Machine learning based rock cross-scale parameter correlation and geometric feature matching method

Abstract

The application discloses a rock cross-scale parameter correlation and geometric feature matching method based on machine learning. The method comprises the following steps: obtaining a rock microstructure digital image; inputting the rock microstructure digital image into a rock microstructure phase segmentation model to obtain a divided rock microstructure phase; constructing a rock micro-mesoscopic parameter correlation model to obtain rock mesoscopic parameters; constructing a rock micro-mesoscopic-macro cross-scale deep learning generalized generative model to match the rock mesoscopic parameters and rock geometric structure features to obtain a rock mesoscopic parameter-geometric feature matching model; constructing a rock mesoscopic-macro parameter correlation model to obtain rock macro parameters; and matching the rock macro parameters and rock geometric structure features based on the rock micro-mesoscopic-macro cross-scale deep learning generalized generative model to obtain a rock macro parameter-geometric feature matching model. The application realizes the cross-scale correlation of rock micro-mesoscopic-macro physical and mechanical parameters and geometric features.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of rock mechanics and machine learning, and in particular to a rock cross-scale parameter correlation and geometric feature matching method and device based on machine learning, a storage medium and an electronic device. BACKGROUND

[0002] The geometric features of the micro material phase structure of rock and its physical and mechanical properties determine the physical and mechanical properties and the fracture characteristics of the whole rock, and play a very important role in revealing the micro-meso-macro full-scale fracture mechanism of rock. In-depth study of the whole process of rock fracture evolution and mechanism can greatly improve the efficiency and technical development of deep resource exploitation (such as shale oil and gas and geothermal energy, etc.), and promote the development of early warning and prevention technology of deep engineering disasters (tunnel collapse and deep buried CO2-nuclear waste material leakage, etc.), which is of great significance to the safety of personnel and property and the sustainable development of national economy.

[0003] At present, modern high-precision imaging technologies with multi-scale imaging resolution, such as X-ray computed tomography topology imaging (XCT), electron backscatter diffraction (EBSD) and focused ion beam scanning electron microscope (FIB-SEM), have greatly promoted the research on the microstructure of geotechnical engineering materials. Such technologies can clearly and intuitively capture the micro-pore, crack and mineral matrix structure characteristics of rock, and further study the micro-mechanical properties and fracture mechanism of rock. However, the study of the geometric structure characteristics and physical and mechanical properties of rock at a specific scale can only reveal the fracture behavior and mechanism of rock at the current scale, which leads to the inability to form an effective unified understanding of the multi-scale rock material micro-meso-macro full-scale fracture mechanism and mechanical properties, and thus cannot be further applied to engineering scale research, which seriously affects the development of deep resource exploitation technology and deep engineering disaster monitoring and prevention technology. The main deficiency of the existing technology is that it cannot establish the correlation characteristics of rock physical and mechanical parameters at the micro-meso-macro scale, which leads to the inability to effectively characterize the heterogeneous characteristics of rock physical and mechanical parameters. SUMMARY

[0004] The present application provides a rock cross-scale parameter correlation and geometric feature matching method and device based on machine learning, a storage medium and an electronic device, which can establish the correlation characteristics of rock physical and mechanical parameters at the micro-meso-macro scale and match the geometric structure characteristics, effectively characterizing the heterogeneous characteristics of rock physical and mechanical parameters.

[0005] The present application provides a rock cross-scale parameter correlation and geometric feature matching method based on machine learning, which comprises:

[0006] Generating a rock microstructure digital image based on a rock micro sample;

[0007] The digital image of the rock microstructure is input into the rock microstructure phase segmentation model to obtain the segmented rock microstructure phases;

[0008] A rock micro-mesoscopic parameter correlation model is constructed, and the rock mesoscopic parameters are obtained based on the rock microstructure phase and the rock micro-mesoscopic parameter correlation model;

[0009] A deep learning generalized generative model for rocks across micro-, meso-, and macro scales is constructed, and the rock meso-parameter parameters are matched with the rock geometric features to obtain a rock meso-parameter parameter-geometric feature matching model.

[0010] A rock micro-macro parameter correlation model is constructed, and the rock macro parameters are obtained based on the rock microstructure phase and the rock micro-macro parameter correlation model.

[0011] Based on the aforementioned rock micro-meso-macro multi-scale deep learning generalized generative model, the rock macroscopic parameters are matched with the rock geometric structure features to obtain a rock macroscopic parameter-geometric feature matching model.

[0012] Furthermore, in the aforementioned machine learning-based method for cross-scale parameter association and geometric feature matching of rocks, the step of inputting the digital image of the rock microstructure into a rock microstructure phase segmentation model to obtain the segmented rock microstructure phases includes:

[0013] The digital image of the rock microstructure is input into the rock microstructure phase segmentation model, and feature extraction, upsampling and convolution operations are performed to obtain the rock microstructure prediction and classification results.

