A method and test device suitable for testing energy of tunnel surrounding rock

Abstract

The application provides a method and a test device suitable for tunnel surrounding rock energy test, wherein the method comprises: performing a loading test on a tunnel surrounding rock test piece, obtaining a stress-strain curve, and determining a mapping relationship between strain and total energy density, elastic strain energy density and plastic dissipation energy density; based on the mapping relationship, a data set is constructed; a strain-energy mapping model is obtained by using a BP neural network; strain characteristic data corresponding to each pixel point in a continuous strain cloud image is input into the trained strain-energy mapping model, and corresponding total energy density values, elastic strain energy density values and plastic dissipation energy density values are output pixel by pixel by the model and are respectively mapped into color values to generate a total strain energy density cloud image, an elastic strain energy density cloud image and a plastic dissipation energy density cloud image. By combining the digital image correlation technology with the strain-energy mapping model, the full-field strain cloud image is converted into an energy cloud image, and direct test support is provided for the energy theory.

Description

Technical Field

[0001] This invention belongs to the field of tunnel engineering technology, specifically relating to a method and testing device suitable for energy testing of surrounding rock in tunnels. Background Technology

[0002] With the rapid development of transportation and underground engineering in my country, tunnel engineering is increasingly extending to deep burial, long distances, and complex geological conditions. Rock deformation control and stability analysis have become core issues in engineering design and construction. Traditional tunnel rock analysis primarily uses stress and displacement as core control indicators. However, energy analysis methods, which only involve scalar calculations, have significant advantages over vector-based stress methods when dealing with complex geological and excavation conditions. They can more intuitively reflect the energy storage, energy release, and failure mechanisms of the surrounding rock, and have gradually become an important direction in tunnel mechanics research.

[0003] Currently, the application of energy methods in tunnel engineering is still mainly based on theoretical derivation and numerical simulation. Indoor model tests are difficult to directly observe and quantify the energy evolution law of surrounding rock, resulting in a lack of experimental verification support for energy theory and limiting its widespread application in engineering practice. In recent years, non-contact strain measurement technologies such as digital image correlation (DIC) have gradually become more widespread, enabling the acquisition of continuous strain across the entire surface of the specimen. However, strain is still a vector physical quantity, which has limitations in the evaluation of overall deformation and energy conversion analysis of surrounding rock.

[0004] Existing tunnel model testing devices primarily focus on loading, support, and deformation monitoring, lacking effective means to convert full-field strain information into energy distribution. This makes it impossible to obtain dynamic evolution cloud maps of the total strain energy, elastic strain energy, and plastic dissipation energy of the surrounding rock during excavation. Furthermore, traditional excavation simulation methods struggle to achieve stable and controllable unloading, easily leading to instantaneous instability of the surrounding rock model and failing to fully capture the continuous changes in strain and energy. Therefore, these problems urgently need to be addressed. Summary of the Invention

[0005] In view of the above-mentioned defects or deficiencies in the prior art, a method and test apparatus suitable for testing the energy of surrounding rock in tunnels are provided.

[0006] Firstly, this application provides a method for energy testing of surrounding rock in tunnels, including... Establish strain-energy mapping relationship: Perform uniaxial or triaxial loading tests on multiple tunnel surrounding rock specimens of the same standard to obtain stress-strain curves. Determine the mapping relationship between strain and total energy density, elastic strain energy density, and plastic dissipation energy density based on the area enclosed by the stress-strain curves. Constructing a training dataset: Based on the mapping relationship, construct a dataset with strain parameters as input and the corresponding total energy density, elastic strain energy density, and plastic dissipation energy density as output; Training the neural network model: The dataset is trained using a BP neural network to obtain a strain-energy mapping model for converting strain data into energy data; Input strain contour map: Obtain continuous strain contour map of the tunnel surrounding rock surface; Point-by-point energy conversion: The strain feature data corresponding to each pixel in the continuous strain cloud map is input into the trained strain-energy mapping model, and the model outputs the corresponding total energy density value, elastic strain energy density value and plastic dissipation energy density value pixel by pixel. Generate energy cloud maps: Map the output total energy density value, elastic strain energy density value and plastic dissipation energy density value to color values ​​respectively to generate total strain energy density cloud maps, elastic strain energy density cloud maps and plastic dissipation energy density cloud maps.

