Bridge bolt group damage identification method based on DIC

Abstract

The invention discloses a bridge bolt group damage identification method based on DIC, and relates to the technical field of bridge damage identification. According to the invention, the whole-field strain monitoring capability is improved, the technology upgrade from point-like monitoring to whole-field monitoring is realized through the DIC technology, the whole-field strain data of the bridge structure can be comprehensively acquired, and the strain change of the bolt group and the surrounding structure can be accurately reflected through the whole-field strain diagram acquired by the DIC equipment. According to the method, the occurrence position of the damage can be captured in real time, the damage evolution process can be analyzed, and the problems that a traditional strain gauge detection method can only provide local strain data and cannot comprehensively reflect strain distribution of a bridge high-strength bolt group, and accurate recognition of the damage is limited are solved.

Description

Technical Field

[0001] The invention relates to the technical field of bridge damage identification, and in particular to a bridge bolt group damage identification method based on DIC. Background Art

[0002] High-strength bolt connection is an important connection method for bridge steel structures, and its mechanical properties directly affect the overall stability and safety of the structure. However, under long-term complex alternating loads and environmental effects (such as vehicle dynamic loads, wind vibration, temperature changes, etc.), high-strength bolt connections are prone to slippage, loosening, fatigue cracks, and even fractures, which pose a threat to the safety and durability of bridges.

[0003] Traditional bolt connection detection and damage monitoring methods include strain gauges, ultrasonic testing, and vibration signal analysis. Although these methods can obtain the stress state or internal defect information of bolts within a certain range, they generally have the following limitations:

[0004] (1) Locality and non-full-field monitoring: Methods such as strain gauges can only obtain local point information of the bolt or connection area, making it difficult to achieve full-field strain distribution monitoring.

[0005] (2) Insufficient dynamic monitoring capabilities: Ultrasonic and vibration analysis methods have limited monitoring capabilities for strain distribution in dynamic environments, especially under alternating loads, which makes it difficult to capture the damage evolution process in real time.

[0006] (3) Weak environmental adaptability: Some methods are sensitive to environmental conditions (such as temperature, humidity, and vibration) and are difficult to adapt to complex working conditions such as bridges.

[0007] Of course, there are also strain gauge detection methods and ultrasonic detection methods in the prior art. The strain gauge detection method: by sticking the strain gauge on the bolt head or the connecting plate, the local strain data is measured to analyze the stress state and damage of the bolt. The ultrasonic detection method: uses the attenuation and reflection characteristics of the signal during the ultrasonic propagation process to detect the damage or looseness inside the bolt.

[0008] However, these two methods still have the following problems:

[0009] 1. The strain gauge detection method can only provide point monitoring data and cannot reflect the full-field strain distribution of the bolt group; the installation position and pasting quality directly affect the detection accuracy; it is not suitable for long-term or dynamic monitoring.

[0010] 2. The ultrasonic detection method is only sensitive to internal defects of bolts, and it is difficult to obtain full-field displacement and strain data; it is greatly affected by material properties and environmental noise; and the detection efficiency is low.

[0011] Therefore, a new solution to the above problems needs to be proposed. Summary of the invention

[0012] The object of the present invention is to provide a method for identifying damage to a bridge bolt group based on DIC. By collecting the full-field strain of the bridge gusset plate under the condition of alternating traffic loads through a DIC device, analyzing the changes in the full-field strain of the gusset plate at different time periods, and identifying damage to bolts or gusset plates, the technical problems raised in the background art are solved.

[0013] To achieve the above object, the present invention provides the following technical solution: A method for identifying damage to a bridge bolt group based on DIC, at least including the following steps:

[0014] S1: Conduct data collection. Under normal bridge operating conditions, start the DIC system to continuously collect image data;

[0015] S2: Conduct theoretical analysis. That is, under cyclic loads, the pre-tightening force of high-strength bolts begins to decrease and loosen, the friction force between the bolts and the gusset plate becomes smaller, and relative movement between the contact surfaces will occur;

[0016] S3: Extract and analyze the full-field strain distribution. Based on the data collected in S1, extract and analyze the full-field strain distribution, set the criteria for damage identification and determination, and obtain the identification result based on the analysis.

[0017] Further, the S1 at least includes the following steps:

[0018] Prepare the DIC device and install it to ensure that the device covers the bolt group in the target area;

[0019] When the bridge is under alternating loads, start the DIC device to collect strain data and record the strain change data under alternating vehicle loads.

