in the wellbores 18A and 18B to acquire various borehole data (e.g., well logs). The downhole equipment 42 may include one or more logging tools 22 that may acquire information associated with the subsurface formations (e.g., rock formations) surrounding the wellbores 18A and 18B. Based on acquired information, certain properties (e.g., lithology) of the subsurface formations may be interpreted by users (e.g., geoscientists) to facilitate decision making related to a well development process (e.g., well construction of the subject well 20C).

[0016] For example, the oil and gas industry often uses wireline logging to obtain a continuous record of physical properties of a subsurface formation (e.g. rock formation). The wireline logging (e g., collected using a wireline logging tool as the one or more logging tools 22) may include measurements and analysis of geophysical data performed as a function of wellbore (e.g., wellbores 18A and 19B) depth. The measurements and the associated analysis may be used to infer further properties of the subsurface formation, such as hydrocarbon saturation and formation pressure, thereby facilitating decisionmaking for further drillings and productions to extract natural resources (e.g., hydrocarbon resources) in close proximity to the subsurface formation.

[0017] The measurements may be recorded either at the wellsite surface 16 or in a wellbore (e.g., wellbore 18A or 18B) in an electronic data format (e.g., well log) and can be provided to users. Well logging may be performed during a drilling process (e.g., Measurements While Drilling (MWD), Logging While Drilling (LWD)) to provide realtime information about the subsurface formations being penetrated by the wellbore (e.g., wellbores 18A and 18B), or after a well reaches a target such that the whole depth of the wellbore (e.g., wellbores 18A and 18B) may be logged. Logs collected during the process of MWD may include gamma-ray logs (e.g., collected using a gamma-ray logging tool as the one or more logging tools 22). The gamma-ray logs may help drillers as well as geoscientists to infer a distribution of a formation of rocks through which the drilling is conducted. One process of collecting gamma-ray values may include a tool deployed within a Bottom Hole Assembly (BHA), or measured at the surface using drilled cuttings.

[0018] Thus, the logging tools 22 may include gamma-ray logging tools that measure naturally occurring gamma radiation that may characterize rocks or sediments associated with the wellbores 18A-18C. The naturally occurring gamma radiation may be emitted primarily from potassium, thorium, and uranium in the structure of clay minerals, radioactive salts in the formation waters, radioactive salts bound to the charged surfaces of clay minerals, potassium, thorium, and uranium associated with feldspars, and radioactive minerals associated with igneous rocks and rock fragments. The gamma-ray response is used for correlation of formations between wells and for estimating volume shale and / or volume clay minerals.

[0019] Logging tools 22 may be run downhole on wirelines 24 into the wellbores 18A and 18B respectively by the surface acquisition systems 12. The logging tools 22 may include any suitable measurement devices (e.g., sensors, meters) capable of acquiring borehole data including measurements of various properties (e.g., velocities, porosity, resistivity, natural gamma-ray, electrical, acoustic, stimulated radioactive responses, electromagnetic, nuclear magnetic resonance, pressure, and so forth) associated with the geological formation 14 and contained fluids. In certain embodiments, the logging tools22 may include data processing components to perform certain pre-processing tasks. In certain embodiments, certain borehole data (e.g., measured log data) from the offset wells 20A and 20B may be used to generate (e.g., using model-based data simulation and prediction) new data (e.g., synthetic log data) for the subject well 20C where no logging tools are deployed.

[0020] Each surface acquisition system 12 may, for example, include a logging system 30 and a deploying system 32, such as a drilling rig, workover rig, platform, derrick, and / or other surface structures. The borehole data (e.g., log data) related to the geological formation 14 surrounding the offset wells 20A and 20B is gathered by the logging tools 22 and transmitted to the logging system 30 via the wirelines 24 and cables 34. Each logging system 30 may include surface equipment 50 configured to collect, store, and / or pre- process the borehole data. Each logging system 30 may communicate with a processing system 56 using certain communication components (e.g., routers, transmitters, and so forth) via data communication lines 52 or wireless connections. The processing system 56 may perform data processing and analysis based on the borehole data and other reference data (e.g., seismic data). Additional details with regard to acquiring the borehole data using the downhole equipment 42, surface equipment 50, and processing system 56 will be discussed below with reference to FIG. 2.

