A rock failure mechanism discrimination method based on three-dimensional digital image and acoustic emission

Abstract

The present application relates to rock failure mechanism identification technical field, a kind of rock failure mechanism discrimination method based on three-dimensional digital image and acoustic emission, comprising: based on the test rock test piece to construct multi-source collaborative test platform, utilize multi-source collaborative test platform to obtain axial loading data set, speckle strain table set and acoustic emission waveform signal set, according to speckle strain table set and axial loading data set calculate shear modulus, based on speckle strain table set and shear modulus to construct global plastic shear strain energy curve and local plastic shear strain energy curve, signal analysis is carried out to acoustic emission waveform signal set, obtain RA variation curve, AF variation curve and AE energy curve.The present application can improve the calculation accuracy of rock energy evolution and the reliability of failure mechanism discrimination, reduce engineering safety risk.

Description

Technical Field

[0001] This invention relates to the field of rock failure mechanism identification technology, and in particular to a method for identifying rock failure mechanisms based on three-dimensional digital images and acoustic emission. Background Technology

[0002] With the continuous growth of my country's national economy, numerous geotechnical engineering projects, such as mines, highways, and water conservancy and hydropower infrastructure, have been constructed on a large scale. However, these facilities are at risk of engineering disasters such as faulting and surface subsidence under the influence of external environmental factors and loads. Among these risks, the deformation and failure of rock mass is a key factor affecting the safety of these projects. The failure of rock mass is a complex process from local to overall failure, and rock mass failure is mainly plastic failure. The evolution of local deformation has extremely important significance for the macroscopic mechanical behavior of rock.

[0003] Traditional techniques primarily employ point-based measurements such as extensometers and resistance strain gauges combined with qualitative analysis of acoustic emission parameters. This is achieved through simple correlation between characteristic parameters such as local strain monitoring and acoustic emission counting. However, this technique has significant drawbacks: it cannot precisely monitor the overall displacement and local crack evolution of the rock surface, leading to inaccurate energy calculations. Furthermore, the lack of a unified energy characterization method makes it difficult to quantitatively correlate acoustic emission energy with deformation energy, thus limiting the accurate identification of damage mechanisms. Summary of the Invention

[0004] This invention provides a method for identifying rock failure mechanisms based on three-dimensional digital images and acoustic emission. Its main purpose is to improve the calculation accuracy of rock energy evolution and the reliability of failure mechanism identification, thereby reducing engineering safety risks.

[0005] To achieve the above objectives, this invention provides a method for identifying rock failure mechanisms based on three-dimensional digital images and acoustic emission, comprising:

[0006] Obtain raw rock specimens and perform speckle pretreatment on the raw rock specimens to obtain test rock specimens, wherein the surface of the test rock specimens includes multiple speckle marks;

[0007] A multi-source collaborative testing platform was constructed based on test rock specimens. The multi-source collaborative testing platform includes: a triaxial experimental system, a 3D-DIC system, and an acoustic emission monitoring system.

[0008] Triaxial compression experiments were conducted on test rock specimens using a multi-source collaborative testing platform to obtain axial loading datasets, speckle strain tables, and acoustic emission waveform signal sets.

[0009] Calculate the shear modulus based on the speckle strain table and the axial loading dataset;

[0010] Global and local plastic shear strain energy curves are constructed based on speckle strain tables and shear modulus.

[0011] Based on the acoustic emission waveform signal set, global plastic shear strain energy curve, and local plastic shear strain energy curve, rock failure mechanism discrimination based on three-dimensional digital image and acoustic emission was completed.

[0012] Optionally, the triaxial compression test on the test rock specimen using a multi-source collaborative testing platform to obtain an axial loading dataset, a speckle strain table set, and an acoustic emission waveform signal set includes:

[0013] The multi-source collaborative testing platform is started synchronously to obtain the startup test platform, and the compression time is recorded;

[0014] Experimental data were acquired from the test rock specimen based on the compression time and the start-up test platform, and axial loading data, speckle strain gauge and acoustic emission waveform signal were obtained.

[0015] The update time is calculated based on the preset detection interval and compression time.

[0016] The update time is used as the compression time, and the process returns to the step of collecting experimental data on the test rock specimen based on the compression time and starting the test platform, until the test rock specimen is destroyed.

[0017] The axial loading data, speckle strain table, and acoustic emission waveform signal are summarized to obtain the axial loading dataset, speckle strain table set, and acoustic emission waveform signal set.

[0018] Optionally, the experimental data acquisition of the test rock specimen based on the compression time and the start-up of the test platform to obtain axial loading data, speckle strain gauges, and acoustic emission waveform signals includes:

[0019] At the compression moment, the test rock specimen is compressed using the triaxial experimental system in the start-up test platform and the preset test confining pressure to obtain axial loading data, where the axial loading data is axial stress;

[0020] During the compression process, the 3D-DIC system in the start-up test platform is used to detect the strain of multiple speckle marks in the test rock specimen to obtain a speckle strain table. The speckle strain table includes multiple speckle strain data, and each speckle strain data corresponds to a speckle mark. The speckle strain data includes: transverse axis strain and longitudinal axis strain.

[0021] During the compression process, the acoustic emission waveform signal of the test rock specimen is acquired by the acoustic emission monitoring system in the start-up test platform.

[0022] Optionally, the step of calculating the shear modulus based on the speckle strain table set and the axial loading dataset includes:

[0023] Extract the speckle strain tables sequentially from the speckle strain tables set, and identify the loading data at the same moment in the axial loading dataset based on the speckle strain tables.

[0024] Calculate the rate of change of horizontal and vertical displacements based on the speckle strain gauge.

[0025] Calculate the overall shear strain based on the rate of change of displacement along the horizontal axis and the rate of change of displacement along the vertical axis.

[0026] Determine the current confining pressure corresponding to the speckle strain table, and calculate the overall shear stress based on the axial stress in the loading data at the same moment and the current confining pressure.

[0027] By summing up the overall shear stress and overall shear strain respectively, we obtain the overall shear stress set and the overall shear strain set;

[0028] The overall stress-strain curve is constructed based on the overall shear stress set and the overall shear strain set. The overall stress-strain curve is then fitted to obtain the shear modulus.

[0029] Optionally, the construction of global plastic shear strain energy curves and local plastic shear strain energy curves based on speckle strain tables and shear modulus includes:

[0030] The speckle strain table is divided into directions to obtain a horizontal axis strain dataset and a vertical axis strain dataset. The horizontal axis strain dataset includes multiple horizontal axis strain data, and the vertical axis strain dataset includes multiple vertical axis strain data. Both the horizontal axis strain data and the vertical axis strain data correspond to a compression time.

[0031] A horizontal strain matrix is ​​constructed based on the horizontal strain dataset, where each column of the horizontal strain matrix represents a horizontal strain data point in the horizontal strain dataset.

[0032] Construct the vertical axis strain matrix based on the vertical axis strain dataset;

[0033] The speckle shear strain matrix is ​​calculated based on the longitudinal strain matrix and the transverse strain matrix, wherein the shear strain matrix includes multiple speckle shear strains;

[0034] An elastic shear strain matrix is ​​constructed using the shear modulus and the overall shear stress set, wherein the elastic shear strain matrix is ​​of the same type as the speckle shear strain matrix;

[0035] Calculate the plastic shear strain matrix based on the elastic shear strain matrix and the speckle shear strain matrix;

[0036] The plastic shear strain column set in the plastic shear strain matrix was identified. The plastic shear strain column set includes multiple plastic shear strain columns, and the plastic shear strain columns in the plastic shear strain column set are arranged in order of acquisition time.

[0037] Global plastic shear strain energy curves are obtained based on the plastic shear strain set;

[0038] The local plastic shear strain energy curve is constructed based on the preset final failure time and speckle strain table.

[0039] Optionally, the step of constructing the elastic shear strain matrix using the shear modulus and the overall shear stress set includes:

[0040] The speckle shear strain is extracted sequentially from the speckle shear strain matrix to determine the detection timestamp of the speckle shear strain.

[0041] Based on the detection timestamp, identify the shear stress at the same moment in the overall shear stress concentration;

[0042] Calculate elastic shear strain based on simultaneous shear stress and shear modulus;

[0043] The elastic shear strain is summarized to obtain the elastic shear strain set, and the elastic shear strain matrix is ​​constructed based on the elastic shear strain set.