[0014] The predicted classification results of the rock microstructure are digitally encoded to obtain the divided rock microstructure phases.

[0015] Furthermore, in the aforementioned machine learning-based method for cross-scale parameter association and geometric feature matching of rocks, the step of inputting the digital image of the rock microstructure into a rock microstructure facies segmentation model, performing feature extraction, upsampling, and convolution operations respectively to obtain the rock microstructure prediction and classification result, is expressed by the following formula:

[0016]

[0017] in, For the first i Layer feature map, For upsampling, , For the side convolution kernel function, For convolution operations and the first layer of the underlying network i Layer feature map, Indicates the first i In each convolutional layer, the input feature map Perform convolution operations. The output is the predicted classification result of rock microstructure.

[0018] Furthermore, in the above-mentioned machine learning-based method for cross-scale parameter association and geometric feature matching of rocks, the rock micro-mesoscopic parameter association model is as follows:

[0019]

[0020] in, For the microscopic parameters of the rock, For the first i The physical parameters of the microstructure phase of the rock itself. For the first i The content of each phase in the microstructure of the rock. This represents the number of phases in the rock's microstructure.

[0021] Furthermore, in the aforementioned machine learning-based method for cross-scale parameter association and geometric feature matching of rocks, the rock micro-macro parameter association model is as follows:

[0022]

[0023]

[0024] in, For parameters characterizing micro- and macro-level heterogeneity, These are macroscopic parameters of the rock.

[0025] Furthermore, the above-mentioned machine learning-based method for cross-scale parameter association and geometric feature matching of rocks further includes:

[0026] Based on the aforementioned micro-macro heterogeneous characterization parameters, a probability density function is calculated, and based on the probability density function, the distribution information and value information of each basic unit of the rock macro-parameter-geometric feature matching model are determined.

[0027] Furthermore, in the above-mentioned machine learning-based method for cross-scale parameter association and geometric feature matching of rocks, the probability density function is:

[0028]

[0029] in, x It is the smallest basic unit of the rock macroscopic parameter-geometric feature matching model.

[0030] Further, the above-mentioned machine learning-based rock cross-scale parameter correlation and geometric feature matching method, wherein the joint loss function of the training process of the rock micro-meso-macro cross-scale deep learning generalized generative model is:

[0031]

[0032] wherein, 、 、 、 are the predicted rock macroscopic parameters, the real rock macroscopic parameters, the predicted rock mesoscopic parameters, and the real rock mesoscopic parameters, respectively.

[0033] The embodiment of the present application also provides a machine learning-based rock cross-scale parameter correlation and geometric feature matching device, comprising:

[0034] A rock microstructure digital image generation module is configured to generate a rock microstructure digital image based on a rock microsample.

[0035] A rock microstructure phase segmentation module is configured to input the rock microstructure digital image into a rock microstructure phase segmentation model to obtain a divided rock microstructure phase.

[0036] A rock mesoscopic parameter calculation module is configured to construct a rock micro-meso parameter correlation model and obtain rock mesoscopic parameters based on the rock microstructure phase and the rock micro-meso parameter correlation model.

[0037] A rock mesoscopic parameter-geometric feature matching module is configured to construct a rock micro-meso-macro cross-scale deep learning generalized generative model, match the rock mesoscopic parameters with rock geometric structure features, and obtain a rock mesoscopic parameter-geometric feature matching model.

[0038] A rock macroscopic parameter calculation module is configured to construct a rock meso-macro parameter correlation model and obtain rock macroscopic parameters based on the rock microstructure phase and the rock meso-macro parameter correlation model.

[0039] A rock macroscopic parameter-geometric feature matching module is configured to match the rock macroscopic parameters with rock geometric structure features based on the rock micro-meso-macro cross-scale deep learning generalized generative model and obtain a rock macroscopic parameter-geometric feature matching model.

[0040] The embodiment of the present application also provides a computer readable storage medium, wherein a plurality of instructions are stored in the computer readable storage medium, and the instructions are suitable for being loaded by a processor to execute any one of the above-mentioned machine learning-based rock cross-scale parameter correlation and geometric feature matching methods.

[0041] The embodiment of the present application also provides an electronic device, comprising a processor and a memory, the processor is electrically connected with the memory, the memory is used for storing instructions and data, and the processor is used for the steps in the rock cross-scale parameter correlation and geometric feature matching method based on machine learning.