[0021] In some embodiments, the surface equipment 50 may include additional systems and devices, for example, an injector for injecting drilling fluid that is circulated in an internal space of a drill string in, for example, wellbore 18A or 18B. Additionally, the surface equipment can include, for example, a shale shaker that operates to collect drilling fluid flowing out from a discharge pipe at the at the wellsite surface 16 from, for example, wellbore 18A or 18B surface. The drilling fluid is charged with drilling residues, known as cuttings. The shale shaker can include, for example, a sieve used to separate the solid drilling cuttings, e.g., rock samples, from the drilling mud. The rock samples obtained may be prepared and imaged so as to be used to evaluate the geologic formation and characterize one or several of its properties, such as its mineralogy, lithology, porosity, density, etc.

[0022] Imaging of the drilling cuttings may be accomplished via, for example, an imaging device, such as any type of optical or electronic microscope, camera, etc. The images obtained by the imaging device may be digital images, which, in some embodiments, can be automatically analyzed. Additionally, as noted above, the shale shaker can separate the solid drilling cuttings from the drilling mud. In some embodiments, the mud gas extracted at the shaker and, for example, at an active pit is transported to a mud logging unit as a wellsite gas analyzer that can operate to measure, for example, continuous total gas, alkanes (e.g., Cl to C5), and alkenes (e.g., ethene and propene), etc. to assist drillers and geologists with safety and formation evaluation. In some embodiments, the cutting image data and / or the data generated by the mud logging unit can be transmitted to, for example, a surface logging system 55. The surface logging system 55 may represent one or more systems to capture or receive one or more of drilling data and drill cuttings formation gas data (e.g., mud gas and drilled cuttings information). This information may additionally be transmitted to the processing system 56, as additionally discussed below with reference to FIG. 2.

[0023] FIG. 2 illustrates a well control system 58 (e.g., that includes the processing system 56) configured to control the wellsite system 10 of FIG. 1. In certain embodiments, the processing system 56 may include one or more analysis modules 60 (e.g., a program of computer-executable instructions and associated data) that may be configured to perform various functions of the embodiments described herein. In certain embodiments, to perform these various functions, the one or more analysis modules 60 may execute on one or more processors 62 of the processing system 56, which may be connected to one or more storage media 64 of the processing system 56. Indeed, in certain embodiments, the one or more analysis modules 60 may be stored in the one or more storage media 64.

[0024] In certain embodiments, the computer-executable instructions of the one or more analysis modules 60, when executed by the one or more processors 62, may cause the one or more processors 62 to generate one or more models (e.g., forward model, inverse model, mechanical model, and so forth). Such models may be used by the processing system 56 to predict values of operational parameters that may or may not be measured (e.g., using gauges, sensors, and so forth) during well operations.

[0025] In certain embodiments, the one or more processors 62 may include a microprocessor, a microcontroller, a processor module or subsystem, a programmable integrated circuit, a programmable gate array, a digital signal processor (DSP), or another control or computing device. In certain embodiments, the one or more processors 62 may include machine learning and / or artificial intelligence (Al) based processors. In certain embodiments, the one or more storage media 64 may be implemented as one or more non- transitory computer-readable or machine-readable storage media. In certain embodiments, the one or more storage media 64 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); or other types of storage devices. Note that the computer-executable instructions and associated data of the analysis module(s) 60 may be provided on one computer- readable or machine-readable storage medium of the storage media 64, or alternatively, may be provided on multiple computer-readable or machine-readable storage media distributed in a large system having possibly plural nodes. Such computer-readable or machine-readable storage medium or media are considered to be part of an article (or article of manufacture), which may refer to any manufactured single component or multiple components. In certain embodiments, the one or more storage media 64 may be located either in the machine running the machine-readable instructions or may be located at a remote site from which machine-readable instructions may be downloaded over a network for execution.

