A high-quality reservoir characterization method based on machine learning

Abstract

This invention provides a method, apparatus, computer-readable storage medium, and electronic device for characterizing high-quality reservoirs based on machine learning. The method includes: determining sensitive logging curves for identifying high-quality reservoirs; obtaining elastic parameter curves by deriving rock physical parameter curves through rock physical modeling based on multi-mineral optimization evaluation; analyzing the probability distribution of sensitive logging curves for high-quality reservoirs based on the correspondence between elastic parameter curves and sensitive logging curves determined through machine learning; obtaining an elastic parameter volume through pre-stack statistical inversion under phase control; then, obtaining the corresponding sensitive logging curve volume using the elastic parameter volume based on the correspondence between the elastic parameter curves and sensitive logging curves; and finally, characterizing the distribution of high-quality reservoirs based on the probability distribution and sensitive logging curve volume of the sensitive logging curves.

Description

Technical Field

[0001] This invention relates to the field of oil and gas exploration well detection technology, and in particular to a machine learning-based high-quality reservoir characterization method, apparatus, computer-readable storage medium, and electronic device suitable for tight sandstone. Background Technology

[0002] For tight sandstone gas reservoirs, high-quality reservoirs are key to high and stable gas well production. Therefore, clarifying their distribution is crucial for the exploration and development of tight oil and gas reservoirs. However, tight sandstone has poor overall physical properties and strong heterogeneity, making it difficult to characterize high-quality reservoirs.

[0003] Currently, commonly used methods for characterizing high-quality reservoirs include seismic attribute analysis and impedance oscillation. Seismic attribute analysis is effective in characterizing sand bodies, but since sand bodies contain tight layers with poor physical properties, attributes cannot characterize the reservoir itself. Impedance oscillation characterizes high-quality reservoirs based on the differences in elastic parameters between the high-quality reservoir and the surrounding rock. Often, the mineral composition and physical properties of tight sandstone high-quality reservoirs and the surrounding rocks are similar, resulting in small differences in elastic parameters. Therefore, there is some overlap, or even severe overlap, between the high-quality reservoir and the surrounding rock on the elastic parameter cross-plot. The reliability of directly using these parameters to characterize high-quality reservoirs needs to be improved.

[0004] To address the aforementioned issues, Li Long et al. (2019) reconstructed GR curves and acoustic impedance curves to predict the sandstone bodies in the Shahe Formation of the Qingshui Depression in the western Liaohe River Basin. However, this method can only characterize the distribution of sandstone, not the distribution of high-quality reservoirs within it. Gan Dayong et al. (2020) obtained acoustic impedance data volumes through well-constrained seismic inversion; then, they conducted geostatistical GR inversion to obtain lithological data volumes, and calculated the lithological data volumes with the acoustic impedance data volumes to eliminate the influence of lithology; finally, they used the relationship between porosity and acoustic impedance to obtain porosity inversion data volumes, characterizing the presence, thickness, and quality of tight sandstone reservoirs in the Shaximiao Formation in the QL area of ​​central Sichuan. Since the correlation between porosity and acoustic impedance in tight sandstone is generally weak, the reliability of porosity volumes obtained using acoustic impedance volumes is relatively low. Sinopec Exploration Company (2017) conducted a method of stepwise approximation and multiple dimensionality reduction to predict the tight sandstone reservoirs and gas-bearing potential of the Xujiahe Formation in northeastern Sichuan. First, impedance inversion bodies and GR inversion bodies were used to distinguish sandstone from mudstone and clarify the distribution of sandstone. Then, elastic parameters such as Poisson's ratio and impedance, or reconstructed parameters, were used to separate reservoirs and tight layers within the sandstone. Finally, elastic parameters such as density and impedance, or partially reconstructed parameters, were used to separate gas layers and water layers. Analysis of the cross-plots of elastic parameters distinguishing between sandstone and mudstone, reservoirs and tight layers, and gas layers and water layers reveals that there is still some overlap between sandstone and mudstone, reservoirs and tight layers, and gas layers and water layers. Therefore, the reliability of this method in characterizing high-quality reservoirs needs to be improved. Summary of the Invention

[0005] To address the aforementioned problems, embodiments of the present invention provide a method, apparatus, computer-readable storage medium, and electronic device for high-quality reservoir characterization based on machine learning.