[0007] Furthermore, Obtaining the continuous strain contour map includes the following steps: Specimen preparation: Continuous tunnel surrounding rock specimens are made using cement mortar or 3D printing technology. Tunnel holes are reserved in the middle of the tunnel surrounding rock specimens, and speckles are made on the surface of the tunnel surrounding rock specimens. Specimen installation: A high-strength airbag is filled into the tunnel cavity, the tunnel surrounding rock specimen is installed in the loading device, a preset stress is applied to simulate the geostress field, and it is left to stand for a predetermined time.

[0008] Furthermore, Obtaining the continuous strain contour map further includes the following steps: Acquire baseline image: Start the image acquisition device and capture the initial speckle field on the surface of the tunnel surrounding rock specimen; Simulated tunnel excavation: The high-strength airbag is deflated, and a series of speckle images of the surface of the tunnel surrounding rock specimen are continuously captured. Strain calculation: The image sequence is imported into the digital image analysis module, and a continuous strain cloud map of the surrounding rock surface after tunnel excavation is obtained through calculation.

[0009] Furthermore, The size of the speckle is calculated using the following formula: (one) Among them, L d Indicates speckle size; P d D represents the number of pixels required for speckle; D represents the image scale. The value of D is calculated using the following formula: (two) Among them, P p Indicates the number of pixels occupied by the subject being photographed; L pThis indicates the actual size of the target being photographed.

[0010] Furthermore, The speckle pattern preparation includes the following steps: A white primer was applied to the surface of the tunnel surrounding rock specimen. The primer was required to be of uniform thickness and color and without obvious brushing texture, forming a thin, uniform white base surface. Apply black ink using dotting or spraying methods to create black speckles on a white background.

[0011] Furthermore, Integrating the stress-strain curve, Calculate the area under the loading curve from zero to the current strain, and use it as the total energy density corresponding to the current strain; Calculate the area enclosed by the unloading curve when unloading from the current stress point, and use it as the elastic strain energy density corresponding to the current strain. The difference between the total energy density and the elastic strain energy density is taken as the plastic dissipation energy density corresponding to the current strain.

[0012] Secondly, this application provides a testing apparatus, including... A reaction frame, the reaction frame being rectangular in shape, having the loading device disposed inside; The loading device includes an H-shaped positioning steel frame and a loading mechanism for applying stress to the tunnel surrounding rock specimen; The end of the positioning steel frame is fixedly installed on the reaction frame, forming a positioning cavity and three loading cavities between the steel frame and the reaction frame; The loading mechanism is located inside the loading cavity, and its output end passes through the positioning steel frame and extends into the positioning cavity.

[0013] Furthermore, The positioning cavity is also equipped with a force-transmitting steel plate; The number of force-transmitting steel plates includes three, each connected to the loading mechanism to disperse stress.

[0014] Furthermore, The image acquisition device is located above the reaction frame and includes a camera; The camera is suspended above the positioning cavity by a bracket, and the shooting direction is downward; The bracket is equipped with a gimbal for mounting the camera.

[0015] Furthermore, It also includes the light source; The number of light sources includes multiple sources; Multiple light sources are located outside the field of view to provide supplemental lighting for the tunnel surrounding rock specimen.

[0016] The advantages and positive effects of this application are: This technical solution combines digital image correlation technology with a strain-energy mapping model to transform the full-field strain cloud map into an energy cloud map, enabling dynamic quantification and visualization of the total energy density, elastic strain energy density, and plastic dissipation energy density of the surrounding rock during tunnel excavation and unloading, thus providing direct experimental support for energy theory. Attached Figure Description

[0017] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 A flowchart of a method for energy testing of surrounding rock in tunnels provided in the embodiments of this application; Figure 2 This is a schematic diagram of the structure of the test apparatus for energy testing of surrounding rock in tunnels, provided in an embodiment of this application. Figure 3 This is a schematic diagram of the image acquisition device structure of the test apparatus for energy testing of surrounding rock in tunnels, provided in an embodiment of this application.