[0026] In certain embodiments, the processor(s) 62 may be connected to a network interface 66 of the processing system 56 to allow the processing system 56 to communicate with multiple downhole sensors 54 and surface sensors 68, as well as communicate with actuators 70 and / or programmable logic controllers (PLCs) 72 of the surface equipment 50, the surface logging system 55, and of the downhole equipment 42 of a Bottom Hole Assembly (BHA). In certain embodiments, the network interface 66 may also facilitate theprocessing system 56 to communicate data to cloud computing resources 74, which may in turn communicate with external computing systems 76 to access and / or to remotely interact with the processing system 56.

[0027] It should be appreciated that the well control system 58 illustrated in FIG. 2 is only one example of a well control system, and that the well control system 58 may have more or fewer components than shown, may combine additional components not depicted in the embodiment of FIG. 2, and / or the well control system 58 may have a different configuration or arrangement of the components depicted in FIG. 2. In addition, the various components illustrated in FIG. 2 may be implemented in hardware, software, or a combination of both hardware and software, including one or more signal processing and / or application specific integrated circuits. Furthermore, the operations of the well control system 58 as described herein may be implemented by running one or more functional modules in an information processing apparatus such as application specific chips, such as application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), programmable logic devices (PLDs), systems on a chip (SOCs), or other appropriate devices. These modules, combinations of these modules, and / or their combination with hardware are all included within the scope of the embodiments described herein.

[0028] In some embodiments, the well control system 58 and, more specifically, the processing system utilizes machine and deep learning to integrate surface data (e.g., drilling, drill cutting, drill formation gas) together with available gamma ray with or without downhole drilling mechanics data to predict reservoir properties such as porosity, saturation, fraction of shale, and permeability. The gamma ray and drilling mechanics data can be acquired via downhole (as described in conjunction with FIG. 1) and / or surface mud logging while drilling (e.g., by using techniques such as X-ray fluorescence). Offset wells (e.g., the offset wells 20A and 20B) from a selected basin are sorted, and machine learning and deep learning is applied to the data to allow prediction of the reservoir properties. A model generated is then applied to new wells to generate reservoir properties while drilling a subject well (e.g., subject well 20C). This model can be applied on a partial or full well as required. The results can be generated at a remote location, or at the wellsite, in real-time, for example, using edge-computing. In one embodiment, real-time automated logquality control and log editing using unsupervised machine learning models us undertaken, which in turn, improve the robustness of the predictions from the method. Prediction uncertainty metrics derived from machine-learning answers provide the confidence needed by the operator to use the machine-learning results with confidence in, for example, an uHT drilling environment where, for example, there is high economic cost for down hole logging tools.

[0029] FIG. 3 illustrates an example of a flow chart (e.g., method) 78 for generation of a machine learning model using one or more algorithms based at least in part on the input data associated with a target well and one or offset wells and training the machine learning model using at least the input data associated with the one or more offset wells to generate a final model. Additional embodiments may further include applying the final model that is generated to new wells to generate reservoir properties, for example, while drilling the well. That is, the final model generated in FIG. 3 can be used to generate reservoir properties in absence of downhole petrophysical tools, due to tools unavailability and / or failure of tools while drilling (e.g., in uHT wells).

[0030] It should be noted that one or more of the steps of flow chart 78 may be performed by the processing system 56. Additionally, one or more of the steps of flow chart 78 can be performed by the computing system 76. For example, in one or more embodiments, processor(s) 62 can operate in conjunction with a deep-learning processor or a neural -network processor and / or, for example, the one or more processor(s) 62 may include machine learning and / or artificial intelligence (Al) based processors. Similarly, the computing system 76 may include one or more processors that can operate in conjunction with a deep-learning processor or a neural -network processor and / or, for example, the one or more processors of the computing system 76 may include machine learning and / or artificial intelligence (Al) based processors. Furthermore, the one or more neural networks can be software-implemented or hardware-implemented in either the processing system 56 or the computing system 76.

