Soil liquefaction surface deformation field measurement system and method based on digital image technology

Abstract

The application discloses a soil liquefaction surface deformation field measuring system based on digital image technology, which comprises a control platform, a servo control and data acquisition system, a workbench, a support table, a soil box, an angle adjusting mechanism, a power loading system and a digital image deformation measuring system. The system can non-contact, full-field and high-resolution acquire the evolution data of the displacement field and the strain field of the soil surface in the vibration process, and can adapt to model tests of different sizes and inclination angles, thereby providing an advanced technical means for in-depth research on the liquefaction mechanism of saturated sand.

Description

Technical Field

[0001] This invention relates to the field of geotechnical engineering testing and measurement technology, specifically to a system and method for measuring the deformation field of soil liquefaction surface based on digital image technology. Background Technology

[0002] Liquefaction of saturated sand caused by earthquakes and the subsequent lateral slippage are the main causes of major earthquake damage, such as pile foundation failure, slope instability, and underground pipeline rupture. In-depth research on its occurrence mechanism and evolution process is crucial for seismic design of engineering projects, and indoor physical model tests are the main means of reproducing and quantifying this process.

[0003] Therefore, the design of the model and the measurement methods of the system become particularly important. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of existing technologies and provide a soil liquefaction surface deformation field measurement system and method based on digital image correlation technology. This system and method can acquire non-contact, full-field, and high-resolution data on the displacement and strain field evolution of the soil surface during vibration, and can adapt to model tests of different sizes and inclination angles, providing advanced technical means for in-depth research on the liquefaction mechanism of saturated sand.

[0005] To achieve the above objectives, the method adopted by the present invention is as follows:

[0006] A soil liquefaction surface deformation field measurement system based on digital image technology includes:

[0007] Control platform;

[0008] A servo control and data acquisition system is connected to the control platform and is equipped with a pore water pressure sensor and an acceleration sensor to synchronously acquire pore water pressure and acceleration data during the experiment; a workbench is laid with horizontal guide rails; a support plate is mounted on the horizontal guide rails by a slider and can slide along the horizontal guide rails;

[0009] A soil box is placed on a support platform, with its bottom center hinged to the top center of the support platform, and at least one side is transparent. An angle adjustment mechanism is located between the soil box and the support platform to adjust the tilt angle of the soil box. A power loading system is connected to the servo control and data acquisition system and is connected to the slider via a rocker arm to drive the slider and the soil box to reciprocate along a horizontal guide rail to simulate horizontal seismic loads. A digital image deformation measurement system includes an image acquisition unit, an illumination unit, and a data processing and analysis unit, with the image acquisition unit facing the transparent surface of the soil box.

[0010] Furthermore, the angle adjustment mechanism includes several liftable supports evenly distributed on the support platform, and the tilt angle of the soil box is changed by adjusting the liftable supports.

[0011] Furthermore, a limiter is also provided at the end of the horizontal guide rail.

[0012] Furthermore, the image acquisition unit includes at least one digital camera; the illumination unit is an LED array light source; and the data processing and analysis unit filters the image sequences acquired by the image acquisition unit and performs PIV calculation.

[0013] Furthermore, the data processing and analysis unit is configured to perform the following steps:

[0014] a. Acquire the image sequence acquired by the image acquisition unit, set a clear image before the vibration begins as the reference image, and use the subsequent frames as the target images;

[0015] b. A multi-level mesh deformation iterative algorithm is used to calculate the displacement field between the reference image and each subsequent target image;

[0016] c. The overall strain and strain increment of the soil are calculated using the Green-Lagrange strain tensor based on the finite deformation theory;

[0017] d. Generate and output displacement contour plots, strain contour plots, and strain increment contour plots.

[0018] Furthermore, the multi-level mesh deformation iterative algorithm in step b includes the following steps:

[0019] b1. Divide the reference image into a uniform initial grid subset, for a center coordinate of ( , The reference subset is traversed within the search region of the target image, and the similarity between the reference subset and the target candidate subset is evaluated by calculating the standardized cross-correlation coefficient C, as follows:

[0020] ;

[0021] in, Represents the grayscale value of pixels in the reference subset; The grayscale value of the corresponding pixel in the target candidate subset; , These are the mean gray values ​​within the subsets; , () is a candidate value for integer pixel displacement;

[0022] , ;

[0023] Where S represents the reference subset, and N is the total number of pixels in subset S;

[0024] By finding the integer pixel displacement that maximizes the C value ( , This yields the integer pixel displacement estimate of the center point of the subset;