[0020] The collected data includes an image sequence, and the corresponding strain data is extracted through an image processing algorithm in the later stage.

[0021] Further, the DIC device at least includes a high-resolution camera, an optical sensor, and a reflection marker to ensure that the device can work stably under high dynamic loads. During the installation process of the DIC, reflection markers need to be installed on the bridge so that the DIC system can accurately track the surface deformation of the object. At the same time, a system with a high-resolution camera needs to be configured to ensure that the shooting angle and field of view can cover the entire bolt group area;

[0022] During use, conduct an initialization test when starting the DIC device to confirm the accuracy and stability of image collection;

[0023] The overall installation position of the DIC device should be selected at the key nodes of the bridge, and the key nodes at least include the bolt group, connection plate, and support structure to ensure that the entire strain monitoring area can be covered.

[0024] Furthermore, S3 at least includes the following steps:

[0025] S3.1: Data extraction. Extract the strain data from the strain contour map output by DIC to ensure that the strain distribution of the entire bolt group and the gusset plate can be obtained. That is, after obtaining a clear image, use the DIC algorithm to process the image data, calculate the displacement field of each point, and then calculate the strain field based on the displacement field. Assume that μ(x,y) and are the displacement components in the horizontal and vertical directions, and adopt the strain calculation formula:

[0026]

[0027] Calculate the strain of each pixel in the image through numerical methods to obtain a complete strain distribution map;

[0028] S3.2: Conduct multi-type data analysis. Conduct loss determination through comprehensive multi-type analysis. The multi-type data analysis includes full-field strain field anomaly analysis, time series analysis for damage identification, and statistical methods.

[0029] Furthermore, the strain field anomaly analysis at least includes the following steps:

[0030] First, set the analysis direction, that is, conduct full-field strain field anomaly analysis;

[0031] The full-field strain data measured by the DIC device is high-resolution displacement gradient information. Structural damage usually causes abnormal local strain distribution. This anomaly can be analyzed through gradients, mutations, discontinuities, and concentrated areas.

[0032] Cracks, looseness, or material damage will cause local strain mutations, that is, the strain in a small area suddenly increases or decreases compared with the surrounding area. Since DIC data is discrete, the gradient value of the grid points can be calculated to quantify the strain change.

[0033] Strain gradient calculation, refer to the following formula:

[0034]

[0035] where G x , G y represent the strain gradients in the x and y directions respectively; ∈ represents the full-field strain; x, y represent spatial coordinates.

[0036] Gradient modulus calculation, refer to the following formula:

[0037]

[0038] where G represents the modulus of the strain gradient; G x , Gy respectively represent the strain gradients in the x and y directions.

[0039] Set a threshold γ. If γ > G (usually the mean + 3 times the standard deviation), there may be damage.

[0040] Regions of high stress concentration tend to cause high strain accumulation, especially at the crack tip or in the damage propagation region.

[0041] Pay attention to the distribution of high principal strain regions. The principal strain can be calculated for quantitative analysis:

[0042]

[0043] where ∈1 and ∈2 respectively represent the maximum and minimum principal strains; ∈ xx , ∈ yy respectively represent the normal strains in the x and y directions; ∈ xy represents the shear strain.

[0044] For the skewness calculation of the strain distribution, refer to the following formula:

[0045]

[0046] where Skewness represents skewness, measuring the symmetry of the data; N represents the total number of strain data points; ∈ i represents the strain value at the i-th measurement point; represents the mean of the strain data; σ represents the standard deviation of the strain data.

[0047] If the skewness is far from 0, it indicates that there is an abnormality in the data distribution.

[0048] Furthermore, the time series analysis for using the alternating load time series data to analyze the damage evolution process at least includes the following steps:

[0049] First, set the analysis direction, that is, the loading analysis:

[0050] When there is no damage, the strain increases linearly with the load; after damage occurs, the strain growth rate changes or shows a sudden change;

[0051] Record the maximum strain ∈ max (t);

[0052] Secondly, perform the incremental calculation, refer to the following formula:

[0053] Δ∈ = ∈ t+1 - ∈ t

[0054] where Δ∈ represents the strain increment between adjacent time steps; ∈ t represents the strain value at the t-th time step; ∈ t+1Represents the strain value at the (t + 1)-th time step.