[0031] In step 80 of flow chart 78, raw data is received from the downhole equipment 42, the surface equipment 50, and / or the surface logging system 55 by the processingsystem 56 at the network interface. This raw data can include, for example, surface data (drilling, drill cuttings, drill formation gas) together with available gamma ray with or without downhole drilling mechanics data.

[0032] One portion of the data received at step 80 is mud gas data while another may be, for example, drilling mechanics data. This data can be raw and can be digitized and stored in time-based and depth-based categories by the processing system 56 or the computing system 76. Alternatively, the data can be received at step 80 as already having been digitized and stored in time-based and depth-based categories. When utilized, the drilling mechanics data may also be made available from the rig instrumentation or measured by measurement while drilling / logging while drilling (MWD / LWD) operations.

[0033] Additional data received at step 80 may be information from drilled cuttings and may be available in description formats. In other embodiments, methods of digitization of the drilled cuttings include can include study of drilled cutting on digital images, X-ray diffraction (XRD), X-ray fluorescence (XRF), thin sections, scanning electron microscopy (SEM), etc.

[0034] While also available, manual description of drilled cuttings can be input data in step 80; however, this data may be subject to inconsistencies due to the perception and experience of individuals. Accordingly, it may be beneficial to provide consistent and repeatable description of the cutting data as processed by machines that require minimum intervention of individuals. One technique to provide a method of repeatably quantifying the cutting structure, grain size, shape and sorting as information from the drilled cutting can be digitized by integrated camera and Artificial Intelligence / Machine Learning to define individual cutting structures, grains size, shape, size and sorting.

[0035] The cutting analysis (e.g., of cutting data) described above taken together with the gamma ray data can be utilized in the prediction of porosity. The gamma ray data can, for example, be acquired in real-time while drilling by downhole logging tools or from the cutting analysis. Volume of shale (Vshale) can be computed from the downhole gamma ray to normalize the data. Thereafter, the grain size, shape and sorting are added to the matrix to decide the pore space available for a fluid / gas to define “porosity.” Each ofthese processes can be undertaken prior to transmission of the data in step 80 or by the processing system 56 or the computing system 76.

[0036] Mud gas data received at step 80 can also be subject to several inconsistencies, which can be normalized for drilling parameters. This normalization allows for removal of subjectivity on the gas data and the resultant data is studied, thereafter, by one or more machine learning algorithms to define the hydrocarbon saturation.

[0037] Mud gas data consist of several individual light hydrocarbon (organics) (Methane (Cl) to Decane (CIO) together with light aromatics, napthanes and cycloalkanes) components and non-hydrocarbons (inorganics) (Hydrogen sulphide (H2S), Carbon Dioxide (CO2), Helium (He), Hydrogen (H2)) and of a total component called “Total Gas.” These can measured by various technologies such as Flame Ionization detector (FID), Thermal conductivity detector (TCD), Mass spectrometer (MS), or other hydrocarbon analysis techniques. The data then acquired is stored in time and depth in a mud logging database and can be provided as data in step 80. Additionally, in some embodiments, the analyzers do not measure the “Total Gas” and instead, it is computed from the measured hydrocarbon components.

[0038] The hydrocarbon portion of the mud gas data is directly related to the hydrocarbon fraction in a drilled rock, i.e., if the hydrocarbon content in the drilled rock increase, the hydrocarbon signal recorded by the surface mud logging equipment will increase. This signal is directly proportional to the volume of the hydrocarbon in the rock and the type of fluid present in the rock. This is the basis of the saturation derived from the mud gas logging.