[0025] b2. The deformation within the subset is described using first-order form functions, as follows:

[0026] ;

[0027] ;

[0028] in,( , () represents the center coordinates of a subset of the reference image; , These are the displacement components of the center point of this subset in the x and y directions; , , , The displacement gradient describes the tensile and shear deformation of the subset; , () are the center coordinates of the target subset;

[0029] Obtain the deformation parameters of each subset ( , , , , , );

[0030] b3. Continuously optimize deformation parameters through iteration ( , , , , , This maximizes the cross-correlation between the deformed target subset image and the reference subset image; and a three-dimensional Gaussian surface is fitted to the calculated cross-correlation peak, repeating the calculation until the displacement change is less than a set threshold, finally outputting the high-precision subpixel displacement of the subset. , ); combine the displacement vectors of all subsets to form a displacement field;

[0031] b4. A multi-level mesh iteration strategy from coarse to fine is adopted. The initial displacement field is obtained on the initial mesh in the first level calculation. In each subsequent level, the mesh size is halved and the displacement field of the previous level is used as the initial guess for iteration. The optimization iteration continues until the mesh size reaches the preset minimum value or the displacement field change converges.

[0032] Furthermore, based on the aforementioned soil liquefaction surface deformation field measurement system using digital image technology, the measurement process includes the following steps:

[0033] S1. Hinge the soil box to the support plate, set the tilt angle of the soil box using the angle adjustment mechanism, place the pore water pressure sensor in the soil box, and fix the acceleration sensor to the support plate.

[0034] S2. Set up the image acquisition unit and the lighting unit, and adjust the parameters of the image acquisition unit to match the estimated maximum deformation rate of the soil.

[0035] S3. Start the control program, synchronously control the power loading system through the servo control and data acquisition system to drive the soil box to vibrate according to the predetermined waveform, trigger the image acquisition unit to continuously shoot at the preset frame rate, acquire image sequences, and simultaneously acquire data from the pore water pressure sensor and the acceleration sensor; S4. Process the image sequences using a multi-level grid deformation iterative algorithm to obtain the displacement field; S5. Calculate the overall strain and strain increment based on the displacement field; S6. Generate displacement cloud map, overall strain cloud map, and strain increment cloud map, and combine the pore water pressure and acceleration data to analyze the deformation evolution law during the soil liquefaction process.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] 1. It abandons traditional contact sensors and uses digital image correlation technology to achieve non-invasive, full-field, continuous dynamic measurement of soil surface deformation field, capturing more comprehensive information;

[0038] 2. By highly integrating dynamic loading, model attitude adjustment, multi-sensor synchronous acquisition, and image acquisition and processing, the experimental process is fully automated and precisely synchronized.

[0039] 3. The design of the bottom hinge of the soil box with adjustable support makes the system able to easily simulate horizontal sites and slope conditions with different inclination angles, improving the flexibility of the test.

[0040] 4. The PIV algorithm, which combines first-order shape function to describe deformation and multi-level grid iteration strategy, can effectively handle the large dynamic range deformation from small deformation to violent flow during soil liquefaction and obtain a high-precision displacement field. At the same time, the strain calculation method based on finite deformation theory is used to more accurately characterize the large deformation process.

[0041] 5. It can generate various cloud maps such as displacement, overall strain, and strain increment, and analyze them together with pore water pressure and acceleration data, providing intuitive and quantitative data support for revealing the entire process of soil liquefaction initiation, deformation localization, shear zone formation and lateral flow slip development. Attached Figure Description

[0042] Figure 1 This is a schematic diagram of the overall structure of the soil liquefaction surface deformation field measurement system based on digital image technology in this invention;

[0043] Figure 2 This is a schematic diagram of the soil box angle adjustment mechanism in this invention;

[0044] Figure 3 This is a schematic diagram of the image acquisition structure in this invention;

[0045] Figure 4 This is a schematic diagram illustrating the displacement deformation principle of the present invention;

[0046] Figure 5 This is a schematic diagram of the multi-level grid iteration process in this invention;

[0047] Figure 6 Here is an example of a 48s displacement cloud map generated in this embodiment;

[0048] Figure 7 This is an example of a 48s overall strain cloud diagram generated in this embodiment;

[0049] Figure 8 This is an example of a 48s strain increment cloud map generated in this embodiment.

