Earthquake response testing device and method for tunnel model test

Abstract

The invention belongs to the technical field of tunnel model tests, and particularly discloses a tunnel model test earthquake response testing device and method, and the device comprises a vibration table which is used for applying an earthquake load to a tunnel model; the model box is fixed on the vibration table, filled with soil and provided with a tunnel supporting structure; the speckles are pasted on the surface of the inner wall of the tunnel supporting structure and used for image recognition; the camera array is fixedly arranged relative to the tunnel supporting structure and is used for continuously shooting deformation images of the speckles; and the image processing unit is in communication connection with the camera array and is used for receiving, splicing and analyzing the deformation images to obtain full-field strain, displacement, deformation and acceleration data of the tunnel supporting structure under the action of the earthquake load. The testing device and method have the advantages of non-contact measurement, full-field continuous deformation monitoring, synchronous acquisition of multi-dimensional parameters and the like, and non-contact high-density seismic response data measurement of the tunnel model supporting structure can be realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of tunnel model test, and particularly relates to a tunnel model test earthquake response testing device and method. BACKGROUND

[0002] In the field of tunnel engineering, it is crucial to study the seismic performance of tunnel structures under seismic loads. At present, scale model test is one of the commonly used research methods. In traditional tests, contact-type measurement elements such as multi-point displacement meters, vibration sensors, acceleration sensors or optical fibers are usually used to collect seismic response data of supporting structures. These contact-type measurement methods have many inherent defects: the measurement elements need to be embedded in the structure or closely attached to the structure surface, and the installation process will destroy the original stress field of the soil, and the mass of the sensor itself will change the local vibration characteristics, resulting in distorted measurement data. The number of measurement points is limited by the sensor arrangement density, and it is difficult to obtain continuous deformation data of the whole structure. Contact-type measurement requires a complex wiring system, which is prone to signal interference in a vibrating environment, affecting the reliability of data acquisition. In addition, the traditional method cannot synchronously obtain multi-dimensional response parameters such as displacement, strain and acceleration, making it difficult to comprehensively evaluate the seismic performance of the structure. The optical measurement method used in the prior art achieves non-contact measurement, but has problems such as insufficient image stitching accuracy and difficulty in three-dimensional deformation reconstruction, which cannot meet the special needs of tunnel ring continuous deformation measurement. The prior art needs to be improved in view of the above problems. SUMMARY

[0003] The purpose of the present application is to provide a tunnel model test earthquake response testing device and method, which has the advantages of non-contact measurement, whole-field continuous deformation monitoring, multi-dimensional parameter synchronous acquisition, etc.

[0004] To achieve the above-mentioned purpose, in one aspect, the present application provides a tunnel model test earthquake response testing device, comprising: a vibration table for applying seismic loads to a tunnel model; a model box fixed on the vibration table, filled with soil inside and provided with a tunnel supporting structure; a speckle pasted on the inner wall surface of the tunnel supporting structure for image recognition; a camera array fixedly arranged relative to the tunnel supporting structure for continuously shooting deformation images of the speckle; and an image processing unit in communication connection with the camera array for receiving, splicing and analyzing the deformation images to obtain whole-field strain, displacement, deformation and acceleration data of the tunnel supporting structure under seismic loads.

[0005] Optionally, the model box is provided with an observation hole at a position corresponding to the tunnel supporting structure, and a transparent sealing plate is arranged at the observation hole to prevent soil from leaking out and allow the camera array to shoot the speckle through the sealing plate.

[0006] Optionally, the camera array comprises a plurality of cameras distributed in an array, the optical axes of the cameras intersect at the same ring section of the tunnel support structure, and the overlapping field of view of adjacent cameras contains at least three common speckles for subsequent image stitching.

[0007] Optionally, the image processing unit comprises an image stitching module for automatically identifying and stitching the common speckles of adjacent images using a multiple three-point positioning method; and a data analysis module for calculating the time sequence displacement field of the speckles based on the DIC algorithm to obtain the strain field and the acceleration field.

[0008] Optionally, the image stitching module only retains three common speckles with a relative error less than 5% for stitching.