[0039] Quality control (QC) of the mud gas data can be automated by utilizing “gas while drilling (GWD)” techniques, this helps in identifying zones where there is inconsistency of the mud gas data in step 82. A QC flag (e.g., from the QC database 84) can be applied, e.g., automatically, and the data can be excluded from learning and prediction. This operates to ensure greater levels of quality of data that is input to the learning algorithm of the machine learning system.

[0040] Mud gas data is also subject to inconsistencies due to the circulation time and the transport properties of hydrocarbon fluid in the mud. This causes a shift of the mud gas data from the reservoir rock. Accordingly, in step 82, the shifts can be identified and shifted in conjunction with the QC database 84 to the correct depths to provide for more reliable input data to the algorithm of the machine learning system.

[0041] Hydrocarbons are continuously liberated from the pore space while drilling, even in the non-reservoir section which creates a background gas level. Recognition of this background gas level can be important to evaluate the elevated signals of the mud gas from reservoirs. Elevated mud gas signals may also come from several non-reservoirs, which may be related to coals, source rocks and produced gasses due to near-balanced or underbalanced drilling conditions. Accordingly, in step 82 in conjunction with the QC database 84, these elevated mud signals can be flagged and can be removed from learning and interpretation.

[0042] Furthermore, hydrocarbons can be artificially generated during drilling and can elevate the gas readings. Recognizing and correcting of these artificially generated hydrocarbons can be useful to reduce the error of learning and predictions. Accordingly, in step 82 in conjunction with the QC database 84, these elevated gas readings can be flagged and can be removed from learning and interpretation.

[0043] The above-mentioned QC steps as part of step 82 can be useful to identify errors that would otherwise be inputted to the learning portion of flow chart 78 and otherwise would be translated to interpretations from Machine learning techniques.

[0044] Once QC process is completed normalization of the mud gas is carried out. The mud gas data changes with different parameters that are used during drilling, namely hole size, rate of penetration, and flow rates. The mud gas data can be normalized for these parameters. In some embodiments, Gaussian smoothing of the mud gas is carried out to remove the artificial spike created due to sudden changes in the drilling parameters. The normalization of the mud gas data can be undertaken for “Total Gas” and “Total Hydrocarbons.” The Total hydrocarbon is the sum of all hydrocarbons recorded by the analyzer.

[0045] Once normalization of mud gas data is completed, the data (as prepared data in step 86 from pre-processing of the data in step 82) is pushed into the learning stage. The learning on the mud gas data is performed together with, for example, the Gamma Ray and known saturations in step 88. That is, one or more machine learning algorithms from step 90 operating in the machine learning stage in step 88 can include running one or more several machine learning algorithms developing the root mean square errors (RMSE) as candidates for a final model in step 92. The machine learning method in step 88 generating candidates in step 92 allow for selection of the best fit or the highest RMSE to be chosen in step 94 for prediction for a particular basin. This results in the final model generated in step 96 that can be applied for new wells.

[0046] FIG. 4 illustrates an example of a flow chart (e.g., method) 98 for utilization of the final model generated in step 96 of FIG. 3. As part of the flow chart 98, step 100 represents learning of wells in a given area, for example, a particular basin. This step 100 may correspond to, for example, one or more of steps 80 and 82 of FIG. 3 and the result of step 100 may be, for example, the prepared data of step 86 of FIG. 3. Step 102 of flow chart 98 may include the machine / deep learning steps previously discussed in, for example, steps 88, 90, and 92 of FIG. 3. The resultant final model in step 104 may correspond to, for example, the final model of step 96 of FIG. 3. In this manner, a portion of FIG. 4 (e.g., steps 100, 102, and 104) may correspond to a model training operation, which can be completed prior to determining and / or generating reservoir properties of a new well, for example, subject well 20C.

