Bridge structure hidden damage identification method
By monitoring and analyzing corrosion images and deformation data of bridge structures, a corrosion prediction model was constructed to simulate the impact of corrosion on bonding strength and identify potential hazard areas. This solved the problem of real-time accuracy in bridge corrosion monitoring, improved bridge safety, and reduced maintenance costs.
Patent Information
- Application Number
- CN202411058470.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-02
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2044-08-02
AI Technical Summary
Existing technologies make it difficult to identify the type and trend of corrosion in bridge structures in real time and accurately, which affects bridge safety and maintenance costs.
By monitoring corrosion images, structural deformation, and environmental data of bridge structures, a corrosion prediction model is constructed to simulate the relationship between corrosion degree and adhesion. The impact of corrosion on adhesion is evaluated by combining finite element analysis, identifying potential hazard areas and formulating maintenance strategies.
It has achieved improved bridge structural safety and reduced maintenance costs. Through precise monitoring and prediction, it has enhanced the safety and lifespan of bridges.
Smart Images

Figure CN119043606A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to a bridge structure damage identification technology, and in particular to a bridge structure hidden damage identification method. Background Art
[0002] In the field of bridge maintenance and monitoring, ensuring structural safety is of vital importance. The corrosion of steel bar surface and its influence on bond strength are key technical difficulties in bridge health monitoring. Since corrosion is not a homogeneous process, the influence of its morphological changes on bond strength is complex and nonlinear, and this change varies with time and environmental factors. It is a challenge to analyze the correlation between steel bar corrosion and bond strength. It is necessary to cover the corrosion morphology on the microscopic scale of the steel bar surface and the bond strength measurement on the macroscopic scale, covering the comprehensive process of corrosion conditions at different stages and their influence on bond strength. In order to ensure the generalization ability of the algorithm, data collection needs to involve different types of bridges, corrosion changes in different environments, and the influence of different material properties. Even with comprehensive data support, the unique conditions of different bridges may still limit the applicability of the algorithm in specific situations, such as special use history, unique structural design or the influence of specific environmental factors. Only by deeply understanding and accurately simulating the mechanism of the influence of corrosion on bond strength can the progress of bridge health monitoring technology be promoted. Summary of the invention
[0003] In order to solve the problems existing in the above-mentioned prior art, the purpose of the present invention is to provide a method for identifying hidden damage of bridge structures. The method for identifying hidden damage of bridge structures solves the problem that traditional bridge corrosion monitoring methods are difficult to accurately identify the type of corrosion and its development trend in real time; through precise monitoring and prediction, the safety and life of the bridge structure are effectively improved, and the maintenance cost is significantly reduced.
[0004] A bridge structure hidden damage identification method according to the present invention comprises the following steps:
[0005] S1. Monitor and collect corrosion image data, structural deformation data, dynamic response data and environmental condition data of bridge structures;
[0006] S2, analyzing the collected steel bar surface corrosion images to identify the type, distribution characteristics and corresponding current corrosion degree of the corrosion;
[0007] S3. Construct a corrosion prediction model based on the corrosion degree under multiple time series;
[0008] S4. Conducting simulation tests on the bonding strength between steel bars and concrete with different preset corrosion degrees, obtaining multiple sets of correlation data between the preset corrosion degrees and the bonding strength, and constructing a correlation analysis model between the steel bar surface corrosion degree and the bonding strength;
[0009] S5, comparing and analyzing the corrosion prediction model with current monitoring data of different bridge structures, evaluating the accuracy of the corrosion prediction model under current environmental conditions, and adjusting the corrosion prediction model;
[0010] S6. Compare and analyze the correlation analysis model with current monitoring data of different bridge structures, evaluate the accuracy of the correlation analysis model under the current corrosion degree, and adjust the corrosion prediction model;
[0011] S7. Based on the corrosion type, distribution characteristics and degree of corrosion, as well as the comparative analysis results of the corrosion prediction model and the current monitoring data, identify the potential risk areas affected by corrosion, and analyze the impact of passing vehicles on these potential risk areas;
[0012] S8. Evaluate the effect of corrosion on adhesion in potential risk areas based on the stress characteristics of the bridge structure. Determine the degree of weakening of adhesion caused by different degrees of corrosion based on the distance from the area to the stress point, material properties and environmental conditions. Analyze historical data and the characteristics of passing vehicles, set a threshold for the impact of adhesion, and classify the affected areas based on this threshold.
[0013] Preferably, the step S1 specifically includes:
[0014] Use image sensors to obtain high-definition image data of the bridge structure surface, segment and extract features from the image, automatically identify and locate the damaged area, and calculate the corrosion area ratio and crack length as the evaluation basis for corrosion monitoring;
[0015] Use strain sensors and accelerometers to collect the strain and vibration acceleration of the bridge structure in real time. According to the change trend and amplitude of the strain and acceleration data, combined with the preset judgment threshold, determine whether the bridge structure has deformed or vibrated abnormally, and obtain monitoring data on the deformation and dynamic response of the bridge structure;
[0016] Temperature, humidity and corrosion sensors are used to monitor the temperature, humidity and corrosive gas concentration of the bridge environment in real time. According to the correlation between environmental parameters and the characteristics of bridge structure materials, a quantitative evaluation model is established to determine the impact of current environmental conditions on the durability of the bridge structure and obtain environmental condition monitoring results.
[0017] Preferably, the step S2 specifically includes:
[0018] Based on the high-definition image data collected from the steel bar surface, image preprocessing technology is used to denoise, enhance and standardize the image to improve the image quality and the recognizability of the steel bar surface corrosion features;
[0019] A semantic segmentation algorithm based on deep learning is used to perform pixel-level segmentation on the rust image, accurately extract the location and boundary information of the rust area, and obtain a binary mask map of the rust area;
[0020] According to the mask map of the corrosion area, the ratio of the corrosion area to the total surface area of the steel bar is calculated as an important indicator to measure the degree of corrosion;
[0021] By extracting the grayscale histogram features of the rusted area, the grayscale distribution of the rusted area is obtained;
[0022] By calculating the local binary pattern features of the rust area, the microstructure and regularity of the rust texture are captured, and these feature parameters are combined into a multidimensional feature vector as the input for subsequent rust type discrimination and degree assessment.
[0023] Preferably, the step S3 specifically includes:
[0024] Obtain historical inspection data of steel structure, extract the corrosion area and maximum corrosion depth at different time nodes, and construct a time series data set of corrosion degree indexed by time;
[0025] Perform a stationarity test on the time series of the degree of corrosion. If it is not stationary, perform differential processing until it passes the stationarity test.
[0026] Adopting the STL decomposition method and based on LOESS fitting, the time series is decomposed into trend term, seasonal term and random term. According to the decomposition results, linear regression or polynomial regression modeling is adopted for trend term, Fourier series or periodic function modeling is adopted for seasonal term, and ARMA modeling is adopted for random term.
[0027] Select the time series prediction model, and determine the hyperparameters of the model based on the stationarity, autocorrelation and trend characteristics of the corrosion degree time series, combined with the ACF and PACF graphs;
[0028] The goodness of fit and predictive performance of different models were evaluated by cross-validation and information criteria, and the optimal model was selected;
[0029] Environmental factors are introduced as covariates and fused with the corrosion degree time series to construct a corrosion prediction model for multivariate time series.
[0030] Preferably, the step S4 specifically includes:
[0031] According to the preset corrosion degree level, the steel bar specimens are artificially corroded by using the electrochemical accelerated corrosion method;
[0032] By controlling the corrosion current density and corrosion time, the degree of corrosion can be regulated;
[0033] During the corrosion process, the mass loss rate and volume expansion rate of the steel bars are regularly monitored to ensure that the preset corrosion degree is reached;
[0034] Simulate the uneven corrosion in the actual environment, prepare steel bar specimens with different corrosion degrees, measure the corrosion area ratio and corrosion depth on the surface, and quantitatively characterize the corrosion degree;
[0035] The steel bar specimens after corrosion treatment were cast with concrete into standard specimens, and the concrete mix ratio and curing conditions similar to the actual structure were adopted to ensure the representativeness of the specimens;
[0036] A control group of uncorroded steel bars and concrete specimens was set up to evaluate the effect of corrosion on the bond strength;
[0037] The pull-out test was carried out on reinforced concrete specimens with different corrosion degrees. The load-displacement curve and steel bar strain distribution during the pull-out process were continuously recorded by displacement sensors and strain gauge testing equipment to obtain the mechanical parameters reflecting the bonding force.