[0042] The present application provides a rock cross-scale parameter correlation and geometric feature matching method based on machine learning, device, storage medium and electronic equipment, the present application first divides rock microstructure phase, constructs micro-mesoscopic parameter correlation model to calculate rock mesoscopic parameter, constructs rock mesoscopic-macroscopic parameter correlation model to calculate rock macroscopic parameter, and then constructs rock micro-mesoscopic-macroscopic cross-scale deep learning generalized generative model, performs rock mesoscopic structure and parameter feature matching to generate rock mesoscopic parameter-geometric feature matching model, and performs rock macroscopic unit and parameter feature matching to generate rock macroscopic parameter-geometric feature matching model. The present application greatly improves the identification efficiency and accuracy of microstructure phase, realizes the cross-scale correlation of rock micro-mesoscopic-macroscopic physical and mechanical parameters, breaks the limitation of rock geometric structure scale, improves the correlation accuracy of heterogeneous rock geometric structure representation and multi-scale physical and mechanical parameters, and significantly improves the accuracy and efficiency of numerical simulation results. BRIEF DESCRIPTION OF DRAWINGS

[0043] The technical scheme and other beneficial effects of the present application will be apparent from the following detailed description of the specific embodiments of the present application combined with the accompanying drawings.

[0044] Figure 1 The flowchart of the rock cross-scale parameter correlation and geometric feature matching method based on machine learning provided by the embodiment of the present application.

[0045] Figure 2 The schematic diagram of the sandstone microstructure segmented by the rock microstructure phase segmentation model provided by the embodiment of the present application.

[0046] Figure 3 The flowchart of matching the rock mesoscopic parameter and rock geometric structure feature provided by the embodiment of the present application.

[0047] Figure 4 The flowchart of matching the rock macroscopic parameter and rock geometric structure feature provided by the embodiment of the present application.

[0048] Figure 5 The structural schematic diagram of the rock cross-scale parameter correlation and geometric feature matching device based on machine learning provided by the embodiment of the present application.

[0049] Figure 6 The structural schematic diagram of the electronic device provided by the embodiment of the present application. DETAILED DESCRIPTION

[0050] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by a person skilled in the art without creative work fall within the protection scope of the present application.

[0051] At present, modern high-precision imaging technologies with multi-scale imaging resolution, such as X-ray computed tomography topology imaging (XCT), electron backscatter diffraction (EBSD) and focused ion beam scanning electron microscope (FIB-SEM), greatly promote the research on the microstructure of geotechnical engineering materials. Such technologies can clearly and intuitively capture the micro-pore, crack and mineral matrix structure characteristics of rocks, and then deeply study the micro-mechanical properties and failure mechanism of rocks. However, the research on the geometric structure characteristics and physical and mechanical properties of rocks at a specific scale can only reveal the failure behavior and mechanism of rocks at the current scale, which leads to the inability to form an effective unified understanding of the multi-scale rock material micro-meso-macro full-scale failure mechanism and mechanical properties, and thus cannot be further applied to engineering scale research, which seriously affects the development of deep resource mining technology and deep engineering disaster monitoring and prevention technology. The main shortcomings are as follows:

[0052] (1) The existing technology cannot establish the correlation characteristics of the physical and mechanical parameters of rocks at the micro-meso-macro scale, which leads to the inability to effectively characterize the heterogeneous characteristics of the physical and mechanical parameters of rocks.

[0053] (2) The existing technology cannot establish the characteristic matching relationship between the irregular geometric structure and the physical and mechanical parameters of rocks at multiple scales, which leads to the inability to accurately establish a multi-scale heterogeneous calculation model consistent with the structure of rocks.

[0054] (3) The existing numerical simulation technology has a large difference between the physical and mechanical parameters of rocks and the geometric model and the actual situation, which leads to low calculation precision and poor accuracy, and cannot provide high-precision construction design parameters for actual engineering.

[0055] To solve the above problems, the present application provides a rock cross-scale parameter correlation and geometric feature matching method and device based on machine learning, a storage medium and an electronic device. The rock cross-scale parameter correlation and geometric feature matching device based on machine learning provided by the present application can be integrated in an electronic device, which can be a terminal, a server or the like. The terminal can include a tablet computer, a notebook computer, a personal computer (PC), a micro processing box or other devices, etc.

[0056] Please refer to Figure 1, Figure 1 The flowchart illustrates a machine learning-based method for cross-scale parameter association and geometric feature matching of rocks, applicable to electronic devices. This method includes the following steps:

[0057] S1, generating digital images of rock microstructure based on rock microsamples.

[0058] Specifically, modern high-precision imaging technology is used to capture and transparently image the microscopic patterns of rocks without damage, generating digital images of the rock's microstructure.