[0009] Further, the vibration table synchronously or individually applies horizontal transverse waves and vertical longitudinal waves, and the amplitude, frequency and duration of each level of load are programmatically set through an external control console to simulate different seismic working conditions.

[0010] Further, the tunnel model test seismic response testing device further comprises a sealing assembly for sealing the opening of the model box.

[0011] Further, the sealing assembly is made of a flexible waterproof material to ensure that the soil does not leak during the test.

[0012] Further, the tunnel model test seismic response testing device further comprises a speckle attaching tool for uniformly attaching speckles to the wall of the tunnel support structure.

[0013] In another aspect, the present application also provides a tunnel model test seismic response testing method, which uses the tunnel model test seismic response testing device according to any one of the above-mentioned solutions, and comprises the following steps:

[0014] S1: installing the tunnel support structure in the model box, filling the soil and fixing the model box to the vibration table;

[0015] S2: attaching speckles to the inner wall of the tunnel support structure and sealing the opening of the model box;

[0016] S3: starting the camera array calibration to ensure that the field of view of each camera covers the entire ring section to be measured;

[0017] S4: applying the first level of seismic load through the vibration table, and continuously shooting speckle images at a set frame rate through the camera array;

[0018] S5: gradually increasing the seismic load and repeatedly shooting until all working conditions are completed;

[0019] S6: transmitting the images collected by each camera to the image processing unit and automatically stitching them into a ring panoramic image using a multiple three-point positioning method.

[0020] S7: calculating the full-field displacement, strain and acceleration response of the tunnel support structure based on the spliced panoramic image sequence.

[0021] Optionally, in step S2, the speckle is a random distribution of high-contrast black and white patterns, with a minimum spot diameter not less than 3 times the camera pixel resolution, to ensure the calculation accuracy of the DIC algorithm.

[0022] Optionally, in step S6, the multiple three-point positioning method comprises:

[0023] automatically identifying all speckle points in the overlapping area of adjacent images;

[0024] iterating all three-point combinations to calculate the spatial coplanar error thereof;

[0025] selecting the first N groups of three points with the smallest error for weighted averaging to obtain the optimal splicing transformation matrix.

[0026] Specifically, since the captured images are spliced into a ring after being captured in sub-regions, it is necessary to identify and splice the common points (speckles) of adjacent two images, and a multiple three-point positioning method is adopted, i.e., three common speckle points in adjacent two images are found, and splicing is performed according to the three-point coplanar principle. Since one-time three-point positioning may have errors, multiple three-point positioning is automatically performed by writing a script, and three common points with a relative error less than 5% are selected for splicing.

[0027] Optionally, in step S7, the displacement increment under each level of load is calculated with the image in the initial unloaded state as a reference; the acceleration field is obtained by time domain difference; and the displacement field and the acceleration field are superimposed on the three-dimensional tunnel model to generate a seismic response cloud chart.

[0028] Optionally, between steps S4 and S5, it further comprises: pausing the shaking table and emptying the soil inside the tunnel support structure to eliminate the constraint of the soil on the free deformation of the support structure, and then resuming the application of the seismic load and continuing to capture

[0029] Compared with the prior art, the present application at least discloses the following beneficial effects:

[0030] The test device and method of the present application can realize non-contact high-density seismic response data measurement of the tunnel model support structure, can analyze the full-field strain, displacement, deformation and acceleration of the support structure under the action of the seismic load, and overcomes the problems of complex operation and large interference in seismic response data measurement of the tunnel model test support structure, and has the advantages of simple operation, fast measurement and accurate measurement under the condition of meeting the measurement requirements. BRIEF DESCRIPTION OF DRAWINGS

[0031] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed in the embodiments. Obviously, the drawings described below only illustrate some of the embodiments of the present application, and for those skilled in the art, other drawings can also be obtained from these drawings without creative labor.

[0032] Figure 1 It is a structural schematic diagram of the testing device in the embodiments of the present application.

[0033] Figure 2 It is a camera shooting schematic diagram of the testing device in the embodiments of the present application in the test process.