[0050] In the diagram: 1. Control platform; 2. Servo control and data acquisition system; 3. Workbench; 5. Pore water pressure sensor; 6. Accelerometer; 7. Horizontal guide rail; 8. Support plate; 9. Slider; 10. Soil box; 11. Rocker arm; 12. Image acquisition unit; 13. Lighting unit; 14. Limiter; 15. Liftable support column. Detailed Implementation

[0051] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments:

[0052] Example 1:

[0053] like Figure 1 As shown, this embodiment provides a soil liquefaction surface deformation field measurement system based on digital image technology, including:

[0054] Control platform 1 is a high-performance computer;

[0055] Servo control and data acquisition system 2 is connected to control platform 1 and is equipped with pore water pressure sensor 5 and acceleration sensor 6 for synchronously acquiring pore water pressure and acceleration data during the test; workbench 3 is on which horizontal guide rail 7 is laid; support plate 8 is mounted on horizontal guide rail 7 by slider 9 and can slide along horizontal guide rail 7, and limiter 14 is also provided at the end of horizontal guide rail 7;

[0056] A soil box 10 is placed on a support platform 8. The bottom center of the soil box 10 is hinged to the top center of the support platform 8, and at least one side is transparent. An angle adjustment mechanism is located between the soil box 10 and the support platform 8. Figure 2 As shown, the angle adjustment mechanism includes several liftable supports 15 evenly distributed on the support platform 8, which change the inclination angle of the soil box 10 by adjusting the liftable supports 15; the power loading system 4 is connected to the servo control and data acquisition system 2, and is connected to the slider 9 through the rocker arm 11, used to drive the slider 9 and the soil box 10 to reciprocate along the horizontal guide rail 7 to simulate horizontal seismic loads; the digital image deformation measurement system includes an image acquisition unit 12, an illumination unit 13, and a data processing and analysis unit, such as Figure 3 As shown, the image acquisition unit 12 is positioned facing the transparent surface of the soil box 10 and includes at least one digital camera. The digital camera uses a global shutter and is connected to the servo control and data acquisition system 2 via an external trigger. The lighting unit 13 is a low-heat, flicker-free LED array light source. The data processing and analysis unit filters the image sequences acquired by the image acquisition unit 12 and performs PIV calculation.

[0057] The data processing and analysis unit is configured to perform the following steps:

[0058] a. Acquire the image sequence acquired by the image acquisition unit 12, set a clear image before the vibration begins as the reference image, and use the subsequent frames as the target images;

[0059] b. A multi-level mesh deformation iterative algorithm is used to calculate the displacement field between the reference image and each subsequent target image, as follows;

[0060] b1. Divide the reference image into a uniform initial grid subset, for a center coordinate of ( , The reference subset is traversed within the search region of the target image, and the similarity between the reference subset and the target candidate subset is evaluated by calculating the standardized cross-correlation coefficient C, as follows:

[0061] ;

[0062] in, Represents the grayscale value of pixels in the reference subset; The grayscale value of the corresponding pixel in the target candidate subset; , These are the mean gray values ​​within the subsets; , () is a candidate value for integer pixel displacement;

[0063] By finding the integer pixel displacement that maximizes the C value ( , This yields the integer pixel displacement estimate of the center point of the subset;

[0064] b2. Use first-order shape functions to describe the deformation within the subset, such as Figure 4 As shown, the shape function defines the coordinates (x, y) of the reference subset and the coordinates of the target subset. , The mapping relationship is as follows:

[0065] ;

[0066] ;

[0067] in,( , () represents the center coordinates of a subset of the reference image; , These are the displacement components of the center point of this subset in the x and y directions; , , , The displacement gradient describes the tensile and shear deformation of the subset; , () are the center coordinates of the target subset;

[0068] Obtain the deformation parameters of each subset ( , , , , , );

[0069] b3. Continuously optimize deformation parameters through iteration ( , , , , , This maximizes the cross-correlation between the deformed target subset image and the reference subset image; and a three-dimensional Gaussian surface is fitted to the calculated cross-correlation peak, repeating the calculation until the displacement change is less than a set threshold, finally outputting the high-precision subpixel displacement of the subset. , ); combine the displacement vectors of all subsets to form a displacement field;

[0070] b4. After forming the initial displacement field, outlier vector removal is performed first, followed by data interpolation and smoothing. Outlier vector removal: The residual between each point in the displacement field and the median displacement vector of its eight neighboring points is calculated using the normalized median test. If the residual of a point is greater than a preset threshold, it is identified as an outlier vector and removed. Data interpolation and smoothing: Data gaps resulting from outlier removal are filled using bilinear interpolation or local polynomial fitting of nearby valid data.