[0044] Optionally, obtaining the global plastic shear strain energy curve based on the plastic shear strain set includes:

[0045] Extract the plastic shear strain columns sequentially from the plastic shear strain column set;

[0046] Identify the preceding shear strain column of the extracted plastic shear strain column in the set of plastic shear strain columns, and calculate the plastic shear strain increment column based on the preceding shear strain column and the extracted plastic shear strain column.

[0047] The plastic shear strain energy increment series is calculated based on the plastic shear strain increment series and the shear stress at the same moment corresponding to the plastic shear strain series.

[0048] The global plastic shear strain energy is obtained by summing the increments of the plastic shear strain energy.

[0049] The global plastic shear strain energies are summarized to obtain the global plastic shear strain energy set, which includes multiple global plastic shear strain energies, and each global plastic shear strain energy represents a different compression time.

[0050] Curve fitting is performed based on the global plastic shear strain energy set to obtain the global plastic shear strain energy curve.

[0051] Optionally, the step of constructing the local plastic shear strain energy curve based on the preset final failure time and speckle strain table includes:

[0052] The failure strain table is extracted from the speckle strain table set based on the final failure time. The failure strain table is the speckle strain table that is the last one in the speckle strain table set.

[0053] A multimodal strain verification model is constructed, wherein the multimodal strain verification model is a neural network model;

[0054] Based on the final destruction time, the destroyed waveform signal is identified in the acoustic emission waveform signal set, and features are extracted from the destroyed waveform signal to obtain the waveform signal feature set;

[0055] The waveform signal feature set and the failure strain gauge are input into the multimodal strain verification model to obtain the accurate values ​​of the strain gauge.

[0056] If the accurate value of the strain gauge is not less than the preset accurate threshold, then multiple speckle strain data in the failure strain gauge are filtered according to the preset maximum strain threshold to obtain multiple failure strain data.

[0057] Local plastic shear strain energy curves were constructed based on multiple failure strain data and speckle strain tables.

[0058] Optionally, the construction of the local plastic shear strain energy curve based on multiple failure strain data and speckle strain tables includes:

[0059] Multiple failure markers corresponding to multiple failure strain data were identified from multiple speckle markers;

[0060] Data filtering of the speckle strain table set based on multiple failure markers yields a failure speckle strain table set.

[0061] Local plastic shear strain energy curves are obtained using a broken speckle strain table.

[0062] To achieve the above objectives, the present invention also provides a rock failure mechanism discrimination system based on three-dimensional digital images and acoustic emission, comprising:

[0063] The rock specimen marking module is used to obtain the original rock specimen, perform speckle pretreatment on the original rock specimen to obtain the test rock specimen, wherein the surface of the test rock specimen includes multiple speckle marks;

[0064] The compressed data acquisition module is used to construct a multi-source collaborative testing platform based on the test rock specimen. The multi-source collaborative testing platform includes a triaxial experimental system, a 3D-DIC system, and an acoustic emission monitoring system. The multi-source collaborative testing platform is used to conduct triaxial compression experiments on the test rock specimen to obtain axial loading datasets, speckle strain tables, and acoustic emission waveform signal sets.

[0065] The shear strain energy analysis module is used to calculate the shear modulus based on the speckle strain table set and the axial loading dataset;

[0066] The shear strain analysis module is used to construct global and local plastic shear strain energy curves based on speckle strain tables and shear modulus.

[0067] To address the above problems, the present invention also provides an electronic device, the electronic device comprising:

[0068] Memory, storing at least one instruction;

[0069] The processor executes the instructions stored in the memory to implement the rock failure mechanism discrimination method based on three-dimensional digital images and acoustic emission described above.

[0070] To address the aforementioned problems, the present invention also provides a computer-readable storage medium storing at least one instruction, which is executed by a processor in an electronic device to implement the rock failure mechanism discrimination method based on three-dimensional digital images and acoustic emission described above.

[0071] To address the problems described in the background art, this invention first constructs a multi-source collaborative testing platform based on test rock specimens. This step, by integrating triaxial loading, optical measurement, and acoustic emission monitoring systems, achieves synchronous data acquisition, overcoming the timing inconsistencies caused by the independent operation of each system in existing technologies. This provides a foundational platform for subsequent energy correlation analysis. Furthermore, this scheme constructs global and local plastic shear strain energy curves based on speckle strain tables and shear modulus. This step, for the first time, uses 3D-DIC full-field displacement data to refine the calculation of plastic shear strain energy, achieving energy evolution characterization from local to global levels. This solves the problem that traditional methods such as extensometers or resistance strain gauges lack the ability to perform such calculations. This invention addresses the issue of monitoring local crack energy and identifying failure mechanisms through the consistency of local curves and acoustic emission (AE) energy. It qualitatively identifies tensile-shear failure modes using RA-AF analysis and, combined with AE energy curves and 3D-DIC energy coupling, reveals the intrinsic relationship between acoustic emission parameters and deformation energy. This overcomes the shortcomings of existing acoustic emission technologies that focus only on counting or amplitude and lack quantitative energy inference. By establishing the aforementioned curves, simultaneous quantitative analysis of 3D-DIC plastic shear strain energy and AE energy is achieved, clarifying the energy distribution law under different confining pressures. This provides a more scientific and accurate evaluation basis for rock engineering safety and overcomes the limitation of existing technologies lacking a unified energy characterization method. Therefore, this invention, for the first time, uses refined quantitative calculation of deformation energy based on full-field rock displacement data and couples it with acoustic emission energy. From an energy perspective, it monitors the entire process of rock deformation and failure from local to global levels, establishes a unified and refined energy characterization method, and thus identifies rock tensile-shear failure modes, providing a fundamental theoretical basis for engineering safety. Attached Figure Description

[0072] Figure 1 This is a flowchart illustrating a rock failure mechanism discrimination method based on three-dimensional digital images and acoustic emission, provided in an embodiment of the present invention.

[0073] Figure 2 This is a functional block diagram of a rock failure mechanism discrimination system based on three-dimensional digital images and acoustic emission, provided in an embodiment of the present invention.

[0074] Figure 3 This is a schematic diagram of the structure of an electronic device for implementing the rock failure mechanism discrimination method based on three-dimensional digital images and acoustic emission, as provided in an embodiment of the present invention.

[0075] Explanation of reference numerals in the attached figures:

[0076] 10. Electronic device; 11. Processor; 12. Memory; 13. Bus.

[0077] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0078] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0079] This application provides a method for identifying rock failure mechanisms based on three-dimensional digital images and acoustic emission. The executing entity of this method includes, but is not limited to, at least one electronic device that can be configured to execute the method provided in this application, such as a server or a terminal. In other words, the method can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster.

[0080] Reference Figure 1 The diagram shown is a flowchart illustrating a rock failure mechanism discrimination method based on three-dimensional digital images and acoustic emission according to an embodiment of the present invention. In this embodiment, the rock failure mechanism discrimination method based on three-dimensional digital images and acoustic emission includes:

[0081] S1. Obtain the original rock specimen and perform speckle pretreatment on the original rock specimen to obtain the test rock specimen. The surface of the test rock specimen includes multiple speckle marks.

[0082] It is clear that the original rock specimen refers to the rock used for subsequent triaxial compression tests. The test rock specimen refers to the original rock specimen after speckle pretreatment. The speckle pretreatment involves uniformly applying black waterproof adhesive to the surface of the original rock specimen, allowing it to air dry, spraying a white base coat, and then using matte black paint to create a uniform speckle pattern (this uniform speckle pattern consists of multiple speckle markers). Optionally, the spraying density during the speckle pretreatment process is 5-10 spots / mm², with a spot diameter of 0.1-0.3 mm. The purpose of this speckle pretreatment is to create a high-contrast speckle field, ensuring that the 3D-DIC system can effectively track the surface displacement and strain of the specimen through image discrimination algorithms, thus guaranteeing the accuracy and reliability of the experimental data. The speckle markers refer to the black spots sprayed on the specimen surface, used as reference points for image discrimination in 3D-DIC monitoring.

