A method and system for evaluating the bearing capacity of RC components

Abstract

The present invention discloses a method and system for evaluating the bearing capacity of RC components, which relates to the technical field of building materials, and includes the following steps: S1, constructing an image data graph based on RC components; S2, performing preprocessing and model training; S3, obtaining the three-dimensional morphology of the RC component to be evaluated; S4, performing mapping reconstruction; S5, constructing a twin model; S6, finite element simulation calculation of bearing performance prediction results; S7, analyzing the performance changes of RC components; S8, evaluating the bearing capacity of RC components. The method and system for evaluating the bearing capacity of RC components numerically calculate the bearing performance of the mapped and reconstructed damaged RC components by using finite element analysis software. The method introduces the reconstructed surface morphology into the finite element model, and the calculation result has higher accuracy, realizing the prediction of the bearing performance and reliability evaluation of the damaged RC components, and solving the problem that the actual engineering and mechanical performance tests cannot be repeated.

Description

Technical Field

[0001] The invention relates to the technical field of building materials, and in particular to a method and system for evaluating the bearing capacity of RC components. Background Art

[0002] Reinforced concrete (RC) components play a vital role in modern construction projects due to their excellent mechanical properties and wide range of applications. RC components not only bear the deadweight and service load of the building, but also must withstand external loads caused by natural factors such as earthquakes and wind. Therefore, ensuring the safety and functionality of RC components is crucial to the overall stability of the building structure. With the increase in the service life of the building and the influence of the external environment, RC components may be damaged. If these damages are not detected and treated in time, they will seriously affect the safety and service life of the structure. Damage detection can detect the damage of components in time, while bearing capacity assessment can quantitatively analyze the bearing performance of damaged components in the current state, providing a scientific basis for subsequent maintenance and reinforcement. Traditional RC component damage detection methods mainly include visual inspection, knock detection, ultrasonic detection, etc. These methods have their own advantages and disadvantages, but they generally have problems of low efficiency, insufficient accuracy or high cost. With the combination of digital twin technology with numerical simulation models and online monitoring data, it provides a new idea and method for RC component damage detection.

[0003] Therefore, studying an effective RC component damage detection and bearing capacity assessment method and system is of great practical significance for ensuring building safety and extending building life. Summary of the invention

[0004] In view of the deficiencies of the prior art, the present invention provides a method and system for evaluating the bearing capacity of RC components to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solution: a method for evaluating the bearing capacity of RC components, comprising the following steps:

[0006] S1, constructing an image data graph based on RC components;

[0007] Collect images of RC components with damage, and use LabelMe to annotate and classify different damage categories in the images, such as cracks and spalling. Then collect public datasets of damage features from multiple data sources. The annotated and classified images and the public datasets together form an image database.

[0008] S2, perform preprocessing and model training;

[0009] First, the RC component images are denoised, and then the CNN model is trained using the image database combined with Python language;

[0010] S3, obtaining the three-dimensional morphology of the RC component to be evaluated;

[0011] Scanning the existing damaged RC components with scanning equipment, analyzing the collected data using the trained CNN model, identifying the damage characteristics of the components, and obtaining the three-dimensional morphology of the RC components;

[0012] S4, performing mapping reconstruction;

[0013] Based on the scanning equipment, the depth information of the damage is obtained, and the point cloud information and coordinate information of the scanned image are exported through the software, and the three-dimensional morphology of the RC component to be evaluated is mapped and reconstructed to obtain the morphology mapping and reconstruction result of the RC component to be evaluated;

[0014] S5. Build a twin model;

[0015] According to the reconstruction results of the three-dimensional morphology mapping of the RC component to be evaluated, a twin model of the finite element geometric model and material model of the RC component is constructed using the finite element analysis method;

[0016] S6. Prediction results of bearing capacity calculated by finite element simulation;

[0017] Using the finite element parametric simulation method, the finite element geometric model of the RC component to be evaluated is simulated and calculated to obtain the bearing performance prediction results of the RC component to be evaluated;

[0018] S7. Analyze the performance changes of RC components;

[0019] By simulating the occurrence and development trend of damage in the twin model, the performance changes of components under different damage states are analyzed;

[0020] S8. Carry out load-bearing capacity assessment of RC components;

[0021] Based on the damage simulation results, the bearing capacity of the component is evaluated to determine whether it meets the safety requirements, including limit state analysis, performance evaluation and risk assessment.

[0022] To further optimize the technical solution, in step S1, the RC component image clearly marks the categories and names of different damages, and the public data set includes the original image and the mask file;

[0023] When annotating with labelme:

[0024] Different damage categories are marked with different colors;

[0025] The images annotated by labelme are json files, and python code is used to convert json files into mask files.