[0047] FIG. 4 additionally includes steps that correspond to deployment of the final model of step 104. For example, these additional steps may include step 106 in which a new well (e.g., subject well 20C) is selected for analysis. As an additional portion of step 106, fluid saturation of the new well is calculated or determined. This determination may be made using, for example, surface logging data. For example, the surface logging data can include, for example, surface data (drilling, drill cuttings, drill formation gas) that is then applied to the final model in step 104 to generate fluid saturation for the new well.

[0048] Thus, in step 106 of flow chart 98, raw data is received, for example, from the downhole equipment 42, the surface equipment 50, and / or the surface logging system 55by the processing system 56 at the network interface. This raw data can include, for example, surface data (drilling, drill cuttings, drill formation gas) together with available gamma ray with or without downhole drilling mechanics data. This data can be raw and can be digitized and stored in time-based and depth-based categories by the processing system 56 or the computing system 76. Alternatively, the data can be received at step 106 as already having been digitized and stored in time-based and depth-based categories. When utilized, the drilling mechanics data may also be made available from the rig instrumentation or measured by measurement while drilling / logging while drilling (MWD / LWD) operations.

[0049] Additional data received at step 106 may be information from drilled cuttings and may be available in description formats. In other embodiments, methods of digitization of the drilled cuttings include can include study of drilled cutting on digital images, X-ray diffraction (XRD), X-ray fluorescence (XRF), thin sections, scanning electron microscopy (SEM), etc.

[0050] While also available, manual description of drilled cuttings can be input data in step 106; however, this data may be subject to inconsistencies due to the perception and experience of individuals. Accordingly, it may be beneficial to provide consistent and repeatable description of the cutting data as processed by machines that require minimum intervention of individuals. One technique to provide a method of repeatably quantifying the cutting structure, grain size, shape and sorting as information from the drilled cutting can be digitized by integrated camera and Artificial Intelligence / Machine Learning to define individual cutting structures, grains size, shape, size and sorting.

[0051] The cutting analysis (e.g., of cutting data) described above can be utilized in the generation of fluid saturation for the new well via use of the final model in step 104. Additionally, when taken together with the gamma ray data, the fluid saturation data determined using the final model can be utilized in the prediction of porosity utilizing the final model in step 104. The gamma ray data can, for example, be acquired in real-time while drilling by downhole logging tools or from the cutting analysis. Volume of shale (Vshale) can be computed from the downhole gamma ray to nomialize the data. Thereafter, the grain size, shape and sorting are added to the matrix to decide the pore space availablefor a fluid / gas to define “porosity.” Each of these processes can be undertaken prior to transmission of the data by the processing system 56 or the computing system 76.

[0052] Mud gas data received at step 106 for use in conjunction with the techniques illustrated in flow chart 98 can also be subject to several inconsistencies, which can be normalized for drilling parameters. This normalization allows for removal of subjectivity on the gas data and the resultant data is studied, thereafter, by one or more machine learning algorithms to define the hydrocarbon saturation.

[0053] Mud gas data consist of several individual light hydrocarbon (organics) (Methane (Cl) to Decane (CIO) together with light aromatics, napthanes and cycloalkanes) components and non-hydrocarbons (inorganics) (Hydrogen sulphide (H2S), Carbon Dioxide (CO2), Helium (He), Hydrogen (H2)) and of a total component called “Total Gas.” These can be measured by various technologies such as Flame Ionization detector (FID), Thermal conductivity detector (TCD), Mass spectrometer (MS), or other hydrocarbon analysis techniques. The data then acquired is stored in time and depth in a mud logging database and can be provided as data in step 106. Additionally, in some embodiments, the analyzers do not measure the “Total Gas” and instead, it is computed from the measured hydrocarbon components.

[0054] The hydrocarbon portion of the mud gas data is directly related to the hydrocarbon fraction in a drilled rock, i.e., if the hydrocarbon content in the drilled rock increase, the hydrocarbon signal recorded by the surface mud logging equipment will increase. This signal is directly proportional to the volume of the hydrocarbon in the rock and the type of fluid present in the rock. This is the basis of the saturation derived from the mud gas logging.