[0083] For example, the preparation method of the above-mentioned test rock specimen is as follows: a piece of sandstone that needs to be damaged is cut into a cylindrical specimen of 25mm×50mm. This 25mm×50mm cylindrical specimen is the original rock specimen. Then, the two ends of the original rock specimen are finely ground with sandpaper to ensure that the dimensional accuracy and flatness of the two ends of the original rock specimen meet the standards of the International Society for Rock Mechanics (ISRM), and to ensure that the geometry and surface quality of the original rock specimen meet the experimental requirements, thereby reducing experimental errors. Furthermore, the original rock specimens after fine grinding underwent the following pretreatment: First, a thin layer of black waterproof adhesive was evenly applied to the surface of the cut cylindrical specimens and allowed to air dry naturally. Then, a layer of white base color was sprayed onto the surface of the cylindrical specimens, followed by a layer of uniform black spots sprayed with matte black paint to form a speckle field. The density of the matte black paint spraying was 5-10 spots / mm², and the spot diameter was 0.1-0.3mm, ensuring that the speckle pattern met the grayscale contrast requirements of the 3D-DIC image discrimination algorithm. After the speckle field was allowed to air dry naturally, the sprayed cylindrical specimens were placed in a vacuum pump and vacuumed to ensure that the speckle field adhered firmly to the surface of the specimens, thus obtaining the test rock specimens.

[0084] S2. Construct a multi-source collaborative testing platform based on the test rock specimen. The multi-source collaborative testing platform includes a triaxial experimental system, a 3D-DIC system, and an acoustic emission monitoring system.

[0085] It should be explained that the multi-source collaborative testing platform refers to a collaborative testing system integrating triaxial loading, optical measurement, and acoustic emission monitoring. This system is used to simultaneously acquire data on the mechanical response, surface deformation, and internal damage of rock specimens during the stress process. The triaxial experimental system refers to a mechanical servo device capable of applying axial pressure and confining pressure to simulate the triaxial stress state of rocks under different geological conditions, and to control the loading rate (e.g., 10⁻⁻⁶). 4The 3D-DIC system achieves progressive failure. The 3D-DIC system refers to a non-contact optical measurement system based on three-dimensional digital image correlation technology. This system can acquire speckle images of the specimen surface from different angles using multiple cameras and calculate the full-field displacement and strain field using image discrimination algorithms. The acoustic emission monitoring system refers to an acoustic emission sensor capable of monitoring the elastic wave signals generated during rock fracture in real time. The setup method for testing rock specimens using the aforementioned multi-source collaborative testing platform is as follows: The rock specimen is placed in the cavity of the triaxial experimental system, and an acoustic emission probe is attached to the surface of the cavity. The cavity with the acoustic emission probe attached is then placed on the triaxial test platform of the triaxial experimental system (the acoustic emission monitoring system receives acoustic emission signals through the acoustic emission probe). The position of the cavity is adjusted so that it is located in the middle of three sets of high-speed cameras (i.e., the 3D-DIC system). Each set of high-speed cameras is a binocular camera to facilitate monitoring the three-dimensional deformation image of the rock specimen. Then, hydrostatic pressure is applied to the triaxial experimental system, and the channels of the acoustic emission monitoring system are adjusted. Once the hydrostatic pressure reaches the preset target value and the acoustic emission signal received by the acoustic emission monitoring system is noise-free, the triaxial experimental system begins loading, the 3D-DIC system records images, and the acoustic emission monitoring system records collision events and maintains the monitoring data. This completes the setup of the aforementioned multi-source collaborative testing platform.

[0086] Furthermore, the above-mentioned calculation of the full-field displacement and strain field using image discrimination algorithms involves: calculating the strain field of the surface speckle image of the acquired rock specimen using the image processing and analysis software XDTIC-3D Digital Speckle Strain Measurement and Analysis System. Specifically, the analysis calculation area to be analyzed in the rock specimen is first determined. Then, a set of binocular cameras (referred to as high-speed camera 1 and high-speed camera 2) capable of capturing the analysis calculation area is selected from the three sets of high-speed cameras of the aforementioned 3D-DIC system. The analysis calculation area is delineated in the surface speckle image acquired by high-speed camera 1, and seed points are selected within this analysis calculation area. Subsequently, a speckle image matching algorithm is used to select the seed points in the surface speckle image acquired by high-speed camera 2. The matching of the seed points is completed, thereby establishing the correspondence between seed points with the same position in the images captured by the binocular camera. Based on the above correspondence, the relevant matching is carried out sequentially on multiple surface speckle images under different states in the analysis and calculation area of ​​high-speed camera 1, and the matching operation of multiple surface speckle images under different states corresponding to high-speed camera 2 is completed simultaneously. Finally, the accurate matching of speckle images in all states of the binocular camera is achieved. Through the above matching operation, combined with the principle of binocular stereo vision and image matching technology, the strain in the three directions of XYZ (i.e., Dis-X, Dis-Y, Dis-Z) of the surface of the analysis and calculation area captured by high-speed camera 1, as well as the overall displacement field (Dis-E) and multiple strain fields, can be calculated by the software XDIC.

[0087] For example, when constructing a multi-source collaborative testing platform, the test rock specimen is first installed on the test bench of the triaxial experimental system, and an acoustic emission probe is attached to the surface of the test rock specimen and ensured good coupling. Then, the confining pressure and loading parameters of the triaxial experimental system, the acquisition frequency of the 3D-DIC system (e.g., 1 frame / s) and the acoustic emission monitoring parameters are set so that each system starts synchronously until the specimen is destroyed, thereby realizing collaborative data acquisition.

[0088] Importantly, in addition to the aforementioned 3D-DIC system, this solution can also incorporate strain gauges and fiber Bragg gratings to detect the deformation of various speckle marks on the surface of the rock specimen. The purpose of introducing multiple strain detection methods is that, since the 3D-DIC system can only detect the surface deformation of the rock specimen and cannot detect internal damage, and during strain detection using the 3D-DIC system, factors such as illumination may cause discrepancies between the detected strain data (i.e., speckle strain gauges) and the actual speckle strain. Therefore, this solution constructs a multimodal strain verification model based on multiple detection methods (i.e., strain gauges and fiber Bragg gratings). This model fits multiple different strain detection methods (including the 3D-DIC system), thereby significantly reducing the deviation between the 3D-DIC system's strain detection and the actual situation. The specific construction method of the multimodal strain verification model will be discussed in detail later.

[0089] S3. Triaxial compression experiments were conducted on the test rock specimens using a multi-source collaborative testing platform to obtain axial loading datasets, speckle strain tables, and acoustic emission waveform signal sets.

[0090] It is clear that the axial loading dataset refers to the set of mechanical data of the test rock specimen recorded by the triaxial testing system during the triaxial compression experiment. The speckle strain table set refers to the set of strain data of the test rock specimen recorded by the 3D-DIC system during the triaxial compression experiment. The acoustic emission waveform signal set refers to the set of acoustic wave data of the test rock specimen recorded by the acoustic emission monitoring system during the triaxial compression experiment.

[0091] In detail, the triaxial compression test on the test rock specimen using the multi-source collaborative testing platform yields an axial loading dataset, a speckle strain table set, and an acoustic emission waveform signal set, including:

[0092] The multi-source collaborative testing platform is started synchronously to obtain the startup test platform, and the compression time is recorded;

[0093] Experimental data were acquired from the test rock specimen based on the compression time and the start-up test platform, and axial loading data, speckle strain gauge and acoustic emission waveform signal were obtained.

[0094] The update time is calculated based on the preset detection interval and compression time.

[0095] The update time is used as the compression time, and the process returns to the step of collecting experimental data on the test rock specimen based on the compression time and starting the test platform, until the test rock specimen is destroyed.

[0096] The axial loading data, speckle strain table, and acoustic emission waveform signal are summarized to obtain the axial loading dataset, speckle strain table set, and acoustic emission waveform signal set.

[0097] Understandably, the "start-up test platform" refers to the multi-source collaborative test platform after startup. When starting the multi-source collaborative test platform, the triaxial experimental system, 3D-DIC system, and acoustic emission monitoring system need to be started simultaneously. The "compression time" refers to the moment when experimental data is subsequently collected from the test rock specimen. This compression time is manually set; for example, if the compression time is set to time A, then at time A, the triaxial experimental system compresses the test rock specimen. The "detection interval" refers to the manually set time interval for experimental data collection. The "update time" refers to the time one detection interval after the compression time. If the test rock specimen is not destroyed, it indicates that the triaxial compression experiment has not yet ended. At this point, experimental data collection needs to be repeated, i.e., the steps of collecting experimental data from the test rock specimen based on the compression time and starting the test platform need to be returned until the test rock specimen is destroyed.