[0026] To further optimize the technical solution, in step S2, during preprocessing, a weighted average method and a histogram equalization method are used to perform grayscale conversion and image enhancement processing on the image;

[0027] The image segmentation algorithm is used to train the CNN model. The image is divided into multiple sub-regions based on the characteristics of color, edge, and grayscale. The edge points of the image are found by using operators to judge the changes in the grayscale values ​​of the image. Different grayscale thresholds are set, and then the grayscale histogram is used to find the segmentation threshold required by the threshold method.

[0028] To further optimize the technical solution, in step S3, the scanning equipment is a three-dimensional non-contact laser scanner and a SLR camera, and by adjusting the scanner position and the focal length of the rubber ring, physical images at different angles and positions are acquired to obtain the three-dimensional morphology of the damaged component to be evaluated.

[0029] To further optimize the technical solution, in step S4, when the depth information of the damage is obtained, the following process is included:

[0030] Post multiple calibration points in advance;

[0031] The scanning device identifies the calibration points and determines the point cloud coordinate data of the calibration points;

[0032] According to the distance between the lens and the component, the specific depth of each point is calculated based on the point cloud coordinate data.

[0033] To further optimize the technical solution, in step S5, the finite element analysis method uses Abaqus to collect information on design parameters, material properties, and historical maintenance records of RC components to provide basic data for model construction;

[0034] The process of building a twin model of an RC component is as follows:

[0035] Data collection: Collect information on design parameters, material properties, and historical maintenance records of RC components;

[0036] Model building: Based on the collected data, the finite element geometric model and material model of the RC component are established;

[0037] Parameter calibration: Calibrate model parameters through comparative testing with actual components;

[0038] Real-time update: Integrate sensor data to update the model in real time to reflect the actual working status of the component.

[0039] To further optimize the technical solution, in step S6, the finite element simulation calculation specifically includes:

[0040] Based on the finite element twin model of the RC component, the material properties are set, the loads and boundary conditions are applied, and the finite element numerical simulation is performed to obtain the prediction results of the bearing capacity of the RC component to be evaluated.

[0041] To further optimize the technical solution, in step S8, when evaluating the bearing capacity of the component, the following specific process is included:

[0042] Preprocessing and analysis: Preprocess the data simulated in the twin model to eliminate noise and outliers, and use statistical methods to analyze the damage mode and development trend of the components;

[0043] Bearing capacity calculation: Based on the material properties, geometric parameters and damage state of the RC component, the bearing capacity is calculated using the structural analysis method. For example, for bending members, the moment-curvature analysis method is used to evaluate its ultimate bearing state;

[0044] Safety factor assessment: Determine the safety factor based on the use environment and load characteristics of the component, and evaluate the safety of the component under different working conditions. The determination of the safety factor complies with relevant industry standards;

[0045] Maintenance and reinforcement suggestions: Based on the load-bearing performance prediction results, suggestions for component maintenance and reinforcement are put forward. For components with insufficient load-bearing capacity, appropriate reinforcement measures are recommended, such as increasing reinforcement and using prestressing technology;

[0046] Long-term monitoring plan: Establish a long-term monitoring plan to regularly conduct damage detection and load-bearing capacity assessment on RC components, grasp the performance changes of components in real time, and provide data support for the full life cycle management of the structure;

[0047] In the calculation of bearing capacity, the change of bearing capacity under the influence of cracks is considered, and a damage-corrected moment-curvature relationship model is designed. The model is as follows:

[0048] ;

[0049] in, is the bending moment after damage correction; is the initial limit bending moment in the undamaged state; is the adjustment factor related to material properties; is the number of cracks in the damaged area; is the total number of cracks in the component; is the damage-corrected curvature; is the initial curvature in the undamaged state; is the adjustment factor related to the geometric characteristics of the component; is the width of damage, such as the width of a crack or spall; is the total width of the component;

[0050] By considering the distribution characteristics of damage, such as the number and width of cracks, the bending moment and curvature are corrected to accurately predict the bearing capacity of flexural members in the damaged state;

[0051] As the damage increases, the bending moment will decrease, while the curvature will increase, reflecting the phenomenon of reduced stiffness and increased deformation capacity of the component;

[0052] When using,

[0053] Extract damage features, including crack quantity and width, from 3D topography reconstruction results;

[0054] Calculate the damage-corrected bending moment using the formula and curvature ;

[0055] During structural analysis, the modified bending moment and curvature are used to evaluate the ultimate bearing capacity of RC members.

[0056] A RC component bearing capacity evaluation system is constructed based on the above RC component bearing capacity evaluation method, and includes a scanning module, an index module, a mapping reconstruction module, a geometric model module, a numerical simulation module and an evaluation module.

