Existing concrete beam bearing capacity utilization rate evaluation and safety rating method and device

Abstract

The application provides a kind of both concrete beam bearing capacity utilization rate evaluation and safety rating method and device, the application obtains the current crack image of beam and extracts static damage characteristics;Obtain the design characteristic parameter containing axial constraint stiffness coefficient, to count the passive constraint effect of surrounding component;When there is historical crack image, calculate the inter-period damage increment feature, input the prediction model jointly with static characteristics and design parameters;Otherwise, only input static characteristics and design parameters;The ratio of current actual load and ultimate peak load, i.e. bearing capacity utilization rate, is output by the model;According to the preset threshold, the corresponding disposal suggestion is output. The method realizes objective, adaptive calibration of bearing capacity evaluation for irregular inspection by quantifying apparent damage, introducing axial constraint effect and using time sequence change of damage.

Description

Technical Field

[0001] This invention relates to the field of civil engineering structural health monitoring technology, and in particular to a method and apparatus for assessing the load-bearing capacity utilization rate and safety rating of existing concrete beams. Background Technology

[0002] Reinforced concrete beams are the main load-bearing components in building structures. Over long periods of use, due to factors such as load variations, material aging, and environmental erosion, beams inevitably suffer damage such as cracks and concrete crushing, leading to a gradual decrease in load-bearing capacity. Accurately assessing the load-bearing capacity utilization rate of in-service beams—that is, the ratio of the current actual load to the ultimate peak load—is a crucial basis for determining the structural safety status and making maintenance decisions.

[0003] Currently, load-bearing capacity assessment mainly relies on in-situ load tests or finite element numerical analysis. In-situ load tests require direct loading onto the structure, which is costly, time-consuming, and may affect the normal use of the building or even cause additional damage to the structure. Finite element analysis, on the other hand, requires the establishment of a detailed structural model, the modeling process is complex, and the selection of material parameters and boundary conditions is highly dependent on the engineer's experience, resulting in greater uncertainty in the assessment results.

[0004] In recent years, computer vision-based crack detection technology has been developed. Deep learning algorithms can automatically identify cracks and measure their geometric parameters from beam surface images, offering advantages such as non-contact operation and high efficiency. However, most of these technologies currently remain at the level of crack identification and parameter measurement, failing to establish a quantitative mapping relationship between apparent damage such as cracks and crushing and internal bearing capacity. Assessment conclusions still rely on the experience and judgment of inspectors, lacking objectivity and consistency.

[0005] Furthermore, beams in actual buildings are not isolated components but are constrained by surrounding floor slabs, columns, and other structural members. This constraint generates passive axial pressure within the beam, significantly altering crack propagation patterns and load-bearing capacity evolution characteristics. Most existing inspection and assessment methods treat beams as unconstrained simply supported members, ignoring this constraint effect and leading to assessment results that deviate from the actual stress state. Simultaneously, existing image-based inspection methods typically rely solely on static damage images obtained from a single inspection, failing to utilize damage changes across different inspection periods, thus making it difficult to accurately reflect the decay trend of the component's safety reserve and the risk of failure.

[0006] In summary, the problems existing in the current technology urgently need to be solved. Summary of the Invention

[0007] This invention provides a method and apparatus for evaluating the load-bearing capacity utilization rate and safety rating of existing concrete beams, in order to overcome the deficiencies in the prior art and achieve an objective load-bearing capacity assessment that is compatible with irregular inspections.

[0008] This invention provides a method for evaluating the load-bearing capacity utilization rate and safety rating of existing concrete beams, including: Obtain crack images of the target reinforced concrete beam in its current state; The crack image is processed to extract static damage feature parameters; The design characteristic parameters of the target reinforced concrete beam are obtained, and the design characteristic parameters include at least the axial constraint stiffness coefficient, which characterizes the axial constraint effect of the surrounding components on the target reinforced concrete beam. Determine if historical crack images exist in the target reinforced concrete beam: If present, the inter-period damage increment features are determined based on the crack image under the current state and the historical crack image. The static damage feature parameters, the design feature parameters, and the inter-period damage increment features are then input into a pre-established prediction model to output the current bearing capacity utilization rate of the target reinforced concrete beam. If it does not exist, the static damage characteristic parameters and the design characteristic parameters are input into the prediction model, and the bearing capacity utilization rate is output. Based on the load-bearing capacity utilization rate, the safety level of the target reinforced concrete beam is determined according to a preset grading threshold, and a treatment suggestion scheme corresponding to the safety level is output. The bearing capacity utilization rate is the ratio of the current actual load to the ultimate peak load of the target reinforced concrete beam.

[0009] According to the present invention, a method for evaluating the load-bearing capacity utilization and safety rating of existing concrete beams, the step of processing the crack image and extracting static damage feature parameters specifically includes: The crack image is preprocessed, including grayscale conversion, filtering and noise reduction, and contrast enhancement. Pixel-level semantic segmentation is performed on the preprocessed image to identify crack areas and concrete crushing areas; Crack feature parameters are extracted from the identified crack regions, including crack width feature values, crack length feature values, crack spacing feature values, and crack penetration ratio feature values; and The crushing feature parameters are extracted from the identified concrete crushing areas, and the crushing feature parameters include the characteristic value of the concrete crushing area.

[0010] According to the present invention, a method for evaluating the load-bearing capacity utilization rate and safety rating of existing concrete beams, wherein obtaining the design characteristic parameters of the target reinforced concrete beam, the design characteristic parameters including at least the axial constraint stiffness coefficient for characterizing the axial constraint effect of surrounding members on the target reinforced concrete beam, specifically includes: The target reinforced concrete beam is placed in an overall structural model that includes surrounding beam-column nodes for loading; Measure the axial force and axial displacement data at the beam end during the loading process, and determine the equivalent axial constraint stiffness based on the ratio of the axial force to the axial displacement. The axial constraint stiffness coefficient is determined based on the equivalent axial constraint stiffness, beam width, and axial compressive strength of concrete. The axial constraint stiffness coefficient represents the degree to which the passive axial pressure generated by the constraint of the surrounding floor slabs and vertical components inhibits the development of cracks in the beam.

[0011] According to the present invention, a method for assessing the load-bearing capacity utilization rate and safety rating of existing reinforced concrete beams includes the following steps: determining the inter-period damage increment characteristics based on crack images in the current state and historical crack images, and inputting the static damage characteristic parameters, the design characteristic parameters, and the inter-period damage increment characteristics into a pre-established prediction model to output the current load-bearing capacity utilization rate of the target reinforced concrete beam. The historical crack images and the crack images in the current state are subjected to viewpoint correction and pixel-level registration to eliminate image deviations caused by differences in shooting viewpoint, distance and lighting conditions. Based on the registered images, the change of the same damage feature parameter within the two detection intervals is calculated to form the inter-period damage increment feature. The inter-period damage increment feature includes at least the crack width increment, crack length increment, crack penetration ratio increment, number of new cracks, and concrete crushing area increment. The degree of damage in the current period, represented by the static damage feature parameters, and the magnitude of damage change, represented by the inter-period damage increment features, are jointly inferred through the prediction model. The prediction model uses the bearing capacity utilization rate obtained based on the static damage characteristics as a benchmark value, uses the damage change amplitude corresponding to the intertemporal damage increment characteristics to quantify and offset the benchmark value, and performs convergence calibration on the prediction confidence interval to output the corrected bearing capacity utilization rate and prediction confidence.

