A digital image correlation-based buckling identification method

Abstract

The application discloses a buckling identification method based on digital image correlation, which comprises the following steps: S1, spraying random speckle or regular pattern on the surface of a test piece and calibrating a digital image correlation measurement system; S2, continuously collecting speckle images on the surface of the test piece through the measurement system; S3, dividing a grid on the collected speckle images; S4, calculating, analyzing and processing the collected speckle images through digital image correlation technology by using analysis software; S5, sequentially obtaining the strain field gradient of adjacent grid nodes, the total gradient module of the strain field in each speckle image and the normalized gradient of the total gradient module G of the strain field according to the full-field strain distribution diagram, and finally obtaining the speckle image serial number of the buckling start; and S6, obtaining the position of the buckling start according to the full-field strain distribution diagram of the corresponding test piece. The application can also determine the occurrence time of local buckling in the slow development area of local buckling by setting a calculation threshold.

Description

Technical Field

[0001] This invention relates to the field of digital image technology, and more specifically, to a buckling recognition method based on digital image correlation. Background Technology

[0002] With rapid economic development, steel structures and steel-concrete composite structures are increasingly widely used in construction engineering, such as steel-concrete composite tubes and steel-plate concrete shear walls. Along with the use of high-strength steel, the problem of local buckling in steel plates has become increasingly prominent. In mechanical property tests of steel components, the initial local buckling cannot be directly observed with the naked eye due to the long development process of buckling deformation. Therefore, it is necessary to use measuring instruments and local buckling identification methods to determine it. These methods mainly include: judgment based on strain sensor-based local area strain measurement results, judgment based on macroscopic load-displacement measurement results, and judgment based on digital image correlation measurement results.

[0003] Local buckling is based on strain sensor measurements of strain in a local area, using strain abrupt changes as a criterion. However, its applicability is limited due to the randomness of the location of local buckling. Existing methods for judging local buckling based on macroscopic load-displacement measurements mainly fall into two categories: one is based on elastic theory analysis, using a "displacement / load-displacement" curve to determine the initial local buckling time, such as the Southwell method; the other is to find abrupt change points in the "load-displacement" curve by observation or derivative calculation, such as finding the first nonlinear point on the load-displacement curve, thereby determining the initial local buckling time. The first type of method generally relies on elastic theory or small deflection theory, but due to initial defects and inelastic buckling problems in actual experiments, its judgment results have certain errors, limiting its applicability.

[0004] In the second type of method, the method of finding the abrupt change point of the curve by observation is affected by human factors; the method of finding the abrupt change point of the curve by calculating the derivative of the macroscopic load-displacement curve is prone to difficulties in obtaining the abrupt change point or the abrupt change point is inaccurate due to the randomness of the initial defects, the occurrence of local buckling at multiple locations and the influence of the development of plastic deformation. Moreover, the location of the critical initial local buckling cannot be determined from the macroscopic curve alone.

[0005] Digital Image Correlation (DIC) is a non-contact, full-field measurement method. It is easy to operate, has a wide measurement range, and requires minimal experimental conditions, and has been increasingly adopted in various disciplines such as civil engineering and aerospace. In mechanical property testing of steel components, DIC can be used to measure the full-field deformation data of the structure. Based on its ability to measure out-of-plane deformation data, it is divided into two-dimensional (2D) and three-dimensional (3D) DIC. 3D DIC can measure out-of-plane deformation data of the specimen, directly reflecting the occurrence of local buckling in the steel plate, but the cost of the equipment is high. 2D DIC can directly measure in-plane deformation data of the specimen, but for 3D deformation, it can only obtain 2D projected deformation data of the deformation area and cannot directly derive out-of-plane deformation data. However, the projected strain field of the component obtained by 2D DIC can still reflect the difference before and after local buckling. Compared with methods that determine the initial local buckling time based on macroscopic load-displacement curves, DIC can track the location, extent, and sequence of local buckling, offering significant advantages.

[0006] However, because the image data obtained by digital image correlation has two-dimensional or three-dimensional characteristics, it is more difficult to determine the initial local buckling point than through macroscopic load-displacement curves. In practice, although the strain field obtained by two-dimensional or three-dimensional digital image correlation can reflect the difference before and after local buckling, the strain field abrupt change in the plane (projected) strain map is not obvious because the development process of local buckling deformation is long, large, and generally irregular.