[0034] Figure 3 It is a three-point positioning splicing image schematic diagram in image processing of the testing device in the embodiments of the present application.

[0035] Figure 4 It is a splicing process and effect schematic diagram in image processing of the testing device in the embodiments of the present application.

[0036] In the figure: 1, soil body; 2, model box; 3, tunnel supporting structure; 4, vibration table; 5, seismic load; 6, camera array; 7, regional image one; 8, regional image two; 9, first group of common speckle points; 10, second group of common speckle points; 11, speckle. DETAILED DESCRIPTION

[0037] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the present application.

[0038] The existing non-contact measurement technology (such as optical measurement) has not been mature in the application in the tunnel model test, especially in the continuous collection and analysis of full-field deformation, strain and acceleration and other data of the supporting structure. The present application aims to overcome the above technical defects, and provides a tunnel model seismic response testing device and method which is simple to operate, fast to measure and accurate in result.

[0039] In order to make the above purposes, features and advantages of the present application more obvious and easy to understand, the present application will be further described in detail below with reference to the drawings and specific embodiments.

[0040] REFERENCE Figure 1As shown, the embodiment provides a tunnel model test seismic response testing device, which includes a shaking table 4, a model box 2, a speckle 11, a camera array 6 and an image processing unit. The shaking table 4 is used to apply seismic load 5 to the tunnel model. The model box 2 is fixed on the shaking table 4, filled with soil 1 inside and provided with a tunnel support structure 3. The speckle 11 is pasted on the inner wall surface of the tunnel support structure 3 for image recognition. The camera array 6 is fixedly arranged relative to the tunnel support structure 3 for continuously shooting deformation images of the speckle 11. The image processing unit is in communication connection with the camera array 6 for receiving, splicing and analyzing the deformation images to obtain full-field strain, displacement, deformation and acceleration data of the tunnel support structure 3 under the action of the seismic load 5.

[0041] The shaking table 4 can adopt an electromagnetic or hydraulic shaking table 4, and the frequency range and amplitude thereof need to meet the seismic simulation requirements. The model box 2 can adopt steel or aluminum alloy materials, and the size thereof is determined according to the test requirements. The internal soil 1 can be sand or clay. The speckle 11 can adopt a high-contrast black-and-white pattern, which is fixed on the inner wall of the tunnel support structure 3 by spraying or pasting. The camera array 6 can include multiple cameras arranged around the tunnel support structure 3 to ensure covering the entire to-be-measured area. The image processing unit can adopt a computer system equipped with professional image processing software to realize image splicing and data analysis functions.

[0042] The above technical solution solves the problem of interference of traditional contact measurement on soil 1 vibration by using a non-contact measurement method. The speckle 11 and the camera array 6 are used for image acquisition, and the image processing technology is combined to obtain full-field response data of the tunnel support structure 3 under the action of the seismic load 5. Compared with the traditional method, this method has the advantages of high measurement accuracy, large data density, simple operation and the like, and provides an effective means for tunnel seismic performance research.

[0043] In a specific embodiment, the model box 2 is provided with an observation hole at a position corresponding to the tunnel support structure 3, and a transparent sealing plate is arranged at the observation hole.

[0044] Specifically, the observation hole is arranged in a region where the side wall of the model box 2 is aligned with the outer contour of the tunnel support structure 3, and the hole diameter size needs to meet the complete shooting requirement of the camera array 6 on the tunnel inner wall speckle 11. The transparent sealing plate adopts high-strength acrylic material, and is sealed and connected with the model box 2 through a rubber gasket and a flange bolt, which can ensure the observation transmittance while preventing soil 1 leakage. As a preferred embodiment, the observation hole can be designed as a rectangular array distribution, and each observation hole is provided with an independently detachable sealing plate, which is convenient for local maintenance or replacement. In addition, the inner surface of the transparent sealing plate can be plated with an anti-fog coating to avoid the influence of water vapor condensation caused by temperature difference on the image acquisition quality during the test.