[0012] According to the present invention, a method for evaluating the load-bearing capacity utilization rate and safety rating of existing concrete beams is provided, wherein the prediction model is pre-established in the following manner: Obtain a time-series image set of a reinforced concrete beam with axial constraints from the initial loading to failure during the stepwise loading process; Static damage characteristic parameters at each load time are extracted from the time series image set, and the inter-period damage increment samples between adjacent time times are calculated. The design characteristic parameters of each beam, the static damage characteristic parameters at each load moment, the incremental damage samples across periods, and the bearing capacity utilization rate at the corresponding moment are correlated to construct a damage characteristic sample dataset. The prediction model is obtained by training a machine learning algorithm using the design feature parameters, static damage features, and inter-period damage increment features as inputs and the bearing capacity utilization rate as output.

[0013] According to the present invention, a method for evaluating the load-bearing capacity utilization rate and safety rating of existing concrete beams is provided, wherein the prediction model is further used for: Obtain the intermediate feature map and corresponding attention weights generated by the prediction model during the inference process of the crack image in the current state; Based on the attention weights, the contribution weight of each image region to the predicted load capacity utilization rate is calculated. The contribution weights are mapped back to the spatial coordinates of the crack image in the current state to generate a damage contribution heatmap. In the damage contribution heatmap, the areas marked with a preset warm color represent the damage areas whose contribution weights to the bearing capacity utilization prediction results meet preset conditions.

[0014] According to the present invention, a method for assessing the load-bearing capacity utilization rate and safety rating of existing concrete beams, the step of determining the safety level of the target reinforced concrete beam based on the load-bearing capacity utilization rate and a preset grading threshold, and outputting a treatment suggestion scheme corresponding to the safety level, specifically includes: When the load-bearing capacity utilization rate is in the first value range, a suggestion for routine inspection without maintenance is output. When the load-bearing capacity utilization rate is in the second numerical range, a suggestion is output to periodically detect and seal cracks whose width exceeds the normal use limit. When the load-bearing capacity utilization rate is in the third value range, a recommendation is given to perform crack grouting or carbon fiber reinforcement within a specified period and to limit the load. When the load-bearing capacity utilization rate is in the fourth value range, the system will immediately recommend setting up temporary support, limiting or stopping use, and initiating replacement or permanent reinforcement solutions within a specified time limit.

[0015] This invention also provides a device for evaluating the load-bearing capacity utilization rate and safety rating of existing concrete beams, comprising: The image acquisition module is used to acquire crack images of the target reinforced concrete beam in its current state. The image processing module is used to process the crack image and extract static damage feature parameters; The parameter acquisition module is used to acquire the design characteristic parameters of the target reinforced concrete beam. The design characteristic parameters include at least the axial constraint stiffness coefficient, which characterizes the axial constraint effect of the surrounding components on the target reinforced concrete beam. The image determination module is used to determine whether historical crack images of the target reinforced concrete beam exist. The first calculation module is used to determine the inter-period damage increment features based on the crack image in the current state and the historical crack image if they exist, and input the static damage feature parameters, the design feature parameters and the inter-period damage increment features into a pre-established prediction model to output the current bearing capacity utilization rate of the target reinforced concrete beam. The second calculation module is used to input the static damage characteristic parameters and the design characteristic parameters into the prediction model if the static damage characteristic parameters do not exist, and output the bearing capacity utilization rate. The result output module is used to determine the safety level of the target reinforced concrete beam according to the bearing capacity utilization rate and a preset grading threshold, and output a treatment suggestion scheme corresponding to the safety level. The bearing capacity utilization rate is the ratio of the current actual load to the ultimate peak load of the target reinforced concrete beam.

[0016] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the existing concrete beam load-bearing capacity utilization rate assessment and safety rating method as described above.

[0017] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the existing concrete beam load-bearing capacity utilization rate assessment and safety rating method as described above.

[0018] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the existing concrete beam load-bearing capacity utilization rate assessment and safety rating method as described above.

[0019] The present invention provides a method and apparatus for evaluating the load-bearing capacity utilization and safety rating of existing concrete beams. By extracting static damage feature parameters from crack images, the apparent damage of the beam is transformed into quantifiable indicators, providing an objective data basis for quantitative evaluation of load-bearing capacity and avoiding the subjectivity of manual visual inspection.

[0020] By introducing the axial constraint stiffness coefficient as a design characteristic parameter, the passive constraint effect of surrounding components on the beam is incorporated into the evaluation, so that the damage-bearing capacity mapping relationship is established on the basis of the actual stress state, reducing the evaluation deviation caused by ignoring the constraint effect.

[0021] By determining the existence of historical crack images and adaptively selecting the input mode, when historical images are available, the method utilizes intertemporal damage increment features to reflect the magnitude of damage changes, calibrating the static feature-based assessment results and improving accuracy. When no historical images are available, the method can still output assessment results based solely on static features, ensuring its usability under single-detection conditions. Switching between the two modes does not require a fixed detection interval, adapting to scenarios of irregular inspections of existing buildings.

[0022] By mapping the load-bearing capacity utilization rate to a safety level according to a preset grading threshold and outputting corresponding disposal suggestions, a direct connection from damage detection to operation and maintenance decision-making is achieved, reducing information loss in intermediate links and reliance on manual judgment. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0024] Figure 1 This is a flowchart illustrating the method for evaluating the load-bearing capacity utilization and safety rating of existing concrete beams provided by the present invention. Figure 2 This is a current crack image of a reinforced concrete frame beam and a corresponding schematic diagram of its load-bearing capacity utilization rate, provided by the present invention. Figure 3 This is a schematic diagram of the structure of the device for evaluating the load-bearing capacity utilization rate and safety rating of existing concrete beams provided by the present invention. Figure 4 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0026] To address the problems in existing technologies, this invention proposes a method for evaluating the load-bearing capacity utilization rate and safety rating of existing concrete beams, aiming to achieve an objective load-bearing capacity assessment that is compatible with irregular inspections. The method for evaluating the load-bearing capacity utilization rate and safety rating of existing concrete beams is described below, as follows: Figure 1 As shown, including but not limited to the following steps: Step 110: Obtain the crack image of the target reinforced concrete beam in its current state.

[0027] In step 110, the side of the target reinforced concrete beam is photographed using an image acquisition device to obtain image data including surface cracks and crushed areas. The acquired images should cover the entire length and height of the beam side to ensure the integrity of the surface damage information. The image acquisition device can be a handheld camera, a wall-climbing robot equipped with a camera module, or a ground-based mobile robot, or other devices with image acquisition capabilities. Uniform lighting conditions should be ensured during acquisition to avoid the loss of local image information due to shadows or reflections.