[0059] In a specific example, a rock microstructure specimen (specimen size: diameter = 5 mm, height = 10 mm) was prepared. Modern high-precision imaging technology (such as X-ray computed tomography topological imaging technology with an imaging resolution of 2 μm) was used to capture the rock microstructure in a non-destructive and refined manner and to create a transparent image, generating a digital image of the rock microstructure.

[0060] S2, input the digital image of the rock microstructure into the rock microstructure phase segmentation model to obtain the segmented rock microstructure phases.

[0061] Specifically, based on the deep learning U-Net segmentation model, a rock microstructure facies segmentation model (FPN-U-Net) based on the feature pyramid algorithm is established.

[0062] In one embodiment, step S2 includes the following steps:

[0063] S21. Input the digital image of the rock microstructure into the rock microstructure phase segmentation model, and perform feature extraction, upsampling and convolution operations to obtain the rock microstructure prediction and classification results.

[0064] Specifically, it is expressed by the following formula:

[0065]

[0066] in, For the first i Layer feature map, For upsampling, , For the side convolution kernel function, For convolution operations and the first layer of the underlying network i Layer feature map, Indicates the first i In each convolutional layer, the input feature map Perform convolution operations. The output is the predicted classification result of rock microstructure.

[0067] The rock microstructure facies segmentation model is iteratively trained based on a fractal dimension loss function. The expression for the fractal dimension loss function is as follows:

[0068]

[0069] in, For the loss function based on fractal dimension, To predict fractal dimension, This represents the true fractal dimension.

[0070] Furthermore, the difference between the predicted rock microstructure classification results and the actual rock microstructure images is measured by calculating a contour similarity parameter, thus improving the accuracy of rock microstructure identification and segmentation. The formula for calculating contour similarity is:

[0071]

[0072] in, For contour similarity parameters, For image dimensions, and The images show the predicted classification results of rock microstructure and the actual classification images of rock microstructure, respectively.

[0073] S22, the predicted classification results of rock microstructure are digitally encoded to obtain the divided rock microstructure phases.

[0074] Figure 2 This is a schematic diagram illustrating the segmentation of sandstone microstructure using a rock microstructure phase segmentation model, as provided in an embodiment of the present invention. Figure 2 As shown, the original sandstone microstructure digital map was segmented using a rock microstructure facies segmentation model to obtain the predicted classification results of the sandstone microstructure. As an example, a total of four microstructure facies were identified, namely 0-porosity facies, 1-feldspar facies, 2-quartz facies, and 3-mica facies.

[0075] S3. Construct a rock micro-mesoscopic parameter correlation model and obtain rock mesoscopic parameters based on rock microstructure phase and rock micro-mesoscopic parameter correlation model.

[0076] The correlation model between the microscopic and mesoscopic parameters of rocks is as follows:

[0077]

[0078] in, For the microscopic parameters of the rock, For the first i The physical parameters of the microstructure phase of the rock itself. For the first i The content of each phase in the microstructure of the rock. The number of rock microstructure phases.

[0079] As a specific example, the micro-meso-elastic modulus of the rock (one of the rock meso-parameters) can be calculated by the rock micro-meso parameter correlation model, and the elastic model and content of each phase are respectively 0-pore phase (0GPa, 0.15), 1-feldspar phase (7GPa, 0.55), 2-quartz phase (8GPa, 0.2), 3-mica phase (15GPa, 0.1), and thus the rock meso-elastic model calculated by the rock micro-meso parameter correlation model is 0*0.15+7*0.55+8*0.2+15*0.1=6.97GPa.

[0080] S4, constructing a rock micro-meso-macro cross-scale deep learning generalized generative model, matching the rock meso-parameters with the rock geometric structure features, and obtaining a rock meso-parameter-geometric feature matching model.

[0081] Figure 3 The flowchart provided by the embodiment of the present application matches the rock meso-parameters with the rock geometric structure features. It can be referred to Figure 3 .

[0082] Specifically, the generalized generative network (GAN) is taken as a basic framework, the feature pyramid (FPN) algorithm is used to identify and extract the rock microstructure features, and the rock micro-meso-macro cross-scale deep learning generalized generative (DL-GAN) model is established.

[0083] Specifically, the rock meso-parameters (as the real rock meso-parameters) and the rock microstructure phases in step S2 are input into the rock micro-meso-macro cross-scale deep learning generalized generative model to obtain the predicted rock meso-parameters. The rock micro-meso-macro cross-scale deep learning generalized generative model is trained by the real rock meso-parameters and the joint loss function, the batch size, the learning rate and the iteration number of the DL-GAN model are adjusted, and the standard deviation, the contour similarity coefficient and the accuracy parameters are analyzed. 30% of the preprocessed rock microstructure digital image data set is taken as an input parameter to obtain the optimal DL-GAN model. 70% of the rock microstructure digital image data set with digital coding is taken as an input parameter, and the optimal DL-GAN model is used for testing and verification to obtain the DL-GAN model with the best generalization ability.