[0007] Therefore, current methods for determining the timing of initial local buckling using digital image correlation (DIR) rely on direct observation, which is highly subjective and forces analysts to focus on the later buckling stages where the local buckling range is larger. However, establishing a digital discrimination method and criteria for local buckling points in steel members based on the planar (projected) strain field obtained through DIR could significantly improve the accuracy of the discrimination results and enable rapid identification.

[0008] No effective solutions have yet been proposed to address the problems in the relevant technologies. Summary of the Invention

[0009] In view of the problems in related technologies, this invention proposes a buckling recognition method based on digital image correlation to overcome the aforementioned technical problems existing in the existing related technologies.

[0010] Therefore, the specific technical solution adopted by the present invention is as follows:

[0011] A buckling recognition method based on digital image correlation, the method comprising the following steps:

[0012] S1. Spray random speckle or regular pattern onto the surface of the test piece, install the test piece onto the testing instrument, and calibrate the digital image-related measurement system;

[0013] S2. Load the test piece and continuously acquire speckle images of the test piece surface through the measurement system;

[0014] S3. Specify the horizontal and vertical spacing of the grid to divide the image into grids.

[0015] S4. Using analysis software, the acquired speckle images are calculated, analyzed, and processed using digital image correlation technology to obtain the full-field strain distribution map;

[0016] S5. Based on the full-field strain distribution map, the strain field gradient of adjacent grid nodes, the total gradient magnitude of the strain field in each speckle image, and the normalized gradient of the total gradient magnitude of the strain field G are obtained in sequence. Finally, the speckle image sequence number at which buckling begins is obtained.

[0017] S6. Based on the speckle image sequence number of the buckling initiation, and according to the full-field strain distribution map of the corresponding test specimen, obtain the location of the buckling initiation.

[0018] Furthermore, the step of sequentially calculating the strain field gradient of adjacent grid nodes, the total strain field gradient magnitude in each speckle image, and the normalized gradient of the total strain field gradient magnitude G based on the full-field strain distribution map, and finally obtaining the speckle image sequence number at the start of buckling, includes the following steps:

[0019] S51. Calculate the strain field gradient of adjacent grid nodes in each speckle image based on the full-field strain distribution map and the strain field gradient calculation formula.

[0020] S52. Obtain the total gradient magnitude of the strain field in each speckle image based on the strain field gradient of adjacent grid nodes in each speckle image.

[0021] S53. Obtain the normalized gradient of the total gradient modulus G of the strain field in each speckle image based on the total gradient modulus of the strain field and the test specimen load, peak load or theoretical bearing capacity corresponding to each speckle image.

[0022] S54. Based on the normalized gradient of the total gradient magnitude G of the strain field in each speckle image, obtain the speckle image number at which buckling begins.

[0023] Furthermore, obtaining the total gradient magnitude of the strain field in each speckle image based on the strain field gradient of adjacent grid nodes in each speckle image includes the following steps:

[0024] S521. Calculate the strain gradient modulus of adjacent grid nodes based on the strain field gradient of adjacent grid nodes in each speckle image and using the formula for calculating the total gradient modulus of the strain field.

[0025] S522. Accumulate the strain gradient modes of all adjacent grid nodes to obtain the total strain field gradient mode in each speckle image.

[0026] Furthermore, obtaining the speckle image sequence number at the start of buckling based on the normalized gradient of the total strain field gradient magnitude G in each speckle image includes the following steps:

[0027] S541. Using the order in which the speckle images were captured as the x-axis and the normalized gradient of the total strain field magnitude G in the speckle images as the y-axis, we obtain... Bar chart;

[0028] S542, Obtain bar chart From negative to positive and remain positive thereafter The speckle image sequence number corresponding to the first occurrence of a value greater than the threshold.

[0029] Furthermore, the measurement system includes a high-speed camera and a light source.

[0030] Furthermore, the full-field strain distribution map is a two-dimensional strain distribution map;

[0031] The two-dimensional strain distribution map includes a two-dimensional transverse strain distribution map and a two-dimensional longitudinal strain distribution map.

[0032] Furthermore, the formula for calculating the strain field gradient is as follows:

[0033]

[0034]

[0035]

[0036]

[0037] In the formula, This represents the absolute value of the lateral strain gradient between adjacent nodes. This represents the absolute value of the transverse strain gradient between longitudinally adjacent nodes; This represents the absolute value of the longitudinal strain gradient between laterally adjacent nodes; Represents the absolute value of the longitudinal strain gradient between adjacent nodes; i represents the transverse grid node code; j represents the longitudinal grid node code; This represents the lateral strain at the grid node with coordinates (i, j); This represents the longitudinal strain at the grid node with coordinate (i, j); Indicates the horizontal spacing between adjacent grid nodes; This indicates the vertical spacing between adjacent grid nodes.