[0006] In a first aspect, embodiments of the present invention provide a high-quality reservoir characterization method based on machine learning, comprising:

[0007] S100, obtain the pre-stack seismic gathers for the study area;

[0008] S200: Obtain the logging curves of the study area; by analyzing the intersection of the logging curves, determine the sensitive logging curves that can be used to identify high-quality reservoirs in the study area.

[0009] S300, determine the mineral composition and content of the study area, and based on the mineral composition and content, perform forward modeling of the rock physical parameter curves of the study area through rock physical modeling, and obtain the elastic parameter curves of the study area based on the rock physical parameter curves.

[0010] S400: Using a deep neural network algorithm, a correspondence is established between the sensitive logging curve and the elastic parameter curve. Based on the correspondence, the corresponding sensitive logging curve is calculated from the rock physical parameter curves of the wells already drilled in the study area. By analyzing the intersection of the sensitive logging curves, the probability distribution of the sensitive logging curves of high-quality reservoirs is determined.

[0011] S500: Using the pre-stack seismic gathers of the study area, the elastic parameter volume of the study area is obtained through pre-stack statistical inversion under phase control. Based on the correspondence between the sensitive logging curves and the elastic parameter curves, the sensitive logging curve volume is inverted using the elastic parameter volume.

[0012] S600, based on the probability distribution of the sensitive logging curves of high-quality reservoirs, determine whether each position on the sensitive logging curve body is a high-quality reservoir, thereby determining the distribution of high-quality reservoirs in the study area.

[0013] According to an embodiment of the present invention, in step S100 above, after obtaining the pre-stack seismic gathers of the study area and before utilizing the pre-stack seismic gathers of the study area, the quality of the pre-stack seismic gathers is optimized.

[0014] According to an embodiment of the present invention, in step S200 above, the sensitive logging curves that can be used to identify high-quality reservoirs in the study area are natural gamma (GR) logging curves and deep lateral resistivity (RD) logging curves.

[0015] According to an embodiment of the present invention, in step S300 above, determining the mineral composition and content of the study area includes:

[0016] A multi-mineral optimization model for the study area was established, and the mineral composition and content of the study area were determined using the multi-mineral optimization model.

[0017] According to an embodiment of the present invention, in step S300 above, the elastic parameter curves include longitudinal wave velocity, transverse wave velocity, Vp / Vs, Lamé coefficient, and Poisson's ratio elastic parameter curves.

[0018] According to an embodiment of the present invention, in step S400 above, the deep neural network algorithm includes a support vector machine deep neural network algorithm based on FVR; the probability distribution of the sensitive logging curve of the high-quality reservoir is a Bayesian probability distribution.

[0019] According to an embodiment of the present invention, in step S500 above, obtaining the elastic parameter volume of the study area by using pre-stack seismic gathers of the study area through pre-stack statistical inversion under phase control includes:

[0020] Using pre-stack seismic gathers in the study area, the distribution of sand-soil ratio was determined by geologically delineated sedimentary microfacies maps and pre-stack deterministic inversion P-wave impedance plane maps derived from sand-soil ratio machine learning.

[0021] Pre-stack statistical inversion was carried out under the constraint of sand-land ratio distribution to obtain the elastic parameter volume of the study area.

[0022] Secondly, the present invention also provides a high-quality reservoir characterization and identification device based on machine learning, characterized in that it comprises:

[0023] The pre-stack data acquisition module is used to acquire pre-stack seismic gathers for the study area.

[0024] The well logging curve analysis module is used to acquire well logging curves in the study area. By analyzing the intersection of the well logging curves, sensitive well logging curves that can be used to identify high-quality reservoirs in the study area are determined.

[0025] The elastic parameter extrapolation module is used to determine the mineral composition and content of the study area. Based on the mineral composition and content, the rock physical parameter curve of the study area is forward modeled through rock physical modeling. Based on the rock physical parameter curve, the elastic parameter curve of the study area is obtained.