[0045] The above specific embodiment realizes non-contact optical observation of the deformation state of the tunnel support structure 3 by the combination design of the observation hole and the transparent sealing plate, while maintaining the sealing of the model box 2. Compared with the prior art scheme of embedding sensors inside the support structure, the measurement element does not interfere with the vibration field of the soil 1, and the soil 1 loss problem caused by the opening is avoided. The high light transmission property of the transparent sealing plate ensures the clarity of the image acquisition, and the anti-fog treatment further improves the data reliability in the complex test environment. The structural design takes into account the observation function and the model integrity, providing stable data source for subsequent image processing.

[0046] In one specific embodiment, the camera array 6 includes a plurality of cameras arranged in an array, the optical axes of each camera intersecting at the same ring section of the tunnel support structure 3, and the overlapping field of view between adjacent cameras containing at least three common speckle points for subsequent image stitching.

[0047] Specifically, the plurality of cameras are arranged in a ring array, and the installation angle is ensured by geometric calibration to ensure that all optical axes intersect at the same ring section center point of the tunnel support structure 3. The field of view overlap rate between adjacent cameras is controlled within the range of 30% to 50%, and at least three non-collinear speckles 11 in the overlap area are determined as common feature points by a calibration plate. As a preferred embodiment, the camera can use a 5mm focal length fixed focus lens, arranged uniformly along the ring at a distance of 200mm, and the field of view angle of each camera covers 60 degrees, forming a seamless ring-shaped observation area. Further, the spatial coordinates of the common speckle points are reconstructed in three dimensions by stereo vision principle, and the position error is controlled within 0.1 pixels.

[0048] The above specific embodiment realizes non-contact optical observation of the deformation state of the tunnel support structure 3 by the combination design of the observation hole and the transparent sealing plate, while maintaining the sealing of the model box 2. Compared with the prior art scheme of embedding sensors inside the support structure, the measurement element does not interfere with the vibration field of the soil 1, and the soil 1 loss problem caused by the opening is avoided. The high light transmission property of the transparent sealing plate ensures the clarity of the image acquisition, and the anti-fog treatment further improves the data reliability in the complex test environment. The structural design takes into account the observation function and the model integrity, providing stable data source for subsequent image processing.

[0049] Further optimization scheme, the image processing unit includes an image stitching module and a data analysis module. The image stitching module uses a multiple three-point positioning method to automatically identify and stitch the common speckle points of adjacent images. The data analysis module calculates the time sequence displacement field of the speckle 11 based on the DIC algorithm to obtain the strain field and acceleration field.

[0050] Specifically, the image stitching module selects the first N groups of three points with the smallest error by automatically identifying all speckle points in the overlapping area of adjacent images, traversing all three-point combinations and calculating their spatial coplanar errors, and then performing a weighted average to obtain the optimal stitching transformation matrix. Figure 3 As shown in FIG. 7, the region image one 7 and the region image two 8 are adjacent images captured by the two-dimensional camera, and there is an overlapping area between the two images, and in the overlapping area, three common speckle points with an error less than 5% are identified, such as the first group of common speckle points 9 on the region image one 7 and the second group of common speckle points 10 on the region image two 8, and then the image stitching module stitches the above common speckle points.

[0051] The specific embodiment solves the problem of interference of traditional contact measurement elements on the vibration of the soil 1 by using a non-contact measurement method. The multiple three-point positioning method is used for image stitching, which can effectively improve the stitching accuracy and stability, and ensure the accuracy of subsequent data analysis. The data analysis module based on the DIC algorithm can realize high-density full-field displacement, strain and acceleration data acquisition, and provide more comprehensive and accurate test data for the seismic response research of the tunnel supporting structure 3.

[0052] As a preferred embodiment, the least squares method can be used to calculate the spatial coplanar error of the speckle points, and the error threshold is determined by multiple tests. In addition, the selection process of the speckle points is automatically completed by the computer vision algorithm, which first extracts all possible common speckle point combinations, and then calculates the relative error one by one and excludes the out-of-limit data. By setting a strict error threshold, the cumulative error problem caused by the identification deviation of the speckle points in the multi-camera image stitching process is effectively solved. The specific working principle is: in the overlapping area of the adjacent camera field of view, only high-precision speckle points are selected as the stitching reference points, thereby ensuring the geometric consistency of the image stitching. Compared with the prior art, this method avoids the subjective error caused by manual intervention, and significantly improves the calculation accuracy of the full-field strain field and displacement field by error control.