[0038] Furthermore, the formula for calculating the total gradient modulus of the strain field is:

[0039]

[0040] In the formula, G represents the total gradient mode of the strain field; This represents the absolute value of the lateral strain gradient between adjacent nodes. This represents the absolute value of the transverse strain gradient between longitudinally adjacent nodes; This represents the absolute value of the longitudinal strain gradient between laterally adjacent nodes; represents the absolute value of the longitudinal strain gradient of adjacent nodes; i represents the transverse grid node code; j represents the longitudinal grid node code.

[0041] Furthermore, the formula for calculating the normalized gradient of the total gradient magnitude G of the strain field is as follows:

[0042]

[0043] In the formula, k represents the sequential number of the speckle image; G represents the normalized gradient of the total gradient magnitude of the strain field in the k-th speckle image; It represents the ratio of the total gradient magnitude G of the strain field of the k-th speckle image to the total gradient magnitude G of the peak strain field of all speckle images in the group; This represents the ratio of the load on the test specimen to the peak load or theoretical bearing capacity corresponding to the k-th speckle image.

[0044] Furthermore, the aforementioned bar chart From negative to positive and remain positive thereafter The time when the speckle image sequence number corresponding to the first occurrence of a value greater than the threshold is the time when the test specimen begins to buckle.

[0045] The beneficial effects of this invention are as follows:

[0046] 1. Unlike methods that rely on observation or measurement of local areas to determine local buckling, the method for local buckling identification based on digital image correlation technology provided by this invention can obtain the full-field strain distribution map of the test specimen through digital image correlation technology, which is not affected by the randomness of local buckling location and has wider applicability.

[0047] 2. Unlike methods that rely on macroscopic load-displacement measurement results for local buckling judgment, the method for local buckling identification based on digital image correlation technology provided by this invention is not affected by the randomness of initial defects or the development of plastic deformation. It can determine the load that causes initial local buckling in the entire field or any defined area, track the location, range, and sequence of local buckling, and has wider applicability.

[0048] 3. The method for local buckling recognition based on digital image correlation technology provided by the present invention can objectively calculate the time of local buckling in areas with obvious local buckling, and can also determine the time of local buckling by setting a calculation threshold in areas with slow local buckling development.

[0049] 4. Unlike methods that rely on human observation to determine local buckling based on deformation results obtained from digital image correlation technology, the method of this invention can quantitatively provide buckling point selection results, effectively reducing the uncertainty and lag of observation results caused by human factors. It is significantly superior to the direct observation method in terms of objectivity and accuracy.

[0050] 5. When there are a large number of images, the method for local buckling identification based on digital image correlation technology provided by this invention can quickly identify the initial local buckling in different regions. Its applicability is also significantly better than the direct observation method. Furthermore, through calculation examples, it is shown that the buckling load obtained by the method of this invention is close to the buckling load obtained by theoretical analysis, with high accuracy, and can identify buckling abrupt changes that cannot be observed by the naked eye.

[0051] 6. Unlike previous two-dimensional digital image correlation techniques which were only used to measure in-plane deformation data, the method for local buckling identification based on digital image correlation techniques provided by this invention can perform calculations based on the measurement results of two-dimensional or three-dimensional digital image correlation techniques, thus expanding the applicable scenarios of two-dimensional digital image correlation techniques.

[0052] 7. In the normalized gradient calculation formula of the total gradient modulus G of the strain field provided by the present invention, by introducing the peak value G of all speckle images and the peak load or theoretical bearing capacity, the normalized gradient of the total gradient modulus G of the strain field is made into a dimensionless quantity. When the test specimen is subjected to displacement loading, the dimensionless method of the present invention can avoid the abnormality of the G gradient value caused by the large difference in the load increment corresponding to the speckle images taken at different loading stages of the test specimen. Moreover, the dimensionless method of the present invention can establish a local buckling identification standard by setting a unified dimensionless threshold, which is suitable for the situation of comparing local buckling between test specimens with different bearing capacities. Attached Figure Description

[0053] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0054] Figure 1 This is a flowchart of a buckling recognition method based on digital image correlation according to an embodiment of the present invention;

[0055] Figure 2 This is a schematic diagram of strain gradient calculation in a buckling recognition method based on digital image correlation according to an embodiment of the present invention;

[0056] Figure 3 This refers to the normalized gradient in a buckling recognition method based on digital image correlation in Embodiment 1 of the present invention. Bar chart;

[0057] Figure 4 This invention provides a method for buckling recognition based on digital image correlation, specifically a method for local buckling recognition based on digital image correlation, which yields a full-field strain distribution map corresponding to a speckle image captured before the buckling occurred.