[0026] The reservoir probability analysis module is used to establish the correspondence between the sensitive logging curve and the elastic parameter curve using a deep neural network algorithm. Based on the correspondence, the corresponding sensitive logging curve is calculated from the rock physical parameter curves of the wells already drilled in the study area. By analyzing the intersection of the sensitive logging curves, the probability distribution of the sensitive logging curves of high-quality reservoirs is determined.

[0027] The sensitive logging curve inversion module is used to obtain the elastic parameter volume of the study area through pre-stack statistical inversion under phase control by utilizing the pre-stack seismic gathers of the study area. Based on the correspondence between the sensitive logging curves and the elastic parameter curves, the sensitive logging curve volume is inverted using the elastic parameter volume.

[0028] The reservoir distribution determination module is used to determine whether each position on the sensitive logging curve body is a high-quality reservoir based on the probability distribution of the sensitive logging curve of high-quality reservoir, thereby determining the distribution of high-quality reservoirs in the study area.

[0029] Thirdly, embodiments of the present invention provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a high-quality reservoir characterization method based on machine learning as described in the first aspect above.

[0030] Fourthly, embodiments of the present invention provide an electronic device comprising:

[0031] processor;

[0032] Memory used to store the processor's executable instructions;

[0033] The processor is configured to execute the instructions to implement a high-quality reservoir characterization method based on machine learning as described in the first aspect above.

[0034] Compared with the prior art, the above-mentioned technical solution of the present invention has the following beneficial effects:

[0035] 1) The present invention provides a high-quality reservoir characterization method based on machine learning, which expands the parameters that distinguish between high-quality tight sandstone reservoirs and surrounding rocks by using machine learning-sensitive logging curves, thereby improving the characterization accuracy of high-quality tight sandstone reservoirs.

[0036] 2) The present invention provides a high-quality reservoir characterization method based on machine learning, which does not require characterizing the sand body first and then the reservoir in the sand body. It can directly determine the sensitive logging curves and elastic parameters of high-quality reservoirs, and thus characterize their distribution.

[0037] 3) The present invention provides a high-quality reservoir characterization method based on machine learning, which can not only accurately characterize high-quality reservoirs, but also directly characterize gas layer distribution based on the determination of gas layer sensitive logging curves and elastic parameters, and has wide applications. Attached Figure Description

[0038] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0039] Figure 1 This is a flowchart of the high-quality reservoir characterization method based on machine learning provided in Embodiment 1 of the present invention;

[0040] Figure 2 This is a schematic diagram of the composition structure of the electronic device according to Embodiment 5 of the present invention. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0042] Example 1

[0043] Considering the difficulty of accurately distinguishing between high-quality reservoirs and surrounding rocks in tight sandstone using elastic parameters, this invention obtains reservoir-sensitive logging curves through elastic parameter machine learning. By combining these sensitive logging curves with elastic parameters, the invention differentiates between reservoirs and surrounding rocks, accurately characterizing the distribution of high-quality reservoirs. This method can be widely applied to the exploration and development of tight oil and gas reservoirs.

[0044] like Figure 1 As shown, the high-quality reservoir characterization method based on machine learning provided in Embodiment 1 of the present invention mainly includes the following: The present invention adopts the support vector machine deep neural network algorithm of FVR, which expands the parameters for identifying high-quality tight sandstone reservoirs and improves the accuracy of reservoir characterization.

[0045] This invention is achieved through the following steps:

[0046] Step 1: Based on the pre-stack seismic gather quality analysis, targeted optimization techniques are adopted to improve the quality of the pre-stack gathers.

[0047] Step 2: Through cross-analysis of logging curves of high-quality reservoirs and surrounding rocks, identify the sensitive logging curves that can distinguish between high-quality reservoirs and surrounding rocks.

[0048] Step 3: Establish a multi-mineral optimization model to clarify the mineral composition and content of the tight sandstone; on this basis, carry out rock physics modeling to accurately derive rock physics parameters such as P-wave and S-wave velocities and density of a single well, and then obtain a variety of elastic parameters.