[0084] The joint loss function of the training process of the rock micro-meso-macro cross-scale deep learning generalized generative model is:

[0085]

[0086] wherein, 、 、 、 are predicted rock macro parameters, real rock macro parameters, predicted rock meso parameters, real rock meso parameters, respectively.

[0087] The standard deviation of the fractal dimension is:

[0088]

[0089] wherein, is a square root function.

[0090] The accuracy parameter is:

[0091]

[0092] wherein, TP, TN, FP, FN are true positive, true negative, false positive, false negative, respectively.

[0093] Then, the rock meso-macro parameter correlation model is constructed, and the rock macro parameters are obtained based on the rock microstructure phase and the rock meso-macro parameter correlation model.

[0094] S5, constructing a rock meso-macro parameter correlation model, obtaining rock macro parameters based on rock microstructure phase and rock meso-macro parameter correlation model.

[0095] The rock meso-macro parameter correlation model is:

[0096]

[0097]

[0098] wherein, is a meso-macro heterogeneity characterization parameter, is a rock macro parameter.

[0099] Further, the method further comprises: calculating a probability density function based on the above meso-macro heterogeneity characterization parameter, and determining distribution information of each basic unit and value information of each unit of the rock macro parameter-geometric feature matching model based on the probability density function.

[0100] The probability density function is:

[0101]

[0102] wherein, x is a smallest basic unit of the rock macro parameter-geometric feature matching model, is a Weibull probability density function.

[0103] S6, based on the rock micro-meso-macro cross-scale deep learning generalized generative model, the rock macroscopic parameter and the rock geometric structure feature are matched, and the rock macroscopic parameter-geometric feature matching model is obtained.

[0104] Figure 4 The flowchart for matching the rock macroscopic parameter and the rock geometric structure feature provided by the embodiment of the present application is shown in Figure 4 . Similarly, the rock macroscopic parameter is also predicted by the rock micro-meso-macro cross-scale deep learning generalized generative model, and then the rock macroscopic parameter and the structure feature in the rock microstructure phase are associated to generate the rock macroscopic parameter-geometric feature matching model.

[0105] The present application first divides the rock microstructure phase, constructs the micro-meso parameter correlation model to calculate the rock mesoscopic parameter, constructs the rock meso-macro parameter correlation model to calculate the rock macroscopic parameter, and then constructs the rock micro-meso-macro cross-scale deep learning generalized generative model to match the rock mesoscopic structure and the parameter feature to generate the rock mesoscopic parameter-geometric feature matching model, and to match the rock macroscopic unit and the parameter feature to generate the rock macroscopic parameter-geometric feature matching model. The present application has the following beneficial technical effects:

[0106] (1) Breaking the rock structure scale limit, accurately correlating the rock micro-meso-macro physical and mechanical parameters;

[0107] (2) The rock heterogeneity distribution shape parameter and scale parameter can be directly obtained, and the rock macroscopic scale parameterization calculation model can be quickly generated;

[0108] (3) The correlation accuracy of the heterogeneous rock geometric structure representation and the multi-scale physical and mechanical parameters is improved, and finally the rock numerical simulation calculation efficiency and accuracy are improved.

[0109] According to the method described in the above embodiment, the present embodiment will be further described from the perspective of the rock cross-scale parameter correlation and geometric feature matching device based on machine learning. The rock cross-scale parameter correlation and geometric feature matching device based on machine learning can be realized as an independent entity, or can be integrated in an electronic device, which can be a terminal, a server, etc. The terminal can include a tablet computer, a notebook computer, a personal computer (PC), a micro processing box, or other devices, etc.

[0110] Please refer to Figure 5 , Figure 5The device for correlating rock cross-scale parameters and matching geometric features based on machine learning provided by the embodiment of the application is specifically described, applied to an electronic device, and can include:

[0111] A rock microstructure digital image generation module is configured to generate a rock microstructure digital image based on a rock microsample;

[0112] A rock microstructure phase segmentation module is configured to input the rock microstructure digital image into a rock microstructure phase segmentation model to obtain a segmented rock microstructure phase;

[0113] A rock meso-scale parameter calculation module is configured to construct a rock micro-meso-scale parameter correlation model and obtain rock meso-scale parameters based on the rock microstructure phase and the rock micro-meso-scale parameter correlation model;

[0114] A rock meso-scale parameter-geometric feature matching module is configured to construct a rock micro-meso-macro-scale deep learning generalized generative model, match the rock meso-scale parameters and rock geometric structure features, and obtain a rock meso-scale parameter-geometric feature matching model;

[0115] A rock macro-scale parameter calculation module is configured to construct a rock meso-macro-scale parameter correlation model and obtain rock macro-scale parameters based on the rock microstructure phase and the rock meso-macro-scale parameter correlation model;

[0116] A rock macro-scale parameter-geometric feature matching module is configured to match the rock macro-scale parameters and rock geometric structure features based on the rock micro-meso-macro-scale deep learning generalized generative model and obtain a rock macro-scale parameter-geometric feature matching model.