[0038] According to the mechanical parameters measured by the pull-out test, the bond force distribution of the reinforced concrete interface under different corrosion degrees is calculated;
[0039] Combined with the volume expansion effect, loose porous characteristics and cross-sectional loss of the corrosion products, the influence of the corrosion products on the bond transfer mechanism is analyzed to identify the key factors that lead to bond degradation due to corrosion.
[0040] These characteristics of the corrosion products were quantitatively evaluated through cross-sectional analysis and microscopic morphology characterization of the drawn specimens;
[0041] Statistical analysis was performed on the bond strength and interfacial stress distribution under different corrosion levels to extract quantitative indicators of bond strength;
[0042] The correlation analysis model between the degree of steel bar surface corrosion and bonding strength is established using multiple linear regression or nonlinear regression methods.
[0043] Preferably, the step S5 specifically includes:
[0044] Extract real-time monitoring data of different bridges, record the area type, traffic flow and load type information of the bridge, and build a bridge corrosion monitoring database;
[0045] The current corrosion state parameters of each bridge in the monitoring database are input into the constructed corrosion prediction model to obtain the predicted value of the corrosion degree of each bridge at the current moment, and compared with the corrosion degree estimated by electrochemical parameters to calculate the prediction error and relative error;
[0046] The prediction errors of each bridge were statistically analyzed, the mean and standard deviation of the prediction errors were calculated, and the distribution histogram and box plot of the prediction errors were drawn to intuitively evaluate the overall prediction accuracy and stability of the corrosion prediction model on different bridges.
[0047] Preferably, the step S6 specifically includes:
[0048] Extract real-time bond monitoring data, including the ultimate bond strength between steel bars and concrete obtained by the pull-out method and the steel bar strain distribution obtained by the steel bar strain method, and invert the bond stress distribution between steel bars and concrete based on the strain distribution;
[0049] For each monitoring section, at least three measuring points are arranged, and at least three specimens are tested in parallel at each measuring point, and the average value is obtained as the representative value of the section;
[0050] The monitoring frequency is determined based on the importance and deterioration rate of the bridge, and the monitoring data are normalized to form a bond data set across bridges and structure types;
[0051] The current bonding force monitoring data is input into the established correlation analysis model between the steel bar surface corrosion degree and bonding force. The model predicts the corresponding theoretical bonding force value according to the current corrosion area ratio and corrosion depth parameters as the bonding force prediction value under the current corrosion degree.
[0052] For each bridge structure, the predicted bond strength value is compared with the measured value, and the difference between the two is calculated as the prediction deviation. The magnitude of the prediction deviation is quantified using indicators such as absolute value and relative percentage.
[0053] Evaluate the accuracy of the correlation analysis model under the current corrosion level and determine whether the prediction accuracy of the model meets the actual application requirements.
[0054] Preferably, the step S7 specifically includes:
[0055] According to the corrosion type identification results, the corrosion of bridge reinforcement is divided into three types: uniform corrosion, local corrosion and pitting corrosion.
[0056] Combining the morphological characteristics, spatial distribution characteristics and causal mechanism of rust, a set of rust type classification standards with multi-index comprehensive evaluation is formulated. The weight of each factor is determined by the hierarchical analysis method. The quantitative index of the rust type is obtained by weighted summation, and the rust type is then divided according to the size of the index value.
[0057] Combined with the analysis of corrosion distribution characteristics, the distribution probability and spatial correlation of various types of corrosion in different parts of the bridge are determined to form a probability model of bridge corrosion type and distribution;
[0058] The corrosion degree assessment method was used to calculate the average corrosion area ratio and maximum corrosion depth of the steel bars in various parts of the bridge. Combined with the corrosion type and distribution probability model, a comprehensive assessment of the overall and local corrosion degree of the bridge was conducted to obtain the spatial distribution map of the bridge corrosion degree.
[0059] Preferably, the step S8 specifically includes:
[0060] According to the stress characteristics of the bridge structure, select the appropriate finite element analysis software, establish a three-dimensional solid model of the bridge structure, and divide the grid reasonably;
[0061] In the bridge corrosion risk area, the area is divided into several sub-areas according to the distribution of steel bar corrosion. The degree of corrosion in each sub-area is approximately uniform. For each sub-area, the equivalent corroded steel bar material parameters are generated by reducing the cross-sectional area and equivalent yield strength of the steel bar according to the cross-sectional loss rate of the corroded steel bar.
[0062] Using parametric modeling technology, the influence area is preset in the geometric model or mesh model. By changing the distance and size of the influence area, different finite element models are automatically generated to simulate the spatial variability of the corrosion area and evaluate the influence of location factors.
[0063] The modified CEB-FIPModel Code 2010 constitutive relationship is used to study the bond-slip behavior of corroded steel bars and concrete.
[0064] When modifying the bond-slip constitutive model, dimensionless parameters reflecting the degree of corrosion are introduced, and quantitative correlations between each parameter and the corrosion parameter are established, and calibration is performed through pull-out test.
[0065] The modified corrosion degradation-bond-slip constitutive model was embedded in the finite element model to evaluate the degree of weakening of the overall bond strength of the bridge due to different corrosion degrees.
[0066] Preferably, the step S8 further includes:
[0067] For the hazardous areas that exceed the bond strength threshold, the finite element analysis method is used to conduct stress analysis and hazard assessment, determine its impact on the overall safety of the bridge structure, calculate the maintenance and repair scope, and formulate corresponding maintenance and repair strategies;
[0068] For non-hazardous areas that do not exceed the bond strength threshold, the long-term benefits after replacement or repair are evaluated and maintenance and repair strategies are formulated based on the structural characteristics of the bridge and monitoring data, combined with cost-benefit analysis.
[0069] The bridge structure hidden damage identification method described in the present invention has the advantages of:
[0070] A bridge structure hidden damage identification method of the present invention solves the problem that the traditional bridge corrosion monitoring method is difficult to identify the corrosion type and its development trend in real time and accurately. Through comprehensive monitoring of the bridge structure, the collected corrosion images are analyzed by image processing and machine learning models, the corrosion type, distribution characteristics and corrosion degree of the steel bar surface are identified, and a corrosion prediction model is constructed through time series data to predict future corrosion conditions; through simulation tests, the correlation data of the corrosion degree and the bond between the steel bar and concrete are obtained, and a correlation analysis model is constructed, and the accuracy of the corrosion prediction model is evaluated and adjusted by comparison and analysis with the current monitoring data; according to the corrosion type, distribution characteristics and corrosion degree, combined with the comparison results of the corrosion prediction model and the current monitoring data, the potential hidden danger areas affected by corrosion are identified, and the impact of traffic on these areas is analyzed; the impact of corrosion on the bond is evaluated by finite element analysis, the bond threshold is set, the stress analysis and hazard assessment are performed on the area exceeding the threshold, and the maintenance and repair strategy is formulated; through accurate monitoring and prediction, the safety and life of the bridge structure are effectively improved, and the maintenance cost is significantly reduced. BRIEF DESCRIPTION OF THE DRAWINGS
[0071] Figure 1 It is a flow chart of a method for identifying hidden damage of a bridge structure described in the present invention. DETAILED DESCRIPTION
[0072] like Figure 1 As shown, a bridge structure hidden damage identification method according to the present invention comprises the following steps:
[0073] S1. Monitor and collect corrosion image data, structural deformation data, dynamic response data and environmental condition data of the bridge structure; specifically, use image sensors to monitor corrosion based on image data, strain sensors and accelerometers to monitor structural deformation and dynamic response, and temperature, humidity and corrosion sensors to monitor environmental conditions;
[0074] S2, analyzing the collected steel bar surface corrosion images to identify the type, distribution characteristics and corresponding current corrosion degree of the corrosion;
[0075] S3. Construct a corrosion prediction model based on the corrosion degree under multiple time series;
[0076] S4. Conducting simulation tests on the bonding strength between steel bars and concrete with different preset corrosion degrees, obtaining multiple sets of correlation data between the preset corrosion degrees and the bonding strength, and constructing a correlation analysis model between the steel bar surface corrosion degree and the bonding strength;
[0077] S5, comparing and analyzing the corrosion prediction model with current monitoring data of different bridge structures, evaluating the accuracy of the corrosion prediction model under current environmental conditions, and adjusting the corrosion prediction model;
[0078] S6. Compare and analyze the correlation analysis model with current monitoring data of different bridge structures, evaluate the accuracy of the correlation analysis model under the current corrosion degree, and adjust the corrosion prediction model;
[0079] S7. Based on the corrosion type, distribution characteristics and degree of corrosion, as well as the comparative analysis results of the corrosion prediction model and the current monitoring data, identify the potential risk areas affected by corrosion, and analyze the impact of passing vehicles on these potential risk areas; analyze the weight, speed and frequency of passing vehicles;
[0080] S8. Evaluate the effect of corrosion on adhesion in potential risk areas based on the stress characteristics of the bridge structure. Determine the degree of weakening of adhesion caused by different degrees of corrosion based on the distance from the area to the stress point, material properties and environmental conditions. Analyze historical data and the characteristics of passing vehicles, set a threshold for the impact of adhesion, and classify the affected areas based on this threshold.