[0049] Step four: Using a deep neural network algorithm based on support vector machines, a nonlinear relationship is established between the sensitive logging curve (determined in step two) and the elastic parameter curve (obtained in step three). Based on this relationship, the sensitive logging curve is calculated from the rock physical parameter curves of the drilled wells. Through the cross-analysis of the sensitive logging curves of high-quality reservoirs and surrounding rocks, the Bayesian probability distribution of high-quality reservoirs is clarified.

[0050] Step 5: Determine the distribution of sand-soil ratio by using the sedimentary microfacies map and the pre-stack deterministic inversion P-wave impedance plane map drawn by machine learning of sand-soil ratio in the well. Under the constraint of the sand-soil ratio map, carry out pre-stack statistical inversion to obtain elastic parameters such as P-wave impedance, S-wave impedance and density. Based on the nonlinear relationship between the sensitive logging curves and elastic parameters determined in Step 4, calculate the sensitive logging curve body of high-quality reservoir.

[0051] Step six: Based on the Bayesian probability distribution of high-quality reservoirs obtained in step four, determine whether each position on the high-quality reservoir sensitive logging curve is a high-quality reservoir, and thus characterize the distribution of high-quality reservoirs.

[0052] Example 2

[0053] The third member of the Xujiahe Formation in the YB area consists of a suite of calcareous sandstone and conglomerate. According to core physical property data, the porosity is mainly distributed between 1% and 3%, indicating a low-porosity and low-permeability reservoir. Among them, the medium- and coarse-grained calcareous sandstone, sandy fine conglomerate, and sand-bearing fine conglomerate have better physical properties and are high-quality reservoirs. However, these high-quality reservoirs are thin and difficult to characterize. This study used the following process to characterize the high-quality reservoirs and achieved good results.

[0054] Step 1 involves analyzing the AVO characteristics, signal-to-noise ratio, multiples, and consistency of amplitude, phase, and energy between full-offset stacked data and pure wave data in the pre-stack CRP gathers of the YB region. This analysis identifies two major problems in the pre-stack gathers of the study area: "spindle-shaped energy distribution" and "insufficient correction for some far-offsets." An optimization strategy of "AVO regularity compensation - automatic leveling - noise reduction filtering" was developed, which effectively improved the quality of the pre-stack gathers.

[0055] Step 2: The Xujiahe Formation III in the study area contains various sandstone types, including medium- to coarse-grained conglomerate, sandy fine conglomerate, sandy fine conglomerate, medium- to coarse-grained calcareous sandstone, fine-grained calcareous sandstone, and siltstone. Among these, the sandy fine conglomerate, sandy fine conglomerate, and medium- to coarse-grained calcareous sandstone exhibit good physical properties, indicating high-quality reservoirs. Based on the core thin sections and physical property analysis data from the core wells, the sandstone types identified by the thin sections from 13 core wells and their corresponding well logging curve values ​​at different depths were statistically analyzed. Various well logging curve cross-plots were then created for each sandstone type, clearly demonstrating that the GR and RD curves can effectively identify the high-quality reservoirs in the Xujiahe Formation III.

[0056] Step three involves establishing a multi-mineral optimization model to clarify the mineral composition of the three tight sandstone sections. Based on this, the Xu-White model is used to plot the P-wave velocity, S-wave velocity, and density curves of the drilled wells, ensuring good consistency with the measured curves. Cross-sectional analysis of various petrophysical parameters for different types of sandstone reveals that petrophysical parameters alone are insufficient for effectively distinguishing high-quality reservoirs.

[0057] Step four involves employing multiple regression and various machine learning methods to construct formulas for calculating GR and RD curves using rock physical parameter curves. The GR and RD curves are then calculated. By evaluating the agreement between the calculated GR and RD curves and the measured curves, the nonlinear relationship between the sensitive logging curves established using the FVR support vector machine deep neural network algorithm and the curves of five elastic parameters—P-wave velocity, S-wave velocity, P-S / S-wave velocity ratio (Vp / Vs), Lamé coefficient, and Poisson's ratio—is optimized. Based on this, GR and RD curves are calculated from the rock physical parameter curves of drilled wells. Through cross-analysis of the GR-RD curves of high-quality reservoirs and surrounding rocks, the Bayesian probability distribution of GR and RD in high-quality reservoirs is clarified.