[0053] The embodiment of the present application also provides a tunnel model test seismic response testing method, which adopts the tunnel model test seismic response testing device disclosed in the above embodiment, and the method comprises the following steps:

[0054] S1: Install the tunnel support structure 3 in the model box 2, fill the soil body 1 and fix the model box 2 to the vibration table 4.

[0055] S2: Paste the speckle 11 on the inner wall of the tunnel support structure 3, and seal the hole of the model box 2.

[0056] S3: Start the camera array 6 calibration to ensure that the field of view of each camera covers the entire ring section to be measured.

[0057] S4: Apply the first level of seismic load 5 through the vibration table 4, while the camera array 6 continuously takes images of the speckle 11 at a set frame rate.

[0058] S5: Increase the seismic load 5 step by step, and repeat the shooting until all working conditions are completed.

[0059] S6: Transfer the images collected by each camera to the image processing unit, and use the multiple three-point positioning method to automatically splice into a ring panoramic image.

[0060] S7: Based on the spliced panoramic image sequence, calculate the full-field displacement, strain and acceleration response of the tunnel support structure 3.

[0061] In one embodiment, the speckle 11 is a randomly distributed high-contrast black and white pattern, with a minimum spot diameter not less than 3 times the camera pixel resolution, to ensure the calculation accuracy of the DIC algorithm. The multiple three-point positioning method includes automatically identifying all speckle points in the overlapping area of adjacent images, calculating the spatial coplanar error of all three-point combinations, selecting the top N three-point combinations with the smallest error for weighted averaging, and obtaining the optimal splicing transformation matrix. As an optional implementation, the displacement increment under each level of load is calculated based on the image in the initial unloaded state; the acceleration field is obtained by time domain difference; and the displacement field and acceleration field are superimposed on the three-dimensional tunnel model to generate a seismic response cloud chart.

[0062] This method uses non-contact image measurement technology to avoid the interference of traditional contact sensors on the vibration of the soil body 1. The use of step loading and panoramic image splicing technology can accurately capture the full-field dynamic response of the tunnel support structure 3 under the action of the seismic load 5. The matching design of the speckle 11 size and the camera resolution ensures the sub-pixel level accuracy of displacement measurement, and the image splicing algorithm based on three-point positioning effectively solves the error accumulation problem of multi-camera field splicing. This method realizes the breakthrough from local measurement to full-field analysis, and provides more reliable data support for the study of tunnel seismic performance.

[0063] In a further optimized scheme, when the speckle 11 is pasted on the inner wall of the tunnel support structure 3, the speckle 11 adopts a randomly distributed high-contrast black and white pattern, with a minimum spot diameter not less than 3 times the camera pixel resolution.

[0064] Further optimization scheme, in step S7, with the initial image in the unloaded state as a reference, the displacement increment under each level of load is calculated; the acceleration field is obtained by time domain difference; the displacement field and the acceleration field are superimposed to the three-dimensional tunnel model to generate the seismic response cloud chart.

[0065] Specifically, the high-contrast black-and-white pattern is made by inkjet printing or laser etching process, and the gray difference between the black and white areas in the pattern needs to be greater than 200 (8-bit gray scale range). The minimum spot diameter is set according to the sampling theorem of digital image correlation algorithm (DIC), and when the speckle 11 feature size is less than 3 times the pixel resolution, it will cause the sub-pixel displacement calculation error to exceed 5%. As a preferred embodiment, the individual speckle 11 in the speckle 11 pattern is approximately circular, and the diameter coefficient of variation is controlled within 15% to ensure the uniformity of the features of the speckle 11 in different areas. Further, the speckle 11 coverage is kept between 40% and 60%, which avoids both insufficient feature point density and recognition interference caused by pattern overlap.