[0081] Furthermore, in this embodiment, step S1 specifically includes:
[0082] Use image sensors to obtain high-definition image data of the bridge structure surface, segment and extract features from the image, automatically identify and locate the damaged area, and calculate the corrosion area ratio and crack length as the evaluation basis for corrosion monitoring; use YOLO or Faster R-CNN image processing algorithms to segment and extract features from the image, focusing on the corrosion and crack damage characteristics;
[0083] Use strain sensors and accelerometers to collect the strain and vibration acceleration of the bridge structure in real time. According to the change trend and amplitude of the strain and acceleration data, combined with the preset judgment threshold, determine whether the bridge structure has deformed or vibrated abnormally, and obtain monitoring data on the deformation and dynamic response of the bridge structure;
[0084] Temperature, humidity and corrosion sensors are used to monitor the temperature, humidity and corrosive gas concentration of the bridge environment in real time. According to the correlation between environmental parameters and the characteristics of bridge structure materials, a quantitative evaluation model is established to determine the impact of current environmental conditions on the durability of the bridge structure and obtain environmental condition monitoring results.
[0085] Furthermore, based on the results of corrosion monitoring, structural deformation monitoring, dynamic response monitoring and environmental condition monitoring, a bridge structure health assessment model was established, and the analytic hierarchy process was used to assign weights to each monitoring indicator, and the comprehensive health index was calculated by weighted summation;
[0086] Based on the comparison between the comprehensive health index and the preset threshold, the health status of the bridge structure is divided into normal, slightly damaged, moderately damaged and severely damaged levels;
[0087] Evaluate the safety risk level based on the importance and service life of the bridge structure;
[0088] Develop corresponding maintenance strategies based on health status and risk level, such as whether local repair, reinforcement, or suspension of use is needed, and dynamically adjust monitoring frequency, such as reducing sampling intervals and increasing sampling points;
[0089] In the early warning mechanism, set multiple levels of warning thresholds. For example, a yellow warning is triggered when the comprehensive health index is lower than 0.7, an orange warning is triggered when it is lower than 0.5, and a red warning is triggered when it is lower than 0.3;
[0090] Warning information includes bridge number, location, health status, risk level, and recommended measures, which will be promptly notified to relevant managers via SMS, email, and APP push;
[0091] Automatically adjust the monitoring frequency according to the warning level. For example, the sampling frequency is increased by 50% during yellow warning, 100% during orange warning, and continuous high-frequency monitoring during red warning.
[0092] The historical monitoring data was mined and analyzed through big data analysis technology, and the long short-term memory neural network (LSTM) model was selected. Time series data including strain, acceleration, temperature, humidity, etc. were input to train the LSTM model and learn the time law of bridge structure performance degradation.
[0093] During the model training process, early stopping and regularization methods are used to prevent overfitting. The trained LSTM model is used to make rolling predictions on the health status of bridge structures in the future (such as 1 month or 1 year), and the maintenance plan is dynamically updated based on the prediction results.
[0094] Use 3D visualization technology to intuitively display monitoring data and analysis results. Based on the BIM model, associate the location of monitoring sensors, real-time data and 3D models to achieve real-time mapping of monitoring data.
[0095] According to the abnormal degree of the monitoring parameters, different color warning levels are set, such as green for normal, yellow for mild abnormality, orange for moderate abnormality, and red for severe abnormality;
[0096] Through the visualization of 3D models, bridge managers can intuitively understand the overall and local health status of the bridge. The 3D model supports interactive operations such as rotation, scaling, and slicing, which facilitates managers to conduct multi-angle and multi-level observation and analysis, and improves the readability and usability of monitoring data.
[0097] Here is an example:
[0098] When applying the YOLO algorithm to identify bridge surface damage, the confidence threshold of damage detection can be set to 0.8, that is, when the confidence of the recognition result is greater than or equal to 0.8, it is considered that there is damage in the area;
[0099] The identified cracks can be divided into mild cracks (less than 10 cm), moderate cracks (10-30 cm) and severe cracks (greater than 30 cm) according to their length;
[0100] When judging the deformation of the bridge structure, the strain threshold can be set to ±1000με and the acceleration threshold to ±0.2g. When the monitoring data exceeds the threshold range, it is judged as abnormal deformation or vibration;
[0101] By establishing a mathematical model of environmental parameters and the degradation rate of bridge structural material performance, for example, for every 10% increase in humidity, the steel corrosion rate increases by 0.05mm / a, the impact of environmental conditions on bridge durability is quantitatively evaluated;
[0102] When calculating the comprehensive health index, the weights of corrosion monitoring, structural deformation monitoring, dynamic response monitoring, and environmental monitoring can be assigned as 0.3, 0.3, 0.2, and 0.2, respectively, which are determined by the AHP method;
[0103] When the warning level is yellow, the monitoring frequency will be increased from once per hour to once every 30 minutes. When the warning level is orange, it will be increased to once every 10 minutes. When the warning level is red, real-time monitoring will be implemented.
[0104] When training the LSTM prediction model, you can select the monitoring data of the past year as input and the health index of the next month as output. Use the gradient descent algorithm to optimize the model parameters, set the number of early stop rounds to 50, and the regularization coefficient to 0.01 to prevent the model from overfitting.
[0105] In 3D visualization, different color mapping rules can be set according to the abnormal degree of the monitoring parameters. For example, if the strain exceeds the normal value by 10%, it will be displayed in yellow; if it exceeds 10%-20%, it will be displayed in orange; if it exceeds 20%, it will be displayed in red;
[0106] Through interactive operations, the local area of the bridge can be enlarged, rotated and sliced for analysis. For example, you can zoom in on the damaged area, rotate to the direction of the maximum crack opening, and slice to analyze the stress distribution inside the beam and column.
[0107] Furthermore, in this embodiment, step S2 specifically includes:
[0108] According to the collected high-definition image data of the steel bar surface, the image preprocessing technology is used to remove noise, enhance and standardize the image, so as to improve the image quality and the recognizability of the corrosion characteristics of the steel bar surface; the high-definition image data of the steel bar surface is collected by an image sensor;
[0109] Use a semantic segmentation algorithm based on deep learning to perform pixel-level segmentation on the rust image, accurately extract the location and boundary information of the rust area, and obtain a binary mask map of the rust area; the semantic segmentation algorithm based on deep learning can be selected from U-Net or DeepLab;
[0110] According to the mask map of the corrosion area, the ratio of the corrosion area to the total surface area of the steel bar is calculated as an important indicator to measure the degree of corrosion;
[0111] By extracting the grayscale histogram features of the rusted area, the grayscale distribution of the rusted area is obtained; the grayscale histogram features include mean, standard deviation, skewness, and kurtosis;
[0112] By calculating the local binary pattern features of the rusted area, the microstructure and regularity of the rust texture are captured, and these characteristic parameters are combined into a multidimensional feature vector as the input for subsequent rust type discrimination and degree assessment;
[0113] In order to train a multi-classification machine learning model, a dataset of labeled rust images is required. Each image must be labeled with the category label of the rust type (e.g. uniform rust, pitting, spotting).
[0114] Using cross-validation, the data set is divided into training set, validation set and test set, and the ratio can be set to 7:2:1;
[0115] Select machine learning algorithms suitable for multi-classification tasks, such as logistic regression, support vector machine, random forest, and train and optimize the models;
[0116] Adjust the model's hyperparameters, such as the regularization coefficient and the number of trees, through the grid search method to improve the classification accuracy;
[0117] Evaluate the performance of the model on the test set, calculate the classification accuracy, precision, recall rate, and F1 score, and select the model with the best performance for the identification of corrosion types;
[0118] If the corrosion type is pitting or spotting, image morphological operations, such as opening and closing operations, are further used to perform morphological analysis on the corrosion area and calculate geometric characteristic parameters such as the number, size, and distribution density of the corrosion points.
[0119] Combining the discrimination results of the corrosion type with the morphological characteristics of the corrosion area, a comprehensive corrosion evaluation index system is constructed. When establishing a mapping model between corrosion characteristic parameters and corrosion degree, a decision tree regression algorithm can be used.
[0120] The decision tree recursively divides the feature space into different regions, each of which corresponds to a predicted value of the degree of corrosion. The training data set can be used to train the decision tree model by taking the proportion of the corrosion area, the average gray value, the texture characteristics, the corrosion type, etc. as input features and the corresponding corrosion level (such as mild, moderate, severe) as the target value.