[0018] The text labels in the figure are as follows: 100-Reaction frame; 110-Positioning steel frame; 120-Loading mechanism; 130-Force transmission steel plate; 200-Camera; 210-Bracket; 220-Light source. Detailed Implementation

[0019] The present application will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.

[0020] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0021] Firstly, regarding the technical problems mentioned in the background section, this application proposes a method for energy testing of surrounding rock in tunnels, comprising the following steps: S100, establish the strain-energy mapping relationship: Uniaxial or triaxial loading tests were conducted on multiple tunnel surrounding rock specimens of the same standard to obtain stress-strain curves. Based on the area enclosed by the stress-strain curves, the mapping relationship between strain and total energy density, elastic strain energy density, and plastic dissipation energy density was determined.

[0022] In this embodiment, multiple standard cubic specimens were fabricated using the same materials as those used in the tunnel surrounding rock model test. The dimensions and curing conditions of these specimens were kept consistent to ensure the representativeness of the test results.

[0023] In this embodiment, a loading device was used to conduct uniaxial compression tests and triaxial compression tests under different confining pressures on these standard specimens. During the loading process, the corresponding data of stress and strain were recorded to generate complete stress-strain curves.

[0024] In this embodiment, for each stress-strain curve, according to the first law of thermodynamics, the total energy density absorbed by the material during deformation is equal to the area enclosed below the stress-strain curve. Therefore, by calculating this area, the mapping relationship from strain value to total energy density is established. For any point on the curve (corresponding to a specific strain value), the area enclosed by the unloading curve from that point back to the zero-stress state can be calculated to obtain the release elastic strain energy density under that strain state. The difference between the total energy density and the elastic strain energy density is the plastic strain energy density dissipated during deformation. By performing the above analysis on multiple stress-strain curves under different confining pressures, a complete nonlinear mapping relationship database covering both elastic and plastic deformation stages is finally established between strain and total energy density, elastic strain energy density, and plastic dissipated energy density.

[0025] S200, Construct the training dataset: Based on the mapping relationship, a dataset is constructed with strain parameters as input and the corresponding total energy density, elastic strain energy density, and plastic dissipation energy density as output.

[0026] In this embodiment, data is organized and formatted based on the mapping relationship established in S100.

[0027] Specifically, the discrete strain values ​​collected during the loading process of each standard specimen are used as input features, and the total energy density, elastic strain energy density, and plastic dissipation energy density corresponding to each strain value are used as output labels to form a set of basic training samples.

[0028] In this embodiment, to improve the generalization ability and prediction accuracy of the neural network, these basic samples are further augmented. For example, interpolation is performed between adjacent data points to generate more virtual samples covering the entire strain range; or, small-amplitude random noise conforming to a Gaussian distribution is added to the original data to simulate measurement errors present in real experiments, thereby enhancing the robustness of the model. After the above processing, a high-quality training dataset containing thousands or even tens of thousands of samples is finally constructed.

[0029] S300, training the neural network model: The dataset is trained using a BP neural network to obtain a strain-energy mapping model for converting strain data into energy data.

[0030] In this embodiment, the back propagation (BP) neural network is used as the core algorithm.

[0031] First, construct a neural network structure with three or more layers: the number of nodes in the input layer corresponds to the number of input strain features; the number of nodes in the output layer is fixed at three, corresponding to the total energy density, elastic strain energy density and plastic dissipation energy density, respectively.

[0032] Then, the training dataset constructed by S200 is randomly divided into training, validation, and test sets according to a certain ratio (e.g., 70% for training, 15% for validation, and 15% for testing). During network training, the Levenberg-Marquardt algorithm or Adam optimizer is used for iterative updates of weights and thresholds to minimize the mean squared error between the predicted output and the true label. Simultaneously, the validation set is used to monitor for overfitting; training is terminated early when the validation set error no longer decreases after several consecutive iterations. Finally, once the model reaches the preset accuracy requirement on an independent test set, it is saved as a stable and accurate strain-energy mapping model.

[0033] S400, Input strain contour plot: Obtain continuous strain contour maps of the tunnel surrounding rock surface.