[0098] Furthermore, since the acoustic emission monitoring system can detect the internal damage of the rock specimen, the strain data detected by the 3D-DIC system is subsequently detected using the acoustic emission waveform signal, so that the strain data detected by the 3D-DIC system can accurately represent the actual strain of the rock specimen.

[0099] In detail, the experimental data acquisition of the test rock specimen based on the compression time and the start-up test platform, obtaining axial loading data, speckle strain gauges, and acoustic emission waveform signals, includes:

[0100] At the compression moment, the test rock specimen is compressed using the triaxial experimental system in the start-up test platform and the preset test confining pressure to obtain axial loading data, where the axial loading data is axial stress;

[0101] During the compression process, the 3D-DIC system in the start-up test platform is used to detect the strain of multiple speckle marks in the test rock specimen to obtain a speckle strain table. The speckle strain table includes multiple speckle strain data, and each speckle strain data corresponds to a speckle mark. The speckle strain data includes: transverse axis strain and longitudinal axis strain.

[0102] During the compression process, the acoustic emission waveform signal of the test rock specimen is acquired by the acoustic emission monitoring system in the start-up test platform.

[0103] It should be explained that the test confining pressure refers to the constant lateral pressure applied in the triaxial experiment, used to simulate the stress state of rocks at different geological depths or environments. Its value is preset by the experiment (e.g., 0 MPa, 5 MPa, or 10 MPa), and the test confining pressure can be changed to obtain experimental data under different confining pressures. The axial loading data refers to the axial stress of the test rock specimen detected at the compression moment. In addition to the axial stress, the axial loading data also includes axial strain. The speckle strain table refers to a structured dataset used to store the strain information of each speckle mark during the entire compression process. It is a collection of multiple speckle strain data, where one speckle strain data corresponds to one speckle mark. The speckle strain data refers to the strain value of a specific speckle mark detected by the 3D-DIC system during compression. The 3D-DIC system detects speckle images by simultaneously capturing them with multiple cameras, comparing the grayscale distribution of the images before and after deformation using a digital image correlation algorithm, calculating the displacement field, and deriving the strain field to obtain the strain value of each speckle mark. The process of obtaining strain values ​​from speckle images is existing technology and will not be elaborated here. The horizontal axis strain refers to the strain component of the speckle mark in the horizontal direction (radial or x-axis). The vertical axis strain refers to the strain component of the speckle mark in the vertical direction (axial or y-axis). The acoustic emission waveform signal refers to the elastic wave signal released when cracks are generated and propagate inside the rock during compression. Its waveform includes characteristic parameters such as rise time, amplitude, frequency, and duration. This acoustic emission waveform signal is collected by the acoustic emission monitoring system.

[0104] S4. Calculate the shear modulus based on the speckle strain table and the axial loading dataset.

[0105] Understandably, the shear modulus refers to a material constant obtained by fitting the ratio of shear stress to shear strain in the elastic stage, which can represent the material's ability to resist shear deformation.

[0106] In detail, the calculation of the shear modulus based on the speckle strain table set and the axial loading dataset includes:

[0107] Extract the speckle strain tables sequentially from the speckle strain tables set, and identify the loading data at the same moment in the axial loading dataset based on the speckle strain tables.

[0108] Calculate the rate of change of horizontal and vertical displacements based on the speckle strain gauge.

[0109] Calculate the overall shear strain based on the rate of change of displacement along the horizontal axis and the rate of change of displacement along the vertical axis.

[0110] Determine the current confining pressure corresponding to the speckle strain table, and calculate the overall shear stress based on the axial stress in the loading data at the same moment and the current confining pressure.

[0111] By summing up the overall shear stress and overall shear strain respectively, we obtain the overall shear stress set and the overall shear strain set;

[0112] The overall stress-strain curve is constructed based on the overall shear stress set and the overall shear strain set. The overall stress-strain curve is then fitted to obtain the shear modulus.

[0113] It should be explained that the simultaneous loading data refers to the axial loading data during the compression process corresponding to the same compression moment as the speckle strain gauge. The transverse displacement change rate refers to the change rate of all transverse strains in the speckle strain gauge, used to represent the deformation of the test rock specimen in the transverse (radial) direction. It is calculated by extracting all transverse strains from the speckle strain gauge and calculating the average value of all transverse strains; this average value is the transverse displacement change rate. The longitudinal displacement change rate refers to the change rate of all longitudinal strains in the speckle strain gauge, used to represent the deformation of the test rock specimen in the longitudinal (axial) direction. Its calculation method is the same as the transverse displacement change rate, and will not be repeated here. The overall shear strain refers to the overall shear strain of the test rock specimen, used to represent the degree of shear deformation of the specimen. The overall shear strain is calculated as the absolute difference between the transverse displacement change rate and the longitudinal displacement change rate. The current confining pressure refers to the test confining pressure at the time the speckle strain gauge was acquired.

[0114] Furthermore, the overall shear stress refers to the magnitude of the shear stress experienced by the specimen, used to represent the shear load. The overall shear stress is calculated as the absolute difference between the axial stress and the current confining pressure. The overall stress-strain curve refers to the curve showing the change between the overall shear stress set and the overall shear strain set, with the horizontal axis representing the overall shear strain and the vertical axis representing the overall shear stress. Fitting the overall stress-strain curve refers to fitting a straight line to the linear elastic phase of the curve, calculating the slope through linear regression or the least squares method; this slope is the shear modulus.

[0115] S5. Construct global plastic shear strain energy curves and local plastic shear strain energy curves based on speckle strain tables and shear modulus.

[0116] It is clear that the global plastic shear strain energy curve refers to the energy change over time obtained by accumulating the plastic shear strain energy increments of all speckles across the entire field. This global plastic shear strain energy curve is used to represent the energy dissipation evolution process of plastic shear deformation in the specimen over the entire range. The local plastic shear strain energy curve refers to the energy curve obtained by selecting speckles that reach the critical strain value at the final moment and accumulating the plastic shear strain energy increments only in these local areas. This local plastic shear strain energy curve is used to represent the energy concentration and evolution behavior of the specimen in local failure areas (such as macroscopic crack penetration areas). By introducing this local plastic shear strain energy curve, the tensile-shear failure mode of the rock can be more accurately identified.

[0117] Specifically, the construction of global and local plastic shear strain energy curves based on speckle strain tables and shear modulus includes:

[0118] The speckle strain table is divided into directions to obtain a horizontal axis strain dataset and a vertical axis strain dataset. The horizontal axis strain dataset includes multiple horizontal axis strain data, and the vertical axis strain dataset includes multiple vertical axis strain data. Both the horizontal axis strain data and the vertical axis strain data correspond to a compression time.

[0119] A horizontal strain matrix is ​​constructed based on the horizontal strain dataset, where each column of the horizontal strain matrix represents a horizontal strain data point in the horizontal strain dataset.

[0120] Construct the vertical axis strain matrix based on the vertical axis strain dataset;

[0121] The speckle shear strain matrix is ​​calculated based on the longitudinal strain matrix and the transverse strain matrix, wherein the shear strain matrix includes multiple speckle shear strains;

[0122] An elastic shear strain matrix is ​​constructed using the shear modulus and the overall shear stress set, wherein the elastic shear strain matrix is ​​of the same type as the speckle shear strain matrix;

[0123] Calculate the plastic shear strain matrix based on the elastic shear strain matrix and the speckle shear strain matrix;

[0124] The plastic shear strain column set in the plastic shear strain matrix was identified. The plastic shear strain column set includes multiple plastic shear strain columns, and the plastic shear strain columns in the plastic shear strain column set are arranged in order of acquisition time.

[0125] Global plastic shear strain energy curves are obtained based on the plastic shear strain set;

[0126] The local plastic shear strain energy curve is constructed based on the preset final failure time and speckle strain table.

[0127] It should be explained that the horizontal axis strain dataset refers to a collection of multiple horizontal axis strain data, specifically the set of all horizontal axis strains from a specific speckle strain table within the speckle strain table set. The vertical axis strain dataset refers to a collection of multiple vertical axis strain data, specifically the set of all vertical axis strains from a specific speckle strain table within the speckle strain table set. The horizontal axis strain matrix refers to a matrix composed of the horizontal axis strain datasets, where each row represents all horizontal axis strains from a single horizontal axis strain dataset, and each column represents the horizontal axis strain of all speckle-marked strains at the same compression moment. The vertical axis strain matrix refers to a matrix composed of the vertical axis strain datasets, which is of the same type as the horizontal axis strain matrix and will not be described further here.