[0121] During the training process, the complexity and generalization ability of the tree are controlled by setting parameters such as the maximum depth of the tree and the minimum number of samples of leaf nodes;
[0122] The mean square error is used as the evaluation index, and the decision tree with the smallest mean square error is selected as the optimal model;
[0123] When applied, the feature parameters of the new corrosion image are input into the decision tree to obtain the corresponding corrosion degree prediction value;
[0124] Here is an example:
[0125] When using the U-Net algorithm for rust area segmentation, the input image size can be set to 512x512, the number of convolution layers of the encoder and decoder can be set to 4 and 4 respectively, the convolution kernel size can be set to 3x3, and the activation function can be set to ReLU;
[0126] The model was trained using the cross entropy loss function and the Adam optimizer, with the learning rate set to 0.001, the batch size to 8, and the number of iterations to 100;
[0127] When extracting the LBP features of the rusted area, 8 sampling points and an LBP operator with a radius of 1 can be selected, and the LBP feature map can be divided into 8x8 grids. The LBP histogram in each grid is calculated, and the histograms are connected into a feature vector with a length of 8x8x256=16384 dimensions.
[0128] For the multi-classification model of rust type, we can select the support vector machine algorithm, use the radial basis kernel function, select the optimal penalty coefficient C and kernel function parameter gamma through 5-fold cross validation, evaluate the model performance on the test set, and the classification accuracy reaches more than 90%;
[0129] When constructing a decision tree model for corrosion assessment, the CART algorithm was used to optimize the maximum depth of the decision tree, the minimum number of split samples and other parameters through grid search, and the mean square error was used as the evaluation index to select the optimal decision tree model.
[0130] Furthermore, the corrosion types on the steel bar surface are identified, including uniform corrosion, local corrosion and pitting corrosion, the depth and area of corrosion are evaluated, the distribution characteristics of corrosion on the steel bar surface are analyzed, the concentrated area and diffusion trend of corrosion are determined, and the current degree of corrosion is quantified;
[0131] Use deep learning-based image segmentation algorithms, such as U-Net or MaskR-CNN, to perform semantic segmentation on the steel bar surface corrosion image to obtain an accurate mask map of the corrosion area, thereby determining the location and range of the corrosion;
[0132] According to the mask image of the corroded area obtained by segmentation, the ratio of the corroded area to the total surface area of the steel bar is calculated to obtain the quantitative evaluation result of the corroded area;
[0133] By establishing a calibration model of grayscale value and corrosion depth, the average depth and maximum depth of the corrosion area are estimated. The texture, color and shape of the corrosion area are extracted;
[0134] For texture features, the gray-level co-occurrence matrix algorithm is used to calculate the gray-level co-occurrence probability matrix at different directions and distances, and to extract texture indicators such as energy, entropy, contrast, and homogeneity;
[0135] For color features, select the HSV color space and calculate the histograms of the three channels H, S, and V. For shape features, calculate the geometric parameters of the rusted area, such as area, perimeter, eccentricity, and convex hull area;
[0136] A high-dimensional feature vector is constructed by integrating multiple features, and machine learning algorithms such as support vector machines or random forests are used for training and optimization to achieve automatic classification of corrosion types, including uniform corrosion, local corrosion, and pitting corrosion;
[0137] Select the optimal model parameters through grid search and cross-validation to improve classification accuracy and generalization ability;
[0138] The steel bar surface is divided into several grid areas, and the corrosion area ratio and average depth in each grid are counted. A corrosion distribution heat map is drawn through a reasonable color mapping scheme to intuitively present the spatial distribution characteristics of corrosion.
[0139] Use K-means or DBSCAN clustering algorithm to cluster the heat map and find the concentrated areas with the highest degree of corrosion, which can provide key focus for subsequent corrosion spread and development analysis;
[0140] At the boundary of the rust concentrated area, the Canny operator is used to perform edge detection and extract the location information of the boundary pixels;
[0141] The grayscale gradient direction and magnitude of the boundary pixels are calculated by Sobel operator, and the gradient direction is quantified into 8 discrete directions. The average gradient magnitude in each direction is counted as the rate of rust diffusion in that direction.
[0142] Comprehensively analyze the diffusion rate of the concentrated corrosion area in different directions, determine the diffusion trend of corrosion, simulate the changes in the corrosion area in the future, and provide a basis for predicting the development of corrosion. Consider multiple indicators such as the proportion of corrosion area, average depth, maximum depth, corrosion type and diffusion trend, and establish a quantitative model of corrosion degree;
[0143] The analytic hierarchy process (AHP) was used to rank the importance and assign weights to each indicator, construct a judgment matrix and conduct consistency tests;
[0144] The fuzzy comprehensive evaluation method (FCE) is used to map the membership function of each index to a unified evaluation scale, and the comprehensive membership of the corrosion degree is obtained through fuzzy operation to determine the corrosion degree level.
[0145] Introducing the time dimension into the corrosion degree assessment process, analyzing the changing trend of the corrosion degree in different time periods, and realizing dynamic corrosion process evaluation;
[0146] Track and analyze the corrosion images of the same steel bar collected at different time points, extract key indicators such as corrosion area ratio and average depth, and construct time series data;
[0147] Here is an example:
[0148] When using U-Net for rust area segmentation, the input image size can be set to 512x512, the encoder and decoder both contain 4 convolutional layers, the convolution kernel size is 3x3, and the number of feature maps is 64, 128, 256, and 512 respectively;
[0149] The Dice coefficient is used as the loss function, and the Adam optimizer is used for training. The learning rate is 0.001, the batch size is 8, and the iteration is 100 epochs. When extracting the GLCM texture features of the rusted area, the distance d = 1, the direction θ = 0°, 45°, 90°, 135°, the grayscale level is 16, and the four indicators of energy, entropy, contrast, and homogeneity are extracted;
[0150] SVM was used to classify the corrosion types, and the RBF kernel function was used. Through 5-fold cross validation, the penalty coefficient C = 10 and the kernel function parameter γ = 0.01 were selected. The classification accuracy on the test set reached 95%;
[0151] When drawing the corrosion distribution heat map, the steel bar surface is divided into 50x50 grids, and the corrosion area ratio in each grid is normalized and mapped to the JET pseudo-color color table. The higher the area ratio, the redder the color.
[0152] The K-means clustering algorithm was used to cluster the corrosion distribution heat map into three categories, corresponding to light, moderate and severe corrosion areas.
[0153] Furthermore, in this embodiment, step S3 specifically includes:
[0154] Obtain historical inspection data of steel structure, extract the corrosion area and maximum corrosion depth at different time nodes, and construct a time series data set of corrosion degree indexed by time;
[0155] Perform a stationarity test on the time series of the degree of corrosion. If it is not stationary, perform differential processing until it passes the stationarity test; if it is not stationary, perform differential processing until it passes the ADF test or KPSS test;
[0156] Adopting the STL decomposition method and based on LOESS fitting, the time series is decomposed into trend term, seasonal term and random term. According to the decomposition results, linear regression or polynomial regression modeling is adopted for trend term, Fourier series or periodic function modeling is adopted for seasonal term, and ARMA modeling is adopted for random term.
[0157] Select a time series prediction model, and determine the hyperparameters of the model based on the stationarity, autocorrelation and trend characteristics of the corrosion degree time series, combined with the ACF and PACF graphs; the time series prediction model can be ARIMA, SARIMA, Holt-Winters; the hyperparameters include the p, d, and q values of the ARIMA model;
[0158] The goodness of fit and predictive performance of different models were evaluated by cross-validation and information criteria, and the optimal model was selected;
[0159] Environmental factors are introduced as covariates, which are integrated with the corrosion degree time series to build a corrosion prediction model for multivariate time series.