[0034] In this embodiment, a physical testing system suitable for a tunnel plane strain model is built and run. A pre-fabricated tunnel surrounding rock model specimen with random speckle patterns on its surface is placed in a loading device. By controlling the vertical and horizontal loading devices, the ground stress is applied to a preset target value to simulate the initial ground stress field. Then, using a high-precision industrial camera and digital image correlation technology, speckle images of the specimen surface before and after the tunnel excavation simulation are continuously acquired. All acquired image sequences are imported into digital image correlation analysis software. After image matching, correlation calculations, and other steps, the full-field displacement field generated on the entire specimen observation surface during the tunnel excavation unloading process is calculated. Further differentiation yields the principal strain or equivalent strain distribution of the entire field, thus obtaining a high-resolution continuous strain contour map. The pixel value or grayscale value of each pixel in this contour map represents a specific strain characteristic data at that physical location.

[0035] S500, Point-by-Point Energy Conversion: The strain feature data corresponding to each pixel in the continuous strain cloud map is input into the trained strain-energy mapping model, and the model outputs the corresponding total energy density value, elastic strain energy density value and plastic dissipation energy density value pixel by pixel.

[0036] In this embodiment, after obtaining the continuous strain contour map, a pre-written data processing program (e.g., based on MATLAB or Python) is launched. The program first reads the image file of the strain contour map and parses it into a two-dimensional matrix, where each element represents the strain data of that pixel. Then, the program iterates through the entire matrix, taking the strain value of each pixel as input and passing it to the strain-energy mapping model trained and fixed in S300. Upon receiving the input, the model immediately performs forward propagation calculations, generating three values ​​in parallel at its output, corresponding to the total energy density, elastic strain energy density, and plastic dissipation energy density under that strain state, respectively. This process is repeated until all pixels in the contour map have been processed. For a contour map with a resolution of 1920×1080 pixels, this step can complete the conversion calculation of all approximately 2 million pixels within a few seconds, demonstrating extremely high efficiency.

[0037] S600, generating energy cloud map: The output total energy density value, elastic strain energy density value, and plastic dissipation energy density value are mapped to color values ​​to generate total strain energy density cloud maps, elastic strain energy density cloud maps, and plastic dissipation energy density cloud maps.

[0038] In this embodiment, the three energy density data matrices obtained by S500, each with the same size as the original strain cloud map, are visualized and rendered. First, the minimum and maximum values ​​in each energy density matrix are determined to set the range of the color mapping scale. Then, a scientifically applicable color mapping table is selected, and each energy density value in the matrix is ​​mapped to a specific color value (RGB triplet) on the color mapping table according to its relative position within the scale range. Finally, the mapped color matrices are recombined into an image, and auxiliary information such as legends and coordinate axes are added to generate the total strain energy density cloud map, elastic strain energy density cloud map, and plastic dissipation energy density cloud map, respectively. These three cloud maps clearly and intuitively demonstrate the spatial redistribution, concentration, and dissipation characteristics of energy inside and around the surrounding rock after tunnel excavation, providing direct experimental data support for analyzing the stability of the surrounding rock.

[0039] In a preferred embodiment, obtaining the continuous strain contour map includes the following steps: S410, Specimen Preparation: Continuous tunnel surrounding rock specimens are fabricated using cement mortar or 3D printing technology. Tunnel holes are pre-reserved in the middle of the tunnel surrounding rock specimens, and speckle patterns are created on the surface of the tunnel surrounding rock specimens.

[0040] In this embodiment, cement mortar is used as the model material. Ordinary silicate cement, standard sand, and water are mixed in a specific mass ratio, with an appropriate amount of water-reducing agent added to improve fluidity and strength. After thorough mixing, the mixture is poured into a custom-made large steel mold. The internal dimensions of the mold are determined according to the experimental requirements, and a cylindrical or arched air bladder or soluble core mold is pre-embedded at the center of the mold to form a reserved tunnel cavity. The diameter or span of the tunnel cavity is determined based on a similarity ratio conversion. During the pouring process, a vibrating table is used to compact the mixture and remove air bubbles. After curing under standard conditions for 28 days, the mold is removed, forming a continuous tunnel surrounding rock specimen.