[0058] Figure 5 This invention provides a buckling recognition method based on digital image correlation, specifically a local buckling recognition method based on digital image correlation, which yields a full-field strain distribution map at the moment of buckling.

[0059] Figure 6 This refers to the normalized gradient in a buckling recognition method based on digital image correlation in Embodiment 2 of the present invention. Bar chart;

[0060] Figure 7 This is an example of a buckling recognition method based on digital image correlation in Embodiment 2 of the present invention, which uses a local buckling recognition method based on digital image correlation to obtain the full-field strain distribution map corresponding to the speckle image captured before the buckling occurred.

[0061] Figure 8 This is an example of a buckling recognition method based on digital image correlation in Embodiment 2 of the present invention, which uses a local buckling recognition method based on digital image correlation to obtain the full-field strain distribution map at the moment of buckling. Detailed Implementation

[0062] To further illustrate the various embodiments, the present invention provides accompanying drawings, which are part of the disclosure of the present invention. These drawings are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these drawings, those skilled in the art should be able to understand other possible implementation methods and the advantages of the present invention.

[0063] According to an embodiment of the present invention, a buckling recognition method based on digital image correlation is provided.

[0064] The present invention will now be further described in conjunction with the accompanying drawings and specific embodiments, such as... Figure 1 As shown, the buckling recognition method based on digital image correlation according to an embodiment of the present invention includes the following steps:

[0065] S1. Spray random speckle or regular pattern onto the surface of the test piece, install the test piece onto the testing instrument, and calibrate the digital image-related measurement system;

[0066] The measurement system includes a high-speed camera and a light source.

[0067] S2. Load the test piece and continuously acquire speckle images of the test piece surface through the measurement system;

[0068] Specifically, the test specimen is loaded, and during the loading process, the measurement system continuously acquires speckle images of the specimen surface and saves relevant test data;

[0069] Specifically, the test-related data includes speckle images acquired by the measurement system and the corresponding loads applied to the test specimens.

[0070] S3. Specify the horizontal and vertical spacing of the grid to divide the image into grids.

[0071] Specifically, the grid is a uniformly divided grid on a two-dimensional speckle image; the grid node spacing is the grid node spacing on a two-dimensional plane. For a three-dimensional deformable surface obtained by the three-dimensional digital image correlation method, it needs to be projected onto a two-dimensional plane before the grid is divided, and the grid node spacing is taken as the grid node spacing on the projected plane.

[0072] S4. Using analysis software and digital image correlation technology, the collected speckle images are calculated, analyzed and processed to obtain a series of full-field strain distribution maps of the test specimens corresponding to each level of load.

[0073] The full-field strain distribution map is a two-dimensional strain distribution map.

[0074] The two-dimensional strain distribution map includes a two-dimensional transverse strain distribution map and a two-dimensional longitudinal strain distribution map;

[0075] Specifically, the full-field strain distribution map is a two-dimensional strain distribution map, which can be obtained from a single speckle image or calculated from three-dimensional deformation measured by three-dimensional digital image correlation technology; the two-dimensional strain distribution map includes a two-dimensional transverse strain distribution map and a two-dimensional longitudinal strain distribution map. By dividing the two-dimensional transverse strain distribution map and the two-dimensional longitudinal strain distribution map into grids, the transverse strain and longitudinal strain of the grid nodes can be obtained respectively for strain field gradient calculation.

[0076] S5. Based on the full-field strain distribution map, the strain field gradient of adjacent grid nodes, the total gradient magnitude of the strain field in each speckle image, and the normalized gradient of the total gradient magnitude of the strain field G are obtained in sequence. Finally, the speckle image sequence number at which buckling begins is obtained.