[0160] Furthermore, the sliding window method is used to divide the corrosion degree time series into training set and test set, and cross-validation is performed. According to the length of the time series and the complexity of the prediction model, the appropriate training sample length (e.g., 1 year) and prediction step length (e.g., 1 month) are selected to improve the timeliness of the prediction while ensuring the stability of the model. The performance of the prediction model is evaluated by using indicators such as root mean square error (RMSE) and mean absolute percentage error (MAPE) to measure the deviation between the predicted value and the actual value;
[0161] Construct prediction intervals to quantify the uncertainty of the prediction results based on confidence levels (e.g., 95%) and error distribution assumptions. Update and tune the prediction model online, using a sliding window approach, fixed training sample length, and regular (e.g., monthly) updates to model parameters, while modifying exogenous variables in the ARIMAX or SARIMAX model to incorporate the latest environmental monitoring data;
[0162] Continuously monitor the prediction performance of the model, calculate the rolling prediction error, and trigger model retraining or model structure adjustment when the prediction error exceeds the preset threshold;
[0163] Compare the prediction results with the actual test data, calculate the residual sequence, and feed the residuals back to the model for dynamic correction, so as to continuously improve the prediction accuracy and reliability and adapt to the non-stationary characteristics of the steel bar corrosion process;
[0164] Here is an example:
[0165] When obtaining historical inspection data of steel structure, we can select the data of the last five years, conduct 12 inspections per year, and construct a monthly corrosion degree time series with a length of 60;
[0166] Perform an ADF test on the time series. If the P value is greater than 0.05, perform the ADF test again through the first-order difference operation until the P value is less than 0.05 to ensure that the time series is stable;
[0167] The STL decomposition method is used to extract the trend term, seasonal term and random term of the time series by setting the seasonal period to 12, the window span of the trend term to 5, and the cutoff frequency of the low-pass filter to 0.5.
[0168] Quadratic polynomial regression was used for trend terms, 6th-order Fourier series expansion was used for seasonal terms, and ARMA (1,1) model was used for random terms.
[0169] When selecting the time series forecasting model, the ARIMA(1,1,1), SARIMA(1,1,1)(1,1,1)_12, and Holt-Winters models were trained and evaluated respectively, and the model selection was performed using the AIC criterion. The results showed that the SARIMA model had the smallest AIC value, so SARIMA(1,1,1)(1,1,1)_12 was selected as the optimal model.
[0170] When introducing environmental factors, the monthly average temperature, monthly average relative humidity and monthly total rainfall related to the corrosion process were selected as covariates, and partial correlation analysis was performed between them and the corrosion degree time series. If the P value of the partial correlation coefficient was less than 0.05, the environmental factor was included in the SARIMAX model to construct the SARIMAX (1,1,1) (1,1,1)_12 model.
[0171] When dividing the training set and the test set, the training sample length is set to 48 and the prediction step length is set to 6. That is, the model is trained with the data of the first four years to predict the degree of corrosion in the second half of the year. The expanding window method is used for cross-validation to obtain a series of prediction error values.
[0172] For the model performance evaluation, the RMSE between the predicted value and the true value was calculated to be 0.02, and the MAPE was 3%, indicating that the prediction model has high accuracy;
[0173] The 95% prediction interval was constructed based on the bootstrap method, and the average interval width was 0.05, reflecting the degree of uncertainty in the prediction results;
[0174] When the model is updated online, whenever new test data arrives, it is added to the training set, and the earliest observation value is removed. The SARIMAX model is retrained and the model parameters are updated.
[0175] When the RMSE of the prediction error exceeds 0.05 for three consecutive months, the model structure adjustment is triggered, the difference order, autoregression order and moving average order are reselected, and new environmental factors are introduced to continuously optimize the prediction performance of the model. Through online updating and feedback correction, the prediction model can adapt to the dynamic changes of the steel bar corrosion process and improve the prediction accuracy and reliability.
[0176] Furthermore, in this embodiment, step S4 specifically includes:
[0177] According to the preset corrosion degree level, the steel bar specimens are artificially corroded by using the electrochemical accelerated corrosion method;
[0178] By controlling the corrosion current density and corrosion time, the degree of corrosion can be regulated; the corrosion current density is 0.1-1mA / cm 2 and corrosion time is 1-30 days;
[0179] During the corrosion process, the mass loss rate and volume expansion rate of the steel bars are regularly monitored to ensure that the preset corrosion degree is reached;
[0180] Simulate the uneven corrosion in the actual environment, prepare steel bar specimens with different corrosion degrees, measure the corrosion area ratio and corrosion depth on the surface, and quantitatively characterize the corrosion degree; simulate the uneven corrosion in the actual environment to pre-create some defects on the steel bar surface, such as scratches and pits, to induce local corrosion;
[0181] The steel bar specimens after corrosion treatment were cast with concrete into standard specimens, and the concrete mix ratio and curing conditions similar to the actual structure were adopted to ensure the representativeness of the specimens;
[0182] A control group of uncorroded steel bars and concrete specimens was set up to evaluate the effect of corrosion on the bond strength;
[0183] Conduct pull-out tests on reinforced concrete specimens with different degrees of corrosion. Use displacement sensors and strain gauge testing equipment to continuously record the load-displacement curve and steel strain distribution during the pull-out process to obtain mechanical parameters reflecting the bond strength. The mechanical parameters reflecting the bond strength include ultimate bond strength and yield bond strength.
[0184] According to the mechanical parameters measured by the pull-out test, the bond force distribution of the reinforced concrete interface under different corrosion degrees is calculated;
[0185] Combined with the volume expansion effect, loose porous characteristics and cross-sectional loss of the corrosion products, the influence of the corrosion products on the bond transfer mechanism is analyzed to identify the key factors that lead to bond degradation due to corrosion.
[0186] Through cross-sectional analysis and microscopic morphology characterization of the drawing specimens, these characteristics of the corrosion products can be quantitatively evaluated; quantitative evaluation of these characteristics of the corrosion products can provide a basis for revealing the corrosion damage mechanism;
[0187] Statistical analysis was performed on the bond strength and interfacial stress distribution under different corrosion levels to extract quantitative indicators of bond strength; quantitative indicators of bond strength include peak bond strength, residual bond strength, and uniformity of stress distribution.
[0188] Use multiple linear regression or nonlinear regression methods to establish a correlation analysis model between the degree of steel bar surface corrosion and bonding strength;
[0189] Taking multiple linear regression as an example, the proportion of rust area, average rust depth, and rust layer thickness were selected as independent variables, and the ultimate bond strength and residual bond strength were selected as dependent variables. The regression equation was fitted by the least squares method.
[0190] During the fitting process, outliers are detected and removed from the sample data to improve the robustness of the regression model;
[0191] Perform model diagnostics, such as residual analysis and multicollinearity tests, to ensure the statistical significance and predictive power of the model;
[0192] For complex nonlinear relationships, polynomial regression, exponential regression, logarithmic regression model forms, or machine learning algorithms such as support vector machines and neural networks can be used for nonlinear modeling;
[0193] Based on the above test data and the associated analysis model, numerical simulation methods, such as finite element analysis, are used to numerically simulate the bonding behavior of the reinforced concrete interface; the degradation of interface properties caused by corrosion, such as reduced stiffness and strength, is considered, and parametric characterization of corrosion damage is achieved in the numerical simulation;
[0194] By comparing the numerical simulation results with the pull-out test results, the rationality of the correlation model is verified, and the experimental phenomena are explained theoretically;
[0195] When applying the correlation analysis model to actual projects, it is necessary to obtain corrosion monitoring data of reinforced concrete structures. On the one hand, by using the structural health monitoring system, corrosion sensors such as linear polarization resistance sensors and electrochemical impedance sensors are deployed at key locations to monitor the corrosion state of the steel bars in real time and obtain data such as the corrosion area ratio and corrosion current density. On the other hand, non-destructive testing technologies such as ultrasonic testing and radar detection are used to conduct regular inspections of the structure to obtain spatial distribution information of steel bar corrosion.
[0196] Input monitoring data and test data into the correlation analysis model to predict the structural bond index, and consider the influence of environmental factors and load factors, calibrate and dynamically update the correlation analysis model. The prediction results can serve as an important basis for structural performance evaluation and maintenance and reinforcement decision-making, providing technical support for the safe operation of existing reinforced concrete structures.
[0197] Here is an example:
[0198] In the electrochemical accelerated corrosion process, HRB400 grade threaded steel bars with a diameter of 16 mm and a length of 500 mm were selected, 3.5% NaCl solution was used as the corrosion medium, and the corrosion current density was set to 0.5 mA / cm 2 The corrosion time is 5 days, and a lightly corroded specimen with a corrosion area ratio of 5% is prepared;
[0199] The corroded steel bars and C30 strength grade concrete were cast into cylindrical tensile specimens with a diameter of 150 mm and a height of 300 mm, and compared with the uncorroded steel bar specimens. The monotonic tensile load was applied by a universal testing machine, and the pulling rate was controlled at 0.1 kN / s. The load and displacement data were collected synchronously to draw the load-displacement curve;
[0200] The ultimate bond strength was identified by the mutation point on the curve, and the average ultimate bond strength of the lightly corroded specimens was calculated to be 5.2MPa, which was 15% lower than that of the uncorroded specimens. Scanning electron microscopy was performed on the cross section of the pull-out specimens, and it was found that a loose layer of corrosion products was attached to the surface of the steel bar, with a thickness of about 120μm. Based on the ultimate bond strength data of specimens with different degrees of corrosion, the least squares method was used to fit the linear regression equation of the corrosion area ratio x and the ultimate bond strength y: y = -0.35x + 6.8, with a correlation coefficient R 2 =0.92, indicating that the two are significantly negatively correlated;
[0201] In the finite element numerical simulation, the cohesive-zone model is used to describe the bond-slip behavior between steel bars and concrete. The bond structural parameters of the interface are set according to the proportion of the corrosion area. For example, under the 5% corrosion state, the bond strength of the interface is reduced by 20% and the fracture energy is reduced by 15%.