[0041] In this embodiment, as an alternative, 3D printing technology can be used when more complex surrounding rock structures need to be simulated. First, a three-dimensional digital model of the surrounding rock and tunnel cavities is created based on the design drawings. Then, a photosensitive resin or gypsum-based powder with similar mechanical properties to the surrounding rock is selected as the printing material. High-precision tunnel surrounding rock specimens are produced layer by layer using powder bonding or photopolymerization molding processes. 3D printing can precisely control the internal structure and cavity morphology of the specimens, avoiding internal defects that may occur with casting methods.

[0042] In this embodiment, to meet the surface feature requirements of digital image correlation technology, a high-contrast, randomly distributed speckle pattern needs to be fabricated on the observation surface of the specimen. The specific steps are as follows: First, apply a white primer. Use matte white acrylic paint and apply 2-3 coats evenly to the surface of the specimen to be observed using an air spray gun or high-density sponge roller. Allow each coat to dry for at least 30 minutes before applying the next coat. The final result should be a uniform film thickness, complete coverage, consistent color, and no obvious brush marks or runs, forming a thin, evenly colored white base. Then, apply black ink using dot or spray techniques to create black speckles on a white base.

[0043] S420, Specimen Installation: A high-strength airbag is filled into the tunnel cavity, the tunnel surrounding rock specimen is installed in the loading device, a preset stress is applied to simulate the geostress field, and the specimen is left to stand for a predetermined time.

[0044] In this embodiment, a high-strength rubber airbag with a shape consistent with the reserved tunnel hole is selected. The airbag should have sufficient compressive strength and good airtightness. The airbag is slowly inserted into the reserved hole in the middle of the specimen, ensuring a tight fit between the outer wall of the airbag and the inner wall of the hole. The airbag's inflation nozzle should extend to the outside of the specimen for easy connection to subsequent inflation / deflation equipment. The airbag is pre-inflated to slightly expand and fill the hole space, but the inflation pressure does not exceed the simulated initial ground stress level (typically less than 0.1 MPa), serving only a supporting function.

[0045] In this embodiment, the tunnel surrounding rock specimen with airbags is stably installed in the loading device. Pressure is applied to three sides of the tunnel surrounding rock specimen; after the loading force in all three directions reaches the preset target value, the loading force is kept constant and left to stand for 5 minutes.

[0046] In a preferred embodiment, obtaining the continuous strain contour map further includes the following steps: S430, acquiring reference image: The image acquisition device is activated to capture the initial speckle field on the surface of the tunnel surrounding rock specimen.

[0047] In this embodiment, after the specimen is installed and left to stand for a predetermined time, it is confirmed that the stress field on the surface of the tunnel surrounding rock specimen has become stable, and at this time, the reference image is acquired.

[0048] In this embodiment, the image acquisition device includes a high-resolution industrial camera, a bracket and pan-tilt unit for fixing the camera, and two LED light sources for supplementary lighting. The camera is mounted on the bracket via the pan-tilt unit, positioned directly above the tunnel rock specimen, with the lens pointing vertically downwards at the surface of the specimen. The camera height and lens focal length are adjusted to ensure the entire observation area of ​​the specimen surface fills the camera's field of view and that the image is clear. The LED light sources are then turned on, and their position and illumination angle are adjusted to ensure uniform illumination of the specimen surface without obvious shadows or reflective areas, thus guaranteeing the quality of subsequent image analysis.

[0049] In this embodiment, the camera is connected to a computer with image acquisition and control software installed via a data cable. The software is opened, the camera device is added, and the camera monitoring screen is accessed. Based on the ambient lighting conditions and the contrast of the speckle pattern on the specimen surface, an appropriate exposure time and shutter speed are set.

[0050] Specifically, by observing the brightness of the image on the monitoring screen, the shutter speed is adjusted to achieve a clear image with moderate brightness and low noise. Several test images are taken to confirm that the image quality meets the requirements.

[0051] In this embodiment, the shooting mode is set to single-shot. After confirming that all parameters are correct, the camera is triggered to capture the first image. This image records the surface speckle field of the tunnel surrounding rock specimen under initial in-situ stress before the tunnel is excavated. This image is named "baseline image" or "reference image" and is automatically saved by the software to the designated project folder. The shooting time is recorded as the zero-time reference for subsequent analysis.