[0128] For example, time The transverse strains of the three speckle markers (ID1, ID2, ID3) were 0.0008, 0.0009, and 0.0007, respectively, over time. The values ​​for time are 0.0010, 0.0011, and 0.0009 respectively, therefore the horizontal axis strain dataset is {time}. [0.0008, 0.0009, 0.0007], Time If the first row of the horizontal axis strain matrix, composed of this horizontal axis strain dataset, is [0.0010, 0.0011, 0.0009], then the first row is [0.0008, 0.0009, 0.0007] (representing the first compression moment). The horizontal strain of all speckle marks), the first column is: [0.0008, 0.0010] (representing the strain of speckle mark ID1 at compression time). and The transverse strain, i.e., the transverse strain sequence of the speckle at different times.

[0129] It is clear that the speckle shear strain matrix refers to a matrix composed of multiple speckle shear strains. These multiple speckle shear strains represent the shear strains at different compression times for different speckle marks, obtained after calculation using the longitudinal and transverse strain matrices. Each speckle shear strain corresponds to one speckle mark and one compression time. The speckle shear strain matrix is ​​constructed as follows: the longitudinal and transverse strains are extracted sequentially from the longitudinal and transverse strain matrices, respectively. The extracted longitudinal and transverse strains have the same position in their respective matrices. The difference between the extracted longitudinal and transverse strains is calculated, and this difference is the speckle shear strain. All speckle shear strains are obtained in this way, resulting in multiple speckle shear strains. The speckle shear strain matrix is ​​then constructed based on the matrix structure of the longitudinal or transverse strain matrix and the multiple speckle shear strains.

[0130] Furthermore, the elastic shear strain matrix refers to a matrix composed of multiple elastic shear strains, where the elastic shear strain refers to the elastic strain of a speckle mark at a certain compression moment. The plastic shear strain matrix refers to a matrix composed of multiple plastic shear strains, where the plastic shear strain refers to the plastic strain of a speckle mark at a certain compression moment. The plastic shear strain matrix is ​​constructed as follows: subtract the elastic shear strain at the corresponding position in the elastic shear strain matrix from the speckle shear strain in the speckle shear strain matrix to obtain the plastic shear strain. Repeat the above steps to obtain multiple plastic shear strains, and construct the plastic shear strain matrix based on the multiple plastic shear strains.

[0131] It should be explained that the plastic shear strain set refers to a collection of multiple plastic shear strain columns, where a plastic shear strain column is a column in the plastic shear strain matrix. Different plastic shear strain columns represent the set of plastic shear strains for each speckle marker at different compression times. The final failure time refers to the compression time at which the tested rock specimen fails.

[0132] In detail, the construction of the elastic shear strain matrix using the shear modulus and the overall shear stress set includes:

[0133] The speckle shear strain is extracted sequentially from the speckle shear strain matrix to determine the detection timestamp of the speckle shear strain.

[0134] Based on the detection timestamp, identify the shear stress at the same moment in the overall shear stress concentration;

[0135] Calculate elastic shear strain based on simultaneous shear stress and shear modulus;

[0136] The elastic shear strain is summarized to obtain the elastic shear strain set, and the elastic shear strain matrix is ​​constructed based on the elastic shear strain set.

[0137] It is clear that the detection timestamp refers to the compression moment when the speckle shear strain is detected. The simultaneous shear stress refers to the overall shear stress corresponding to the detection timestamp. The calculation of elastic shear strain based on the simultaneous shear stress and shear modulus means dividing the simultaneous shear stress by the shear modulus, and the result is the elastic shear strain. The construction of the elastic shear strain matrix based on the elastic shear strain set means constructing the elastic shear strain set into a matrix form according to the structure of the speckle shear strain matrix, and this matrix is ​​the elastic shear strain matrix.

[0138] Specifically, the acquisition of the global plastic shear strain energy curve based on the plastic shear strain set includes:

[0139] Extract the plastic shear strain columns sequentially from the plastic shear strain column set;

[0140] Identify the preceding shear strain column of the extracted plastic shear strain column in the set of plastic shear strain columns, and calculate the plastic shear strain increment column based on the preceding shear strain column and the extracted plastic shear strain column.

[0141] The plastic shear strain energy increment series is calculated based on the plastic shear strain increment series and the shear stress at the same moment corresponding to the plastic shear strain series.

[0142] The global plastic shear strain energy is obtained by summing the increments of the plastic shear strain energy.

[0143] The global plastic shear strain energies are summarized to obtain the global plastic shear strain energy set, which includes multiple global plastic shear strain energies, and each global plastic shear strain energy represents a different compression time.

[0144] Curve fitting is performed based on the global plastic shear strain energy set to obtain the global plastic shear strain energy curve.

[0145] It should be explained that the preceding shear strain column refers to the plastic shear strain column that is adjacent to and precedes the extracted plastic shear strain column in terms of acquisition time. For example, if the extracted plastic shear strain column is the third plastic shear strain column in the set of plastic shear strain columns, then the preceding shear strain column corresponding to this plastic shear strain column is the second plastic shear strain column in the set of plastic shear strain columns. If the extracted plastic shear strain column is the first plastic shear strain column in the set of plastic shear strain columns, then the subsequent operation steps of this plastic shear strain column are skipped, and the second plastic shear strain column is extracted.

[0146] Furthermore, the plastic shear strain increment series refers to a set of multiple plastic shear strain increments, wherein each plastic shear strain increment is the value obtained by subtracting the plastic shear strain at the corresponding position in the preceding shear strain series from a single plastic shear strain in the extracted plastic shear strain series. For example, the extracted plastic shear strain series is... The preceding shear strain is listed as: [ ], then the plastic shear strain increment is listed as [ , , ],in, , and All are plastic shear strain increments in the plastic shear strain increment column. The plastic shear strain increment column refers to a set of multiple plastic shear strain energy increments, each corresponding one-to-one with a plastic shear strain increment. The plastic shear strain energy increment represents the energy change of a speckle mark due to plastic shear deformation within a specific time interval. The calculation method for this plastic shear strain energy increment is: multiply the shear stress corresponding to the plastic shear strain column at the same instant by the corresponding plastic shear strain increment in the plastic shear strain increment column; the result is the plastic shear strain energy increment. For example, the plastic shear strain increment column is […]. , , The shear stress corresponding to the increment of plastic shear strain at the same instant is: The increment of plastic shear strain energy is then listed as: [ , , ],in, , and All represent the plastic shear strain energy increments in the plastic shear strain energy increment column. The summation operation refers to adding all the plastic shear strain energy increments in the plastic shear strain energy increment column, and the result is the global plastic shear strain energy. This global plastic shear strain energy represents the cumulative value of the plastic shear strain energy increments of all speckle-marked samples during a certain period of the triaxial compression experiment. The curve fitting mentioned above refers to fitting the discrete global plastic shear strain energy data points into a continuous curve using mathematical algorithms to analyze the energy evolution trend. Curve fitting can be performed using methods such as least squares method and linear regression.

[0147] Specifically, the step of constructing a local plastic shear strain energy curve based on a preset final failure time and a speckle strain table includes:

[0148] The failure strain table is extracted from the speckle strain table set based on the final failure time. The failure strain table is the speckle strain table that is the last one in the speckle strain table set.

[0149] A multimodal strain verification model is constructed, wherein the multimodal strain verification model is a neural network model;

[0150] Based on the final destruction time, the destroyed waveform signal is identified in the acoustic emission waveform signal set, and features are extracted from the destroyed waveform signal to obtain the waveform signal feature set;

[0151] The waveform signal feature set and the failure strain gauge are input into the multimodal strain verification model to obtain the accurate values ​​of the strain gauge.

[0152] If the accurate value of the strain gauge is not less than the preset accurate threshold, then multiple speckle strain data in the failure strain gauge are filtered according to the preset maximum strain threshold to obtain multiple failure strain data.

[0153] Local plastic shear strain energy curves were constructed based on multiple failure strain data and speckle strain tables.