[0057] Further optimizing the technical solution, in the system:

[0058] A scanning module is used to scan the surface features of the RC component to be evaluated to obtain the three-dimensional morphology of the damaged RC component to be evaluated;

[0059] An index module is used to determine damage evaluation indexes according to the three-dimensional morphology of the RC component to be evaluated;

[0060] A mapping reconstruction module is used to map and reconstruct the RC component to be evaluated based on the surface random characterization model and damage evaluation index;

[0061] A geometric model module, used for establishing a finite element geometric model of the RC component to be evaluated according to the mapping reconstruction result of the RC component to be evaluated;

[0062] A numerical simulation module is used to perform finite element simulation calculations on the finite element geometric model of the RC component to be evaluated using a finite element parametric simulation method to obtain a prediction result of the bearing performance of the RC component to be evaluated;

[0063] An evaluation module, used to evaluate the bearing performance of the RC component to be evaluated according to the bearing performance prediction result of the RC component to be evaluated, and obtain the bearing performance evaluation result of the RC component;

[0064] The evaluation module is provided with a bearing capacity reduction coefficient model based on damage depth. When evaluating the damage depth of RC components, the model designs a damage depth reduction coefficient. , used to describe the influence of damage depth on the bearing capacity of components. The model combines damage depth, location and geometric characteristics of components to accurately describe the changes in bearing capacity;

[0065] The damage depth affects the reduction factor formula, as shown below:

[0066] ;

[0067] in,

[0068] is the damage depth reduction factor, which ranges from 0 to 1, with 1 indicating no damage and 0 indicating complete failure;

[0069] Adjustment parameters related to material properties and structure type;

[0070] is the average depth of the damaged area;

[0071] is the total thickness of the component;

[0072] is the position correction factor, which is used to describe the influence of damage position on bearing capacity;

[0073] It is the distance between the damaged area and the supporting point or the key stress point;

[0074] is the total length of the component;

[0075] By combining the damage depth and location, its impact is quantified as a reduction factor , used to correct the original bearing capacity during bearing capacity assessment;

[0076] When the damage depth is large or the damaged area is close to the critical stress point, the reduction factor will be significantly reduced, reflecting that the bearing capacity of the structure has been severely weakened;

[0077] When this model is used,

[0078] Obtaining damage depth information through 3D non-contact laser scanning and location of the damaged area ;

[0079] The damage depth reduction factor is calculated according to the formula ;

[0080] When evaluating the carrying capacity, the initial carrying capacity Multiply To obtain the load-bearing capacity after damage .

[0081] Compared with the prior art, the present invention provides a method and system for evaluating the bearing capacity of RC components, which has the following beneficial effects:

[0082] 1. The damage detection of the present invention is simple and efficient. Damage detection based on digital twin technology can provide more comprehensive and in-depth structural status information. At the same time, through integrated learning and data fusion technology, the accuracy and reliability of damage detection are improved. In addition, the twin model can also be used to predict the remaining service life of components and provide support for maintenance and repair decisions.

[0083] 2. The present invention provides all-round and scalable data analysis to achieve long-term tracking and monitoring of the overall damage of RC component structures. It is a useful supplement to local and short-term damage diagnosis technology, greatly expands the connotation of the RC component detection field, and improves the possibility of predictive evaluation.

[0084] 3. The present invention is based on statistical analysis of data and evaluation indexes to establish a characterization model of RC components with different damages and meeting certain statistical characteristics. Batch mapping and reconstruction of RC components can be achieved based on limited morphological data, providing data support for subsequent structural performance prediction and reliability evaluation.

[0085] 4. The present invention uses finite element analysis software to perform numerical calculation of the bearing performance of the mapped and reconstructed damaged RC components, introduces the reconstructed surface morphology into the finite element model, and the calculation results have higher accuracy, realizing the prediction of the bearing performance and reliability evaluation of the damaged RC components, solving the problem that actual engineering and mechanical performance tests cannot be repeated.

[0086] 5. The present invention can further enrich the performance evaluation methods in the field of engineering structure durability and safety, and provide a technical basis for engineering technicians engaged in the detection, evaluation and reinforcement of existing engineering structures. BRIEF DESCRIPTION OF THE DRAWINGS

[0087] Figure 1 A schematic diagram of a flow chart of a method for evaluating the bearing capacity of a RC component proposed by the present invention;

[0088] Figure 2 A schematic diagram of a flow chart of evaluating the bearing capacity of a component in a method for evaluating the bearing capacity of an RC component proposed by the present invention;

[0089] Figure 3 This is a structural schematic diagram of a RC component bearing capacity assessment system proposed by the present invention. DETAILED DESCRIPTION

[0090] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0091] Embodiment 1:

[0092] See also Figure 1 , a method for evaluating the bearing capacity of RC components, uses a large number of data sets to train a refined numerical model of RC components, simulates its response under different working conditions, and provides a theoretical basis for damage detection. In addition, it combines advanced algorithms such as deep learning to analyze monitoring data, automatically identify damage characteristics, and improve the accuracy and efficiency of detection. Subsequently, the digital twin technology is used to build a twin model, and based on the damage simulation results, the bearing capacity of the component is evaluated to determine whether it meets the safety requirements.