[0052] S440, Simulated tunnel excavation: The high-strength airbag is deflated, and a series of speckle images of the surface of the tunnel surrounding rock specimen are continuously captured.

[0053] In this embodiment, after the baseline image is acquired, a tunnel excavation simulation operation is immediately performed. Specifically, an airbag is deflated to simulate the unloading process of the surrounding rock caused by tunnel excavation.

[0054] In this embodiment, an inflation pipeline connected to a high-strength airbag inside a pre-reserved tunnel opening is located, and a pressure relief valve is installed on this pipeline. The operator opens the valve slowly or quickly according to a preset excavation rate to deflate the airbag. The deflation rate can be adjusted according to the experimental purpose.

[0055] In this embodiment, the camera's continuous shooting mode is activated simultaneously with the start of gas release. The shooting mode is pre-switched to continuous shooting in the image acquisition software, and an appropriate frame rate (e.g., 10 frames / second or higher) is set to ensure complete capture of the entire process of the surrounding rock from unloading to deformation stabilization. From the start of gas release, the camera continuously captures a sequence of speckle images of the specimen surface at the set frame rate until the surrounding rock deformation stabilizes. Throughout the shooting process, the light source brightness is kept constant to avoid interference from ambient light. Simultaneously, the specimen surface condition is observed in real-time via a computer monitor, and any abnormal phenomena (such as crack appearance, large local deformation, etc.) and their corresponding time points are recorded.

[0056] In this embodiment, after the deflation is completed, the airbag is completely depressurized, and the tunnel cavity loses its internal support, simulating the process of unloading the surrounding rock onto the free face after tunnel excavation. The speckle pattern on the specimen surface shifts and deforms with the deformation of the surrounding rock, and these changes are continuously recorded by the camera to form a complete image sequence.

[0057] S450, calculate strain: The image sequence is imported into the digital image analysis module, and a continuous strain cloud map of the surrounding rock surface after tunnel excavation is obtained through calculation.

[0058] In this embodiment, after image acquisition is completed, the acquired image sequence is imported into the digital image analysis module for processing. The digital image analysis module uses commercial software (such as the XTXIC system) or open-source software based on digital image-related technologies.

[0059] The specific steps are as follows: First, create a new project in the digital image analysis software and import the reference image acquired by the S430 and all the sequence images acquired by the S440. The software sets the reference image as the reference frame and each of the remaining images as a deformation frame.

[0060] Next, a region of interest (ROI) is defined on the specimen surface. This ROI typically encompasses the entire specimen surface excluding edge portions to eliminate boundary effects. Within the ROI, computational parameters are set, including subset size, step size, and correlation criteria. The selection of subset size and step size needs to be adjusted based on speckle quality and expected deformation gradient to ensure computational accuracy and efficiency. Zero-mean normalized cross-correlation is chosen as the correlation matching criterion to improve robustness to different illumination conditions.

[0061] Next, system calibration is performed. Two-dimensional scaling parameters are set by placing a ruler of known length on the specimen surface or by using a positioning steel frame of known dimensions in the loading device. These parameters are used to convert pixel displacements in the image into actual physical displacements.

[0062] After completing the parameter settings, the calculation is started. The software performs subset-by-subset correlation matching between each frame of the deformed image and the reference image to calculate the displacement field in the X and Y directions for each calculation point. Then, by numerically differentiating the displacement field, the strain components (including normal strain and shear strain) of each pixel are obtained. Depending on the analysis requirements, the software can output the principal strain contour map, the maximum shear strain contour map, or the equivalent strain contour map.

[0063] After the calculation is completed, the software interface displays a continuous strain contour map of the surrounding rock surface after tunnel excavation. This map, presented as colored contour lines, clearly shows the areas of strain concentration, strain distribution range, and strain evolution over time caused by the unloading during tunnel excavation. Users can export this continuous strain contour map as an image file or data matrix for subsequent energy conversion analysis.