[0154] It is clear that the failure strain table refers to the speckle strain table corresponding to the final failure time. The maximum strain threshold is a manually set strain critical value used to distinguish between failure and non-failure regions. If the horizontal or vertical strain of the speckle mark is greater than the maximum strain threshold, it indicates that the speckle mark has undergone significant plastic deformation or failure at the final failure time, and the speckle mark can be recorded as a failure mark. The process involves filtering multiple speckle strain data in the failure strain table according to the preset maximum strain threshold: speckle strain data are extracted sequentially from multiple speckle strain data. If a horizontal or vertical strain greater than the maximum strain threshold appears in the speckle strain data, the speckle strain data is recorded as failure strain data. The failure strain data are then summarized to obtain multiple failure strain data.

[0155] Furthermore, the multimodal strain verification model refers to a pre-constructed neural network model for verifying the accuracy of the failure strain table. Optionally, a long short-term memory network can be used as this multimodal strain verification model. If the strain data in the failure strain table has a large error, it will lead to a large error in the subsequent local plastic shear strain energy curve. The accuracy value of the strain table refers to the probability that the model training label corresponding to the failure strain table is accurate in DIC detection. If this probability is not less than the manually set minimum probability (i.e., the accuracy threshold), it indicates that the strain data in the failure strain table is accurate and can be used to construct the local plastic shear strain energy curve. Otherwise, the accuracy value of this strain table needs to be uploaded to the relevant operators for decision-making (such as changing the lighting environment and re-conducting the experiment). The aforementioned waveform signal feature set refers to the set of waveform features that can represent the failure waveform signal. This waveform signal feature set includes, but is not limited to: time-domain features (such as skewness, kurtosis, waveform factor, impulse factor, etc.), frequency-domain features (such as center frequency, peak frequency, etc.), and time-frequency-domain features (such as wavelet energy coefficients extracted through wavelet transform, wavelet packet node energy, etc.).

[0156] Importantly, the construction method of the aforementioned multimodal strain verification model is as follows: Multimodal strain detection is performed on the rock specimen. The multimodal detection methods include, but are not limited to, strain gauges, DIC (Diverterless Computation), and fiber Bragg gratings, thereby obtaining strain data corresponding to multiple different detection methods. During this detection process, acoustic emission waveform signals of the rock specimen are acquired in real time, and feature extraction is performed on the acoustic emission waveform signals to obtain a waveform feature set for subsequent training (the extraction method is the same as for the waveform signal feature set). The average strain data corresponding to multiple different detection methods is calculated to obtain the average strain data. Then, the strain data deviation between the average strain data and the strain data acquired through DIC is calculated (using methods such as root mean square error). If the strain data deviation is greater than the manually set maximum deviation, the inaccurate DIC detection is recorded as the model training label; otherwise, the accurate DIC detection is recorded as the model training label. The strain data corresponding to DIC, the model training labels, and the waveform feature set used for training are merged to obtain training data. The above steps are repeated to obtain a large amount of training data. The selected neural network is then subjected to supervised training using this large amount of training data to obtain the multimodal strain verification model.

[0157] Specifically, the construction of the local plastic shear strain energy curve based on multiple failure strain data and speckle strain tables includes:

[0158] Multiple failure markers corresponding to multiple failure strain data were identified from multiple speckle markers;

[0159] Data filtering of the speckle strain table set based on multiple failure markers yields a failure speckle strain table set.

[0160] Local plastic shear strain energy curves are obtained using a broken speckle strain table.

[0161] Understandably, the failure marker refers to the speckle mark corresponding to the failure strain data. The data filtering of the speckle strain table based on multiple failure markers means: only retaining data related to failure markers in the speckle strain table; the retained speckle strain table is the failure speckle strain table. The failure speckle strain table refers to the speckle strain table after data filtering. The method for obtaining the local plastic shear strain energy curve is the same as the method for obtaining the global plastic shear strain energy curve, and will not be repeated here.

[0162] Importantly, in addition to obtaining the aforementioned global and local plastic shear strain energy curves, this scheme also performs signal analysis on the acoustic emission waveform signal set to obtain the RA variation curve, AF variation curve, and AE energy curve. The RA variation curve is a curve constructed based on the RA value set, representing the change of the RA value (the ratio of rise time to amplitude) over time. The AF variation curve is a curve constructed based on the AF value set, representing the change of the AF value (the ratio of ring count to duration) over time. The AE energy curve is a curve constructed based on the AE energy set, representing the change of cumulative acoustic emission energy over time.

[0163] In detail, the signal analysis of the acoustic emission waveform signal set to obtain the RA variation curve, AF variation curve, and AE energy curve includes:

[0164] The acoustic emission waveform signals are extracted sequentially from the acoustic emission waveform signal set;

[0165] The acoustic emission waveform signal is analyzed to obtain the rise time, signal amplitude, ring count, and duration.

[0166] Acquiring AE energy based on acoustic emission waveform signals;

[0167] The RA value is calculated based on the rise time and signal amplitude, and the AF value is calculated based on the ring count and duration.

[0168] The RA values, AF values, and AE energies are summarized to obtain the RA value set, AF value set, and AE energy set, respectively. Curve fitting is then performed based on the RA value set, AF value set, and AE energy set to obtain the RA change curve, AF change curve, and AE energy curve.

[0169] Understandably, the AE energy refers to the elastic wave energy released during rock fracturing in the acoustic emission signal. The RA value refers to the ratio of rise time to signal amplitude. The AF value refers to the ratio of ring count to duration. Because the acoustic emission signals released during tensile and shear fracturing differ significantly—tensile fracturing releases greater longitudinal wave energy, with a shorter rise time and higher frequency—while shear fracturing releases greater transverse wave energy, with a longer rise time and lower frequency—RA and AF values ​​are introduced for qualitative analysis of the rock fracturing mechanism: a larger RA value and a smaller AF value during fracturing indicate that shear failure is dominant; conversely, tensile failure is dominant.

[0170] S6. Based on the acoustic emission waveform signal set, global plastic shear strain energy curve and local plastic shear strain energy curve, complete the rock failure mechanism discrimination based on three-dimensional digital image and acoustic emission.

[0171] It should be explained that the RA and AF variation curves reflect the distribution of acoustic emission (RA-AF) values ​​from the start of loading to final failure. Personnel can use these curves to identify the rock failure mechanism, such as tensile or shear failure. Furthermore, by comparing the AE energy curve, the global plastic shear strain energy curve, and the local plastic shear strain energy curve, personnel can conduct a more accurate analysis of the rock failure mechanism.

[0172] To address the problems described in the background art, this invention first constructs a multi-source collaborative testing platform based on test rock specimens. This step, by integrating triaxial loading, optical measurement, and acoustic emission monitoring systems, achieves synchronous data acquisition, overcoming the timing inconsistencies caused by the independent operation of each system in existing technologies. This provides a foundational platform for subsequent energy correlation analysis. Furthermore, this scheme constructs global and local plastic shear strain energy curves based on speckle strain tables and shear modulus. This step, for the first time, uses 3D-DIC full-field displacement data to refine the calculation of plastic shear strain energy, achieving energy evolution characterization from local to global levels. This solves the problem that traditional methods such as extensometers or resistance strain gauges lack the ability to perform such calculations. This invention addresses the issue of monitoring local crack energy and identifying failure mechanisms through the consistency of local curves and acoustic emission (AE) energy. It qualitatively identifies tensile-shear failure modes using RA-AF analysis and, combined with AE energy curves and 3D-DIC energy coupling, reveals the intrinsic relationship between acoustic emission parameters and deformation energy. This overcomes the shortcomings of existing acoustic emission technologies that focus only on counting or amplitude and lack quantitative energy inference. By establishing the aforementioned curves, simultaneous quantitative analysis of 3D-DIC plastic shear strain energy and AE energy is achieved, clarifying the energy distribution law under different confining pressures. This provides a more scientific and accurate evaluation basis for rock engineering safety and overcomes the limitation of existing technologies lacking a unified energy characterization method. Therefore, this invention, for the first time, uses refined quantitative calculation of deformation energy based on full-field rock displacement data and couples it with acoustic emission energy. From an energy perspective, it monitors the entire process of rock deformation and failure from local to global levels, establishes a unified and refined energy characterization method, and thus identifies rock tensile-shear failure modes, providing a fundamental theoretical basis for engineering safety.

[0173] like Figure 2 The diagram shown is a functional block diagram of a rock failure mechanism discrimination system based on three-dimensional digital images and acoustic emission provided in an embodiment of the present invention.