[0093] The method comprises the following steps:

[0094] S1. Constructing image data graph based on RC components

[0095] Collect images of RC components with damage, and use LabelMe to annotate and classify different damage categories in the images, such as cracks and spalling. Then collect public datasets of damage features from multiple data sources. The annotated and classified images and the public datasets together form an image database.

[0096] In this embodiment, in step S1, the RC component image clearly marks the categories and names of different damages, and the public data set includes the original image and mask file;

[0097] When annotating with labelme:

[0098] Different damage categories are marked with different colors;

[0099] The images annotated by labelme are json files, and python code is used to convert json files into mask files.

[0100] First switch to the labelme installation environment, and then use the code labelme_json_to_dataset to convert the json file into a mask file. Note that the json absolute path cannot contain Chinese paths or spaces, otherwise it cannot be converted.

[0101] S2. Preprocessing and model training

[0102] The RC component images are first denoised, and then the CNN model is trained using the image database combined with the Python language.

[0103] Specifically, the training of the CNN model requires the use of deep learning algorithms, the use of pycharm: Python IDE development platform, combined with Anaconda's integrated python database, to build a pytorch deep learning framework to train the neural network model. The CNN model also requires the use of GPU to install cuda and cudnn.

[0104] In this embodiment, in step S2, during preprocessing, the weighted average method and the histogram equalization method are used to perform grayscale conversion and image enhancement processing on the image;

[0105] The image segmentation algorithm is used to train the CNN model. The image is divided into multiple sub-regions based on the characteristics of color, edge, and grayscale. The edge points of the image are found by using operators to judge the changes in the grayscale values ​​of the image. Different grayscale thresholds are set, and then the grayscale histogram is used to find the segmentation threshold required by the threshold method.

[0106] S3. Obtain the three-dimensional morphology of the RC component to be evaluated

[0107] The existing damaged RC components are scanned based on the scanning equipment, and the collected data are analyzed using the trained CNN model to identify the damage characteristics of the components and obtain the three-dimensional morphology of the RC components, wherein the morphological information of the damaged RC components to be evaluated is the three-dimensional point cloud coordinate data of the damaged RC components to be evaluated.

[0108] In this embodiment, in step S3, the scanning equipment is a three-dimensional non-contact laser scanner and a SLR camera, and by adjusting the scanner position and the focal length of the rubber ring, physical images at different angles and positions are acquired to obtain the three-dimensional morphology of the damaged component to be evaluated.

[0109] S4. Mapping reconstruction

[0110] The depth information of the damage is obtained based on the scanning equipment, and the point cloud information and coordinate information of the scanned image are exported through the software. The three-dimensional morphology of the RC component to be evaluated is mapped and reconstructed to obtain the morphology mapping and reconstruction result of the RC component to be evaluated.

[0111] In this embodiment, in step S4, the acquisition of the damage depth information includes the following process:

[0112] Post multiple calibration points in advance;

[0113] The scanning device identifies the calibration points and determines the point cloud coordinate data of the calibration points;

[0114] According to the distance between the lens and the component, the specific depth of each point is calculated based on the point cloud coordinate data.

[0115] The point cloud information and coordinate information are obtained by scanning to obtain the average value and standard deviation of the damage depth, and determine the damage depth power spectrum of the damaged RC component. According to the damage depth spectrum of the damaged RC component, the point cloud coordinate data is calculated using a two-dimensional random concordance function; then the specific depth is calculated through a series of conversion formulas based on the distance between the real lens and the damaged RC component.

[0116] S5. Build a twin model

[0117] According to the three-dimensional morphology mapping reconstruction results of the RC component to be evaluated, the finite element analysis method is used to construct a twin model of the finite element geometric model and material model of the RC component.

[0118] In this embodiment, in step S5, the finite element analysis method uses Abaqus to collect information on design parameters, material properties, and historical maintenance records of RC components to provide basic data for model construction;

[0119] The process of building a twin model of an RC component is as follows:

[0120] Data collection: Collect information on design parameters, material properties, and historical maintenance records of RC components;

[0121] Model building: Based on the collected data, the finite element geometric model and material model of the RC component are established;

[0122] Parameter calibration: Calibrate model parameters through comparative testing with actual components;

[0123] Real-time update: Integrate sensor data to update the model in real time to reflect the actual working status of the component.