[0202] The pull-out load-displacement curve obtained by numerical simulation is in good agreement with the test results, which verifies the rationality of the bond-slip model.
[0203] This method was applied to reinforced concrete sheet piles in a port project. The instantaneous corrosion rate of the steel bars was monitored using embedded corrosion sensors, and the average corrosion area of the steel bars was found to be 3% through radar nondestructive testing. The ultimate bond strength of the sheet piles was predicted to decrease by 10% by substituting it into the correlation analysis model. Based on this, the bearing capacity of the sheet piles was verified and the safety was evaluated.
[0204] Furthermore, in this embodiment, step S5 specifically includes:
[0205] Extract the real-time monitoring data of different bridges, record the regional type, traffic flow and load type information of the bridge, and build a bridge corrosion monitoring database; specifically, extract the real-time monitoring data of different bridges, including key factors affecting corrosion, such as ambient temperature, humidity, chloride ion concentration, and electrochemical parameters of the steel surface, such as natural corrosion potential and polarization resistance; for ambient temperature and humidity, select high-precision sensors with a measurement range of -20℃60℃ and 0100%RH, a resolution of 0.1℃ and 0.1%RH, and collect data every 1 hour; for chloride ion concentration, select electrochemical sensors, such as Ag / AgCl reference electrodes, with a measurement range of 0.001~5mol / L, and collect data every 24 hours; for electrochemical parameters of steel bars, use linear polarization method or electrochemical impedance spectroscopy to measure natural corrosion potential and polarization resistance, and estimate corrosion current density according to the Stern-Geary formula, and collect data every 7 days;
[0206] The current corrosion state parameters of each bridge in the monitoring database are input into the constructed corrosion prediction model to obtain the predicted value of the corrosion degree of each bridge at the current moment, and compared with the corrosion degree estimated by electrochemical parameters, and the prediction error and relative error are calculated; the prediction error of each bridge is statistically analyzed, the mean and standard deviation of the prediction error are calculated, and the distribution histogram and box plot of the prediction error are drawn to intuitively evaluate the overall prediction accuracy and stability of the corrosion prediction model on different bridges;
[0207] Furthermore, the Pearson correlation coefficient, Spearman rank correlation coefficient, partial correlation analysis, and scatter plot matrix method were used to analyze the correlation between the prediction deviation and factors such as bridge environmental conditions, structural characteristics, and load conditions, quantitatively and qualitatively evaluate the degree of influence of each factor on the prediction deviation, and determine the applicable conditions and limitations of the corrosion prediction model in practical applications;
[0208] For bridges with large prediction deviations, historical monitoring data and existing inspection data are extracted to modify and optimize the internal structure and parameters of the corrosion prediction model.
[0209] Combined with the results of correlation analysis, the input features of the model are adjusted in a targeted manner, the weight coefficients of each feature are modified, and nonlinear terms are added to improve the model's expressiveness and fitting accuracy. The optimized corrosion prediction model is reapplied to the current monitoring data of each bridge. The monitoring data of different bridges are randomly divided into k mutually exclusive subsets using the k-fold cross-validation method. Each time, k-1 subsets are selected as training sets, and the remaining 1 subset is used as a test set. The experiment is repeated k times, and the average value of the k experimental results is taken as the performance indicator of the optimized model.
[0210] The prediction accuracy improvement effect before and after model optimization is evaluated through root mean square error, mean absolute percentage error, and determination coefficient, and the effectiveness of model optimization is objectively evaluated;
[0211] The monitoring data, model prediction results and model optimization of each bridge are integrated to establish a knowledge base and case library for bridge corrosion prediction; the key influencing factors of corrosion prediction of different types of bridges, the applicable conditions of prediction models, and the selection principles of optimization methods are summarized to form a set of standardized corrosion prediction model application and update processes;
[0212] Embed the process into the bridge health monitoring system to realize the automation and intelligent application of the corrosion prediction model. Through continuous data accumulation and model optimization, the accuracy and reliability of corrosion prediction are continuously improved, providing a scientific basis for bridge maintenance decisions and safety assessments.
[0213] Here is an example:
[0214] In the health monitoring system of a cross-sea bridge, 20 temperature and humidity sensors and 10 chloride ion concentration sensors were deployed, with data collection frequencies of 1 hour and 1 day respectively. At the same time, electrochemical sensors were installed at 200 key parts of the bridge to collect the natural corrosion potential and polarization resistance data of the steel bars every 7 days. The monitoring data showed that the temperature range of the bridge environment was -5℃~40℃, the humidity range was 50%~90%, and the chloride ion concentration range was 0.01~0.5mol / L;
[0215] According to the Stern-Geary formula, the average corrosion current density of the steel bar is estimated to be 1.5 μA / cm 2 ;
[0216] The above monitoring parameters were input into the constructed corrosion prediction model, and the corrosion area percentage of the bridge in the current state was obtained to be 2%, while the actual corrosion area percentage detected was 2.5%, with a relative error of 20%;
[0217] By statistically analyzing the forecast errors at 100 historical time points, we found that the mean of the forecast errors was 0.5% and the standard deviation was 0.3%. By drawing a distribution histogram of the forecast errors, we found that the errors were mainly concentrated between 0.2% and 0.8%, showing a normal distribution.
[0218] The Pearson correlation coefficient between the prediction error and the ambient temperature was further calculated to be 0.8, and the Spearman rank correlation coefficient between the prediction error and the chloride ion concentration was 0.6, indicating that there was a significant positive correlation between the prediction error and the environmental factors.
[0219] When other factors remain unchanged, the prediction error increases by 0.2% for every 10℃ increase in ambient temperature; the prediction error increases by 0.15% for every 0.1mol / L increase in chloride ion concentration;
[0220] In view of the fact that the prediction error of the bridge exceeded 0.5%, the ambient temperature and chloride ion concentration were used as new model input features, and the model was retrained using the support vector machine algorithm. Under 5-fold cross validation, the root mean square error was reduced by 20%, the mean absolute percentage error was reduced by 15%, and the determination coefficient was increased by 0.1, indicating that the prediction accuracy of the model was significantly improved;
[0221] Based on the above analysis results, a bridge corrosion prediction knowledge base containing 1,000 actual engineering cases was established, covering the value ranges of key influencing factors of different types of bridges, the applicable conditions of the prediction model, and the selection principles of the optimization algorithm. A set of standardized application processes for bridge corrosion prediction was formed and embedded in the bridge health monitoring system, realizing the automation and intelligent application of the corrosion prediction model.
[0222] Furthermore, through comparative analysis, the deviation between the corrosion degree predicted by the adhesion change model under the current environmental conditions and the current corrosion degree was identified, and the influence of temperature and humidity on the corrosion degree was analyzed;
[0223] The established bond strength variation model is used to input the currently measured steel bar corrosion area ratio, average corrosion depth and other parameters to predict the bond strength between steel bars and concrete under the current environmental conditions, and the bond strength is compared with the bond strength estimated by theoretical calculation or empirical formula to calculate the absolute value and relative percentage of the prediction deviation.
[0224] Obtain the predicted and estimated values of adhesion over a period of history (e.g., 1 year), draw a time series curve of the predicted deviation, analyze the changing trend, fluctuation cycle, and mutation time point of the predicted deviation, and compare it with the ambient temperature and humidity data of the same period to identify the correlation between the predicted deviation and the change of environmental factors;
[0225] Calculate the statistical indicators of the prediction deviation, such as mean, standard deviation, skewness, and kurtosis. Determine the distribution type and dispersion of the prediction deviation through the numerical value and change trend of the indicators. Calculate the Pearson correlation coefficient or Spearman rank correlation coefficient between the prediction deviation and the ambient temperature and humidity to quantitatively measure the correlation strength and direction between the prediction deviation and the environmental factors, and provide a basis for subsequent model correction.
[0226] Statistical methods such as partial correlation analysis and multiple regression analysis were used to quantitatively evaluate the influence of temperature and humidity on prediction deviation when other factors were kept constant. The partial correlation coefficient and regression coefficient of temperature and humidity were obtained, and a quantitative correlation model between prediction deviation and environmental factors was established. On this basis, the original adhesion change model was revised and optimized.