[0028] Step 120: Process the crack image and extract static damage feature parameters.

[0029] In step 120, the crack image obtained in step 110 is digitally processed to extract static damage feature parameters reflecting the current damage state of the beam. These static damage feature parameters include two categories: crack feature parameters and crushing feature parameters. Crack feature parameters describe the geometric morphology of cracks on the beam surface, while crushing feature parameters describe the extent and degree of crushing and spalling in the concrete compression zone. The extracted static damage feature parameters are output in numerical form for use in subsequent steps.

[0030] Step 130: Obtain the design characteristic parameters of the target reinforced concrete beam. The design characteristic parameters include at least the axial constraint stiffness coefficient, which characterizes the axial constraint effect of the surrounding components on the target reinforced concrete beam.

[0031] In step 130, the design characteristic parameters of the target reinforced concrete beam are obtained, including at least the axial restraint stiffness coefficient. The axial restraint stiffness coefficient characterizes the strength of the passive axial restraint effect exerted on the target beam by surrounding floor slabs, columns, and other structural members in the actual structure. This parameter can be obtained through structural analysis calculations or experimental measurements. In addition to the axial restraint stiffness coefficient, the design characteristic parameters may also include other geometric and mechanical parameters reflecting the beam's stress characteristics. These design characteristic parameters will serve as one of the inputs to the prediction model, enabling the model to assess the load-bearing capacity utilization rate while considering the actual structural constraints.

[0032] Step 140: Determine whether there are historical crack images of the target reinforced concrete beam.

[0033] In step 140, it is determined whether the target reinforced concrete beam contains historical crack images acquired at a previous detection time. Historical crack images refer to crack images taken on the same beam at a time earlier than the current detection time. The result of this determination will determine the input mode selection for subsequent steps: if historical crack images exist, proceed to step 150; if not, proceed to step 160.

[0034] Step 150: If it exists, determine the inter-period damage increment features based on the crack image in the current state and the historical crack image, and input the static damage feature parameters, the design feature parameters and the inter-period damage increment features into the pre-established prediction model to output the current bearing capacity utilization rate of the target reinforced concrete beam.

[0035] In step 150, when step 140 determines the existence of historical crack images, inter-period damage increment features are determined based on the current crack image and historical crack images. These inter-period damage increment features characterize the change in the same damage parameter between two detection intervals. The static damage feature parameters extracted in step 120, the design feature parameters obtained in step 130, and these inter-period damage increment features are jointly input into a pre-established prediction model. The prediction model infers based on the above inputs and outputs the current bearing capacity utilization rate of the target reinforced concrete beam, which is defined as the ratio of the beam's current actual load to its ultimate peak load.

[0036] Step 160: If not, input the static damage characteristic parameters and the design characteristic parameters into the prediction model, and output the bearing capacity utilization rate. The bearing capacity utilization rate is the ratio of the current actual load to the ultimate peak load of the target reinforced concrete beam.

[0037] In step 160, when step 140 determines that no historical crack images exist, the static damage feature parameters extracted in step 120 and the design feature parameters obtained in step 130 are input into the prediction model. The prediction model then outputs the current bearing capacity utilization rate of the target reinforced concrete beam based on the static damage level and the design feature parameters. The bearing capacity utilization rate is the ratio of the current actual load to the ultimate peak load, ranging from 0% to 100%. A higher value indicates less remaining bearing capacity reserve in the beam, and the safety state is closer to the critical point.

[0038] Step 170: Based on the bearing capacity utilization rate, determine the safety level of the target reinforced concrete beam according to the preset grading threshold, and output the treatment suggestion scheme corresponding to the safety level.

[0039] In step 170, the load-bearing capacity utilization rate output in step 150 or 160 is compared with a preset grading threshold to determine the safety level of the target reinforced concrete beam. Different safety levels correspond to different structural safety states, and corresponding treatment recommendations are preset. Based on the determined safety level, the corresponding treatment recommendations are output to provide a basis for structural operation, maintenance, and repair / reinforcement decisions.

[0040] As a further optional embodiment, the step of processing the crack image and extracting static damage feature parameters specifically includes: The crack image is preprocessed, including grayscale conversion, filtering and noise reduction, and contrast enhancement. Pixel-level semantic segmentation is performed on the preprocessed image to identify crack areas and concrete crushing areas; Crack feature parameters are extracted from the identified crack regions, including crack width feature values, crack length feature values, crack spacing feature values, and crack penetration ratio feature values; and The crushing feature parameters are extracted from the identified concrete crushing areas, and the crushing feature parameters include the characteristic value of the concrete crushing area.

[0041] In this embodiment, the preprocessing of the crack image is performed sequentially in the order of grayscale conversion, denoising and filtering, and contrast enhancement. Grayscale conversion converts the color image into a single-channel grayscale image, reducing the computational load of subsequent processing. Denoising and filtering can employ Gaussian filtering or median filtering to suppress noise interference introduced during image acquisition while preserving crack edge information as much as possible. Contrast enhancement can utilize histogram equalization to increase the grayscale difference between the crack area and the background area, facilitating accurate differentiation between damaged and intact areas by subsequent segmentation algorithms.

[0042] Pixel-level semantic segmentation employs a deep learning-based image segmentation network to classify each pixel in the preprocessed image, assigning it to a crack region, a concrete crushing region, or the background region. Commonly used segmentation networks include Mask R-CNN, U-Net, or the DeepLab series, which can be selected based on actual accuracy and efficiency requirements. The output of the segmentation network is a mask image of the same size as the original image, where crack regions and crushing regions are identified with different label values.

[0043] After obtaining the mask of the crack region, crack feature parameters are extracted. Crack width features include the maximum and average crack width, which can be calculated from the pixel distance along the normal direction of the crack region edge. Crack length features are obtained by counting pixels in the crack skeleton and converting the data to obtain the actual length. Crack spacing features are obtained by calculating the horizontal distance between adjacent crack skeletons at the bottom of the beam, 1 / 3 of the beam height, and 2 / 3 of the beam height. Crack penetration ratio features are the ratio of the sum of the projected lengths of all cracks along the beam height to the beam section height. After obtaining the mask of the crushed region, the ratio of the crushed region pixel area to the total pixel area of ​​the beam side is calculated as the crushed area feature; this feature is zero when there is no crushed region in the segmentation result.

[0044] As a further optional embodiment, the step of obtaining the design characteristic parameters of the target reinforced concrete beam, wherein the design characteristic parameters include at least the step of an axial constraint stiffness coefficient used to characterize the axial constraint effect of surrounding members on the target reinforced concrete beam, specifically includes: The target reinforced concrete beam is placed in an overall structural model that includes surrounding beam-column nodes for loading; Measure the axial force and axial displacement data at the beam end during the loading process, and determine the equivalent axial constraint stiffness based on the ratio of the axial force to the axial displacement. The axial constraint stiffness coefficient is determined based on the equivalent axial constraint stiffness, beam width, and axial compressive strength of concrete. The axial constraint stiffness coefficient represents the degree to which the passive axial pressure generated by the constraint of the surrounding floor slabs and vertical components inhibits the development of cracks in the beam.