[0124] S6. Prediction results of bearing performance by finite element simulation

[0125] The finite element parametric simulation method is used to perform finite element simulation calculation on the finite element geometric model of the RC component to be evaluated, and the bearing performance prediction result of the RC component to be evaluated is obtained.

[0126] In this embodiment, in step S6, the finite element simulation calculation specifically includes:

[0127] Based on the finite element twin model of the RC component, the material properties are set, the loads and boundary conditions are applied, and the finite element numerical simulation is performed to obtain the prediction results of the bearing capacity of the RC component to be evaluated.

[0128] S7. Analyze the performance changes of RC components

[0129] By simulating the occurrence and development trend of damage in the twin model, the performance changes of components under different damage states are analyzed.

[0130] In this embodiment, the bearing performance results of the component under different simulated damage states are obtained in Abaqus, and compared with the bearing performance results of the component under different actual damage states, the actual bearing performance results of the component to be evaluated are obtained.

[0131] S8. Evaluate the bearing capacity of RC components

[0132] Based on the damage simulation results, the bearing capacity of the component is evaluated to determine whether it meets the safety requirements, including limit state analysis, performance evaluation and risk assessment.

[0133] like Figure 2 As shown, in this embodiment, in step S8, when evaluating the bearing capacity of the component, the following specific process is included:

[0134] Preprocessing and analysis: Preprocess the data simulated in the twin model to eliminate noise and outliers, and use statistical methods to analyze the damage mode and development trend of the components;

[0135] Bearing capacity calculation: Based on the material properties, geometric parameters and damage state of the RC component, the bearing capacity is calculated using the structural analysis method. For example, for bending members, the moment-curvature analysis method is used to evaluate its ultimate bearing state;

[0136] Safety factor assessment: Determine the safety factor based on the use environment and load characteristics of the component, and evaluate the safety of the component under different working conditions. The determination of the safety factor complies with relevant industry standards;

[0137] Maintenance and reinforcement suggestions: Based on the load-bearing performance prediction results, suggestions for component maintenance and reinforcement are put forward. For components with insufficient load-bearing capacity, appropriate reinforcement measures are recommended, such as increasing reinforcement and using prestressing technology;

[0138] Long-term monitoring plan: Establish a long-term monitoring plan to regularly conduct damage detection and load-bearing capacity assessment on RC components, grasp the performance changes of components in real time, and provide data support for the full life cycle management of the structure.

[0139] In the calculation of bearing capacity, the change of bearing capacity under the influence of cracks is considered, and a damage-corrected moment-curvature relationship model is designed. The model is as follows:

[0140] ;

[0141] in, is the bending moment after damage correction; is the initial limit bending moment in the undamaged state; is the adjustment factor related to material properties; is the number of cracks in the damaged area; is the total number of cracks in the component; is the damage-corrected curvature; is the initial curvature in the undamaged state; is the adjustment factor related to the geometric characteristics of the component; is the width of damage, such as the width of a crack or spall; is the total width of the component;

[0142] By considering the distribution characteristics of damage, such as the number and width of cracks, the bending moment and curvature are corrected to accurately predict the bearing capacity of flexural members in the damaged state;

[0143] As the damage increases, the bending moment will decrease, while the curvature will increase, reflecting the phenomenon of reduced stiffness and increased deformation capacity of the component;

[0144] When using,

[0145] Extract damage features, including crack quantity and width, from 3D topography reconstruction results;

[0146] Calculate the damage-corrected bending moment using the formula and curvature ;

[0147] During structural analysis, the modified bending moment and curvature are used to evaluate the ultimate bearing capacity of RC members.

[0148] Embodiment 2:

[0149] See also Figure 3 A RC component bearing capacity evaluation system is constructed based on the RC component bearing capacity evaluation method described in Example 1, and includes a scanning module, an indicator module, a mapping reconstruction module, a geometric model module, a numerical simulation module and an evaluation module.

[0150] In the system:

[0151] The scanning module is used to scan the surface features of the RC component to be evaluated, and obtain the three-dimensional morphology of the damaged RC component to be evaluated.

[0152] The index module is used to determine the damage evaluation index according to the three-dimensional morphology of the RC component to be evaluated.

[0153] The mapping reconstruction module is used to map and reconstruct the RC component to be evaluated according to the surface random characterization model and the damage evaluation index.

[0154] The geometric model module is used to establish a finite element geometric model of the RC component to be evaluated according to the mapping reconstruction result of the RC component to be evaluated.

[0155] The numerical simulation module is used to perform finite element simulation calculation on the finite element geometric model of the RC component to be evaluated using a finite element parametric simulation method to obtain a prediction result of the bearing performance of the RC component to be evaluated.

[0156] The evaluation module is used to evaluate the bearing performance of the RC component to be evaluated according to the bearing performance prediction result of the RC component to be evaluated, and obtain the bearing performance evaluation result of the RC component.