[0227] Determine the magnitude and direction of the impact of temperature and humidity on the prediction deviation based on the quantitative correlation model, and compare the influence weights of the two factors; select a suitable function form (such as linear function, quadratic function, exponential function), introduce temperature and humidity into the adhesion change model, use regression analysis and other methods to estimate the function coefficients and other model parameters, and perform statistical tests and diagnoses to ensure that the revised model is statistically reasonable;
[0228] Use the revised model to re-predict historical data, calculate the prediction deviation, compare it with the prediction deviation before correction, and evaluate the effect of model correction. By repeatedly iterating the above steps, the internal structure and parameters of the adhesion change model are continuously optimized to improve the model's expressiveness and scope of application;
[0229] The sensitivity analysis method is used to evaluate the sensitivity of the modified bond force variation model to temperature and humidity changes, that is, the temperature and humidity values are changed respectively, the change range of the model prediction deviation is observed, and the adaptability and stability of the model to environmental factors are determined; in engineering applications, the temperature and humidity data of the environment in which the reinforced concrete structure is located are collected in real time and input into the modified bond force variation model to obtain the bond force prediction value considering the environmental influence;
[0230] According to the quantitative correlation model between environmental factors and prediction deviation, the prediction deviation compensation value under the current environmental conditions is calculated, and the predicted bond strength value is calibrated to obtain a more reliable estimated bond strength value. However, the influence of other factors (such as load level and component size) needs to be considered when applying the estimated value.
[0231] For the prediction of adhesion under different environmental conditions (such as high temperature and high humidity, low temperature and low humidity), a stratified sampling method is adopted to select representative reinforced concrete structures (such as bridges, docks, buildings) and carry out key test observations; formulate detailed test plans and operating procedures, clarify the purpose, content, method, frequency, and division of labor of the test, and ensure the standardization of the test process and the reliability of the data;
[0232] During the test, the changes in environmental conditions were strictly controlled, the interference of other factors was minimized, and the observation data were timely collated, analyzed and fed back to verify and optimize the adhesion change model, and also provide experience and guidance for other similar projects;
[0233] Here is an example:
[0234] Taking the caisson wharf in a seaport project as an example, the established bonding force variation model is used, and the currently measured average steel corrosion area ratio is 3%, and the average corrosion depth is 0.2mm. It is predicted that under the current environmental conditions (temperature 25°C, relative humidity 75%), the bonding strength between steel bars and concrete is 8.5MPa. According to the empirical formula of the port industry standard JTS167-4-2019, the actual bonding strength is 7.8MPa, the absolute value of the prediction deviation is 0.7MPa, and the relative error is 9.0%. Through the time series analysis of the prediction deviation in the past year, it is found that the mean of the prediction deviation is 0.6MPa, the standard deviation is 0.3MPa, the skewness is 0.5, and the kurtosis is 2.8, which basically obeys the normal distribution.
[0235] The Pearson correlation coefficient between the prediction deviation and the ambient temperature was 0.7 (P<0.01), and the Spearman rank correlation coefficient between the prediction deviation and the relative humidity was 0.6 (P<0.01), showing a significant positive correlation.
[0236] Further partial correlation analysis was used to control other factors unchanged. For every 1°C increase in temperature, the prediction deviation increased by 0.05MPa; for every 1% increase in relative humidity, the prediction deviation increased by 0.03MPa.
[0237] On this basis, the quadratic function form is selected, and the temperature t and relative humidity h are introduced into the adhesion change model to obtain the revised model: y = a0 + a1x1 + a2x2 + b1t + b2t 2 +c1 h+c2h 2, Where y is the predicted value of bonding strength, x1 is the proportion of corrosion area, x2 is the average corrosion depth, and a0, a1, a2, b1, b2, c1, and c2 are parameters to be estimated;
[0238] Using the data from the most recent year for parameter estimation, the root mean square error of the revised model on the training set was 0.4MPa, and the coefficient of determination was 0.85. The prediction accuracy was significantly better than before the revision.
[0239] Using sensitivity analysis, it was found that within the common range of temperature 20-30°C and relative humidity 60%-80%, the prediction deviation of the revised model was less than 0.5MPa, indicating that the model has good adaptability to environmental factors;
[0240] In the subsequent wharf maintenance, the ambient temperature and humidity are monitored in real time and input into the revised adhesion change model. Combined with the predicted deviation compensation value of 0.2MPa under the current environmental conditions, the predicted adhesion strength value after dynamic calibration is obtained, which provides an important reference for safety assessment and service life prediction.
[0241] Furthermore, in this embodiment, step S6 specifically includes:
[0242] Extract real-time bond monitoring data, including the ultimate bond strength between steel bars and concrete obtained by the pull-out method and the steel bar strain distribution obtained by the steel bar strain method, and invert the bond stress distribution between steel bars and concrete based on the strain distribution;
[0243] For each monitoring section, at least three measuring points are arranged, and at least three specimens are tested in parallel at each measuring point, and the average value is obtained as the representative value of the section;
[0244] The monitoring frequency is determined based on the importance and deterioration rate of the bridge, and the monitoring data are normalized to form a bond data set across bridges and structure types;
[0245] The current bonding force monitoring data is input into the established correlation analysis model between the steel bar surface corrosion degree and bonding force. The model predicts the corresponding theoretical bonding force value according to the current corrosion area ratio and corrosion depth parameters as the bonding force prediction value under the current corrosion degree.
[0246] For each bridge structure, the predicted bond strength value is compared with the measured value, and the difference between the two is calculated as the prediction deviation. The magnitude of the prediction deviation is quantified using indicators such as absolute value and relative percentage.
[0247] Evaluate the accuracy of the correlation analysis model at the current corrosion level to determine whether the model's prediction accuracy meets the actual application requirements. When evaluating the accuracy of the correlation analysis model at the current corrosion level, special attention should be paid to the deviation between the predicted value and the measured value;
[0248] According to the evaluation results, if the prediction accuracy of the model is not ideal, it is necessary to adjust the corrosion prediction model in time, including optimizing model parameters and introducing more influencing factors to improve the prediction performance of the model under different corrosion degrees. At the same time, the prediction deviations of different bridge structures are summarized and analyzed to obtain the average prediction deviation and dispersion of the correlation analysis model under different structural types and different corrosion degrees;
[0249] Using statistical methods such as hypothesis testing, under a given confidence level (e.g., 95%), determine whether the prediction deviation of the correlation analysis model is significantly different from zero under the current corrosion degree condition, that is, whether there is a systematic deviation between the predicted value of the model and the measured value;
[0250] According to the distribution type and sample size of the deviation data, select the appropriate hypothesis testing method, such as the single-sample t-test and the Wilcoxon signed-rank test. Before the test, the deviation data needs to be tested for normality (such as the Shapiro-Wilk test) and heteroskedasticity test (such as the Levene test) to determine the applicable test method;
[0251] If there is a significant deviation, it means that the model has a prediction deviation under the current working conditions, and the model needs to be adjusted and corrected; Based on the existing theoretical research results and a large amount of test data, the key factors affecting the relationship between the degree of steel corrosion and the bond strength are analyzed, in addition to the proportion of the corrosion area and the corrosion depth, such as the thickness of the rust layer, the type of corrosion products, and the strength of concrete, and these factors are used as new model inputs;
[0252] Data mining methods such as association rule mining, decision tree, and random forest are used to explore the association rules between various influencing factors and the prediction deviation of adhesion. For example, after discretizing each factor and the prediction deviation according to a certain threshold, the Aprili and FP-growth algorithms are used to mine the frequent item sets and association rules between each factor and the prediction deviation.
[0253] Use the mined association rules to guide the structural optimization of the association analysis model, such as adjusting the model's input features and embedding physical mechanisms according to the rules to improve the model's prediction accuracy and applicability;
[0254] For different types of bridge structures, a corrosion-bond strength correlation analysis sub-model based on structural characteristics is constructed. For example, for prestressed concrete bridges and steel-concrete composite bridges, correlation analysis models suitable for specific structures are trained and optimized respectively.