[0045] In this embodiment, the axial constraint stiffness coefficient is obtained through a loading test of the overall structural model. The overall structural model refers to a test specimen made by combining the target beam with its surrounding columns, floor slabs, and other components according to the connection relationship in the actual structure, which can simulate the constraint effect of the beam-column-floor slab node in the actual structure. During the test, a vertical load is applied to the target beam in the specimen. During the bending deformation process, the two ends of the beam generate passive constraint axial forces due to the restriction of the surrounding columns and floor slabs.

[0046] During loading, force sensors (built-in before concrete pouring) and displacement sensors (which can be installed after concrete pouring) are placed at the beam ends to measure and record the axial force and displacement data at the beam ends in real time. Based on the measured axial force and displacement data at the beam ends, the ratio between the two is calculated to obtain the equivalent axial constraint stiffness. This equivalent axial constraint stiffness reflects the magnitude of the constraint axial force exerted by the surrounding components on the unit axial deformation of the beam ends.

[0047] After obtaining the equivalent axial restraint stiffness, the axial restraint stiffness coefficient is calculated using the following formula, combining the beam width and concrete axial compressive strength of the target beam: Dividing the equivalent axial restraint stiffness by the product of the beam width and concrete axial compressive strength yields the dimensionless axial restraint stiffness coefficient. A larger coefficient indicates stronger restraint from surrounding members on the beam, greater passive axial pressure within the beam, and a more significant inhibitory effect on crack propagation. Using this coefficient as one of the input features of the prediction model allows the model to learn a damage-bearing capacity mapping relationship based on the actual structural constraints.

[0048] Specifically, the formula for calculating the axial constraint stiffness coefficient is as follows: η = K a / ( b × f c ) in, K a The equivalent axial constraint stiffness at the beam end (can be obtained by extracting the reaction force by applying a unit axial displacement to the finite element model). b For the width of the beam, f c This refers to the axial compressive strength of concrete.

[0049] As a further optional embodiment, the step of determining the inter-period damage increment characteristics based on the crack image in the current state and the historical crack image, and inputting the static damage feature parameters, the design feature parameters, and the inter-period damage increment characteristics into a pre-established prediction model to output the current bearing capacity utilization rate of the target reinforced concrete beam, specifically includes: The historical crack images and the crack images in the current state are subjected to viewpoint correction and pixel-level registration to eliminate image deviations caused by differences in shooting viewpoint, distance and lighting conditions. Based on the registered images, the change of the same damage feature parameter within the two detection intervals is calculated to form the inter-period damage increment feature. The inter-period damage increment feature includes at least the crack width increment, crack length increment, crack penetration ratio increment, number of new cracks, and concrete crushing area increment. The degree of damage in the current period, represented by the static damage feature parameters, and the magnitude of damage change, represented by the inter-period damage increment features, are jointly inferred through the prediction model. The prediction model uses the bearing capacity utilization rate obtained based on the static damage characteristics as a benchmark value, uses the damage change amplitude corresponding to the intertemporal damage increment characteristics to quantify and offset the benchmark value, and performs convergence calibration on the prediction confidence interval to output the corrected bearing capacity utilization rate and prediction confidence.

[0050] In this embodiment, firstly, perspective correction and pixel-level registration are performed on the historical crack image and the current crack image. Since the shooting angle, distance, and lighting conditions of the two detections are unlikely to be completely identical, the same physical location in the two images may exhibit rotation, scaling, translation, and brightness differences. Perspective correction extracts feature points from both images and calculates the homography matrix or affine transformation matrix to transform the historical image to a perspective and scale consistent with the current image. Pixel-level registration further employs similarity measures such as mutual information or normalized cross-correlation to perform local fine-tuning of the two images, ensuring that the same crack or crushing area is located at the same pixel coordinate position in both images.

[0051] After image registration, damage characteristic parameters such as crack width, crack length, crack penetration ratio, number of cracks, and concrete crushing area are extracted from the two registered images. The intertemporal increments of each damage parameter are obtained by subtracting the corresponding historical feature values ​​from the current feature value, forming the intertemporal damage increment features. Specifically, the crack width increment reflects the degree of crack opening expansion, the crack length and penetration ratio increments reflect the degree of crack extension along the beam height, the number of newly added cracks reflects the number of new cracks appearing within the two detection intervals, and the crushing area increment reflects the development of the concrete spalling area.

[0052] Static damage characteristic parameters, design characteristic parameters, and inter-period damage increment characteristics are jointly input into the prediction model for inference. The inference mechanism of the prediction model is as follows: First, based on the static damage characteristic parameters and design characteristic parameters, a preliminary bearing capacity utilization benchmark value is output; then, the benchmark value is quantitatively offset and corrected using the damage change amplitude represented by the inter-period damage increment characteristics. The correction logic is: the larger the damage change amplitude, the faster the component's safety reserve decays, and the bearing capacity utilization rate should shift towards a more dangerous direction. At the same time, due to the introduction of inter-period time series information, the model has a more sufficient basis for perceiving the beam damage evolution trend, reducing the uncertainty of the prediction. The model performs convergence calibration on the prediction confidence interval, outputting a narrower confidence interval width than when using only static features, resulting in a higher prediction confidence level.

[0053] As a further optional embodiment, the prediction model is pre-established in the following manner: Obtain a time-series image set of a reinforced concrete beam with axial constraints from the initial loading to failure during the stepwise loading process; Static damage characteristic parameters at each load time are extracted from the time series image set, and the inter-period damage increment samples between adjacent time times are calculated. The design characteristic parameters of each beam, the static damage characteristic parameters at each load moment, the incremental damage samples across periods, and the bearing capacity utilization rate at the corresponding moment are correlated to construct a damage characteristic sample dataset. The prediction model is obtained by training a machine learning algorithm using the design feature parameters, static damage features, and inter-period damage increment features as inputs and the bearing capacity utilization rate as output.

[0054] In this embodiment, the establishment of the prediction model is divided into four stages: data acquisition, feature extraction, dataset construction, and model training.

[0055] During the data acquisition phase, a step-by-step loading test was conducted on reinforced concrete beam specimens with axial constraints. The specimens were fabricated according to the beam-column-slab connection method in actual structures, ensuring that the beam ends were naturally constrained by surrounding components during loading. Loading was performed using a displacement-controlled method, gradually increasing the mid-span deflection of the beam with a fixed displacement step size. Loading was paused after each load level stabilized, and images of cracks on the beam's sides were captured until beam failure occurred. This yielded a time-series image set documenting the entire process from the initial intact state under loading, cracking, crack development, crushing, to final failure, with each image corresponding to a known load level.