[0071] b5, such as Figure 5 As shown, a multi-level mesh iteration strategy from coarse to fine is adopted. The first level calculates the initial displacement field on the initial mesh. In each subsequent level, the mesh size is halved, and the previous level displacement field is used as the initial guess for iteration. The optimization iteration continues until the mesh size reaches the preset minimum value or the displacement field change converges.

[0072] c. The overall strain and strain increment of the soil are calculated using the Green-Lagrange strain tensor based on the finite deformation theory;

[0073] The overall strain reflects the cumulative deformation of the soil relative to its initial state, as follows:

[0074] ;

[0075] ;

[0076] ;

[0077] in, , , These are the horizontal normal strain, vertical normal strain, and shear strain, respectively. The first-order partial derivatives in the formula can be obtained by calculating the displacement field using the central difference method.

[0078] d. Generate and output displacement contour plots, strain contour plots, and strain increment contour plots.

[0079] Example 2:

[0080] This embodiment, based on the soil liquefaction surface deformation field measurement system using digital image technology in Embodiment 1, provides the measurement steps of this system as follows:

[0081] S1. Hinge the soil box 10 to the support plate 8, set the tilt angle of the soil box 10 using the angle adjustment mechanism, place the pore water pressure sensor 5 in the soil box, and fix the acceleration sensor 6 to the support plate 8.

[0082] S2. Set up the image acquisition unit 12 and the lighting unit 13, and adjust the parameters of the image acquisition unit 12 to match the estimated maximum deformation rate of the soil.

[0083] S3. Start the control program and synchronously control the power loading system 4 to drive the soil box 10 to vibrate according to a predetermined waveform through the servo control and data acquisition system 2. Trigger the image acquisition unit 12 to continuously shoot at a preset frame rate to acquire image sequences, and simultaneously acquire data from the pore water pressure sensor 5 and the acceleration sensor 6; S4. Process the image sequences using a multi-level grid deformation iteration algorithm to obtain the displacement field; S5. Calculate the overall strain and strain increment based on the displacement field; S6. Generate displacement cloud map, overall strain cloud map and strain increment cloud map, and combine pore water pressure and acceleration data to analyze the deformation evolution law during the soil liquefaction process.

[0084] like Figure 6 , 7 As shown in Figures 8 and 9, displacement contour maps, overall strain contour maps, and strain increment contour maps were obtained based on Examples 1 and 2. The general displacement contour map shows the displacement distribution at a certain moment during the experiment, with color gradients (e.g., from blue to red) indicating the displacement change from small to large. The overall strain contour map shows the cumulative overall strain distribution at different moments relative to the initial state of the experiment; high strain regions (e.g., red bands) can be used to identify potential sliding surfaces or shear bands that may form during vibration liquefaction. The strain increment contour map shows the distribution of newly generated strain increments over different time intervals, reflecting the instantaneous rate of deformation development.

[0085] Displacement contour plot analysis reveals a significant increase in soil surface displacement between 12 and 18 seconds after the start of the test, indicating rapid soil movement during this phase. Overall strain contour plot analysis, showing cumulative strain distribution from the initial moment to 12, 15, and 18 seconds, indicates a continuous increase in strain value over time, reflecting a gradual accumulation of deformation. Strain increment contour plot analysis reveals a significant increase in strain increment during the 9-12 and 12-15 second periods, indicating a faster internal deformation rate within the soil. Combining time-history data such as pore water pressure and acceleration allows for a comprehensive analysis of the deformation evolution during soil liquefaction.

[0086] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any other way. Any modifications or equivalent changes made based on the technical essence of the present invention shall still fall within the scope of protection claimed by the present invention.

Claims

1. A soil liquefaction surface deformation field measurement system based on digital image technology, characterized in that, include: Control platform; A servo control and data acquisition system is connected to the control platform and is equipped with a pore water pressure sensor and an acceleration sensor to synchronously acquire pore water pressure and acceleration data during the experiment; a workbench is laid with horizontal guide rails; a support plate is mounted on the horizontal guide rails by a slider and can slide along the horizontal guide rails; A soil box is placed on a support platform, with its bottom center hinged to the top center of the support platform, and at least one side is transparent. An angle adjustment mechanism is located between the soil box and the support platform to adjust the tilt angle of the soil box. A power loading system is connected to the servo control and data acquisition system and is connected to the slider via a rocker arm to drive the slider and the soil box to reciprocate along a horizontal guide rail to simulate horizontal seismic loads. A digital image deformation measurement system includes an image acquisition unit, an illumination unit, and a data processing and analysis unit, with the image acquisition unit facing the transparent surface of the soil box.