[0064] In a preferred embodiment, the size of the speckle is calculated using the following formula: (one) Among them, L d Indicates speckle size; P d D represents the number of pixels required for speckle; D represents the image scale. The value of D is calculated using the following formula: (two) Among them, P p Indicates the number of pixels occupied by the subject being photographed; L p This indicates the actual size of the target being photographed.

[0065] Secondly, this application provides a testing apparatus, including a reaction frame 100, which is rectangular and has the loading device inside; the loading device includes an H-shaped positioning steel frame 110 and a loading mechanism 120 for applying stress to the tunnel surrounding rock specimen; the end of the positioning steel frame 110 is fixedly installed on the reaction frame 100, forming a positioning cavity and three loading cavities between the positioning steel frame 110 and the reaction frame 100; the loading mechanism 120 is located in the loading cavity, and its output end passes through the positioning steel frame 110 and extends into the positioning cavity.

[0066] In this embodiment, the reaction frame 100 is placed flat on the support platform of the laboratory, serving as the supporting skeleton of the entire device. It has sufficient rigidity and strength to withstand the huge reaction force generated during the loading process. The loading device is used to apply preset vertical and horizontal loads to the tunnel surrounding rock specimen placed therein, simulating the geostress field of the tunnel.

[0067] It should be noted that after the tunnel surrounding rock specimen is installed, the tunnel holes inside extend vertically, that is, the tunnel axis is perpendicular to the horizontal plane.

[0068] In a preferred embodiment, the positioning cavity is further provided with a force transmission steel plate 130; the number of force transmission steel plates 130 includes three, which are respectively connected to the loading mechanism 120 to disperse stress.

[0069] In this embodiment, the core function of the force-transmitting steel plate 130 is to convert the concentrated force generated by the loading mechanism 120 into a uniformly distributed pressure. Its working principle is as follows: When the output end of the loading mechanism 120 extends forward, it first pushes the force transmission steel plate 130 towards the specimen until the bearing surface of the force transmission steel plate 130 is completely in contact with the side of the tunnel surrounding rock specimen. As the loading force increases, the concentrated force applied at the output end is transmitted to the force transmission steel plate 130 through the connection. Due to the high rigidity and large bearing area of ​​the force transmission steel plate 130, this concentrated force causes stress redistribution inside the steel plate, ultimately forming an approximately uniform compressive stress on the contact surface between the steel plate and the specimen.

[0070] In this embodiment, the loading directions of the three force-transmitting steel plates 130 are located on the same horizontal plane. Combined with the vertical installation method of the tunnel surrounding rock specimen, the two sides and the top of the tunnel surrounding rock specimen can be loaded respectively.

[0071] In a preferred embodiment, the image acquisition device is located above the reaction frame 100 and includes a camera 200; the camera 200 is suspended above the positioning cavity via a bracket 210 and the shooting direction is downward; the bracket 210 is provided with a gimbal for mounting the camera 200.

[0072] In a preferred embodiment, a light source 220 is also included; the number of light sources 220 includes a plurality of them; the plurality of light sources 220 are respectively located outside the field of view for providing supplemental lighting for the tunnel surrounding rock specimen.

[0073] The above description is merely a preferred embodiment of this application and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in this application is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the inventive concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions disclosed in this application.

Claims

1. A method for energy testing of surrounding rock in tunnels, characterized in that, Includes the following steps: Establish strain-energy mapping relationship: Perform uniaxial or triaxial loading tests on multiple tunnel surrounding rock specimens of the same standard to obtain stress-strain curves. Determine the mapping relationship between strain and total energy density, elastic strain energy density, and plastic dissipation energy density based on the area enclosed by the stress-strain curves. Constructing a training dataset: Based on the mapping relationship, construct a dataset with strain parameters as input and the corresponding total energy density, elastic strain energy density, and plastic dissipation energy density as output; Training the neural network model: The dataset is trained using a BP neural network to obtain a strain-energy mapping model for converting strain data into energy data; Input strain contour map: Obtain continuous strain contour map of the tunnel surrounding rock surface; Point-by-point energy conversion: The strain feature data corresponding to each pixel in the continuous strain cloud map is input into the trained strain-energy mapping model, and the model outputs the corresponding total energy density value, elastic strain energy density value and plastic dissipation energy density value pixel by pixel. Generate energy cloud maps: Map the output total energy density value, elastic strain energy density value and plastic dissipation energy density value to color values ​​respectively to generate total strain energy density cloud maps, elastic strain energy density cloud maps and plastic dissipation energy density cloud maps.