[0077] The step of sequentially calculating the strain field gradient of adjacent grid nodes, the total strain field gradient magnitude in each speckle image, and the normalized gradient of the total strain field gradient magnitude G based on the full-field strain distribution map, and finally obtaining the speckle image sequence number at the start of buckling, includes the following steps:

[0078] S51. Calculate the strain field gradient of adjacent grid nodes in each speckle image based on the full-field strain distribution map and the strain field gradient calculation formula.

[0079] Specifically, based on the full-field strain distribution map, the absolute value of the strain field gradient of adjacent grid nodes in each speckle image is calculated according to the strain field gradient calculation formula. This includes the absolute value of the lateral strain gradient of the laterally adjacent nodes. The absolute value of the longitudinal strain gradient of the lateral adjacent nodes The absolute value of the transverse strain gradient of longitudinally adjacent nodes The absolute value of the longitudinal strain gradient of adjacent nodes ;

[0080] like Figure 2 As shown, with Taking the calculation as an example, (i, j) and (i+1, j) are the numbers of two horizontally adjacent grid nodes in the full-field strain distribution map, then... The difference in lateral strain between points (i, j) and (i+1, j) in the figure is divided by the lateral distance between points (i, j) and (i+1, j). .

[0081] The formula for calculating the strain field gradient is as follows:

[0082]

[0083]

[0084]

[0085]

[0086] In the formula, This represents the absolute value of the lateral strain gradient between adjacent nodes. This represents the absolute value of the transverse strain gradient between longitudinally adjacent nodes; This represents the absolute value of the longitudinal strain gradient between laterally adjacent nodes; The absolute value of the longitudinal strain gradient between adjacent nodes is represented by i = 1, 2, ..., n, which represents the transverse grid node code; j = 1, 2, ..., m, which represents the longitudinal grid node code. This represents the lateral strain at the grid node with coordinates (i, j); This represents the longitudinal strain at the grid node with coordinate (i, j); Indicates the horizontal spacing between adjacent grid nodes; This indicates the vertical spacing between adjacent grid nodes.

[0087] S52. Obtain the total gradient magnitude of the strain field in each speckle image based on the strain field gradient of adjacent grid nodes in each speckle image.

[0088] The step of obtaining the total gradient magnitude of the strain field in each speckle image based on the strain field gradient of adjacent grid nodes in each speckle image includes the following steps:

[0089] S521. Calculate the strain gradient modulus of adjacent grid nodes based on the strain field gradient of adjacent grid nodes in each speckle image and using the formula for calculating the total gradient modulus of the strain field.

[0090] The formula for calculating the total gradient modulus of the strain field is as follows:

[0091]

[0092] In the formula, G represents the total gradient mode of the strain field; This represents the absolute value of the lateral strain gradient between adjacent nodes. This represents the absolute value of the transverse strain gradient between longitudinally adjacent nodes; This represents the absolute value of the longitudinal strain gradient between laterally adjacent nodes; Represents the absolute value of the longitudinal strain gradient between adjacent nodes; i represents the transverse grid node code; j represents the longitudinal grid node code;

[0093] Specifically, by controlling the range of values ​​for i and j, the speckle image can be divided into regions, and calculations can be performed on each region separately. Based on this, the time point at which the initial local buckling occurs in each region can be determined.

[0094] S522. Accumulate the strain gradient modes of all adjacent grid nodes to obtain the total strain field gradient mode G in each speckle image.

[0095] S53. Based on the total strain field gradient modulus G and the test specimen load, peak load, or theoretical bearing capacity corresponding to each speckle image, obtain the normalized gradient of G in each speckle image using the formula for calculating the normalized gradient of the total strain field gradient modulus G. ;

[0096] The formula for calculating the normalized gradient of the total gradient magnitude G of the strain field is as follows:

[0097]

[0098] In the formula, k=1,2……, which represents the sequential number of the speckle image, i.e., the number of photos taken; the initial speckle image is the 0th speckle image; G represents the normalized gradient of the total gradient magnitude of the strain field in the k-th speckle image; It represents the ratio of the total gradient magnitude G of the strain field of the k-th speckle image to the total gradient magnitude G of the peak strain field of all speckle images in the group; This represents the ratio of the load on the test specimen to the peak load or theoretical bearing capacity corresponding to the k-th speckle image.

[0099] S54. Based on the normalized gradient of the total gradient magnitude G of the strain field in each speckle image, obtain the speckle image number in which buckling begins.

