A box dimension-based crack detection and damage assessment method for brittle materials

Abstract

The application discloses a box dimension-based crack detection and damage evaluation method for brittle materials. The application adopts a DIC technology (digital image correlation technology) to calculate the change information of the material surface displacement field, obtains a principal strain diagram, and then combines a fractal theory to calculate the box dimension of the material surface principal strain field, and quantifies the damage degree of the material based on the change of the box dimension. The application can observe the development process of the crack of the material specimen from nothing to through the specimen, can obtain the crack growth path, can quantify the development of the crack through the specific box dimension, can evaluate the damage degree of the structure, can realize the early warning of the large crack expansion behavior, and has important practical engineering significance for the health detection and risk evaluation of the brittle material structure such as the steel fiber reinforced concrete.

Description

Technical Field

[0001] This invention relates to the field of material damage detection technology, specifically to a method for crack detection and damage assessment of brittle materials based on box-dimensionality. Background Technology

[0002] Concrete not only possesses excellent physical properties such as compressive and impact resistance, but is also easy to mold and economical, making it the most widely used and consumed building material in the world. To address the relatively low tensile strength of concrete, steel fibers are typically added to create high-performance steel fiber reinforced concrete, which boasts superior combined compressive and tensile mechanical properties.

[0003] Steel fiber reinforced concrete (SFR) is a typical heterogeneous multiphase composite material. Its macroscopic failure is a cumulative process of the propagation and evolution of various microscopic damages, such as internal matrix cracks, aggregate-matrix interface damage, and fiber pull-out. At the microscopic level, numerous disordered micro-defects are distributed among the mortar, cement, and fibers. Under external loads, the propagation of these micro-defects (microcracks and pores) is random and irregular, and the crack development is closely related to the mechanical properties of fiber-reinforced concrete. Therefore, analyzing the crack development in concrete is of great significance; a better assessment of the degree of failure in steel fiber reinforced concrete provides an important reference for the damage detection of in-service steel fiber reinforced concrete buildings.

[0004] When macroscopic cracks appear on the surface of concrete, it indicates that the structure has fractured. Current research often employs acoustic emission technology to study the damage mechanism of steel fiber reinforced concrete, but these methods cannot detect the development of cracks, cannot provide quantitative assessments, and cannot offer early warnings for large crack activity. Summary of the Invention

[0005] In view of this, the present invention provides a method for crack detection and damage assessment of brittle materials based on box dimension, which can detect and quantify the development state of cracks and realize early warning of large crack activity.

[0006] The present invention provides a method for material crack detection and damage assessment based on box-dimensionality, comprising:

[0007] Step 1: Draw several speckles on the surface of the material to be tested in the testing area, and take pictures of the testing area of ​​the material to be tested using a high-speed camera;

[0008] Step 2: Using the first hypersonic image as a reference image, use DIC technology to track and calculate the displacement field change information between the speckle on each hypersonic image and the reference image, calculate the strain field of each hypersonic image, and obtain the master strain map;

[0009] Step 3: Use box counting to determine the box dimension of the principal strain map of each hyperscopy image obtained in Step 2;

[0010] Step 4: Draw a "box dimension-time" relationship diagram; based on the diagram, assess the damage to the material: if the box dimension is 1, it indicates that the material under test is not damaged; if the box dimension is 2, it indicates that the material under test is completely damaged; if the box dimension increases in the "box dimension-time" relationship diagram, it indicates that the degree of damage to the material under test is gradually increasing; the larger the box dimension, the greater the degree of damage to the material under test; when the box dimension reaches its peak, it means that the material under test is about to break.

[0011] Preferably, in step 1, a matte white paint is first sprayed onto the middle area of ​​the material to be tested, and then a black oil-based pen is used to create speckled patterns.

[0012] Ideally, the density of the speckle should be 50%.

[0013] Preferably, in step 2, a hyperphotograph is selected for every 10% increase in peak load before the peak and a hyperphotograph is selected every 50 seconds after the peak. The strain field is calculated using DIC technology to obtain the master strain map, and then step 3 is executed.

[0014] Preferably, in step 2, the strain field and principal strain band of each hypersonic image are calculated using Ncorr software; then, image processing methods are used to remove the strain region in the image except for the principal strain band to obtain the principal strain map, and step 3 is executed.