[0066] Specifically, the images of all areas in the monitored ring are obtained according to the above test method, and the image stitching algorithm of three-point positioning is used to obtain the complete image of the monitored ring, as shown in Figure 4 Finally, the continuous result cloud chart under the action of the seismic load 5 is obtained according to the shooting order, and the analysis of the result cloud chart data can be carried out.

[0067] The above technical scheme solves the displacement field calculation error problem caused by too small feature size of the speckle 11 in the traditional method by quantifying the matching relationship between the speckle 11 size and the camera resolution. Experimental data show that when the speckle 11 diameter reaches 3 times the pixel resolution, the displacement measurement accuracy of the DIC algorithm can be improved to 0.02 pixels, and the strain calculation error is reduced to less than 3%. This size constraint ensures that the speckle 11 pattern remains identifiable even under large deformation under high-frequency excitation of the vibration table 4. Compared with the prior art using random size speckle 11, the present scheme establishes a quantitative relationship between the speckle 11 size and the measurement accuracy, providing a reliable image feature basis for non-contact measurement.

[0068] Further optimization scheme, between steps S4 and S5, also includes: pausing the vibration table 4 and emptying the soil 1 inside the tunnel supporting structure 3 to eliminate the constraint of the soil 1 on the free deformation of the supporting structure, and then restoring the seismic load 5 and continuing to shoot.

[0069] Specifically, the operation of emptying the soil body 1 can be achieved by the following ways: setting an openable soil discharge valve at the bottom of the model box 2, quickly discharging the soil body 1 by mechanical or pneumatic means; or using a vacuum suction device to suck out the loose soil body 1 from the observation hole. Among them, the soil discharge process needs to be linked with the control system of the shaking table 4 to ensure that it is executed after the shaking table 4 completely stops. Further, in order to maintain the consistency of the test environment, the bulk density of the emptied soil body 1 can be compensated by the counterweight to avoid the influence of the sudden change of the overall mass of the model box 2 on the loading accuracy of the shaking table 4.

[0070] As a preferred embodiment, the soil body 1 emptying operation should be completed within 30 seconds to reduce the impact of test interruption on data continuity.

[0071] As a preferred embodiment, the soil body 1 emptying operation is realized through the pre-set soil discharge channel of the model box 2, which is equipped with an airtight valve to ensure the sealing during the emptying process. After emptying, the supporting structure is in a state of no soil body 1 constraint, and at this time, the application of the seismic load 5 can obtain the speckle 11 image of the supporting structure in the free deformation state. In this step, the interval time between soil body 1 emptying and loading recovery is controlled within 5 minutes to avoid the influence of environmental factors on the measurement results.

[0072] As a preferred embodiment, the soil body 1 emptying uses a negative pressure suction device, and a filter screen is arranged at the end of the suction pipe to prevent soil body 1 particles from entering the equipment. The emptying degree is monitored in real time by a weight sensor, and when the residual amount of the soil body 1 is less than 2% of the total mass, it is determined that the emptying is completed. Before the loading is recovered, the focal length of the camera array 6 needs to be adjusted to compensate for the imaging error caused by the change of the light path after the soil body 1 is removed.

[0073] The above technical solution effectively solves the problem of distortion of the real deformation data of the supporting structure caused by the constraint of the soil body 1 in the traditional test by introducing the soil body 1 emptying step. Its working principle is that: in the process of staged loading, the confining pressure effect of the soil body 1 on the supporting structure is removed stage by stage, so that the supporting structure deforms freely under the condition of no soil body 1 constraint, thereby accurately capturing the pure mechanical response of the supporting structure in the process of seismic wave transmission. Compared with the prior art, this method can separate the coupling effect of the soil body 1 and the structure, significantly improve the reliability of the displacement field and strain field data, and is especially suitable for quantitative analysis of the seismic performance of the supporting structure itself.

[0074] In the description of the present application, it should be understood that the orientations or positional relationships indicated by the terms "longitudinal", "transverse", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like are based on the orientations or positional relationships shown in the drawings, and are only for the convenience of describing the present application, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application.

[0075] The above-described embodiments are merely intended to describe the preferred modes of the present application, and are not intended to limit the scope of the present application. Various modifications and improvements of the present application made by those skilled in the art based on the above-described embodiments should fall within the scope of the present application defined by the claims.