[0100] The step of obtaining the speckle image sequence number at the start of buckling based on the normalized gradient of the total gradient magnitude G of the strain field in each speckle image includes the following steps:

[0101] S541. Using the order in which the speckle images were captured as the x-axis and the normalized gradient of the total strain field magnitude G in the speckle images as the y-axis, we obtain... Bar chart;

[0102] S542, Obtain bar chart From negative to positive and remain positive thereafter The first speckle image number corresponding to the value exceeding the threshold;

[0103] in, bar chart From negative to positive and remain positive thereafter The time when the speckle image sequence number first exceeds the threshold is the time when the specimen buckling begins. For The threshold value reached is preferably 1;

[0104] Specifically, the time point or loading step corresponding to the speckle image number is the time or loading step at which the buckling of the test specimen begins, and the test specimen load corresponding to the speckle image number is the load at which the test specimen experiences initial local buckling.

[0105] S6. Based on the speckle image sequence number of the buckling initiation, and according to the full-field strain distribution map of the corresponding test specimen, obtain the location of the buckling initiation.

[0106] The specific embodiments of the present invention will be further described below with reference to examples:

[0107] Example 1

[0108] This embodiment takes the static loading of a square-section hollow steel tube test specimen as an example. The square steel tube is a Q235B seamless steel tube, 900mm high, 300mm wide, and 6mm thick, with a width-to-thickness ratio of 50. A local buckling recognition method based on digital image correlation is used, which includes the following steps:

[0109] Step 1: Process the test surface of the test piece. Spray white paint on the test surface of the test piece and use a speckle tool to mark the test surface of the test piece. Adjust the camera parameters and light source so that the speckle image can clearly show the black spots on the test surface of the test piece. Prepare a calibration plate and take a speckle image with the calibration plate attached to the surface of the component to calculate the relationship between pixels and actual distance.

[0110] Step 2: Install the test specimen onto the testing instrument and load the specimen. Simultaneously, continuously acquire speckle images of the specimen surface using a camera. In the early stages, the image interval is controlled by the axial load increment, with the load interval being one-twentieth of the specimen's bearing capacity, which is 100kN in this test example. When the specimen approaches yield, switch to controlling the image interval by displacement increment, with a displacement interval of 0.2mm. After loading is completed, save the images from the camera and record the corresponding load.

[0111] Step 3: Using analysis software, the speckle images acquired by the measurement system are processed using digital image correlation technology. The grid is divided with a 10px grid width, and the full-field strain distribution map under different shooting sequences is calculated.

[0112] like Figures 4-5 As shown, Figure 4 In this embodiment, the local buckling recognition method based on digital image correlation yields the full-field strain distribution map corresponding to the speckle image captured before the buckling occurred. Figure 5 The digital image-related local buckling recognition method in this embodiment generates a full-field strain distribution map at the moment of buckling occurrence;

[0113] Step 4: Calculate the total gradient magnitude G of the strain field as described in this invention. Since the loading direction of the test specimen in this embodiment is longitudinal, the strain change of the test specimen is mainly longitudinal, and the longitudinal strain change of adjacent grid nodes in the lateral direction is small. Therefore, the calculation formula of the total gradient magnitude G of the strain field is simplified. After ignoring the lateral strain gradient and the longitudinal strain gradient of adjacent grid nodes in the lateral direction, the total gradient magnitude G of the strain field for each speckle image is calculated. The simplified calculation formula of G is as follows:

[0114]

[0115] Step 5: Based on the calculated absolute value G of the longitudinal strain field gradient in each speckle image, the test specimen load corresponding to each speckle image, and the theoretical bearing capacity of the test specimen, obtain the normalized gradient G in each speckle image according to the normalized gradient calculation formula of the total strain field gradient magnitude G. The calculation formula is as follows:

[0116]

[0117] Step 6: Using the order in which the speckle images were captured as the x-axis, normalize the gradient using the total gradient magnitude G of the strain field in each speckle image. Plot the vertical axis Bar chart;

[0118] like Figures 3-5 As shown, Figure 3 After the 23rd speckle image in the middle It changes from negative to positive and remains positive, while also combining... Figure 4 and Figure 5 It can be seen that the strain field at the moment of buckling occurrence obtained by the method of the present invention in this embodiment does have a large abrupt change, which verifies the effectiveness of the digital image-related local buckling recognition method.

[0119] Step 7: Observe the strain distribution diagram of the entire field of the specimen at this moment, as shown below. Figure 5 As shown, it can be determined that the specimen buckled locally at two locations: approximately 250 mm from the upper end plate and approximately 300 mm from the lower end plate.