[0174] The rock failure mechanism discrimination system 100 based on three-dimensional digital imaging and acoustic emission described in this invention can be installed in an electronic device. Depending on the functions implemented, the rock failure mechanism discrimination system 100 based on three-dimensional digital imaging and acoustic emission may include a rock specimen marking module 101, a compressed data acquisition module 102, a shear strain energy analysis module 103, and a shear strain energy analysis module 104. The module described in this invention can also be called a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, and is stored in the memory of the electronic device.

[0175] The rock specimen marking module 101 is used to obtain the original rock specimen, perform speckle pretreatment on the original rock specimen, and obtain the test rock specimen, wherein the surface of the test rock specimen includes multiple speckle marks.

[0176] The compressed data acquisition module 102 is used to construct a multi-source collaborative testing platform based on the test rock specimen. The multi-source collaborative testing platform includes a triaxial experimental system, a 3D-DIC system, and an acoustic emission monitoring system. The multi-source collaborative testing platform is used to conduct triaxial compression experiments on the test rock specimen to obtain axial loading dataset, speckle strain table set, and acoustic emission waveform signal set.

[0177] The shear strain energy analysis module 103 is used to calculate the shear modulus based on the speckle strain table set and the axial loading dataset.

[0178] The shear stress analysis module 104 is used to construct global plastic shear stress energy curves and local plastic shear stress energy curves based on the speckle strain table and shear modulus.

[0179] In detail, the modules in the rock failure mechanism discrimination system 100 based on three-dimensional digital images and acoustic emission described in this embodiment of the invention employ the same methods as described above. Figure 1 The method described above is the same as the rock failure mechanism discrimination method based on three-dimensional digital images and acoustic emission, and can produce the same technical effect, so it will not be elaborated here.

[0180] like Figure 3 The diagram shown is a schematic representation of an electronic device for implementing a rock failure mechanism discrimination method based on three-dimensional digital images and acoustic emission, according to an embodiment of the present invention.

[0181] The electronic device 1 may include a processor 10, a memory 11 and a bus 12, and may also include a computer program stored in the memory 11 and capable of running on the processor 10, such as a rock failure mechanism discrimination method program based on three-dimensional digital images and acoustic emission.

[0182] The memory 11 includes at least one type of readable storage medium, such as flash memory, portable hard drive, multimedia card, card-type memory (e.g., SD or DX memory), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 11 can be an internal storage unit of the electronic device 1, such as the portable hard drive of the electronic device 1. In other embodiments, the memory 11 can be an external storage device of the electronic device 1, such as a plug-in portable hard drive, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the electronic device 1. Furthermore, the memory 11 includes both internal storage units and external storage devices of the electronic device 1. The memory 11 can be used not only to store application software and various types of data installed on the electronic device 1, such as the code of a rock failure mechanism discrimination method program based on three-dimensional digital images and acoustic emission, but also to temporarily store data that has been output or will be output.

[0183] In some embodiments, the processor 10 may be composed of integrated circuits, such as a single packaged integrated circuit or multiple integrated circuits with the same or different functions, including combinations of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor 10 is the control unit of the electronic device, connecting various components of the entire electronic device through various interfaces and lines. It executes programs or modules stored in the memory 11 (e.g., a rock failure mechanism discrimination method program based on three-dimensional digital images and acoustic emissions), and calls data stored in the memory 11 to perform various functions of the electronic device 1 and process data.

[0184] The bus 12 can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus 12 can be divided into an address bus, a data bus, a control bus, etc. The bus 12 is configured to realize the connection and communication between the memory 11 and at least one processor 10, etc.

[0185] Figure 3 Only electronic devices with components are shown; it will be understood by those skilled in the art that... Figure 3The structure shown does not constitute a limitation on the electronic device 1, and may include fewer or more components than shown, or combine certain components, or have different component arrangements.

[0186] For example, although not shown, the electronic device 1 may also include a power supply (such as a battery) to power the various components. Preferably, the power supply can be logically connected to the at least one processor 10 through a power management system, thereby enabling functions such as charging management, discharging management, and power consumption management through the power management system. The power supply may also include one or more DC or AC power supplies, recharging systems, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components. The electronic device 1 may also include various sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be described in detail here.

[0187] Furthermore, the electronic device 1 may also include a network interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a Wi-Fi interface, a Bluetooth interface, etc.), which is typically used to establish communication connections between the electronic device 1 and other electronic devices.

[0188] Optionally, the electronic device 1 may further include a user interface, which may be a display, an input unit (such as a keyboard), and optionally, a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen, etc. The display may also be appropriately referred to as a screen or display unit, used to display information processed in the electronic device 1 and to display a visual user interface.

[0189] The program for identifying rock failure mechanisms based on three-dimensional digital images and acoustic emissions, stored in the memory 11 of the electronic device 1, is a combination of multiple instructions. When run in the processor 10, it can achieve the following:

[0190] Obtain raw rock specimens and perform speckle pretreatment on the raw rock specimens to obtain test rock specimens, wherein the surface of the test rock specimens includes multiple speckle marks;

[0191] A multi-source collaborative testing platform was constructed based on test rock specimens. The multi-source collaborative testing platform includes: a triaxial experimental system, a 3D-DIC system, and an acoustic emission monitoring system.

[0192] Triaxial compression experiments were conducted on test rock specimens using a multi-source collaborative testing platform to obtain axial loading datasets, speckle strain tables, and acoustic emission waveform signal sets.

[0193] Calculate the shear modulus based on the speckle strain table and the axial loading dataset;

[0194] Global and local plastic shear strain energy curves are constructed based on speckle strain tables and shear modulus.

[0195] Based on the acoustic emission waveform signal set, global plastic shear strain energy curve, and local plastic shear strain energy curve, rock failure mechanism discrimination based on three-dimensional digital image and acoustic emission was completed.

[0196] Specifically, the processor 10's implementation method for the above instructions can be found in [reference needed]. Figures 1 to 3 The descriptions of the relevant steps in the corresponding embodiments are not repeated here.

[0197] Furthermore, if the modules / units integrated in the electronic device 1 are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. The computer-readable storage medium can be volatile or non-volatile. For example, the computer-readable medium may include: any entity or system capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, or a read-only memory (ROM).

[0198] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor of an electronic device, can perform the following:

[0199] Obtain raw rock specimens and perform speckle pretreatment on the raw rock specimens to obtain test rock specimens, wherein the surface of the test rock specimens includes multiple speckle marks;

[0200] A multi-source collaborative testing platform was constructed based on test rock specimens. The multi-source collaborative testing platform includes: a triaxial experimental system, a 3D-DIC system, and an acoustic emission monitoring system.

[0201] Triaxial compression experiments were conducted on test rock specimens using a multi-source collaborative testing platform to obtain axial loading datasets, speckle strain tables, and acoustic emission waveform signal sets.

[0202] Calculate the shear modulus based on the speckle strain table and the axial loading dataset;

[0203] Global and local plastic shear strain energy curves are constructed based on speckle strain tables and shear modulus.

[0204] Based on the acoustic emission waveform signal set, global plastic shear strain energy curve, and local plastic shear strain energy curve, rock failure mechanism discrimination based on three-dimensional digital image and acoustic emission was completed.

[0205] In the embodiments provided by this invention, it should be understood that the disclosed devices, systems, and methods can be implemented in other ways. For example, the system embodiments described above are merely illustrative, and actual implementations may have other classification methods.

[0206] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0207] Furthermore, the functional modules in the various embodiments of the present invention 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 in the form of hardware plus software functional modules.