[0015] Preferably, step 3 specifically involves: for each principal strain diagram obtained in step 2,

[0016] S31, cover the main strain band region of the main strain map with a square grid with side length r, and count the number of grids containing strain information, denoted as N(r);

[0017] S32, change the lattice size r, repeat S31, and obtain the corresponding N(r);

[0018] S33, repeat S32 to obtain N(r) corresponding to different r; perform linear fitting on lnN(r)-lnr, and the slope of the fitted line is the box dimension of the principal strain diagram.

[0019] Preferably, the material to be tested is steel fiber reinforced concrete, concrete, rock, or ceramic.

[0020] Preferably, the material to be tested is steel fiber reinforced concrete, and the material to be tested will break when the box-count reaches 1.7.

[0021] Beneficial effects:

[0022] (1) This invention uses DIC (Digital Image Correlation) technology to calculate the changes in the displacement field on the material surface, obtains the principal strain map, and then combines fractal theory to calculate the box dimension of the principal strain field on the material surface. Based on the changes in the box dimension, the degree of material damage is quantified. This invention can observe the development process of internal cracks in material specimens from scratch to penetration, obtain the crack growth path, and assess the degree of structural damage by quantifying crack development through specific box dimensions. It can achieve early warning of the propagation behavior of large cracks and has important practical engineering significance for the health detection and risk assessment of steel fiber reinforced concrete structures.

[0023] (2) This invention can be used not only for crack development and damage assessment of steel fiber reinforced concrete structures, but also for damage assessment of brittle materials such as concrete, rock, and ceramics. Attached Figure Description

[0024] Figure 1 This is a flowchart illustrating a specific embodiment of the present invention;

[0025] Figure 2 This is a schematic diagram of a three-point bend experiment in a specific embodiment of the present invention;

[0026] Figure 3 This is a strain field diagram in the observation region of the peak point in a specific embodiment of the present invention;

[0027] Figure 4 This is a schematic diagram of the principal strain at the peak point after Photoshop processing in a specific embodiment of the present invention;

[0028] Figure 5 This is a schematic diagram of the peak point lnN(r)-lnr fitting result in a specific embodiment of the present invention;

[0029] Figure 6 This is a diagram illustrating the strain growth process as determined in a specific embodiment of the present invention;

[0030] Figure 7 This is a graph showing the relationship between load and box dimension and time in a specific embodiment of the present invention. Detailed Implementation

[0031] The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0032] This invention provides a method for material crack detection and damage assessment based on box-dimensionality, such as... Figure 1 As shown, it includes the following steps:

[0033] Step 1: Draw several scattered spots on the test area of ​​the material surface. You can first spray matte white paint on the test area of ​​the specimen, and then use a black oil-based marker to randomly dot the spots, with a spot density of about 50%.

[0034] Step 2: Place the specimen on the three-point bend testing machine and aim the high-speed camera at the testing area of ​​the specimen. Set the control parameters of the testing machine, adjust the focal length and aperture of the high-speed camera lens, and set appropriate resolution and frame rate.

[0035] Step 3: Simultaneously turn on the testing machine and the high-speed camera to conduct a three-point bend test.

[0036] Step 4: Using the hyperphotograph image of the specimen before deformation as a reference image, the DIC technique is used to track and calculate the spatial changes between the speckle patterns on the reference image and the hyperphotograph image of the specimen after deformation, so as to obtain the strain field and principal strain band of each hyperphotograph image.

[0037] Step 5: Since the fracture path of the specimen is mainly manifested in the principal strain zone, image processing software such as Photoshop is used to process the image, removing strain areas outside the principal strain zone to obtain the principal strain map. Images of the principal strain field of the specimen at the initial concentration, expansion, and steady-state stages of the principal strain are extracted.

[0038] Step 6: Use the box count method to determine the box dimension of the principal strain diagram from Step 5.

[0039] The formula for calculating the box dimension D is:

[0040]

[0041] Where r is the mesh side length and N(r) is the number of meshes containing strain information.