[0058] Step 5: Determine the distribution of sand-soil ratio by using the sedimentary microfacies map and the pre-stack deterministic inversion P-wave impedance plane map drawn by the sand-soil ratio machine learning in the well. Under the constraint of sand-soil ratio, carry out pre-stack statistical inversion to obtain P-wave, S-wave impedance and density volume. Based on the nonlinear relationship between the sensitive curve and elastic parameters obtained in step (4), invert the GR and RD logging curves.

[0059] Step 6: Based on the Bayesian probability distribution of high-quality reservoirs obtained in step (4), determine whether each position on the GR and RD curves is a high-quality reservoir, and thus clarify the distribution of high-quality reservoirs.

[0060] GR and RD logging curves are effective at identifying high-quality reservoirs in the third member of the Xujiahe Formation in the YB area, hence these two logging curves were chosen for inversion. Sensitive curves for high-quality reservoirs in other areas may differ from those in the YB area; therefore, appropriate curves (parameters) can be selected as the object of machine learning to characterize high-quality reservoirs.

[0061] This invention expands the parameters for identifying high-quality reservoirs and surrounding rocks in tight sandstone, improving the accuracy of high-quality reservoir characterization in tight sandstone. The agreement rate with high-quality reservoir interpretations from 72 wells in the study area reaches 86%. The inverted high-quality reservoir thicknesses in different sections are basically consistent with the surface high-quality reservoir thicknesses.

[0062] Example 3

[0063] The following are embodiments of the apparatus of the present invention, which can be used to execute embodiments of the method of the present invention. For details not disclosed in the embodiments of the apparatus of the present invention, please refer to the embodiments of the method of the present invention.

[0064] This embodiment provides a high-quality reservoir characterization device based on machine learning, comprising:

[0065] The pre-stack data acquisition module is used to acquire pre-stack seismic gathers for the study area.

[0066] The well logging curve analysis module is used to acquire well logging curves in the study area. By analyzing the intersection of the well logging curves, sensitive well logging curves that can be used to identify high-quality reservoirs in the study area are determined.

[0067] The elastic parameter extrapolation module is used to determine the mineral composition and content of the study area. Based on the mineral composition and content, the rock physical parameter curve of the study area is forward modeled through rock physical modeling. Based on the rock physical parameter curve, the elastic parameter curve of the study area is obtained.

[0068] The reservoir probability analysis module is used to establish the correspondence between the sensitive logging curve and the elastic parameter curve using a deep neural network algorithm. Based on the correspondence, the corresponding sensitive logging curve is calculated from the rock physical parameter curves of the wells already drilled in the study area. By analyzing the intersection of the sensitive logging curves, the probability distribution of the sensitive logging curves of high-quality reservoirs is determined.

[0069] The sensitive logging curve inversion module is used to obtain the elastic parameter volume of the study area through pre-stack statistical inversion under phase control by utilizing the pre-stack seismic gathers of the study area. Based on the correspondence between the sensitive logging curves and the elastic parameter curves, the sensitive logging curve volume is inverted using the elastic parameter volume.

[0070] The reservoir distribution determination module is used to determine whether each position on the sensitive logging curve body is a high-quality reservoir based on the probability distribution of the sensitive logging curve of high-quality reservoir, thereby characterizing the distribution of high-quality reservoirs in the study area.

[0071] Example 4

[0072] This embodiment provides a computer-readable medium storing a computer program that, when executed by a processor, implements the various steps of a machine learning-based high-quality reservoir characterization method as described in the above embodiment.

[0073] It should be noted that all or part of the processes in the methods of the above embodiments of the present invention can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. Of course, there are other readable storage media, such as quantum memories, graphene memories, etc. It should be noted that the content included in the computer-readable medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electrical carrier signals and telecommunication signals.