[0056] In the feature extraction stage, for each image in the time-series image set, the static damage characteristic parameters at that loading moment are extracted using the same method as in the aforementioned embodiments. These parameters include crack width characteristic values, crack length characteristic values, crack spacing characteristic values, crack penetration ratio characteristic values, and crushing area characteristic values. Simultaneously, after registering images from two adjacent loading moments, the changes in each damage characteristic parameter are calculated to form inter-period damage increment samples. Each set of increment samples corresponds to the static damage characteristics and bearing capacity utilization rate at the next loading moment.

[0057] During the dataset construction phase, the design characteristic parameters of each test beam (including axial constraint stiffness coefficient, shear span ratio, bending-shear ratio, etc.), static damage characteristic parameters at each loading moment, corresponding inter-period damage increment samples, and bearing capacity utilization rate at that moment are correlated to form a structured damage characteristic sample dataset. The bearing capacity utilization rate is determined based on the ratio of the experimentally measured load value to the ultimate peak load of the beam, and is between 0% and 100%. Multiple sets of adjacent time data for the same beam are included in the dataset, enabling the model to learn the correspondence between the degree of damage, the magnitude of damage change, and the bearing capacity decay throughout the entire damage evolution process.

[0058] During the model training phase, design feature parameters, static damage feature parameters, and inter-period damage increment feature parameters are used as input features, and bearing capacity utilization rate is used as the output label. Machine learning algorithms are employed to train the prediction model. Optional algorithms include ensemble learning algorithms such as Extreme Random Forest, XGBoost, and LightGBM, or deep neural network algorithms. During training, the dataset is divided into training, validation, and test sets. Hyperparameter tuning is performed using the validation set, and model accuracy is evaluated using the test set. The trained prediction model supports both single-frame and multi-frame input modes: in single-frame mode, only static damage features and design feature parameters are input; in multi-frame mode, inter-period damage increment features are additionally input, and the model uses the magnitude of damage changes to calibrate the prediction results.

[0059] As a further optional embodiment, the prediction model is also used for: Obtain the intermediate feature map and corresponding attention weights generated by the prediction model during the inference process of the crack image in the current state; Based on the attention weights, the contribution weight of each image region to the predicted load capacity utilization rate is calculated. The contribution weights are mapped back to the spatial coordinates of the crack image in the current state to generate a damage contribution heatmap. In the damage contribution heatmap, the areas marked with a preset warm color represent the damage areas whose contribution weights to the bearing capacity utilization prediction results meet preset conditions.

[0060] In this embodiment, the prediction model generates a damage contribution heatmap to explain the prediction basis while outputting the load-bearing capacity utilization rate. When the prediction model adopts a convolutional neural network architecture, the intermediate feature map output by the last convolutional layer can be obtained during the forward inference process, and the gradient of each channel of the feature map relative to the load-bearing capacity utilization rate output can be calculated. The gradient global average pooling result is used as the attention weight of each feature channel. When the prediction model adopts a Transformer architecture, the attention weight matrix between each image block in the attention layer can be directly extracted. Based on the obtained attention weights, the channels of each intermediate feature map or each image block are weighted and summed to obtain the contribution weight of each pixel position or image block to the load-bearing capacity utilization rate prediction result.

[0061] The contribution weights are mapped back to the spatial coordinates of the original crack image through upsampling or interpolation, forming a heat map with the same size as the original image. This heat map is then rendered in pseudo-color, with warm tones indicating high-contribution areas and cool tones indicating low-contribution areas. Warm tones are typically red, orange, or yellow. The areas marked are the damage areas whose contribution weights to the load-bearing capacity utilization prediction meet preset conditions, indicating that cracks or crushing features in these areas have a significant impact on the model's load-bearing capacity judgment. Engineers can observe the distribution of warm tones in the heat map to verify whether the model's prediction basis is consistent with key damage locations identified in engineering experience. If warm tones are concentrated on a main crack or a crushing and spalling area, it indicates that the model's judgment logic conforms to the structural stress law, increasing confidence in the assessment results. If warm tones are distributed in image noise or shadow areas, it suggests that the model or input data needs to be reviewed. Simultaneously, the high-contribution damage areas marked on the heat map can serve as reference locations for priority reinforcement, guiding targeted reinforcement measures on-site.

[0062] As a further optional embodiment, the step of determining the safety level of the target reinforced concrete beam according to the bearing capacity utilization rate and a preset grading threshold, and outputting a treatment suggestion scheme corresponding to the safety level, specifically includes: When the load-bearing capacity utilization rate is in the first value range, a suggestion for routine inspection without maintenance is output. When the load-bearing capacity utilization rate is in the second numerical range, a suggestion is output to periodically detect and seal cracks whose width exceeds the normal use limit. When the load-bearing capacity utilization rate is in the third value range, a recommendation is given to perform crack grouting or carbon fiber reinforcement within a specified period and to limit the load. When the load-bearing capacity utilization rate is in the fourth value range, the system will immediately recommend setting up temporary support, limiting or stopping use, and initiating replacement or permanent reinforcement solutions within a specified time limit.

[0063] In this embodiment, a preset grading threshold divides the range of load-bearing capacity utilization rate into four consecutive numerical intervals, forming four progressive safety levels. The lower the load-bearing capacity utilization rate, the more sufficient the remaining load-bearing capacity reserve of the beam and the better the safety status; the higher the load-bearing capacity utilization rate, the closer the current load on the beam is to its ultimate bearing capacity and the greater the safety risk.

[0064] When the load-bearing capacity utilization rate is in the first range, it indicates that the beam damage is minor, the current load is far below the ultimate peak load, and the structure is in a safe state. The recommended action at this time is routine inspection without the need for maintenance. A visual inspection according to the normal inspection plan is sufficient, without incurring additional maintenance work.

[0065] When the load-bearing capacity utilization rate is in the second range, it indicates that the beam has suffered some damage and its load-bearing capacity has decreased, but the safety margin is still acceptable. The recommended course of action at this point is to conduct regular inspections and focus on sealing cracks that exceed the normal service life limit. Sealing cracks prevents the intrusion of moisture and corrosive media, slowing down the rate of damage development.

[0066] When the load-bearing capacity utilization rate is in the third range, it indicates that the beam is significantly damaged and its load-bearing capacity has decreased significantly, requiring proactive reinforcement measures. The recommended course of action at this point is to complete reinforcement work such as crack grouting or carbon fiber reinforcement within a specified timeframe, and to restrict the load used until reinforcement is completed. Reinforcement measures can restore some load-bearing capacity reserves, while restricting the load used is a necessary means to control risks before reinforcement.

[0067] When the load-bearing capacity utilization rate is in the fourth range, it indicates that the beam is severely damaged, the current load is close to or may exceed the remaining ultimate bearing capacity, and there is a risk of sudden failure. The recommended course of action at this time is to immediately install temporary supports to ensure safety, restrict or stop the use of the area, and initiate component replacement or permanent reinforcement within the specified time limit to fundamentally eliminate the safety hazard.