[0055] If the Δ∈ of the local area significantly deviates from the overall trend, it may be the damage location.

[0056] Structural damage will cause the strain after loading and unloading to not fully recover, that is, the residual strain increases;

[0057] Record the strain data in the loading-unloading cycle;

[0058] Then perform the residual strain calculation, referring to the following formula:

[0059] ∈ res = ∈ max - ∈ unload

[0060] Where ∈ res Represents the residual strain; ∈ max Represents the strain at maximum loading; ∈ unload Represents the strain after unloading.

[0061] If ∈ res Increases with the number of cycles, there may be damage.

[0062] Furthermore, the step of using statistical methods to analyze the full-field strain data measured by DIC to find the abnormal area at least includes the following steps:

[0063] First, set the analysis direction, that is, mean and standard deviation analysis;

[0064] Calculate the mean strain of all measurement points And the standard deviation σ ∈ ;

[0065] Set the abnormal threshold:

[0066]

[0067] Where ∈ represents the strain value of a single measurement point; Represents the mean of the full-field strain; σ ∈ Represents the standard deviation of the strain.

[0068] Under normal circumstances, 99.7% of the data points will fall within If the strain values of some measurement points exceed this range, it means that their strain levels are abnormally increased or decreased compared to other areas, which often indicates an increased probability of local structural damage;

[0069] Secondly, perform probability distribution analysis:

[0070] Under normal circumstances: The strain data often follows a Gaussian distribution;

[0071] Under damaged conditions: the strain distribution shows a long tail or bimodal phenomenon;

[0072] Calculate the probability density function (PDF) of the strain values;

[0073] Then use the K-S test to determine whether the data conforms to the normal distribution, see the following formula:

[0074]

[0075] where D represents the K-S statistic, which measures the maximum deviation between two distributions; represents taking the maximum difference corresponding to all x values; F n (x) represents the cumulative distribution function of the observed data; F(x) represents the normal distribution function under ideal conditions.

[0076] If the value of D is too large, it indicates that the data distribution is abnormal.

[0077] Compared with the prior art, the beneficial effects of the present invention are:

[0078] 1. The ability of full-field strain monitoring has been improved. Through the DIC technology, the present invention has realized the technological upgrade from point monitoring to full-field monitoring, and can comprehensively collect the full-field strain data of the bridge structure. The full-field strain map obtained by the DIC device can accurately reflect the strain changes of the bolt group and its surrounding structures. It can not only capture the occurrence location of damage in real time, but also analyze the damage evolution process, thus avoiding the traditional strain gauge detection method that can only provide local strain data and is difficult to comprehensively reflect the strain distribution of the high-strength bolt group of the bridge, which limits the accurate identification of damage;

[0079] 2. The ability of dynamic monitoring has been enhanced. Through the design of the technical solution, the present invention can effectively respond to the dynamic changes of the bridge structure under alternating loads. Especially under the influence of factors such as vehicle loads, wind vibrations, and temperature changes, it can monitor the whole process of damage evolution. However, the traditional strain gauge or ultrasonic detection method has weak monitoring ability in a dynamic environment and is difficult to capture the dynamic evolution process of damage in real time. Moreover, the present invention can provide an accurate assessment of the damage evolution trend by real-time collecting strain data;

[0080] 3. It has better high-precision damage identification and evaluation capabilities. Through multi-level and multi-type data analysis, this method does not solely rely on single strain data analysis but adopts various data analysis methods such as time series and frequency domain analysis, stress difference calculation, spatial correlation analysis, etc. This makes damage identification more comprehensive and improves accuracy. Through data comparison and analysis, the present invention can accurately identify the type, location, and degree of damage, especially showing extremely high precision in the identification of common damage forms such as high-strength bolt loosening and cracks. Compared with traditional damage detection methods, the present invention can comprehensively evaluate damage by combining strain data, reducing misjudgments caused by inaccurate damage evaluation;

[0081] 4. It can reduce maintenance costs. Through the full-field dynamic monitoring method implemented by the present invention, damage areas can be detected and accurately located at the initial stage of damage. Compared with traditional manual detection and partial detection methods, it greatly improves the detection efficiency and accuracy. Through precise damage identification, unnecessary over-maintenance can be avoided, while reducing the labor intensity and time cost of maintenance personnel;

[0082] 5. The present invention promotes the upgrade of bridge monitoring technology. The present invention has made technological innovations in strain monitoring technology, upgrading the traditional local monitoring method to full-field dynamic monitoring, which not only improves the monitoring accuracy but also the monitoring efficiency. With the development of intelligent bridge management technology, the present invention provides a new solution for the field of bridge monitoring and will become an important technology in the fields of intelligent transportation and infrastructure construction. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0084] Figure 1 It is the layout diagram of the DIC device of the present invention;

[0085] Figure 2 It is the flow schematic diagram of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0086] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments.