[0157] The evaluation module is provided with a bearing capacity reduction coefficient model based on damage depth. When evaluating the damage depth of RC components, the model designs a damage depth reduction coefficient. , used to describe the influence of damage depth on the bearing capacity of components. The model combines damage depth, location and geometric characteristics of components to accurately describe the changes in bearing capacity;

[0158] The damage depth affects the reduction factor formula, as shown below:

[0159] ;

[0160] in,

[0161] is the damage depth reduction factor, which ranges from 0 to 1, with 1 indicating no damage and 0 indicating complete failure;

[0162] Adjustment parameters related to material properties and structure type;

[0163] is the average depth of the damaged area;

[0164] is the total thickness of the component;

[0165] is the position correction factor, which is used to describe the influence of damage position on bearing capacity;

[0166] It is the distance between the damaged area and the supporting point or the key stress point;

[0167] is the total length of the component;

[0168] By combining the damage depth and location, its impact is quantified as a reduction factor , used to correct the original bearing capacity during bearing capacity assessment;

[0169] When the damage depth is large or the damaged area is close to the critical stress point, the reduction factor will be significantly reduced, reflecting that the bearing capacity of the structure has been severely weakened;

[0170] When this model is used,

[0171] Obtaining damage depth information through 3D non-contact laser scanning and location of the damaged area ;

[0172] The damage depth reduction factor is calculated according to the formula ;

[0173] When evaluating the carrying capacity, the initial carrying capacity Multiply To obtain the load-bearing capacity after damage .

[0174] Assume that the total thickness of a RC member is The average depth of the damaged area is 200 mm. is 50 mm, and the distance between the damage and the support point is 2 m, the total length of the component is 6 m. and To calculate the damage depth effect reduction factor:

[0175] ;

[0176] This means that the bearing capacity of RC components is approximately 79.14% of the original capacity when the damage depth is taken into account. The exponential function describes the influence of damage depth and position on the bearing capacity, which can more flexibly adapt to the complex damage conditions of different components. It can be effectively combined with the results of 3D morphology reconstruction and damage feature analysis to provide reliable data support for bearing capacity prediction.

[0177] In summary, by searching for public data sets and annotating damage pictures in advance, a high-precision CNN model that meets the needs is trained in combination with a deep learning algorithm; the damaged components are scanned by a non-contact 3D scanner and a SLR camera to obtain morphology information, depth information and point cloud coordinates; the twin model is established by mapping the reconstruction results through morphology information, and the finite element analysis is used to establish the twin model; the finite element parametric simulation method is used to perform finite element simulation calculations on the finite element geometric model of the component to be evaluated; the occurrence and development process of damage is simulated in the model, and the performance changes of the components under different damage states are analyzed; based on the damage simulation results, the bearing capacity of the components in the embodiment is evaluated to determine whether they meet the safety requirements. The present invention scans the damage features of the damaged RC component to be evaluated, thereby realizing non-destructive acquisition of damage morphology data and avoiding damage to the original structure; batch mapping and reconstruction of damaged components can be realized based on finite morphology data, providing data support for subsequent structural performance prediction and reliability evaluation; finite element analysis software is used to perform numerical calculation of the bearing performance of the mapped and reconstructed damaged RC component, and the method introduces the reconstructed surface morphology into the finite element model, so that the calculation result has higher accuracy, realizes the bearing performance prediction and reliability evaluation of the damaged RC component, and solves the problem that actual engineering and mechanical performance tests cannot be repeated; and the performance evaluation means in the field of engineering structure durability and safety can be further enriched.

[0178] The beneficial effects of the present invention are:

[0179] 1. The damage detection of the present invention is simple and efficient. Damage detection based on digital twin technology can provide more comprehensive and in-depth structural status information. At the same time, through integrated learning and data fusion technology, the accuracy and reliability of damage detection are improved. In addition, the twin model can also be used to predict the remaining service life of components and provide support for maintenance and repair decisions.

[0180] 2. The present invention provides all-round and scalable data analysis to achieve long-term tracking and monitoring of the overall damage of RC component structures. It is a useful supplement to local and short-term damage diagnosis technology, greatly expands the connotation of the RC component detection field, and improves the possibility of predictive evaluation.

[0181] 3. The present invention is based on statistical analysis of data and evaluation indexes to establish a characterization model of RC components with different damages and meeting certain statistical characteristics. Batch mapping and reconstruction of RC components can be achieved based on limited morphological data, providing data support for subsequent structural performance prediction and reliability evaluation.

[0182] 4. The present invention uses finite element analysis software to perform numerical calculation of the bearing performance of the mapped and reconstructed damaged RC components, introduces the reconstructed surface morphology into the finite element model, and the calculation results have higher accuracy, realizing the prediction of the bearing performance and reliability evaluation of the damaged RC components, solving the problem that actual engineering and mechanical performance tests cannot be repeated.