[0068] The corresponding treatment measures for the four numerical ranges are progressively tiered, forming a complete hierarchical response system from preventative maintenance to emergency intervention. This ensures that beams with varying degrees of damage receive treatment solutions that match their risk levels. The specific thresholds for each numerical range can be adjusted based on factors such as structural importance and subsequent service life to adapt to the safety management needs of different engineering scenarios.

[0069] Preferably, the first numerical range is a load-bearing capacity utilization rate of less than 25%, the second numerical range is a load-bearing capacity utilization rate of greater than or equal to 25% and less than 50%, the third numerical range is a load-bearing capacity utilization rate of greater than or equal to 50% and less than 75%, and the fourth numerical range is a load-bearing capacity utilization rate of greater than or equal to 75%.

[0070] When the load-bearing capacity utilization rate is less than 25%, the current load on the beam is less than one-quarter of its ultimate bearing capacity, with sufficient remaining safety reserve, minor crack development, and no crushing of the concrete or only minor surface damage. At this point, the structure is in a safe service condition and can be inspected at the normal frequency without requiring any maintenance.

[0071] When the load-bearing capacity utilization rate is between 25% and 50%, visible cracks have appeared in the beams, and the width of some cracks may approach or exceed the limits specified in the serviceability limit state. However, the overall structure still has good load-bearing capacity. At this time, the frequency of inspection should be increased, and cracks exceeding the limit should be treated with surface sealing or injection sealing materials to prevent moisture and corrosive media from penetrating along the cracks and causing steel corrosion and accelerated damage.

[0072] When the load-bearing capacity utilization rate is between 50% and 75%, the beam damage is more significant, with larger crack widths and some cracks extending to the upper part of the beam. Concrete peeling or slight crushing may occur in the compression zone. The remaining load-bearing capacity reserve is relatively limited, and proactive reinforcement measures need to be taken within a limited time. These measures include grouting cracks to restore the integrity of the cross-section, bonding carbon fiber cloth to the bottom or sides of the beam to improve bending or shear capacity, and limiting the applied load to reduce risks before reinforcement is completed.

[0073] When the load-bearing capacity utilization rate reaches or exceeds 75%, the beam is severely damaged, with dense and wide cracks, some of which may extend through the beam height. Significant crushing and spalling of the concrete in the compression zone is possible. The structure is approaching its ultimate load-bearing state, posing a risk of sudden failure. At this point, a temporary support system should be immediately installed to bear part of the load, the use of the area should be restricted or suspended, and a component replacement or permanent reinforcement plan should be developed and implemented within a specified timeframe to fundamentally eliminate the safety hazard.

[0074] Furthermore, the aforementioned grading thresholds can be adjusted based on the structural importance level, remaining service life, or the owner's specific requirements for safety reserves. The adjustment method involves multiplying each threshold by a preset adjustment coefficient. For example, for a structure with a first-level importance level, each threshold can be multiplied by an adjustment coefficient of 0.9, adjusting the thresholds to 22.5%, 45%, and 67.5% respectively, to trigger early warnings and interventions. For structures with a short remaining service life or those about to be demolished, each threshold can be multiplied by an adjustment coefficient of 1.1, appropriately relaxing the assessment criteria and avoiding unnecessary maintenance expenditures. The adjustment coefficient can be selected within the range of 0.7 to 1.3 based on engineering experience. After adjustment, the thresholds at each level still maintain an increasing relationship, with the lower limit of the first level not lower than 0% and the upper limit of the fourth level not higher than 100%.

[0075] In summary, a minimal embodiment of the present invention is as follows: For example, a reinforced concrete frame beam with a span of 4m on the third floor of an existing office building has a cross-sectional dimension of 200mm × 400mm and a concrete strength grade of C30. The beam ends are integrally cast to the frame columns and the upper floor slab. An assessment of the load-bearing capacity utilization rate of this beam is required. No historical monitoring images are available on-site. The current crack images of the reinforced concrete frame beam and the corresponding load-bearing capacity utilization rate are as follows: Figure 2 As shown, Figure 2 The left figure is a schematic diagram of the overall structural model including the target beam and surrounding beam-column nodes. The vertical members on the left and right sides are frame columns, and the horizontal member in the middle is the target reinforced concrete beam. The beam ends are integrally cast and connected to the columns to form axial constraints on the beam. In the load-bearing capacity utilization rate coordinate system, the vertical axis is the vertical load value P (kN) applied during the step-by-step loading process, and the horizontal axis is the vertical displacement value δ (mm) at the loading point or in the middle of the span.

[0076] Step 1: Obtain crack images A handheld digital camera was used to photograph one side of the beam under natural lighting conditions. The camera's optical axis was kept perpendicular to the beam's side, and the field of view covered the entire height of the beam from bottom to top and the entire length from end to mid-span. The resulting images clearly show the direction of cracks and the condition of the concrete surface, with no obvious reflections or shadows.

[0077] Step 2: Extract static damage feature parameters The captured crack images were sequentially processed using grayscale conversion, median filtering for noise reduction, and histogram equalization for contrast enhancement. A U-Net semantic segmentation network was then used to perform pixel-level classification of the preprocessed images, identifying crack areas and concrete crushing areas. The segmentation results showed multiple vertical bending cracks at the bottom mid-span of the beam, while no obvious crushing was observed in the compression zone.

[0078] The following static damage characteristic parameters were extracted from the crack area: the maximum crack width was 0.55 mm, the average crack width was 0.28 mm, the maximum crack length was 275 mm after calibration and conversion, the average spacing between adjacent cracks at the bottom of the beam was 120 mm, and the ratio of the sum of the projected lengths of all cracks in the beam height direction to the beam section height, i.e., the crack penetration ratio, was 0.65. The crushing area characteristic value was set to 0 because no crushing was detected.

[0079] Step 3: Obtain design feature parameters Based on the structural drawings and material measurement data of the beam, the following design characteristic parameters were obtained: the equivalent axial constraint stiffness at the beam end was calculated using a finite element model including adjacent columns and floor slabs, and the axial constraint stiffness coefficient η = 32.1 was calculated; the shear span ratio λ = 5.56 (the ratio of the shear span length of 2000mm to the effective section height of 360mm is 5.56); the bending-shear ratio m = 0.38 (the bending capacity and shear capacity were calculated according to the formula in the code).

[0080] Step 4: Determine if the historical image exists and select the input mode. Upon reviewing the inspection records, it was determined that this was the first inspection of the beam, and no historical crack images were available. Therefore, the single-frame input mode was activated, and the static damage feature parameters and design feature parameters were input into the prediction model.

[0081] Step 5: Predictive Model Inference and Output The prediction model uses a pre-trained XGBoost model. The model inputs are: maximum crack width 0.55 mm, average crack width 0.28 mm, maximum crack length 275 mm, average crack spacing 120 mm, crack penetration ratio 0.65, crush area ratio 0, axial constraint stiffness coefficient 32.1, shear span ratio 5.56, and bending-shear ratio 0.38.