[0087] The present invention provides a method for identifying damage to bridge bolt groups based on DIC. Aiming at the full-field strain monitoring and damage identification of high-strength bolt groups, a new solution with non-contact, dynamic real-time and high precision is proposed, breaking through the following limitations of the existing technology:

[0088] (1) Realize the technological upgrade from point monitoring to full-field monitoring, and comprehensively obtain the strain distribution and deformation characteristics on the surface of the bolt group.

[0089] (2) Adapt to the dynamic load environment under the actual working conditions of the bridge, and realize the whole-process damage evolution monitoring and evaluation under alternating loads.

[0090] (3) Through data comparison and analysis, accurately identify the type, location and degree of damage, and provide the ability to predict damage trends and give safety warnings.

[0091] Refer to Figure 2 , the method for identifying damage to bridge bolt groups based on DIC at least includes the following steps:

[0092] S1: Conduct data acquisition. Under normal bridge operation conditions, start the DIC system to continuously acquire image data;

[0093] S2: Conduct theoretical analysis. That is, under cyclic loads, the pre-tightening force of high-strength bolts begins to decrease and loosen, the friction force with the gusset plate becomes smaller, and relative movement between the contact surfaces will occur;

[0094] S3: Extract and analyze the full-field strain distribution. According to the data acquired in S1, extract and analyze the full-field strain distribution, set the criteria for damage identification and determination, and obtain the identification result based on the analysis.

[0095] S1 at least includes the following steps:

[0096] Prepare the DIC equipment and install it to ensure that the equipment covers the bolt group in the target area;

[0097] When the bridge is under alternating loads, start the DIC equipment to collect strain data, and record the strain change data under alternating vehicle loads.

[0098] The data collected includes an image sequence, and the corresponding strain data is extracted through an image processing algorithm in the later stage.

[0099] Refer to Figure 1 , the DIC equipment at least includes a high-resolution camera, an optical sensor and a reflection marker, ensuring that the equipment can work stably under high dynamic loads. During the installation of DIC, reflection markers need to be installed on the bridge so that the DIC system can accurately track the surface deformation of the object. At the same time, a system with a high-resolution camera needs to be configured to ensure that the shooting angle and field of view can cover the entire bolt group area;

[0100] During use, when starting the DIC device, an initialization test is performed to confirm the accuracy and stability of image acquisition;

[0101] The overall installation position of the DIC device should be selected at the key nodes of the bridge. The key nodes should at least include bolt groups, connecting plates and support structures to ensure that the entire strain monitoring area can be covered.

[0102] S3 at least includes the following steps:

[0103] S3.1: Data extraction. The strain data extracted from the strain nephogram output by the DIC is ensured to be able to obtain the strain distribution of the entire bolt group and the gusset plate. That is, after obtaining a clear image, the DIC algorithm is used to process the image data, calculate the displacement field of each point, and then, based on the displacement field, calculate the strain field. Assuming that μ(x,y) and are the displacement components in the horizontal and vertical directions respectively, the strain calculation formula is adopted:

[0104]

[0105] The strain of each pixel in the image is calculated by numerical methods to obtain a complete strain distribution map;

[0106] S3.2: Conduct multi-type data analysis, and comprehensively conduct loss determination through multi-type analysis. The multi-type data analysis includes full-field strain field anomaly analysis, time series analysis for damage identification, and statistical methods.

[0107] The strain field anomaly analysis at least includes the following steps:

[0108] First, set the analysis direction, that is, conduct full-field strain field anomaly analysis;

[0109] The full-field strain data measured by the DIC device is high-resolution displacement gradient information. Structural damage usually causes abnormal distribution of local strain. This anomaly can be analyzed through gradients, mutations, discontinuities and concentrated areas;

[0110] Cracks, looseness or material damage will cause local strain mutations, that is, the strain in a small area suddenly increases or decreases compared with the surrounding area. Since the DIC data is discrete, the gradient value of the grid points can be calculated to quantify the strain change;

[0111] The strain gradient calculation is referred to the following formula:

[0112]

[0113] where G x , G yrespectively represent the strain gradients in the x and y directions; ∈ represents the full-field strain; x, y represent the spatial coordinates.