[0120] Example 2

[0121] The difference between this embodiment and Embodiment 1 is that, during the loading process of the test specimen, water is injected and pressurized onto the steel pipe test specimen to consider the effect of lateral compression on the steel pipe. In the test, the deformation of the steel pipe consists of local buckling and overall outward bulging deformation, resulting in less pronounced local buckling deformation than in Embodiment 1. When using the method of this invention to identify buckling in Embodiment 2, according to... The occurrence of local buckling is determined by the first value exceeding the threshold of 1.

[0122] like Figure 6As shown, after loading the test specimen and processing by the analysis software, the normalized gradient in this embodiment is obtained. Bar chart;

[0123] like Figure 7 As shown, in this embodiment, the local buckling recognition method based on digital image correlation yields the full-field strain distribution map corresponding to the speckle image captured before the buckling occurred.

[0124] like Figure 8 As shown, the local buckling recognition method based on digital image correlation in this embodiment yields the full-field strain distribution map at the moment of buckling.

[0125] visible, Figure 6 After the 11th speckle image The value is greater than 1 for the first time, and at the same time combined with Figure 7 and Figure 8 As can be seen, in this embodiment, the strain field at the moment of buckling occurs is found to have a large abrupt change by the local buckling identification method of this digital image correlation, which verifies the effectiveness of the local buckling identification method of this digital image correlation.

[0126] Observe the strain distribution diagram of the test piece at this moment as follows: Figure 8 As shown, it can be determined that the specimen experienced local buckling at a distance of approximately 300 mm from the upper end plate and 300 mm from the lower end plate.

[0127] In summary, by utilizing the above-mentioned technical solution of the present invention, unlike methods that determine local buckling based on observation or measurement of local areas, the method for local buckling identification based on digital image correlation technology provided by the present invention can obtain the full-field strain distribution map of the test specimen through digital image correlation technology, and is not affected by the randomness of local buckling location, thus having wider applicability.

[0128] Unlike methods that rely on macroscopic load-displacement measurement results to determine local buckling, the method for local buckling identification based on digital image correlation technology provided in this invention is not affected by the randomness of initial defects or the development of plastic deformation. It can determine the load that causes initial local buckling in the entire field or any defined area, track the location, extent, and sequence of local buckling, and has wider applicability.

[0129] The method for local buckling recognition based on digital image correlation technology provided by this invention can objectively calculate the time of local buckling in areas with obvious local buckling, and can also determine the time of local buckling by setting a calculation threshold in areas with slow local buckling development.

[0130] Unlike methods that rely on human observation to determine local buckling based on deformation results obtained from digital image correlation techniques, the method of this invention can quantitatively provide buckling point selection results, effectively reducing the uncertainty and lag of observation results caused by human factors. It is significantly superior to the direct observation method in terms of objectivity and accuracy.

[0131] When there are a large number of images, the method for local buckling identification based on digital image correlation technology provided by this invention can quickly identify the initial local buckling in different regions. Its applicability is also significantly better than the direct observation method. Furthermore, through calculation examples, it is shown that the buckling load obtained by the method of this invention is close to the buckling load obtained by theoretical analysis, with high accuracy, and can identify buckling abrupt changes that cannot be observed by the naked eye.

[0132] Unlike previous two-dimensional digital image correlation techniques, which were only used to measure in-plane deformation data, the method for local buckling identification based on digital image correlation techniques provided in this invention can perform calculations based on the measurement results of two-dimensional or three-dimensional digital image correlation techniques, thus expanding the applicable scenarios of two-dimensional digital image correlation techniques.

[0133] The normalized gradient calculation formula for the total gradient modulus G of the strain field provided by this invention incorporates the peak value G of all speckle images and the peak load or theoretical bearing capacity, making the calculated normalized gradient of the total gradient modulus G of the strain field dimensionless. When the test specimen is subjected to displacement loading, the dimensionless method of this invention can avoid the abnormality of the G gradient value caused by the large difference in the load increment corresponding to the speckle images taken at different loading stages. Furthermore, the dimensionless method of this invention can establish a local buckling identification standard by setting a unified dimensionless threshold, which is suitable for comparing local buckling between test specimens with different bearing capacities.