[0082] The model outputs that the current load-bearing capacity utilization rate of the beam is 42%, and the prediction confidence interval is [36%, 48%].

[0083] Step Six: Safety Classification and Handling Recommendations The load-bearing capacity utilization rate of 42% falls into the second value range (25% ≤ utilization rate < 50%), corresponding to the second safety level. The system outputs the following recommended treatment: regular inspection, and sealing of cracks exceeding the normal service limit. It is recommended that the inspection cycle not exceed one year.

[0084] At this point, the load-bearing capacity utilization rate assessment and performance-based classification of the beam are completed. The entire process only requires a current crack image and known design parameters, and can provide quantitative assessment results and clear treatment recommendations without the need for historical data.

[0085] The following describes the device for evaluating the load-bearing capacity utilization rate and safety rating of existing concrete beams provided by this invention, such as... Figure 3 As shown, the device for assessing the utilization rate of the load-bearing capacity of existing concrete beams and the method ... described above can be referred to in correspondence.

[0086] A device for assessing the load-bearing capacity utilization rate and safety rating of existing concrete beams includes: Image acquisition module 310 is used to acquire crack images of the target reinforced concrete beam in its current state; Image processing module 320 is used to process the crack image and extract static damage feature parameters; The parameter acquisition module 330 is used to acquire the design characteristic parameters of the target reinforced concrete beam. The design characteristic parameters include at least the axial constraint stiffness coefficient, which characterizes the axial constraint effect of the surrounding components on the target reinforced concrete beam. Image determination module 340 is used to determine whether there are historical crack images of the target reinforced concrete beam. The first calculation module 350 is used to determine the inter-period damage increment features based on the crack image in the current state and the historical crack image if they exist, and input the static damage feature parameters, the design feature parameters and the inter-period damage increment features into a pre-established prediction model to output the current bearing capacity utilization rate of the target reinforced concrete beam. The second calculation module 360 ​​is used to input the static damage characteristic parameters and the design characteristic parameters into the prediction model if they do not exist, and output the bearing capacity utilization rate and the corresponding prediction confidence level. The result output module 370 is used to determine the safety level of the target reinforced concrete beam according to the bearing capacity utilization rate and a preset grading threshold, and output a treatment suggestion scheme corresponding to the safety level. The bearing capacity utilization rate is the ratio of the current actual load to the ultimate peak load of the target reinforced concrete beam.

[0087] Figure 4 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 4 As shown, the electronic device may include: a processor 410, a communications interface 420, a memory 430, and a communication bus 440, wherein the processor 410, the communications interface 420, and the memory 430 communicate with each other via the communication bus 440. The processor 410 can call logical instructions in the memory 430 to execute a method for assessing the utilization rate of the existing concrete beam's bearing capacity and evaluating its safety. This method includes: Obtain crack images of the target reinforced concrete beam in its current state; The crack image is processed to extract static damage feature parameters; The design characteristic parameters of the target reinforced concrete beam are obtained, and the design characteristic parameters include at least the axial constraint stiffness coefficient, which characterizes the axial constraint effect of the surrounding components on the target reinforced concrete beam. Determine if historical crack images exist in the target reinforced concrete beam: If present, the inter-period damage increment features are determined based on the crack image under the current state and the historical crack image. The static damage feature parameters, the design feature parameters, and the inter-period damage increment features are then input into a pre-established prediction model to output the current bearing capacity utilization rate of the target reinforced concrete beam. If it does not exist, the static damage characteristic parameters and the design characteristic parameters are input into the prediction model, and the bearing capacity utilization rate is output. Based on the load-bearing capacity utilization rate, the safety level of the target reinforced concrete beam is determined according to a preset grading threshold, and a treatment suggestion scheme corresponding to the safety level is output. The bearing capacity utilization rate is the ratio of the current actual load to the ultimate peak load of the target reinforced concrete beam.

[0088] Furthermore, the logical instructions in the aforementioned memory 430 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0089] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to execute the existing concrete beam bearing capacity utilization rate assessment and safety rating method provided by the above methods, the method including: Obtain crack images of the target reinforced concrete beam in its current state; The crack image is processed to extract static damage feature parameters; The design characteristic parameters of the target reinforced concrete beam are obtained, and the design characteristic parameters include at least the axial constraint stiffness coefficient, which characterizes the axial constraint effect of the surrounding components on the target reinforced concrete beam. Determine if historical crack images exist in the target reinforced concrete beam: If present, the inter-period damage increment features are determined based on the crack image under the current state and the historical crack image. The static damage feature parameters, the design feature parameters, and the inter-period damage increment features are then input into a pre-established prediction model to output the current bearing capacity utilization rate of the target reinforced concrete beam. If it does not exist, the static damage characteristic parameters and the design characteristic parameters are input into the prediction model, and the bearing capacity utilization rate is output. Based on the load-bearing capacity utilization rate, the safety level of the target reinforced concrete beam is determined according to a preset grading threshold, and a treatment suggestion scheme corresponding to the safety level is output. The bearing capacity utilization rate is the ratio of the current actual load to the ultimate peak load of the target reinforced concrete beam.

[0090] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to perform the existing concrete beam load-bearing capacity utilization rate assessment and safety rating method provided by the methods described above, the method comprising: Obtain crack images of the target reinforced concrete beam in its current state; The crack image is processed to extract static damage feature parameters; The design characteristic parameters of the target reinforced concrete beam are obtained, and the design characteristic parameters include at least the axial constraint stiffness coefficient, which characterizes the axial constraint effect of the surrounding components on the target reinforced concrete beam. Determine if historical crack images exist in the target reinforced concrete beam: If present, the inter-period damage increment features are determined based on the crack image under the current state and the historical crack image. The static damage feature parameters, the design feature parameters, and the inter-period damage increment features are then input into a pre-established prediction model to output the current bearing capacity utilization rate of the target reinforced concrete beam. If it does not exist, the static damage characteristic parameters and the design characteristic parameters are input into the prediction model, and the bearing capacity utilization rate is output. Based on the load-bearing capacity utilization rate, the safety level of the target reinforced concrete beam is determined according to a preset grading threshold, and a treatment suggestion scheme corresponding to the safety level is output. The bearing capacity utilization rate is the ratio of the current actual load to the ultimate peak load of the target reinforced concrete beam.