[0114] For the calculation of the gradient magnitude, refer to the following formula:

[0115]

[0116] where G represents the magnitude of the strain gradient; G x , G y respectively represent the strain gradients in the x and y directions.

[0117] Set a threshold γ. If γ > G (usually the mean + 3 times the standard deviation), there may be damage.

[0118] Regions of high stress concentration often cause high strain accumulation, especially at the crack tip or in the damage propagation region.

[0119] By paying attention to the distribution of high principal strain regions, the principal strain can be calculated for quantitative analysis:

[0120]

[0121] where ∈1 and ∈2 respectively represent the maximum and minimum principal strains; ∈ xx , ∈ yy respectively represent the normal strains in the x and y directions; ∈ xy represents the shear strain.

[0122] For the calculation of the skewness of the strain distribution, refer to the following formula:

[0123]

[0124] where Skewness represents the skewness, measuring the symmetry of the data; N represents the total number of strain data points; ∈ i represents the strain value at the i-th measurement point; represents the mean of the strain data; σ represents the standard deviation of the strain data.

[0125] If the skewness is far from 0, it indicates that there is an abnormality in the data distribution.

[0126] The time series analysis for analyzing the damage evolution process using the alternating load time series data includes at least the following steps:

[0127] First, set the analysis direction, that is, the loading analysis:

[0128] When there is no damage, the strain increases linearly with the load; after damage occurs, the strain growth rate changes or shows a mutation;

[0129] Record the maximum strain ∈ max (t);

[0130] Next, perform incremental calculation, refer to the following formula:

[0131] Δ∈ = ∈ t+1 - ∈ t

[0132] where Δ∈ represents the strain increment at adjacent time steps; ∈ t represents the strain value at the t-th time step; ∈ t+1 represents the strain value at the (t + 1)-th time step.

[0133] If the Δ∈ in a local area significantly deviates from the overall trend, it may be the damage location.

[0134] Structural damage will cause the strain after loading and unloading to not fully recover, that is, the residual strain increases;

[0135] Record the strain data during the loading-unloading cycle;

[0136] Then perform residual strain calculation, refer to the following formula:

[0137] ∈ res = ∈ max - ∈ unload

[0138] where ∈ res represents the residual strain; ∈ max represents the strain at maximum loading; ∈ unload represents the strain after unloading.

[0139] If ∈ res increases with the number of cycles, there may be damage.

[0140] Using statistical methods to analyze the full-field strain data measured by DIC and finding abnormal areas includes at least the following steps:

[0141] First, set the analysis direction, that is, mean and standard deviation analysis;

[0142] Calculate the mean strain of all measurement points and the standard deviation σ ∈ ;

[0143] Set the abnormal threshold:

[0144]

[0145] where ∈ represents the strain value of a single measurement point; represents the mean of the full-field strain; σ ∈ represents the standard deviation of the strain.

[0146] Under normal circumstances, 99.7% of the data points will fall within If the strain values at certain measurement points exceed this range within a certain limit, it means that their strain levels increase or decrease abnormally compared to other areas, which often indicates an increased probability of local damage to the structure;

[0147] Secondly, perform probability distribution analysis:

[0148] Under normal circumstances: Strain data often follows a Gaussian distribution;

[0149] In the case of damage: The strain distribution shows a long tail or bimodal phenomenon;

[0150] Calculate the probability density function (PDF) of the strain value;

[0151] Then use the K-S test to determine whether the data conforms to a normal distribution. Refer to the following formula:

[0152]

[0153] Where D represents the K-S statistic, which measures the maximum deviation between two distributions; represents taking the maximum difference corresponding to all x values; F n (x) represents the cumulative distribution function of the observed data; F(x) represents the normal distribution function in the ideal case.

[0154] Among them, if the D value is too large, it indicates that the data distribution is abnormal.

[0155] To sum up:

[0156] Perform multi-type data analysis, and comprehensively use the full-field strain field anomaly analysis, time series analysis, and statistical methods for damage determination.