[0134] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A buckling recognition method based on digital image correlation, characterized in that, The identification method includes the following steps: S1. Spray random speckle or regular pattern onto the surface of the test piece, install the test piece onto the testing instrument, and calibrate the digital image-related measurement system; S2. Load the test piece and continuously acquire speckle images of the test piece surface through the measurement system; S3. Specify the horizontal and vertical spacing of the grid to divide the image into grids. S4. Using analysis software, the acquired speckle images are calculated, analyzed, and processed using digital image correlation technology to obtain the full-field strain distribution map; S5. Calculate the strain field gradient of adjacent grid nodes in each speckle image based on the full-field strain distribution map and the strain field gradient calculation formula; calculate the strain gradient magnitude of adjacent grid nodes based on the strain field gradient of adjacent grid nodes in each speckle image and the strain gradient magnitude calculation formula; sum the strain gradient magnitudes of all adjacent grid nodes to obtain the total strain field gradient magnitude in each speckle image; obtain the normalized gradient of the total strain field gradient magnitude G in each speckle image based on the total strain field gradient magnitude and the test specimen load, peak load, or theoretical bearing capacity corresponding to each speckle image; obtain the speckle image sequence number where buckling begins based on the normalized gradient of the total strain field gradient magnitude G in each speckle image. S6. Based on the speckle image sequence number of the buckling initiation, and according to the full-field strain distribution map of the corresponding test specimen, obtain the location of the buckling initiation.

2. The buckling recognition method based on digital image correlation according to claim 1, characterized in that, The step of obtaining the speckle image sequence number of buckling initiation based on the normalized gradient of the total gradient magnitude G of the strain field in each speckle image includes the following steps: S541. Using the order in which the speckle images were captured as the x-axis and the normalized gradient of the total strain field magnitude G in the speckle images as the y-axis, we obtain... Bar chart; S542, Obtain bar chart From negative to positive and remain positive thereafter The speckle image sequence number corresponding to the first occurrence of a value greater than the threshold.

3. The buckling recognition method based on digital image correlation according to claim 1, characterized in that, The measurement system includes a high-speed camera and a light source.

4. The buckling recognition method based on digital image correlation according to claim 1, characterized in that, The full-field strain distribution diagram is a two-dimensional strain distribution diagram; The two-dimensional strain distribution map includes a two-dimensional transverse strain distribution map and a two-dimensional longitudinal strain distribution map.

5. The buckling recognition method based on digital image correlation according to claim 2, characterized in that, The formula for calculating the strain field gradient is: In the formula, This represents the absolute value of the lateral strain gradient between adjacent nodes. This represents the absolute value of the transverse strain gradient between longitudinally adjacent nodes; This represents the absolute value of the longitudinal strain gradient between laterally adjacent nodes; Represents the absolute value of the longitudinal strain gradient between adjacent nodes; i represents the transverse grid node code; j represents the longitudinal grid node code; This represents the lateral strain at the grid node with coordinates (i, j); This represents the longitudinal strain at the grid node with coordinate (i, j); Indicates the horizontal spacing between adjacent grid nodes; This indicates the vertical spacing between adjacent grid nodes.

6. The buckling recognition method based on digital image correlation according to claim 2, characterized in that, The formula for calculating the total gradient modulus of the strain field is: In the formula, G represents the total gradient mode of the strain field; This represents the absolute value of the lateral strain gradient between adjacent nodes. This represents the absolute value of the transverse strain gradient between longitudinally adjacent nodes; This represents the absolute value of the longitudinal strain gradient between laterally adjacent nodes; represents the absolute value of the longitudinal strain gradient between adjacent nodes; i represents the transverse grid node code, i=1, 2...n; j represents the longitudinal grid node code, j=1, 2...m.

7. The buckling recognition method based on digital image correlation according to claim 2, characterized in that, The formula for calculating the normalized gradient of the total gradient magnitude G of the strain field is as follows: In the formula, k represents the sequential number of the speckle image; G represents the normalized gradient of the total gradient magnitude of the strain field in the k-th speckle image; It represents the ratio of the total gradient magnitude G of the strain field of the k-th speckle image to the total gradient magnitude G of the peak strain field of all speckle images in the group; This represents the ratio of the load on the test specimen to the peak load or theoretical bearing capacity corresponding to the k-th speckle image.

8. The buckling recognition method based on digital image correlation according to claim 7, characterized in that, The bar chart From negative to positive and remain positive thereafter The time when the speckle image sequence number corresponding to the first occurrence of a value greater than the threshold is the time when the test specimen begins to buckle.

Images