[0183] 5. The present invention can further enrich the performance evaluation methods in the field of engineering structure durability and safety, and provide a technical basis for engineering technicians engaged in the detection, evaluation and reinforcement of existing engineering structures.

[0184] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0185] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for evaluating the bearing capacity of RC components, characterized in that: The following steps are involved: S1, constructing an image data graph based on RC components; Collect images of RC components with damage, and use LabelMe to annotate and classify different damage categories in the images, such as cracks and spalling. Then collect public datasets of damage features from multiple data sources. The annotated and classified images and the public datasets together form an image database. S2, perform preprocessing and model training; First, the RC component images are denoised, and then the CNN model is trained using the image database combined with Python language; S3, obtaining the three-dimensional morphology of the RC component to be evaluated; Scanning the existing damaged RC components with scanning equipment, analyzing the collected data using the trained CNN model, identifying the damage characteristics of the components, and obtaining the three-dimensional morphology of the RC components; S4, performing mapping reconstruction; Based on the scanning equipment, the depth information of the damage is obtained, and the point cloud information and coordinate information of the scanned image are exported through the software, and the three-dimensional morphology of the RC component to be evaluated is mapped and reconstructed to obtain the morphology mapping and reconstruction result of the RC component to be evaluated; S5. Build a twin model; According to the reconstruction results of the three-dimensional morphology mapping of the RC component to be evaluated, a twin model of the finite element geometric model and material model of the RC component is constructed using the finite element analysis method; S6. Prediction results of bearing capacity calculated by finite element simulation; Using the finite element parametric simulation method, the finite element geometric model of the RC component to be evaluated is simulated and calculated to obtain the bearing performance prediction results of the RC component to be evaluated; S7. Analyze the performance changes of RC components; By simulating the occurrence and development trend of damage in the twin model, the performance changes of components under different damage states are analyzed; S8. Carry out load-bearing capacity assessment of RC components; Based on the damage simulation results, the bearing capacity of the component is evaluated to determine whether it meets the safety requirements, including limit state analysis, performance evaluation and risk assessment.

2. A method for evaluating the bearing capacity of RC components according to claim 1, characterized in that: In step S1, the RC component image clearly marks the categories and names of different damages, and the public data set includes the original image and mask file; When annotating with labelme: Different damage categories are marked with different colors; The images annotated by labelme are json files, and python code is used to convert json files into mask files.

3. A method for evaluating the bearing capacity of RC components according to claim 1, characterized in that: In step S2, during preprocessing, the weighted average method and histogram equalization method are used to perform grayscale conversion and image enhancement processing on the image; The image segmentation algorithm is used to train the CNN model. The image is divided into multiple sub-regions based on the characteristics of color, edge, and grayscale. The edge points of the image are found by using operators to judge the changes in the grayscale values ​​of the image. Different grayscale thresholds are set, and then the grayscale histogram is used to find the segmentation threshold required by the threshold method.

4. A method for evaluating the bearing capacity of RC components according to claim 1, characterized in that: In step S3, the scanning equipment is a three-dimensional non-contact laser scanner and a single-lens reflex camera. By adjusting the scanner position and the focal length of the rubber ring, physical images at different angles and positions are acquired to obtain the three-dimensional morphology of the damaged component to be evaluated.

5. A method for evaluating the bearing capacity of RC components according to claim 1, characterized in that: In step S4, the acquisition of the damage depth information includes the following process: Post multiple calibration points in advance; The scanning device identifies the calibration points and determines the point cloud coordinate data of the calibration points; According to the distance between the lens and the component, the specific depth of each point is calculated based on the point cloud coordinate data.

6. A method for evaluating the bearing capacity of RC components according to claim 1, characterized in that: In step S5, the finite element analysis method uses Abaqus to collect information on design parameters, material properties, and historical maintenance records of RC components to provide basic data for model construction; The process of building a twin model of an RC component is as follows: Data collection: Collect information on design parameters, material properties, and historical maintenance records of RC components; Model building: Based on the collected data, the finite element geometric model and material model of the RC component are established; Parameter calibration: Calibrate model parameters through comparative testing with actual components; Real-time update: Integrate sensor data to update the model in real time to reflect the actual working status of the component.

7. A method for evaluating the bearing capacity of RC components according to claim 1, characterized in that: In step S6, the finite element simulation calculation specifically includes: Based on the finite element twin model of the RC component, the material properties are set, the loads and boundary conditions are applied, and the finite element numerical simulation is performed to obtain the prediction results of the bearing capacity of the RC component to be evaluated.