[0091] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0092] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0093] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for evaluating the load-bearing capacity utilization rate and safety rating of existing concrete beams, characterized in that, include: Obtain crack images of the target reinforced concrete beam in its current state; The crack image is processed to extract static damage feature parameters; The design characteristic parameters of the target reinforced concrete beam are obtained, and the design characteristic parameters include at least the axial constraint stiffness coefficient, which characterizes the axial constraint effect of the surrounding components on the target reinforced concrete beam. Determine if historical crack images exist in the target reinforced concrete beam: If present, the inter-period damage increment features are determined based on the crack image under the current state and the historical crack image. The static damage feature parameters, the design feature parameters, and the inter-period damage increment features are then input into a pre-established prediction model to output the current bearing capacity utilization rate of the target reinforced concrete beam. If it does not exist, the static damage characteristic parameters and the design characteristic parameters are input into the prediction model, and the bearing capacity utilization rate is output. Based on the load-bearing capacity utilization rate, the safety level of the target reinforced concrete beam is determined according to a preset grading threshold, and a treatment suggestion scheme corresponding to the safety level is output. The bearing capacity utilization rate is the ratio of the current actual load to the ultimate peak load of the target reinforced concrete beam. The step of obtaining the design characteristic parameters of the target reinforced concrete beam, wherein the design characteristic parameters include at least the axial constraint stiffness coefficient used to characterize the axial constraint effect of the surrounding members on the target reinforced concrete beam, specifically includes: The target reinforced concrete beam is placed in an overall structural model that includes surrounding beam-column nodes for loading; Measure the axial force and axial displacement data at the beam end during the loading process, and determine the equivalent axial constraint stiffness based on the ratio of the axial force to the axial displacement. The axial constraint stiffness coefficient is determined based on the equivalent axial constraint stiffness, beam width, and axial compressive strength of concrete. The axial constraint stiffness coefficient characterizes the degree to which the passive axial pressure generated by the constraint of the surrounding floor slabs and vertical members inhibits the development of cracks in the beam. The step of determining the inter-period damage increment characteristics based on the crack image under the current state and the historical crack image, and inputting the static damage feature parameters, the design feature parameters, and the inter-period damage increment characteristics into a pre-established prediction model to output the current bearing capacity utilization rate of the target reinforced concrete beam, specifically includes: The historical crack images and the crack images in the current state are subjected to viewpoint correction and pixel-level registration to eliminate image deviations caused by differences in shooting viewpoint, distance and lighting conditions. Based on the registered images, the change of the same damage feature parameter within the two detection intervals is calculated to form the inter-period damage increment feature. The inter-period damage increment feature includes at least the crack width increment, crack length increment, crack penetration ratio increment, number of new cracks, and concrete crushing area increment. The degree of damage in the current period, represented by the static damage feature parameters, and the magnitude of damage change, represented by the inter-period damage increment features, are jointly inferred through the prediction model. The prediction model is used to take the bearing capacity utilization rate obtained based on the static damage characteristics as the benchmark value, use the damage change amplitude corresponding to the inter-period damage increment characteristics to quantify and offset the benchmark value, and perform convergence calibration on the prediction confidence interval to output the corrected bearing capacity utilization rate and prediction confidence. The prediction model is pre-established in the following manner: Obtain a time-series image set of a reinforced concrete beam with axial constraints from the initial loading to failure during the stepwise loading process; Static damage characteristic parameters at each load time are extracted from the time series image set, and the inter-period damage increment samples between adjacent time times are calculated. The design characteristic parameters of each beam, the static damage characteristic parameters at each load moment, the incremental damage samples across periods, and the bearing capacity utilization rate at the corresponding moment are correlated to construct a damage characteristic sample dataset. The prediction model is obtained by training a machine learning algorithm using the design feature parameters, static damage features, and inter-period damage increment features as inputs and the bearing capacity utilization rate as output.

2. The method for evaluating the load-bearing capacity utilization rate and safety rating of existing concrete beams according to claim 1, characterized in that, The step of processing the crack image and extracting static damage feature parameters specifically includes: The crack image is preprocessed, including grayscale conversion, filtering and noise reduction, and contrast enhancement. Pixel-level semantic segmentation is performed on the preprocessed image to identify crack areas and concrete crushing areas; Crack feature parameters are extracted from the identified crack regions, including crack width feature values, crack length feature values, crack spacing feature values, and crack penetration ratio feature values; and The crushing feature parameters are extracted from the identified concrete crushing areas, and the crushing feature parameters include the characteristic value of the concrete crushing area.

3. The method for evaluating the load-bearing capacity utilization rate and safety rating of existing concrete beams according to claim 1, characterized in that, The prediction model is also used for: Obtain the intermediate feature map and corresponding attention weights generated by the prediction model during the inference process of the crack image in the current state; Based on the attention weights, the contribution weight of each image region to the predicted load capacity utilization rate is calculated. The contribution weights are mapped back to the spatial coordinates of the crack image in the current state to generate a damage contribution heatmap. In the damage contribution heatmap, the areas marked with a preset warm color represent the damage areas whose contribution weights to the bearing capacity utilization prediction results meet preset conditions.

4. The method for evaluating the load-bearing capacity utilization rate and safety rating of existing concrete beams according to claim 1, characterized in that, The step of determining the safety level of the target reinforced concrete beam according to the bearing capacity utilization rate and a preset grading threshold, and outputting a treatment suggestion plan corresponding to the safety level, specifically includes: When the load-bearing capacity utilization rate is in the first value range, a suggestion for routine inspection without maintenance is output. When the load-bearing capacity utilization rate is in the second numerical range, a suggestion is output to periodically detect and seal cracks whose width exceeds the normal use limit. When the load-bearing capacity utilization rate is in the third value range, a recommendation is given to perform crack grouting or carbon fiber reinforcement within a specified period and to limit the load. When the load-bearing capacity utilization rate is in the fourth value range, the system will immediately recommend setting up temporary support, limiting or stopping use, and initiating replacement or permanent reinforcement solutions within a specified time limit.

5. A device for evaluating the load-bearing capacity utilization rate and safety rating of existing concrete beams, used to implement the method for evaluating the load-bearing capacity utilization rate and safety rating of existing concrete beams as described in any one of claims 1 to 4, characterized in that, include: The image acquisition module is used to acquire crack images of the target reinforced concrete beam in its current state. The image processing module is used to process the crack image and extract static damage feature parameters; The parameter acquisition module is used to acquire the design characteristic parameters of the target reinforced concrete beam. The design characteristic parameters include at least the axial constraint stiffness coefficient, which characterizes the axial constraint effect of the surrounding components on the target reinforced concrete beam. The image determination module is used to determine whether historical crack images of the target reinforced concrete beam exist. The first calculation module is used to determine the inter-period damage increment features based on the crack image in the current state and the historical crack image if they exist, and input the static damage feature parameters, the design feature parameters and the inter-period damage increment features into a pre-established prediction model to output the current bearing capacity utilization rate of the target reinforced concrete beam. The second calculation module is used to input the static damage characteristic parameters and the design characteristic parameters into the prediction model if the static damage characteristic parameters do not exist, and output the bearing capacity utilization rate. The result output module is used to determine the safety level of the target reinforced concrete beam according to the bearing capacity utilization rate and a preset grading threshold, and output a treatment suggestion scheme corresponding to the safety level. The bearing capacity utilization rate is the ratio of the current actual load to the ultimate peak load of the target reinforced concrete beam.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method for evaluating the load-bearing capacity utilization and safety rating of existing concrete beams as described in any one of claims 1 to 4.

7. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the method for evaluating the load-bearing capacity utilization and safety rating of existing concrete beams as described in any one of claims 1 to 4.

Images