[0117] In specific implementation, each of the above modules and / or units can be implemented as an independent entity, or can be combined as the same or several entities, and the specific implementation of each of the above modules and / or units can refer to the method embodiments above, and the beneficial effects that can be achieved can also refer to the beneficial effects in the method embodiments above, which will not be described here again.

[0118] In addition, the embodiment of the application further provides an electronic device, which can be a computer, a tablet computer or the like. The electronic device can implement the steps in any embodiment of the method for correlating rock cross-scale parameters and matching geometric features based on machine learning provided by the embodiment of the application, and thus can achieve the beneficial effects of any method for correlating rock cross-scale parameters and matching geometric features based on machine learning provided by the embodiment of the application. Details can refer to the embodiments above, which will not be described here again.

[0119] Figure 6A specific structural block diagram of an electronic device provided by an embodiment of the present application is shown, which can be used to implement the machine learning based rock cross-scale parameter correlation and geometric feature matching method provided in the above embodiments. The electronic device 500 can be a terminal, a server, or the like, wherein the terminal can include a tablet computer, a notebook computer, a personal computer (PC), a micro processing box, or other devices, etc.

[0120] The RF circuit 510 is configured to receive and send electromagnetic waves, and to convert the electromagnetic waves and electrical signals to each other, so as to communicate with a communication network or other devices. The RF circuit 510 can include various existing circuit elements for performing these functions, such as an antenna, a radio frequency transceiver, a digital signal processor, an encryption / decryption chip, a subscriber identity module (SIM) card, a memory, and the like. The RF circuit 510 can communicate with various networks, such as the Internet, an intranet, a wireless network, or other devices through the wireless network. The wireless network can include a cellular telephone network, a wireless local area network or metropolitan area network. The wireless network can use various communication standards, protocols and technologies, including but not limited to Global System for Mobile Communication (GSM), Enhanced Data GSM Environment (EDGE), Wideband Code Division Multiple Access (WCDMA), Code Division Access (CDMA), Time Division Multiple Access (TDMA), Wireless Fidelity (Wi-Fi) (such as Institute of Electrical and Electronics Engineers (IEEE) 802.11a, 802.11b, 802.11g and / or 802.11n standards), Voice over Internet Protocol (VoIP), Worldwide Interoperability for Microwave Access (Wi-Max), other protocols for email, instant messaging and short message service, and any other suitable communication protocol, even including those not yet developed as of the date of this application.

[0121] The memory 520 can be used to store software programs and modules, such as the corresponding program instructions / modules in the above-described embodiments, and the processor 580 can execute various functions and data processing by running the software programs and modules stored in the memory 520, i.e., realize functions such as front camera shooting, processing of the shot image, and switching of display color of the display content on the display screen. The memory 520 can include a high-speed random access memory, and can further include a non-volatile memory such as one or more magnetic storage devices, flash memories, or other non-volatile solid-state memories. In some examples, the memory 520 can further include memories disposed remotely with respect to the processor 580, which can be connected to the electronic device 500 through a network. Examples of the network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0122] The input unit 530 can be used to receive inputted digital or character information, and generate a keyboard, a mouse, and the like related to user settings and function control.

[0123] The display unit 540 can be used to display information inputted by the user or provided to the user, and various graphical user interfaces which can be constituted by graphics, text, icons, video, and any combination thereof. The display unit 540 can include a display panel 541, which can be configured in the form of an LCD (Liquid Crystal Display), an OLED (Organic Light-Emitting Diode), or the like.

[0124] The audio circuit 560, the speaker 561, and the microphone 562 can provide an audio interface between the user and the electronic device 500. The audio circuit 560 can convert received audio data into an electrical signal, transmit the electrical signal to the speaker 561, and convert the electrical signal into a sound signal outputted by the speaker 561; on the other hand, the microphone 562 can convert a sound signal collected into an electrical signal, and the audio circuit 560 can convert the electrical signal into audio data, output the audio data to the processor 580 for processing, and then transmit the audio data to another terminal through the RF circuit 510, or output the audio data to the memory 520 for further processing. The audio circuit 560 can further include an earphone jack to provide communication of an external earphone with the electronic device 500.