2. The soil liquefaction surface deformation field measurement system based on digital image technology according to claim 1, characterized in that, The angle adjustment mechanism includes several liftable supports evenly distributed on the support platform, and the tilt angle of the soil box can be changed by adjusting the liftable supports.

3. The soil liquefaction surface deformation field measurement system based on digital image technology according to claim 1, characterized in that, A limiter is also provided at the end of the horizontal guide rail.

4. The soil liquefaction surface deformation field measurement system based on digital image technology according to claim 1, characterized in that, The image acquisition unit includes at least one digital camera; the illumination unit is an LED array light source; the data processing and analysis unit filters the image sequences acquired by the image acquisition unit and performs PIV calculation.

5. A soil liquefaction surface deformation field measurement system based on digital image technology according to claim 4, characterized in that, The data processing and analysis unit is configured to perform the following steps: a. Acquire the image sequence acquired by the image acquisition unit, set a clear image before the vibration begins as the reference image, and use the subsequent frames as the target images; b. A multi-level mesh deformation iterative algorithm is used to calculate the displacement field between the reference image and each subsequent target image; c. The overall strain and strain increment of the soil are calculated using the Green-Lagrange strain tensor based on the finite deformation theory; d. Generate and output displacement contour plots, strain contour plots, and strain increment contour plots.

6. The soil liquefaction surface deformation field measurement system based on digital image technology according to claim 5, characterized in that, The multi-level mesh deformation iterative algorithm in step b includes the following steps: b1. Divide the reference image into a uniform initial grid subset, for a center coordinate of ( , The reference subset is traversed within the search region of the target image, and the similarity between the reference subset and the target candidate subset is evaluated by calculating the standardized cross-correlation coefficient C, as follows: ； in, Represents the grayscale value of pixels in the reference subset; The grayscale value of the corresponding pixel in the target candidate subset; , These are the average gray levels within the subsets; , ) is a candidate value for integer pixel displacement; ， ； Where S represents the reference subset, and N is the total number of pixels in subset S; By finding the integer pixel displacement that maximizes the C value ( , This yields the integer pixel displacement estimate of the center point of the subset; b2. The deformation within the subset is described using first-order form functions, as follows: ； ； in,( , () represents the center coordinates of a subset of the reference image; , These are the displacement components of the center point of this subset in the x and y directions; , , , The displacement gradient describes the tensile and shear deformation of the subset; , () are the center coordinates of the target subset; Obtain the deformation parameters of each subset ( , , , , , ); b3. Continuously optimize deformation parameters through iteration ( , , , , , This maximizes the cross-correlation between the deformed target subset image and the reference subset image; and a three-dimensional Gaussian surface is fitted to the calculated cross-correlation peak, repeating the calculation until the displacement change is less than a set threshold, finally outputting the high-precision subpixel displacement of the subset. , ); combine the displacement vectors of all subsets to form a displacement field; b4. A multi-level mesh iteration strategy from coarse to fine is adopted. The initial displacement field is obtained on the initial mesh in the first level calculation. In each subsequent level, the mesh size is halved and the displacement field of the previous level is used as the initial guess for iteration. The optimization iteration continues until the mesh size reaches the preset minimum value or the displacement field change converges.

7. A soil liquefaction surface deformation field measurement system based on digital image technology according to claim 6, characterized in that, Using it for measurement involves the following steps: S1. Hinge the soil box to the support plate, set the tilt angle of the soil box using the angle adjustment mechanism, place the pore water pressure sensor in the soil box, and fix the acceleration sensor to the support plate. S2. Set up the image acquisition unit and the lighting unit, and adjust the parameters of the image acquisition unit to match the estimated maximum deformation rate of the soil. S3. Start the control program, synchronously control the power loading system through the servo control and data acquisition system to drive the soil box to vibrate according to the predetermined waveform, trigger the image acquisition unit to continuously shoot at the preset frame rate, acquire image sequences, and simultaneously acquire data from the pore water pressure sensor and the acceleration sensor; S4. Process the image sequences using a multi-level grid deformation iterative algorithm to obtain the displacement field; S5. Calculate the overall strain and strain increment based on the displacement field; S6. Generate displacement cloud map, overall strain cloud map, and strain increment cloud map, and combine the pore water pressure and acceleration data to analyze the deformation evolution law during the soil liquefaction process.

Images