[0074] Example 5

[0075] Figure 2 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention. Figure 2 As shown, at the hardware level, this electronic device includes a processor, and optionally also includes an internal bus, a network interface, and memory. The memory may include main memory, such as high-speed random-access memory (RAM), or it may include non-volatile memory, such as at least one disk drive. Of course, this electronic device may also include other hardware required for other business operations.

[0076] The processor, network interface, and memory can be interconnected via an internal bus, which can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. This bus can be categorized as an address bus, data bus, control bus, etc. For ease of illustration, only line segments are used in the diagram, but this does not imply that there is only one bus or one type of bus.

[0077] A memory is used to store programs. Specifically, the program may include program code, which includes computer operation instructions. The memory may include main memory and non-volatile memory, and provides instructions and data to the processor. The processor reads the corresponding computer program from the non-volatile memory into main memory and then runs it. The processor executes the program stored in the memory to perform all the steps in the aforementioned high-quality reservoir characterization method based on machine learning.

[0078] The communication bus mentioned above can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into address bus, data bus, control bus, etc. For ease of illustration, only one thick line is used to represent it in the diagram, but this does not mean that there is only one bus or one type of bus. The communication interface is used for communication between the above electronic devices and other devices.

[0079] A bus, including hardware, software, or both, is used to couple the aforementioned components together. For example, a bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, a bus may include one or more buses. Although specific buses are described and illustrated in embodiments of the invention, the invention contemplates any suitable bus or interconnect.

[0080] The memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0081] The memory may include a large-capacity storage device for data or instructions. For example, and not limitingly, the memory may include a hard disk drive (HDD), a floppy disk drive, flash memory, an optical disk drive, a magneto-optical disk drive, magnetic tape, or a Universal Serial Bus (USB) drive, or a combination of two or more of these. Where suitable, the memory may include removable or non-removable (or fixed) media. In a particular embodiment, the memory is a non-volatile solid-state memory. In a particular embodiment, the memory includes a read-only memory (ROM). Where suitable, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically rewritable ROM (EAROM), or flash memory, or a combination of two or more of these.

[0082] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0083] It should be noted that those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this invention. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0084] The apparatus, device, system, module, or unit described in the above embodiments can be implemented by a computer chip or entity, or by a product with a certain function. A typical implementation device is a computer. Specifically, the computer can be, for example, a personal computer, a laptop computer, an in-vehicle human-machine interaction device, a cellular phone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or any combination of these devices.

[0085] While this invention provides the method operation steps as described in the embodiments or flowcharts, more or fewer operation steps may be included based on conventional or non-inventive means. The order of steps listed in the embodiments is merely one possible execution order among many and does not represent the only execution order. In actual devices or terminal products, the methods shown in the embodiments or drawings can be executed sequentially or in parallel (e.g., in a parallel processor or multi-threaded processing environment, or even a distributed data processing environment).

[0086] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0087] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0088] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0089] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0090] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of apparatus, electronic devices, and readable storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0091] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention are included within the scope of protection of the present invention.

Claims

1. A method for machine learning based quality reservoir characterization, characterized in that, Includes the following steps: S100, obtain the pre-stack seismic gathers for the study area; S200: Obtain logging curves for the study area; analyze the differences between logging curves of high-quality reservoirs and non-reservoir reservoirs by logging curve intersection; and determine the sensitive logging curves that can be used to identify high-quality reservoirs in the study area. S300, determine the mineral composition and content of the study area, and based on the mineral composition and content, perform forward modeling of the rock physical parameter curves of the study area through rock physical modeling, and obtain the elastic parameter curves of the study area based on the rock physical parameter curves. S400: Using a deep neural network algorithm, a correspondence is established between the sensitive logging curve and the elastic parameter curve. Based on the correspondence, the corresponding sensitive logging curve is calculated from the elastic parameter curve of the drilled well in the study area. Through the cross-sectional analysis of the calculated sensitive logging curves, the probability distribution of the sensitive logging curve of the high-quality reservoir is determined. S500: Using the pre-stack seismic gathers of the study area, the elastic parameter volume of the study area is obtained through pre-stack statistical inversion under phase control. Based on the correspondence between the sensitive logging curves and the elastic parameter curves, the sensitive logging curve volume is inverted using the elastic parameter volume. S600, based on the probability distribution of the sensitive logging curves of high-quality reservoirs, determine whether each position on the sensitive logging curve body is a high-quality reservoir, thereby determining the distribution of high-quality reservoirs in the study area.