[0125] The electronic device 500 can help the user to receive requests, transmit information, and the like through the transmission module 570 (e.g., a Wi-Fi module), which provides the user with wireless broadband Internet access. Although the transmission module 570 is shown, it can be understood that it does not belong to the essential components of the electronic device 500, and can be omitted as needed without changing the essence of the application.

[0126] The processor 580 is a control center of the electronic device 500 that uses various interfaces and lines to connect the entire mobile phone with various parts, performs various functions of the electronic device 500 and processes data by running or executing software programs and / or modules stored in the memory 520 and calling data stored in the memory 520, thereby monitoring the entire electronic device. Optionally, the processor 580 can include one or more processing cores; in some embodiments, the processor 580 can integrate an application processor and a modem processor, in which the application processor mainly processes an operating system, a user interface, and an application program, and the modem processor mainly processes wireless communication. Understandably, the above-mentioned modem processor can also not be integrated into the processor 580.

[0127] The electronic device 500 further includes a power supply 590 (such as a battery) for supplying power to various components, and in some embodiments, the power supply can be logically connected to the processor 580 through a power management system, so as to realize functions such as management of charging, discharging, and power consumption management through the power management system. The power supply 590 can also include one or more than one direct or alternating current power supply, a recharging system, a power failure detection circuit, a power converter or inverter, a power state indicator, and any other components.

[0128] Although not shown, the electronic device 500 further includes a camera (such as a front camera or a rear camera), a Bluetooth module, and the like, which are not described here in detail. Specifically, in the present embodiment, the display unit of the electronic device is a touch screen display, and the mobile terminal further includes a memory and one or more than one program, wherein the one or more than one program is stored in the memory and is configured to be executed by the one or more than one processor, and the one or more than one program includes instructions for performing the following operations:

[0129] Generating a rock microstructure digital image based on a rock micro sample;

[0130] Inputting the rock microstructure digital image into a rock microstructure phase segmentation model to obtain a divided rock microstructure phase;

[0131] Constructing a rock micro-meso parameter correlation model, and obtaining a rock meso parameter based on the rock microstructure phase and the rock micro-meso parameter correlation model;

[0132] Constructing a rock micro-meso-macro cross-scale deep learning generalized generative model, matching the rock meso parameter with a rock geometric structure feature to obtain a rock meso parameter-geometric feature matching model;

[0133] constructing a rock meso-macro parameter correlation model, obtaining rock macro parameters based on the rock microstructure phases and the rock meso-macro parameter correlation model;

[0134] Based on the rock micro-meso-macro scale deep learning generalized generative model, the rock macro parameters are matched with the rock geometric structure characteristics to obtain a rock macro parameter-geometric feature matching model.

[0135] In specific implementation, the above various modules can be implemented as independent entities, or can be combined as the same or several entities, and the specific implementation of the above various modules can be referred to the method embodiments above, which will not be described here.

[0136] Those skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructions, or by related hardware controlled by the instructions, which can be stored in a computer readable storage medium and loaded and executed by a processor. Therefore, the embodiment of the present application provides a storage medium, which stores a plurality of instructions, which can be loaded by a processor to execute the steps of any embodiment of the rock cross-scale parameter correlation and geometric feature matching method based on machine learning provided by the embodiment of the present application.

[0137] The computer readable storage medium can include read only memory (ROM), random access memory (RAM), magnetic disk or optical disk, etc.

[0138] Since the instructions stored in the storage medium can execute the steps in any embodiment of the rock cross-scale parameter correlation and geometric feature matching method based on machine learning provided by the embodiment of the present application, the beneficial effects that can be achieved by any rock cross-scale parameter correlation and geometric feature matching method based on machine learning provided by the embodiment of the present application can be achieved, which will be described in detail in the foregoing embodiments, and will not be described here.

[0139] The rock cross-scale parameter correlation and geometric feature matching method, device, storage medium and electronic equipment based on machine learning provided by the embodiment of the present application are described in detail, and the principle and implementation manner of the present application are described by applying specific examples; the above embodiment is only used to help understand the method and its core idea; meanwhile, for those skilled in the art, according to the idea of the present application, the specific implementation manner and application range will be changed, and the above description should not be understood as the limitation of the present application.