[0157] Strain field anomaly analysis:

[0158] Identify damage through gradients, mutations, discontinuities, and high-strain concentration regions.

[0159] Calculate the strain gradient and its modulus. If it exceeds the mean + 3 times the standard deviation, there may be damage.

[0160] Calculate the principal strain distribution and identify high principal strain regions.

[0161] Calculate the skewness. Deviation from 0 indicates abnormal data.

[0162] Time series analysis:

[0163] Step-by-step loading analysis: When there is no damage, the strain increases linearly with the load. After damage, the growth rate changes or shows a mutation.

[0164] Calculate the strain increment Δ∈. If the increment in a local area significantly deviates from the trend, there may be damage.

[0165] Residual Strain Analysis: Record the loading-unloading cyclic strain and calculate ∈ res = ∈ max - ∈ unload . If it increases with the number of cycles, it indicates damage.

[0166] Statistical Method Analysis:

[0167] Mean and Standard Deviation Analysis: Calculate the mean of the measurement points and the standard deviation σ ∈ . If , there may be damage.

[0168] Probability Distribution Analysis: Under normal circumstances, the strain follows a Gaussian distribution. When damaged, long tails or bimodal phenomena may occur.

[0169] Use the K-S test to detect abnormal distributions. If the statistic D is too large, it indicates that the data is abnormal and there may be damage.

[0170] This method combines strain field, time series and statistical analysis to improve the accuracy and reliability of damage identification. For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and can be implemented in other specific forms without departing from the spirit or basic characteristics of the present invention. Therefore, from any point of view, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be included in the present invention. Any reference signs in the claims should not be regarded as limiting the claimed claims.

Claims

1. A DIC-based damage identification method for bridge bolt groups, characterized in that: It at least includes the following steps: S1: Conduct data acquisition. By starting the DIC system to continuously acquire image data under normal bridge operation conditions; S2: Conduct theoretical analysis. That is, under cyclic loading, the pre-tightening force of high-strength bolts begins to decrease and loosening occurs. The frictional force between the bolt and the gusset plate becomes smaller, and relative movement between the contact surfaces will occur; S3: Extract and analyze the full-field strain distribution. Based on the data acquired in S1, extract and analyze the full-field strain distribution, set the criteria for damage identification and determination, and obtain the identification results based on the analysis.

2. The method for identifying damage to a bridge bolt group based on DIC according to claim 1, wherein: The S1 at least includes the following steps: Prepare the DIC equipment and install it to ensure that the equipment covers the bolt group in the target area; When the bridge is under alternating loads, start the DIC equipment to collect strain data and record the strain change data under alternating vehicle loads; The acquired data includes an image sequence, and the corresponding strain data is extracted through image processing algorithms in the later stage.

3. The method for identifying the damage of bridge bolt groups based on DIC according to claim 2, wherein: The DIC equipment at least includes a high-resolution camera, an optical sensor, and a reflection marker to ensure that the equipment can work stably under high dynamic loads. During the installation process of the DIC, reflection markers need to be installed on the bridge so that the DIC system can accurately track the surface deformation of the object. At the same time, a system with a high-resolution camera needs to be configured to ensure that the shooting angle and field of view can cover the entire bolt group area; During use, conduct an initialization test when starting the DIC equipment to confirm the accuracy and stability of image acquisition; The overall installation position of the DIC equipment should be selected at the key nodes of the bridge. The key nodes at least include the bolt group, connection plate, and support structure to ensure that the entire strain monitoring area can be covered.

4. The method for identifying damage of bridge bolt groups based on DIC according to claim 1, characterized in that: The S3 at least includes the following steps: S3.1: Data extraction. Extract the strain data from the strain nephogram output by the DIC to ensure that the strain distribution of the entire bolt group and gusset plate can be obtained. That is, after obtaining a clear image, use the DIC algorithm to process the image data, calculate the displacement field of each point, and then calculate the strain field based on the displacement field. Assume that μ(x,y) and θ(x,y) are the displacement components in the horizontal and vertical directions, and use the strain calculation formula: Calculate the strain of each pixel in the image through numerical methods to obtain a complete strain distribution map; S3.2: Conduct multi-type data analysis. Conduct loss determination through comprehensive multi-type analysis. The multi-type data analysis includes full-field strain field anomaly analysis, time series analysis for damage identification, and statistical methods.