[0055] Quality control of the mud gas data can be automated by utilizing “gas while drilling (GWD)” techniques, this helps in identifying zones where there is inconsistency of the mud gas data. A QC flag can be applied, e.g., automatically, and the data can be excluded from learning and prediction. This operates to ensure greater levels of quality of data that is input to the learning algorithm of the machine learning system.

[0056] Mud gas data is also subject to inconsistencies due to the circulation time and the transport properties of hydrocarbon fluid in the mud. This causes a shift of the mud gas data from the reservoir rock. Accordingly, the shifts can be identified and shifted in conjunction with a QC database to the correct depths to provide for more reliable input data to the algorithm of the machine learning system.

[0057] Hydrocarbons are continuously liberated from the pore space while drilling, even in the non-reservoir section which creates a background gas level. Recognition of this background gas level can be important to evaluate the elevated signals of the mud gas from reservoirs. Elevated mud gas signals may also come from several non-reservoirs, which may be related to coals, source rocks and produced gasses due to near-balanced or underbalanced drilling conditions. Accordingly, these elevated mud signals can be flagged and can be removed from learning and interpretation.

[0058] Furthermore, hydrocarbons can be artificially generated during drilling and can elevate the gas readings. Recognizing and correcting of these artificially generated hydrocarbons can be useful to reduce the error of learning and predictions. Accordingly, these elevated gas readings can be flagged and can be removed from learning and interpretation.

[0059] The above-mentioned QC steps as part of step 106 can be useful to identify errors that would otherwise be inputted to the final model in step 104 and otherwise would be translated to interpretations from the final model.

[0060] Once QC process is completed normalization of the mud gas is carried out. The mud gas data changes with different parameters that are used during drilling, namely hole size, rate of penetration, and flow rates. The mud gas data can be normalized for these parameters. In some embodiments, Gaussian smoothing of the mud gas is carried out to remove the artificial spike created due to sudden changes in the drilling parameters. The normalization of the mud gas data can be undertaken for “Total Gas” and “Total Hydrocarbons.” The Total hydrocarbon is the sum of all hydrocarbons recorded by the analyzer.

[0061] Once normalization of mud gas data is completed, the data is applied to the final model in step 104. Additionally, Gamma Ray and known saturations can additionally be applied in 104 to the final model. That is, one or more machine learning algorithms operating can include running one or more several machine learning algorithms as candidates for reservoir properties in step 108. The final model in step 104 receiving the inputs discussed above, operates in step 106 to generate results for reservoir properties as digital data in step 108, which can allow for selection of the best fit to be chosen in step 108 for prediction of reservoir properties of a particular well and / or basin. In this manner, the final model of step 104 is applied to characterize new wells through the use of a fluid saturation determination from surface logging data to derive, for example, reserve size and permeability (e.g., by use of fluid saturation and porosity determined from surface logging data).

[0062] This written description uses examples to disclose the subject matter, including the best mode, and also to enable any person skilled in the art to practice the subject matter, including making and using any devices or systems and performing any incorporated methods. The patentable scope of the subject matter is defined by the claims and may include other examples that occur to those skilled in the art. Such other examples are intended to be within the scope of the claims if they have structural elements that do not differ from the literal language of the claims, or if they include equivalent structural elements with insubstantial differences from the literal language of the claims.

[0063] The techniques presented and claimed herein are referenced and applied to material objects and concrete examples of a practical nature that demonstrably improve the present technical field and, as such, are not abstract, intangible, or purely theoretical. Further, if any claims appended to the end of this specification contain one or more elements designated as “means for [perform] ing [a function]...” or “step for [perform] ing [a function]...”, it is intended that such elements are to be interpreted under 35 U.S.C. 112(f). However, for any claims containing elements designated in any other manner, it is intended that such elements are not to be interpreted under 35 U.S.C. 112(f).