Claims

1. A tunnel model test seismic response testing device, characterized in that, The utility model relates to a tunnel seismic response test system, comprising: a vibration table (4) for applying seismic load (5) to a tunnel model; a model box (2) fixed on the vibration table (4), filled with soil (1) and provided with a tunnel support structure (3); speckles (11) attached to the inner wall surface of the tunnel support structure (3) for image recognition; a camera array (6) fixed relative to the tunnel support structure (3) for continuously shooting deformation images of the speckles (11); an image processing unit in communication with the camera array (6) for receiving, stitching and analyzing the deformation images to obtain full-field strain, displacement, deformation and acceleration data of the tunnel support structure (3) under the action of the seismic load (5).

2. The tunnel model test seismic response testing device according to claim 1, wherein, The model box (2) is provided with an observation hole at a position corresponding to the tunnel support structure (3), and a transparent sealing plate is arranged at the observation hole.

3. The tunnel model test seismic response testing device according to claim 1, wherein, The camera array (6) comprises a plurality of cameras arranged in an array, the optical axes of the cameras intersect at the same ring section of the tunnel support structure (3), and the overlapping field of view of adjacent cameras contains at least three common speckle points for subsequent image stitching.

4. The test apparatus for testing seismic response of a tunnel model according to claim 1, wherein The image processing unit comprises: an image stitching module for automatically identifying and stitching the common speckle points of adjacent images using a multiple three-point positioning method; a data analysis module for calculating the time-series displacement field of the speckles (11) based on the DIC algorithm to obtain the strain field and acceleration field.

5. The apparatus for testing seismic response of a tunnel model according to claim 4, wherein The image stitching module only retains the three common speckle points with a relative error less than 5% for stitching.

6. A method for testing seismic response of a tunnel model, using the device for testing seismic response of a tunnel model according to any one of claims 1 to 5, characterized in that, The utility model relates to a tunnel seismic response test system, comprising: S1: installing the tunnel support structure (3) in the model box (2), filling the soil (1) and fixing the model box (2) on the vibration table (4); S2: attaching the speckles (11) to the inner wall of the tunnel support structure (3) and sealing the hole of the model box (2); S3: starting camera array (6) calibration to ensure that the field of view of each camera covers the entire ring section to be measured; S4: applying the first level of seismic load (5) through the vibration table (4), and continuously shooting speckle (11) images at a set frame rate through the camera array (6); S5: gradually increasing the seismic load (5) and repeatedly shooting until all working conditions are completed; S6: transmitting the images collected by each camera to the image processing unit and automatically stitching them into a ring panoramic image using a multiple three-point positioning method; S7: calculating the full-field displacement, strain and acceleration response of the tunnel support structure (3) based on the stitched panoramic image sequence.

7. The tunnel model test seismic response testing method according to claim 6, characterized in that, In step S2, the speckles (11) are randomly distributed high-contrast black and white patterns with a minimum speckle diameter not less than 3 times the camera pixel resolution to ensure the calculation accuracy of the DIC algorithm.

8. The tunnel model test seismic response testing method according to claim 6, characterized in that, In step S6, the multiple three-point positioning method comprises: automatically identifying all speckle points in the overlapping area of adjacent images; iterating all three-point combinations to calculate their spatial coplanar error; selecting the top N three-point combinations with the smallest error for weighted averaging to obtain the optimal stitching transformation matrix.

9. The tunnel model test seismic response testing method according to claim 6, characterized in that, In step S7, the displacement increment under each level of load is calculated with the image in the initial unloaded state as the reference; the acceleration field is obtained through time domain difference; and the displacement field and acceleration field are superimposed on the three-dimensional tunnel model to generate a seismic response cloud chart.

10. The tunnel model test seismic response testing method according to claim 6, characterized in that, Between steps S4 and S5, it is also provided to pause the shaking table (4) and to empty the soil mass (1) inside the tunnel support structure (3) to eliminate the constraints of the soil mass (1) to the free deformation of the support structure, and then to resume the application of the seismic load (5) and continue the shooting.

Images