Claims

1. A method for cross-scale parameter correlation and geometric feature matching of rocks based on machine learning, characterized in that, The method includes: Digital images of rock microstructure generated from rock micro samples; The digital image of the rock microstructure is input into the rock microstructure phase segmentation model to obtain the segmented rock microstructure phases; A rock micro-mesoscopic parameter correlation model is constructed, and the rock mesoscopic parameters are obtained based on the rock microstructure phase and the rock micro-mesoscopic parameter correlation model; The rock micro-mesoscopic parameter correlation model is as follows: in, For the microscopic parameters of the rock, For the first i The physical parameters of the microstructure phase of the rock itself. For the first i The content of each phase in the microstructure of the rock. The number of phases in the rock's microstructure; A deep learning generalized generative model for rocks across micro-, meso-, and macro scales is constructed, and the rock meso-parameter parameters are matched with the rock geometric features to obtain a rock meso-parameter parameter-geometric feature matching model. A rock micro-macro parameter correlation model is constructed, and the rock macro parameters are obtained based on the rock microstructure phase and the rock micro-macro parameter correlation model. The rock micro-macro parameter correlation model is as follows: in, For parameters characterizing micro- and macro-level heterogeneity, Macroscopic parameters of rocks; Based on the aforementioned rock micro-meso-macro multi-scale deep learning generalized generative model, the rock macroscopic parameters are matched with the rock geometric structure features to obtain a rock macroscopic parameter-geometric feature matching model.

2. The method for cross-scale parameter association and geometric feature matching of rocks based on machine learning according to claim 1, characterized in that, The step of inputting the digital image of the rock microstructure into the rock microstructure phase segmentation model to obtain the segmented rock microstructure phases includes: The digital image of the rock microstructure is input into the rock microstructure phase segmentation model, and feature extraction, upsampling and convolution operations are performed to obtain the rock microstructure prediction and classification results. The predicted classification results of the rock microstructure are digitally encoded to obtain the divided rock microstructure phases.

3. The method for cross-scale parameter association and geometric feature matching of rocks based on machine learning according to claim 2, characterized in that, The digital image of the rock microstructure is input into the rock microstructure phase segmentation model, and feature extraction, upsampling, and convolution operations are performed to obtain the rock microstructure prediction and classification result, which is expressed by the following formula: in, For the first i Layer feature map, For upsampling, For convolution kernel function, For the side convolution kernel function, For convolution operations and the first layer of the underlying network i Layer feature map, Indicates the first i In each convolutional layer, the input feature map Perform convolution operations. The output is the predicted classification result of rock microstructure.

4. The method for cross-scale parameter association and geometric feature matching of rocks based on machine learning according to claim 1, characterized in that, The method further includes: Based on the aforementioned micro-macro heterogeneous characterization parameters, a probability density function is calculated, and based on the probability density function, the distribution information and value information of each basic unit of the rock macro-parameter-geometric feature matching model are determined.

5. The method for cross-scale parameter association and geometric feature matching of rocks based on machine learning according to claim 4, characterized in that, The probability density function is: in, x It is the smallest basic unit of the rock macroscopic parameter-geometric feature matching model.

6. The method for cross-scale parameter association and geometric feature matching of rocks based on machine learning according to claim 1, characterized in that, The joint loss function for the training process of the aforementioned rock micro-meso-macro multi-scale deep learning generalized generative model is: in, , , , These are, respectively, the predicted macroscopic parameters of rocks, the actual macroscopic parameters of rocks, the predicted microscopic parameters of rocks, and the actual microscopic parameters of rocks.

7. A machine learning-based device for cross-scale parameter association and geometric feature matching of rocks, wherein the machine learning-based device for cross-scale parameter association and geometric feature matching of rocks is used to implement the machine learning-based method for cross-scale parameter association and geometric feature matching of rocks according to claim 1, characterized in that, include: A digital image generation module for rock microstructure is used to generate digital images of rock microstructure based on rock microsamples. The rock microstructure phase segmentation module is used to input the digital image of the rock microstructure into the rock microstructure phase segmentation model to obtain the segmented rock microstructure phases; The rock microstructure parameter calculation module is used to construct a rock microstructure-microstructure parameter correlation model and obtain rock microstructure parameters based on the rock microstructure phase and the rock microstructure-microstructure parameter correlation model. The rock micro-parameter-geometric feature matching module is used to construct a rock micro-micro-macro-scale deep learning generalized generative model, which matches the rock micro-parameters with the rock geometric structure features to obtain the rock micro-parameter-geometric feature matching model. The rock macroscopic parameter calculation module is used to construct a rock micro-macroscopic parameter correlation model and obtain the rock macroscopic parameters based on the rock microstructure phase and the rock micro-macroscopic parameter correlation model. The rock macroscopic parameter-geometric feature matching module is used to match the rock macroscopic parameters with the rock geometric structure features based on the rock micro-micro-macro-scale deep learning generalized generative model, so as to obtain the rock macroscopic parameter-geometric feature matching model.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a plurality of instructions adapted for loading by a processor to execute the machine learning-based rock cross-scale parameter association and geometric feature matching method according to any one of claims 1 to 6.

Images