5. The method for identifying damage of bridge bolt groups based on DIC according to claim 4, characterized in that: The strain field anomaly analysis at least includes the following steps: First, set the analysis direction, that is, conduct full-field strain field anomaly analysis; The full-field strain data measured by the DIC equipment is high-resolution displacement gradient information. Structural damage usually causes abnormal distribution of local strain, and this abnormal distribution can be analyzed through gradients, mutations, discontinuities, and concentrated areas; Cracks, looseness, or material damage will cause local strain mutations, that is, the strain in a small area suddenly increases or decreases compared with the surrounding area; Since the DIC data is discrete, the strain change can be quantified by calculating the gradient value of the grid points; Strain gradient calculation. Refer to the following formula: where G x , G y represent the strain gradients in the x- and y-directions, respectively; ∈ represents the full-field strain; x, y represent the spatial coordinates; The calculation of the gradient modulus is shown in the following formula: where G represents the magnitude of the strain gradient; G x , G t respectively represent the strain gradients in the x and y directions; Set a threshold γ. If γ > G, usually the mean + 3 times the standard deviation, there may be damage; High stress concentration areas often cause high strain accumulation, especially at the crack tip or the damage propagation area; Pay attention to the distribution of high principal strain areas and quantitatively analyze by calculating the principal strain: where, ∈1 and ∈2 represent the maximum and minimum principal strains respectively; ∈ xx , ∈ yy represent the normal strains in the x and y directions respectively; ∈ xy represents the shear strain; The calculation of the skewness of the strain distribution is shown in the following formula: Among them, Skewness represents skewness and measures data symmetry; N represents the total number of strain data points; ∈ i represents the strain value of the i-th measurement point; represents the mean value of the strain data; σ represents the standard deviation of the strain data; If the skewness is far from 0, it indicates that the data distribution is abnormal.

6. The method for identifying damage of bridge bolt groups based on DIC according to claim 4, characterized in that: The time series analysis of using the alternating load time series data to analyze the damage evolution process includes at least the following steps: First, set the analysis direction, that is, the loading analysis: When there is no damage, the strain increases linearly with the load; after damage occurs, the strain growth rate changes or shows a sudden change; Record the maximum strain ∈ max (t) at different time steps; Secondly, perform the increment calculation, shown in the following formula: Δ∈=∈ t+1 -∈ t Among them, Δ∈ represents the strain increment of adjacent time steps; ∈ t represents the strain value at the t-th time step; ∈ t+1 represents the strain value at the (t + 1)-th time step; If the Δ∈ of a local area significantly deviates from the overall trend, the probability of the damage location increases; Structural damage will cause the strain after loading and unloading to not fully recover, that is, the residual strain increases; Record the strain data in the loading-unloading cycle; Then perform the residual strain calculation, shown in the following formula: ∈ res ＝∈ max -∈ unload where ∈ res represents the residual strain; ∈ max represents the strain at maximum loading; ∈ unload represents the strain after unloading. If ∈ res As the number of cycles increases, damage may occur.

7. The method for identifying damage of bridge bolt groups based on DIC according to claim 4, characterized in that: The above statistical method analyzes the full-field strain data measured by DIC. Finding the abnormal area includes at least the following steps: First, set the analysis direction, that is, the mean and standard deviation analysis; Calculate the mean strain of all measurement points and the standard deviation σ ∈ ; Set the abnormal threshold: Among them, ∈ represents the strain value of a single measurement point; represents the mean value of the full-field strain; σ ∈ represents the standard deviation of the strain; Under normal circumstances, 99.7% of the data points will fall within If the strain values of some measurement points exceed this range, it means that their strain levels increase or decrease abnormally compared to other areas, which often indicates an increased probability of local structural damage; Secondly, perform the probability distribution analysis: Under normal circumstances: the strain data often follows a Gaussian distribution; Under damaged conditions: the strain distribution shows a long tail or bimodal phenomenon; Calculate the probability density function of the strain value; Then use the K-S test to judge whether the data conforms to the normal distribution, shown in the following formula: Among them, D represents the K-S statistic, which measures the maximum deviation between two distributions; denotes taking the maximum difference corresponding to all x values; F n (x) represents the cumulative distribution function of the observed data; F(x) represents the normal distribution function under ideal conditions; Among them, if the D value is too large, it indicates that the data distribution is abnormal.

Images