Claims

CLAIMSWhat is claimed is:

1. A method comprising: receiving first raw data corresponding to at least one attribute of a well; applying a pre-processing operation to correct at least one error in the first raw data to generate prepared data; transmitting the prepared data to a machine learning system; training the machine learning system into a trained machine learning system using the prepared data; and generating a final model via the trained machine learning system, wherein the final model operates to generate a characterization of a reservoir in a subsurface region of Earth when second raw data is input to the final model.

2. The method of claim 1, further comprising generating, at the trained machine learning system, a first candidate model based on the prepared data.

3. The method of claim 2, further comprising generating, at the trained machine learning system, a second candidate model based on the prepared data.

4. The method of claim 3, further comprising selecting one of the first candidate model and the second candidate model as the final model.

5. The method of claim 1, further comprising gathering the first raw data from instrumentation or equipment disposed at a surface region of Earth.

6. The method of claim 1, further comprising gathering at least a portion of the first raw data from a gamma-ray logging tool.

7. The method of claim 1, further comprising gathering a first portion of the second raw data from instrumentation or equipment disposed at a surface region of Earth.

8. The method of claim 7, further comprising gathering at least a second portion of the second raw data from a gamma-ray logging tool.

9. A tangible and non-transitory machine readable medium, comprising instructions to cause a processing system to: receive first raw data corresponding to at least one attribute of a well; apply a pre-processing operation to correct at least one error in the first raw data to generate prepared data; transmit the prepared data to a machine learning system; train the machine learning system into a trained machine learning system using the prepared data; and generate a final model via the trained machine learning system, wherein the final model operates to generate a characterization of a reservoir in a subsurface region of Earth when second raw data is input to the final model.

10. The tangible and non-transitory machine readable medium of claim 9, further comprising instructions to cause the processing system to generate, via the trained machine learning system, a first candidate model based on the prepared data.

11. The tangible and non-transitory machine readable medium of claim 10, further comprising instructions to cause the processing system to generate, via the trained machine learning system, a second candidate model based on the prepared data.

12. The tangible and non-transitory machine readable medium of claim 11, further comprising instructions to cause the processing system to select one of the first candidate model and the second candidate model as the final model.

13. The tangible and non-transitory machine readable medium of claim 9, further comprising instructions to cause the processing system to receive the first raw data from instrumentation or equipment disposed at a surface region of Earth.

14. The tangible and non-transitory machine readable medium of claim 9, further comprising instructions to cause the processing system to receive at least a portion of the first raw data from a gamma-ray logging tool.

15. The tangible and non-transitory machine readable medium of claim 9, further comprising instructions to cause the processing system to receive a first portion of the second raw data from instrumentation or equipment disposed at a surface region of Earth.

16. The tangible and non-transitory machine readable medium of claim 15, further comprising instructions to cause the processing system to receive at least a second portion of the second raw data from a gamma-ray logging tool.

17. A m ethod compri si ng : receiving first raw data corresponding to at least one first attribute of a well; applying a pre-processing operation to correct at least one error in the first raw data to generate first prepared data; transmitting the first prepared data to a trained machine learning system; receiving second raw data corresponding to at least one second attribute of the well; applying a second pre-processing operation to correct at least one second error in the second raw data to generate second prepared data; transmitting the second prepared data to the trained machine learning system; and applying the first prepared data and the second prepared data to a model of the trained machine learning system to generate a characterization of a reservoir comprising the well in a subsurface region of Earth.

18. The method of claim 17, further comprising gathering the first raw data from instrumentation or equipment disposed at a surface region of Earth.

19. The method of claim 17, further comprising gathering the second raw data from a gamma-ray logging tool disposed in the well.

20. The method of claim 17, further comprising generating a prediction of porosity of the well via the model of the trained machine learning system based on the first prepared data and the second prepared data.