2. The high-quality reservoir characterization method based on machine learning as described in claim 1, characterized in that, In step S100, after obtaining the pre-stack seismic gathers of the study area and before utilizing the pre-stack seismic gathers of the study area, the quality of the pre-stack seismic gathers is optimized.

3. The high-quality reservoir characterization method based on machine learning as described in claim 1, characterized in that, In step S200, the sensitive logging curves that can be used to identify high-quality reservoirs in the study area are natural gamma (GR) logging curves and deep lateral resistivity (RD) logging curves.

4. The high-quality reservoir characterization method based on machine learning as described in claim 1, characterized in that, In step S300, determining the mineral composition and content of the study area includes: A multi-mineral optimization model for the study area was established, and the mineral composition and content of the study area were determined using the multi-mineral optimization model.

5. The high-quality reservoir characterization method based on machine learning as described in claim 1, characterized in that, In step S300, the elastic parameter curves include the elastic parameter curves of longitudinal wave velocity, transverse wave velocity, longitudinal wave to transverse wave velocity ratio, Lamé coefficient, and Poisson's ratio.

6. The high-quality reservoir characterization method based on machine learning as described in claim 1, characterized in that, In step S400, the deep neural network algorithm includes a deep neural network algorithm based on support vector machines; the probability distribution of the sensitive logging curve of the high-quality reservoir is a Bayesian probability distribution.

7. The high-quality reservoir characterization method based on machine learning as described in claim 1, characterized in that, In step S500, the process of obtaining the elastic parameter volume of the study area by using pre-stack seismic gathers of the study area through pre-stack statistical inversion under phase control includes: Using pre-stack seismic gathers in the study area, the distribution of sand-soil ratio was determined by geologically delineated sedimentary microfacies maps and pre-stack deterministic inversion P-wave impedance plane maps derived from sand-soil ratio machine learning. Pre-stack statistical inversion was carried out under the constraint of sand-land ratio distribution to obtain the elastic parameter volume of the study area.

8. A high-quality reservoir characterization device based on machine learning, characterized in that, include: The pre-stack data acquisition module is used to acquire pre-stack seismic gathers for the study area. The well logging curve analysis module is used to acquire well logging curves in the study area. By analyzing the intersection of the well logging curves, sensitive well logging curves that can be used to identify high-quality reservoirs in the study area are determined. The elastic parameter extrapolation module is used to determine the mineral composition and content of the study area. Based on the mineral composition and content, the rock physical parameter curve of the study area is forward modeled through rock physical modeling. Based on the rock physical parameter curve, the elastic parameter curve of the study area is obtained. The reservoir probability analysis module is used to establish the correspondence between the sensitive logging curve and the elastic parameter curve using a deep neural network algorithm. Based on the correspondence, the corresponding sensitive logging curve is calculated from the rock physical parameter curves of the wells already drilled in the study area. By analyzing the intersection of the sensitive logging curves, the probability distribution of the sensitive logging curves of high-quality reservoirs is determined. The sensitive logging curve inversion module is used to obtain the elastic parameter volume of the study area through pre-stack statistical inversion under phase control by utilizing the pre-stack seismic gathers of the study area. Based on the correspondence between the sensitive logging curve and the elastic parameter curve, the sensitive logging curve volume is inverted using the elastic parameter volume. The reservoir distribution determination module is used to determine whether each position on the sensitive logging curve body is a high-quality reservoir based on the probability distribution of the sensitive logging curve of high-quality reservoir, thereby characterizing the distribution of high-quality reservoirs in the study area.

9. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed by a processor, implements a high-quality reservoir characterization method based on machine learning as described in any one of claims 1 to 7.

10. An electronic device comprising: processor; Memory used to store the processor's executable instructions; The processor is configured to execute the instructions to implement a high-quality reservoir characterization method based on machine learning as described in any one of claims 1 to 7.

Images