8. A method for evaluating the bearing capacity of RC components according to claim 1, characterized in that: In step S8, when evaluating the bearing capacity of the component, the following specific process is included: Preprocessing and analysis: Preprocess the data simulated in the twin model to eliminate noise and outliers, and use statistical methods to analyze the damage mode and development trend of the components; Bearing capacity calculation: Based on the material properties, geometric parameters and damage state of the RC component, the bearing capacity is calculated using the structural analysis method. For example, for bending members, the moment-curvature analysis method is used to evaluate its ultimate bearing state; Safety factor assessment: Determine the safety factor based on the use environment and load characteristics of the component, and evaluate the safety of the component under different working conditions. The determination of the safety factor complies with relevant industry standards; Maintenance and reinforcement suggestions: Based on the load-bearing performance prediction results, suggestions for component maintenance and reinforcement are put forward. For components with insufficient load-bearing capacity, appropriate reinforcement measures are recommended, such as increasing reinforcement and using prestressing technology; Long-term monitoring plan: Establish a long-term monitoring plan to regularly conduct damage detection and load-bearing capacity assessment on RC components, grasp the performance changes of components in real time, and provide data support for the full life cycle management of the structure; In the calculation of bearing capacity, the change of bearing capacity under the influence of cracks is considered, and a damage-corrected moment-curvature relationship model is designed. The model is as follows: ； in, is the bending moment after damage correction; is the initial limit bending moment in the undamaged state; is the adjustment factor related to material properties; is the number of cracks in the damaged area; is the total number of cracks in the component; is the damage-corrected curvature; is the initial curvature in the undamaged state; is the adjustment factor related to the geometric characteristics of the component; is the width of damage, such as the width of a crack or spall; is the total width of the component; By considering the distribution characteristics of damage, such as the number and width of cracks, the bending moment and curvature are corrected to accurately predict the bearing capacity of flexural members in the damaged state; As the damage increases, the bending moment will decrease, while the curvature will increase, reflecting the phenomenon of reduced stiffness and increased deformation capacity of the component; When using, Extract damage features, including crack quantity and width, from 3D topography reconstruction results; Calculate the damage-corrected bending moment using the formula and curvature ; During structural analysis, the modified bending moment and curvature are used to evaluate the ultimate bearing capacity of RC members.

9. A RC component bearing capacity assessment system, constructed based on the RC component bearing capacity assessment method according to any one of claims 1 to 8, characterized in that: It includes a scanning module, an indicator module, a mapping and reconstruction module, a geometric model module, a numerical simulation module and an evaluation module.

10. A RC component bearing capacity assessment system according to claim 9, characterized in that: In the system: A scanning module is used to scan the surface features of the RC component to be evaluated to obtain the three-dimensional morphology of the damaged RC component to be evaluated; An index module is used to determine damage evaluation indexes according to the three-dimensional morphology of the RC component to be evaluated; A mapping reconstruction module is used to map and reconstruct the RC component to be evaluated based on the surface random characterization model and damage evaluation index; A geometric model module, used for establishing a finite element geometric model of the RC component to be evaluated according to the mapping reconstruction result of the RC component to be evaluated; A numerical simulation module is used to perform finite element simulation calculations on the finite element geometric model of the RC component to be evaluated using a finite element parametric simulation method to obtain a prediction result of the bearing performance of the RC component to be evaluated; An evaluation module, used to evaluate the bearing performance of the RC component to be evaluated according to the bearing performance prediction result of the RC component to be evaluated, and obtain the bearing performance evaluation result of the RC component; The evaluation module is provided with a bearing capacity reduction coefficient model based on damage depth. When evaluating the damage depth of RC components, the model designs a damage depth reduction coefficient. , used to describe the influence of damage depth on the bearing capacity of components. The model combines damage depth, location and geometric characteristics of components to accurately describe the changes in bearing capacity; The damage depth affects the reduction factor formula, as shown below: ； in, is the damage depth reduction factor, which ranges from 0 to 1, with 1 indicating no damage and 0 indicating complete failure; Adjustment parameters related to material properties and structure type; is the average depth of the damaged area; is the total thickness of the component; is the position correction factor, which is used to describe the influence of damage position on bearing capacity; It is the distance between the damaged area and the supporting point or the key stress point; is the total length of the component; By combining the damage depth and location, its impact is quantified as a reduction factor , used to correct the original bearing capacity during bearing capacity assessment; When the damage depth is large or the damaged area is close to the critical stress point, the reduction factor will be significantly reduced, reflecting that the bearing capacity of the structure has been severely weakened; When this model is used, Obtaining damage depth information through 3D non-contact laser scanning and location of the damaged area ; The damage depth reduction factor is calculated according to the formula ; When evaluating the carrying capacity, the initial carrying capacity Multiply To obtain the load-bearing capacity after damage .

Images