[0208] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0209] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A rock failure mechanism discrimination method based on three-dimensional digital images and acoustic emission, characterized in that, The method comprises: An original rock sample is obtained, and speckle pretreatment is performed on the original rock sample to obtain a test rock sample, wherein the surface of the test rock sample comprises a plurality of speckle marks; A multi-source collaborative test platform is constructed based on the test rock sample, wherein the multi-source collaborative test platform comprises a triaxial test system, a 3D-DIC system and an acoustic emission monitoring system; A triaxial compression test is performed on the test rock sample by using the multi-source collaborative test platform to obtain an axial loading data set, a speckle strain table set and an acoustic emission waveform signal set; A shear modulus is calculated according to the speckle strain table set and the axial loading data set; A global plastic shear strain energy curve and a local plastic shear strain energy curve are constructed based on the speckle strain table set and the shear modulus; The global plastic shear strain energy curve and the local plastic shear strain energy curve are constructed based on the speckle strain table set and the shear modulus, comprising: The speckle strain table set is directionally divided to obtain a horizontal axis strain data set and a vertical axis strain data set, wherein the horizontal axis strain data set comprises a plurality of horizontal axis strain data, the vertical axis strain data set comprises a plurality of vertical axis strain data, and the horizontal axis strain data and the vertical axis strain data each correspond to a compression time; A horizontal axis strain matrix is constructed based on the horizontal axis strain data set, wherein each row of the horizontal axis strain matrix represents a horizontal axis strain data in the horizontal axis strain data set; A vertical axis strain matrix is constructed based on the vertical axis strain data set; A speckle shear strain matrix is calculated according to the vertical axis strain matrix and the horizontal axis strain matrix, wherein the speckle shear strain matrix comprises a plurality of speckle shear strains; An elastic shear strain matrix is constructed by using the shear modulus and a global shear stress set, wherein the elastic shear strain matrix is of the same type as the speckle shear strain matrix; A plastic shear strain matrix is calculated according to the elastic shear strain matrix and the speckle shear strain matrix; A plastic shear strain column set in the plastic shear strain matrix is confirmed, wherein the plastic shear strain column set comprises a plurality of plastic shear strain columns, and the plastic shear strain columns in the plastic shear strain column set are arranged in order of time from early to late; A global plastic shear strain energy curve is obtained based on the plastic shear strain column set; A local plastic shear strain energy curve is constructed according to a preset final failure time and the speckle strain table set; The local plastic shear strain energy curve is constructed according to the preset final failure time and the speckle strain table set, comprising: A failure strain table is extracted from the speckle strain table set based on the final failure time, wherein the failure strain table is the last speckle strain table in the speckle strain table set; A multi-modal strain verification model is constructed, wherein the multi-modal strain verification model is a neural network model; A failure waveform signal is identified from the acoustic emission waveform signal set according to the final failure time, and a feature set of the failure waveform signal is obtained by feature extraction on the failure waveform signal; The feature set of the waveform signal and the failure strain table are input into the multi-modal strain verification model to obtain a strain table accurate value; If the strain table accurate value is not less than a preset accurate threshold value, a plurality of failure strain data are obtained by screening a plurality of speckle strain data in the failure strain table according to a preset maximum strain threshold value; A local plastic shear strain energy curve is constructed based on the plurality of failure strain data and the speckle strain table set. Based on the acoustic emission waveform signal set, the global plastic shear strain energy curve and the local plastic shear strain energy curve, rock failure mechanism discrimination based on three-dimensional digital image and acoustic emission is completed.

2. The rock failure mechanism discrimination method based on three-dimensional digital images and acoustic emissions according to claim 1, characterized in that, The triaxial compression experiment of the test rock sample is performed by using the multi-source collaborative test platform to obtain an axial loading data set, a speckle strain table set and an acoustic emission waveform signal set, including: Synchronously starting the multi-source collaborative test platform to obtain a starting test platform and record a compression time; Based on the compression time and the starting test platform, experimental data of the test rock sample is collected to obtain axial loading data, a speckle strain table and acoustic emission waveform signals; Based on a preset detection interval and the compression time, an update time is calculated; The update time is taken as the compression time, and the step of collecting experimental data of the test rock sample based on the compression time and the starting test platform is returned until the test rock sample is damaged; The axial loading data, the speckle strain table and the acoustic emission waveform signals are respectively summarized to obtain the axial loading data set, the speckle strain table set and the acoustic emission waveform signal set.

3. The rock failure mechanism discrimination method based on three-dimensional digital images and acoustic emissions according to claim 2, characterized in that, The experimental data of the test rock sample is collected based on the compression time and the starting test platform to obtain axial loading data, a speckle strain table and acoustic emission waveform signals, including: At the compression time, the test rock sample is compressed by using a triaxial experiment system in the starting test platform and a preset test confining pressure to obtain axial loading data, wherein the axial loading data is axial stress; In the compression process, the speckle strain table is obtained by using a 3D-DIC system in the starting test platform to detect the strain of a plurality of speckle markers in the test rock sample, wherein the speckle strain table includes a plurality of speckle strain data, and each speckle strain data corresponds to a speckle marker, and the speckle strain data includes transverse axis strain and longitudinal axis strain; In the compression process, the acoustic emission waveform signals of the test rock sample are obtained according to an acoustic emission monitoring system in the starting test platform.

4. The rock failure mechanism discrimination method based on three-dimensional digital images and acoustic emissions according to claim 3, characterized by, The shear modulus is calculated based on the speckle strain table set and the axial loading data set, including: In the speckle strain table set, the speckle strain tables are sequentially extracted, and the same time loading data is confirmed from the axial loading data set according to the speckle strain table; The transverse axis displacement change rate and the longitudinal axis displacement change rate are calculated based on the speckle strain table; The overall shear strain is calculated according to the transverse axis displacement change rate and the longitudinal axis displacement change rate; The current confining pressure corresponding to the speckle strain table is determined, and the overall shear stress is calculated according to the axial stress in the same time loading data and the current confining pressure; The overall shear stress and the overall shear strain are respectively summarized to obtain an overall shear stress set and an overall shear strain set; The overall stress strain curve is constructed according to the overall shear stress set and the overall shear strain set, and the shear modulus is obtained by fitting the overall stress strain curve.

5. The rock failure mechanism discrimination method based on three-dimensional digital images and acoustic emissions according to claim 4, characterized in that, The elastic shear strain matrix is constructed by using the shear modulus and the overall shear stress set, including: In the speckle shear strain matrix, the speckle shear strains are sequentially extracted, and the detection time stamp of the speckle shear strain is confirmed; The same time shear stress is identified in the overall shear stress set according to the detection time stamp; The elastic shear strain is calculated based on the same time shear stress and the shear modulus; The elastic shear strain sets are aggregated to obtain an elastic shear strain matrix based on the elastic shear strain sets.

6. The rock failure mechanism discrimination method based on three-dimensional digital images and acoustic emissions according to claim 5, characterized in that, The global plastic shear strain energy curve is obtained based on the plastic shear strain column set, and the obtaining includes: A plastic shear strain column is extracted from the plastic shear strain column set; A previous shear strain column of the extracted plastic shear strain column is identified in the plastic shear strain column set, and a plastic shear strain increment column is calculated according to the previous shear strain column and the extracted plastic shear strain column; A plastic shear strain energy increment column is calculated based on the plastic shear strain increment column and a shear stress corresponding to the plastic shear strain column at the same time; The plastic shear strain energy increment column is summed to obtain global plastic shear strain energy; The global plastic shear strain energy is aggregated to obtain a global plastic shear strain energy set, wherein the global plastic shear strain energy set includes a plurality of global plastic shear strain energies, and each global plastic shear strain energy represents a different compression time; Curve fitting is performed based on the global plastic shear strain energy set to obtain a global plastic shear strain energy curve.

7. The rock failure mechanism discrimination method based on three-dimensional digital images and acoustic emissions according to claim 6, characterized in that, The local plastic shear strain energy curve is obtained based on the plurality of damage strain data and the speckle strain table set, and the obtaining includes: A plurality of damage markers corresponding to the plurality of damage strain data are identified in the plurality of speckle markers; The speckle strain table set is data-filtered based on the plurality of damage markers to obtain a damage speckle strain table set; The local plastic shear strain energy curve is obtained using the damage speckle strain table set.

8. A system for using the rock failure mechanism discrimination method based on three-dimensional digital images and acoustic emission according to any one of claims 1 to 7, characterized in that, The system includes: A rock sample marking module for obtaining an original rock sample, pre-processing the original rock sample to obtain a test rock sample, wherein the surface of the test rock sample includes a plurality of speckle markers; A compression data acquisition module for constructing a multi-source collaborative test platform based on the test rock sample, wherein the multi-source collaborative test platform includes a triaxial test system, a 3D-DIC system, and an acoustic emission monitoring system, and the test rock sample is subjected to a triaxial compression experiment using the multi-source collaborative test platform to obtain an axial loading data set, a speckle strain table set, and an acoustic emission waveform signal set; A shear strain energy analysis module for calculating a shear modulus based on the speckle strain table set and the axial loading data set; A shear energy analysis module for constructing a global plastic shear strain energy curve and a local plastic shear strain energy curve based on the speckle strain table set and the shear modulus.

Images