[0042] In practical applications, the grid side length r cannot be zero. Therefore, this invention employs the slope method: First, a square grid with side length r is used to cover the image obtained in step five, and the number of grids containing strain information is counted, denoted as N(r). Then, the grid size r is changed to alter the grid density, and the total number of grids covering strain information is recorded, resulting in a double logarithmic lnN(r)-lnr plot. The least squares method is used to fit a straight line to the data points in the double logarithmic lnN(r)-lnr plot, and the slope of the fitted line is the box dimension.

[0043] Step 7: Draw a graph showing the relationship between box dimension, load, and time to determine the relationship between the degree of material damage and box dimension. Based on the graph, perform crack detection and damage assessment on the material.

[0044] Example 1

[0045] In this embodiment, for a sample containing 15 kg / m 3 Three-point bending tests were conducted on steel fiber reinforced concrete beam specimens. The concrete beam specimens measured 400mm × 100mm × 100mm. A pre-fabricated notch measuring 0.5mm × 100mm × 25mm was cut in the middle of the beam.

[0046] Step 1, such as Figure 2 As shown, a matte white paint was first sprayed on the 80×100mm area in the middle of the steel fiber reinforced concrete beam specimen, and then a black oil-based pen was used to dot the pattern to ensure that the density of the dots was about 50%.

[0047] Step 2: Place the specimen on the testing machine. Set the testing machine to displacement control, and the indenter speed to 0.05 mm / min. Place the high-speed camera on a tripod in front of the testing machine, and adjust the camera height and lens angle to ensure that the camera optical axis is perpendicular to the speckled surface of the specimen, avoiding experimental errors caused by lens distortion. Adjust the lens focal length to ensure the image is as clear as possible. Connect the high-speed camera to the control microcomputer using a data cable for data transmission. In this embodiment, the resolution is set to 512×512, and the frame rate is 24fps. Additionally, supplement the lighting source as needed for the experiment.

[0048] Step 3: Simultaneously turn on the control switches of the testing machine and the high-speed camera, and turn off the testing machine after the high-speed camera's storage space is full.

[0049] Step 4: Based on the load curve of the testing machine, in this embodiment, a hyperphotograph is selected every 10% increase in peak load before the peak of the load curve, and a hyperphotograph is selected every 50 seconds after the peak. The first hyperphotograph is set as the reference image, and the strain field of the other hyperphotographs is calculated using Ncorr software.

[0050] Step 5: Use Photoshop to delete the strain areas (excluding the main strain band) from the strain field image obtained in Step 4, and export the image with a resolution of 1840×1200 and JPG format. In this embodiment, the peak point is taken as an example. Figure 3 For peak full-field strain, Figure 4 The master strain diagram remaining after processing with Photoshop software.

[0051] Step 6: Taking peak strain as an example, the first time, the mesh size is set to 20×20 resolution for coverage, and the number of boxes containing strain information is counted as 289, i.e., N(r) = 289. The second time, the mesh size is set to 10×10, and N(r) is counted as 996. Then, a plot of ln N(r) - ln r is drawn, and linear fitting is performed using the least squares method.

[0052] In this embodiment, the formula for the box-dimensional fitting curve is:

[0053]

[0054] Where D is the box dimension and t is time.

[0055] The fitting results are as follows Figure 5 As shown, the slope of the line is -1.7144, so the box dimension is 1.7144.

[0056] Step 7: In this embodiment, strain concentration begins at approximately 70%, followed by rapid development. After the peak load, strain growth slows down, and the strain width gradually increases. The strain variation is consistent with the actual crack variation. (Appendix) Figure 6 As shown, in this embodiment, 12 principal strain growth moments are selected to measure the box dimension.

[0057] This invention provides a highly effective method for quantifying the degree of damage to steel fiber reinforced concrete structures based on box-counting theory. When the specimen structure shows no damage, the box-counting dimension is 1; when the specimen structure is completely damaged, the box-counting dimension is 2. In this embodiment, the relationship between experimental load, box-counting dimension, and time is as follows: Figure 7 As shown, the fitted box-time curve correlates well with the failure of the concrete structure. The fitted box-time curve is mainly divided into two parts: the pre-peak part and the post-peak part. In the pre-peak stage, when the principal strain field is concentrated, the box-time number is approximately 1.3, and then increases rapidly with the growth of the principal strain band, tending to be consistent with the increase of the load. In the pre-peak stage, the increase in box-time number indicates that the degree of damage is gradually increasing. The box-time number of the strain field of the concrete beam is 1.714 at the peak. After the peak, the principal strain (crack) increases slowly, and the box-time number also increases slowly. The degree of failure of the specimen is positively correlated with the box-time value. In this embodiment, when the box-time number reaches 1.7, it means that the specimen is about to fracture.