2. The method for energy testing of surrounding rock in tunnels according to claim 1, characterized in that, Obtaining the continuous strain contour map includes the following steps: Specimen preparation: Continuous tunnel surrounding rock specimens are made using cement mortar or 3D printing technology. Tunnel holes are reserved in the middle of the tunnel surrounding rock specimens, and speckles are made on the surface of the tunnel surrounding rock specimens. Specimen installation: A high-strength airbag is filled into the tunnel cavity, the tunnel surrounding rock specimen is installed in the loading device, a preset stress is applied to simulate the geostress field, and it is left to stand for a predetermined time.

3. The method for energy testing of surrounding rock in tunnels according to claim 2, characterized in that, Obtaining the continuous strain contour map further includes the following steps: Acquire baseline image: Start the image acquisition device and capture the initial speckle field on the surface of the tunnel surrounding rock specimen; Simulated tunnel excavation: The high-strength airbag is deflated, and a series of speckle images of the surface of the tunnel surrounding rock specimen are continuously captured. Strain calculation: The image sequence is imported into the digital image analysis module, and a continuous strain cloud map of the surrounding rock surface after tunnel excavation is obtained through calculation.

4. The method for energy testing of surrounding rock in tunnels according to claim 2, characterized in that, The size of the speckle is calculated using the following formula: Among them, L d Indicates speckle size; P d D represents the number of pixels required for speckle; D represents the image scale. The value of D is calculated using the following formula: Among them, P p Indicates the number of pixels occupied by the subject being photographed; L p This indicates the actual size of the target being photographed.

5. The method for energy testing of surrounding rock in tunnels according to claim 2, characterized in that, The speckle pattern preparation includes the following steps: A white primer was applied to the surface of the tunnel surrounding rock specimen. The primer was required to be of uniform thickness and color and without obvious brushing texture, forming a thin, uniform white base surface. Apply black ink using dotting or spraying methods to create black speckles on a white background.

6. The method for energy testing of surrounding rock in tunnels according to claim 1, characterized in that, Integrating the stress-strain curve, Calculate the area under the loading curve from zero to the current strain, and use it as the total energy density corresponding to the current strain; Calculate the area enclosed by the unloading curve when unloading from the current stress point, and use it as the elastic strain energy density corresponding to the current strain. The difference between the total energy density and the elastic strain energy density is taken as the plastic dissipation energy density corresponding to the current strain.

7. A testing apparatus, applicable to the method for energy testing of surrounding rock in tunnels as described in claim 3, characterized in that, include: A reaction frame (100) is rectangular and has the loading device inside; The loading device includes an H-shaped positioning steel frame (110) and a loading mechanism (120) for applying stress to the tunnel surrounding rock specimen. The end of the positioning steel frame (110) is fixedly installed on the reaction frame (100), forming a positioning cavity and three loading cavities between the positioning steel frame (100) and the reaction frame (100); The loading mechanism (120) is located inside the loading cavity, and its output end passes through the positioning steel frame (110) and extends into the positioning cavity.

8. The testing apparatus according to claim 7, characterized in that, The positioning cavity is also equipped with a force-transmitting steel plate (130). The number of the force-transmitting steel plates (130) includes three, which are respectively connected to the loading mechanism (120) to disperse stress.

9. The testing apparatus according to claim 7, characterized in that, The image acquisition device is located above the reaction frame (100) and includes a camera (200). The camera (200) is suspended above the positioning cavity by a bracket (210) and the shooting direction is downward; The bracket (210) is equipped with a gimbal for mounting a camera (200).

10. The testing apparatus according to claim 9, characterized in that, It also includes a light source (220); The number of light sources (220) includes multiple; Multiple light sources (220) are located outside the field of view for providing supplemental lighting to the tunnel surrounding rock specimen.

Images