[0058] The present invention also extends formula (2) to:

[0059]

[0060] In the formula, parameter D0 is related to the box-count of the principal strain, and it is advisable to take the box-count of the maximum principal strain; t0 is the moment of initial strain concentration; w is a parameter related to the steel fiber content, w>0, and the larger the steel fiber content, the larger the value of w. The value of w determines the box-count of the pre-peak stage, and the larger the value of w, the smaller the box-count of the pre-peak stage.

[0061] The results of this embodiment demonstrate that the box-count method can be used to assess the degree of damage to high-performance steel fiber reinforced concrete. The relationship between the degree of damage and the box-count is summarized as follows:

[0062] 1) In the pre-peak stage (when the specimen is not destroyed), the higher the box dimension, the higher the probability of specimen destruction; when the box dimension reaches 1.7, it indicates that the specimen is about to be destroyed.

[0063] 2) In the post-peak stage (when the specimen has been destroyed), the box dimension is positively correlated with the degree of destruction; the higher the box dimension, the higher the degree of destruction.

[0064] In summary, the above are merely preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for crack detection and damage assessment of brittle materials based on box-counting dimension, characterized in that, The method comprises the following steps: Step 1: draw a plurality of speckle points on the surface detection area of the material to be detected, and take pictures of the detection area of the material to be detected by using a high-speed camera; Step 2: take the first high-speed picture as a reference picture, track and calculate the displacement field change information between the speckles in each high-speed picture and the reference picture by using DIC technology, calculate the strain field of each high-speed picture, and obtain a principal strain map; Step 3: determine the box dimension of the principal strain map of each high-speed picture obtained in step 2 by using a box counting method; Step 4: draw a "box dimension-time" relationship diagram, and evaluate the damage of the material based on the relationship diagram: if the box dimension is 1, it indicates that the material to be detected has no damage; if the box dimension is 2, it indicates that the material to be detected is completely damaged; if the box dimension increases in the "box dimension-time" relationship diagram, it indicates that the damage degree of the material to be detected is gradually increasing; the greater the box dimension, the greater the damage degree of the material to be detected; when the box dimension reaches a peak value, it means that the material to be detected will be broken.

2. The method of claim 1, wherein, In step 1, first spray matt white paint in the middle area of the material to be detected, and then draw speckles by using a black oil-based pen.

3. The method of claim 1 or 2, wherein, The density of the speckles is 50%.

4. The method of claim 1, wherein, In step 2, select a high-speed picture every 10% of the peak load in the pre-peak stage, select a high-speed picture every 50s in the post-peak stage, calculate the strain field by using DIC technology, obtain a principal strain map, and execute step 3.

5. The method of claim 1 or 4, wherein, In step 2, calculate the strain field and the principal strain band of each high-speed picture by using Ncorr software, remove the strain area other than the principal strain band in the picture by using an image processing method, obtain a principal strain map, and execute step 3.

6. The method of claim 1, wherein, In step 3, for each principal strain map obtained in step 2, S31: cover the principal strain band area of the principal strain map with a square grid with a side length of r, and count the number of grids containing strain information, denoted as N(r); S32: change the grid size r, and repeat S31 to obtain the corresponding N(r); S33: repeat S32 to obtain N(r) corresponding to different r; perform linear fitting on lnN(r)-lnr, and the slope of the fitted line is the box dimension of the principal strain map.

7. The method of claim 1, wherein, The material to be detected is steel fiber reinforced concrete, concrete, rock or ceramic.

8. The method of claim 1, wherein, When the material to be detected is steel fiber reinforced concrete, the curve formula of the box dimension fitting is: wherein the parameter D0 is related to the box dimension of the principal strain; t0 is the time in the initial strain set; w is a parameter related to the content of the steel fiber.

9. The method of claim 1 or 8, wherein, When the material to be detected is steel fiber reinforced concrete, the material to be detected will be broken when the box dimension reaches 1.7.

Images