[0255] In the process of constructing the sub-model, the overall correlation analysis model optimized in steps 3-5 is fully utilized, and the input, output, and algorithm of the sub-model are adjusted and improved in a targeted manner in combination with the monitoring data of various bridge structures;
[0256] Through model fusion, ensemble learning and other methods, each sub-model is combined with the optimized overall model to form a comprehensive, cross-structure applicable association analysis model to improve the applicability and generalization ability of the model;
[0257] Establish a mechanism for continuous optimization and updating of the model, regularly collect the bonding monitoring data and corrosion status assessment data of each bridge structure, add them to the existing training data set, and iteratively update the structure and parameters of the correlation analysis model in combination with the latest theoretical research results; formulate standards and trigger conditions for model updates, such as triggering model updates when the amount of newly collected monitoring data reaches a certain scale (for example, an increase of 20%), or when the distribution characteristics of the new data change significantly (for example, the mean and variance exceed the original range);
[0258] Here is an example:
[0259] In the health monitoring of a cross-sea bridge, distributed fiber optic sensing technology was used to deploy fiber optic Bragg grating sensors in 800 monitoring sections throughout the bridge to collect steel bar strain data in real time. The wedge-shaped pressure block method was used every six months to conduct on-site testing of the bond performance between steel bars and concrete in 100 representative sections to obtain indicators such as the ultimate bond strength.
[0260] The correlation analysis model constructed by BP neural network predicts the theoretical value of the bonding strength of the whole bridge under the current corrosion state (average corrosion area 3.2%, average corrosion depth 0.25mm), and compares it with the measured value. It is found that the mean of the predicted deviation is 0.42MPa and the relative error is 8.1%. The single sample t test (confidence level 95%) shows that the mean deviation is significantly greater than 0, that is, the model has a systematic overestimation phenomenon;
[0261] The April association rule mining algorithm was further used to find that when the corrosion depth was greater than 0.3 mm and the concrete strength was less than 40 MPa, the probability of the prediction deviation exceeding 0.5 MPa reached 65%. Based on this, the quantitative index of concrete strength was introduced into the model, and the quantitative relationship between corrosion state, concrete strength and bond strength was constructed through multivariate nonlinear regression.
[0262] A sub-model was trained for the prestressed concrete T-beam of the bridge superstructure. The model described the influence of the corrosion difference between the web and the bottom plate of the T-beam on the bonding performance through piecewise linear fitting. Finally, the stacking fusion strategy was adopted to weightedly combine the sub-model with the overall model, which significantly improved the prediction accuracy of this type of bridge (the relative error was reduced to 5.3%).
[0263] In the following 6 months, 1,000 sets of bonding performance monitoring data were added, of which 120 sets came from a newly built steel-concrete composite beam cable-stayed bridge; when the amount of new data exceeded the threshold (15%), the incremental learning process was automatically triggered, and the model was updated online through the Mini-batch gradient descent algorithm to stabilize the performance indicators within the warning threshold (relative error 6%, absolute error 0.5MPa); the model knowledge base was continuously enriched to form a knowledge set containing 20 rules and 5 cases to guide the prediction and evaluation of bonding performance of similar bridges.
[0264] Furthermore, in this embodiment, step S7 specifically includes:
[0265] According to the corrosion type identification results, the corrosion of bridge reinforcement is divided into three types: uniform corrosion, local corrosion and pitting corrosion.
[0266] Combining the morphological characteristics, spatial distribution characteristics and causal mechanism of rust, a set of rust type classification standards with multi-index comprehensive evaluation is formulated. The weight of each factor is determined by the hierarchical analysis method. The quantitative index of the rust type is obtained by weighted summation, and the rust type is then divided according to the size of the index value.
[0267] Combined with the analysis of corrosion distribution characteristics, the distribution probability and spatial correlation of various types of corrosion in different parts of the bridge are determined to form a probability model of bridge corrosion type and distribution; various types of corrosion in different parts of the bridge include bridge decks, piers, and bearings;
[0268] The corrosion degree assessment method is used to calculate the average corrosion area ratio and maximum corrosion depth of the steel bars in each part of the bridge. Combined with the corrosion type and distribution probability model, the overall and local corrosion degrees of the bridge are comprehensively assessed to obtain a spatial distribution map of the corrosion degree of the bridge. The corrosion degree spatial distribution map is input into the established corrosion prediction model. The model predicts the development trend of the bridge corrosion degree within a certain time range (e.g., 1 year, 5 years) based on environmental parameters (temperature, humidity) and time series data, and obtains a dynamically updated corrosion degree prediction distribution map.
[0269] Compare and analyze the spatial distribution of rust degree reflected by the predicted distribution map of rust degree and the real-time monitoring data. Use spatial statistical methods such as Kriging interpolation to convert the discrete monitoring point data into raster data with the same resolution as the predicted distribution map. Use the Pearson correlation coefficient to measure the linear correlation strength between the two raster data.
[0270] Using the moving window method, a window of a certain size (for example, 3x3) is selected on the prediction map and the monitoring map, the mean and variance of the grid value differences in the two windows are calculated, and then the difference levels are divided according to the size of the mean and variance; the spatial distribution map of the local difference index is obtained by traversing the entire area through the moving window;
[0271] Identify areas with significant differences, severe corrosion types (such as pitting), and large distribution range (such as continuous distribution) as potential risk areas affected by corrosion; obtain vehicle data on the bridge where the risk area is located, including vehicle weight (axle weight), vehicle speed, and passing frequency, and establish a load action model for the risk area;
[0272] The refined finite element analysis method is used to calculate the stress distribution and deformation characteristics of the potential danger area at different corrosion degrees, and the structural response model of the potential danger area is established; the load action model and structural response model are used to analyze the correlation mechanism between vehicle load and corrosion development, and quantitatively evaluate the impact of vehicle load on corrosion aging;
[0273] By establishing a finite element model that couples corrosion extension and structural stress, the corrosion parameters of the potential danger area are set in the model, and vehicle loads under different working conditions are applied; by controlling variables, other parameters are fixed and only the value of vehicle weight (or vehicle speed, traffic frequency) is changed, and the change law of response indicators such as stress level and plastic zone range in the corrosion area is observed. The relative contribution and sensitivity of vehicle weight (or vehicle speed, traffic frequency) to corrosion are quantitatively evaluated by response surface method, sensitivity coefficient and other methods, and the most significant controlling factors affecting corrosion aging are identified, providing a basis for subsequent use and maintenance management;
[0274] Dynamically evaluate the corrosion risk level of the potential danger area by integrating the corrosion development prediction, load effect analysis and structural performance assessment results of the potential danger area; formulate targeted safety warning and maintenance reinforcement measures, and dynamically optimize the timing and frequency of implementation of the measures according to vehicle load characteristics and traffic flow data to minimize the development of corrosion and extend the service life of the bridge; during the implementation process, continuously monitor the corrosion aging process and vehicle load effects, dynamically update the corrosion prediction model and risk assessment model, continuously improve the maintenance decision-making plan, and form a long-term mechanism for bridge corrosion risk assessment and maintenance decision-making based on the vehicle-bridge coupling mechanism;
[0275] Here is an example:
[0276] Taking a prestressed concrete continuous rigid frame bridge on a highway as an example, through on-site investigation and testing, it was found that the bridge deck, piers, cap beams and other parts had different degrees of steel corrosion problems. According to the appearance characteristics of corrosion, the fuzzy comprehensive evaluation method was used to select four indicators: corrosion color, corrosion area, rust layer thickness, and rust pit density. The weight coefficients were assigned as 0.2, 0.3, 0.3, and 0.2 respectively, and the corrosion types were divided into mild corrosion (membership degree 0.8), moderate corrosion (membership degree 0.6), and severe corrosion (membership degree 0.9);
[0277] Using GIS spatial interpolation method, the distribution probability map of various types of corrosion was generated, and it was found that the bridge deck and cap beam were mainly moderately corroded, while the bridge piers were mainly lightly corroded.
[0278] The average cross-sectional loss rate of the corroded steel bars was measured by the Archimedean method to be 8%, and the local maximum corrosion depth was 2 mm. The results were input into the established grey Markov chain prediction model, and it was predicted that in the next five years, under the current climate conditions, the corrosion area ratio will increase to 12%, and the maximum corrosion depth will reach 3.5 mm.
[0279] Using the kriging method, the data of 20 corrosion monitoring points on the bridge were interpolated to a 100m×100m grid and compared with the prediction results. The Malangu Nice difference index was calculated to be 0.28, indicating that the spatial distribution difference between the two is small.
[0280] In the mid-span and near the support, the local difference index exceeds 0.5, and the predicted value is significantly higher than the monitored value, which is judged to be a potential corrosion area. The vehicle automatic detection system is deployed in the potential corrosion area, and the vehicle weight distribution is obtained as a normal distribution N (30t, 5t), the average vehicle speed is 80km / h, and the average daily traffic volume is 10,000 vehicles;
[0281] The corrosion parameters of the potential danger area were set in the finite element model, and different vehicle weights (20t, 30t, 40t) and vehicle speeds (60km / h, 80km / h, 100km / h) were applied to calculate the deflection and stress response of the bridge under normal service and severe corrosion conditions, and fit the response surface equation. Through variance analysis, the sensitivity coefficients of vehicle weight and vehicle speed were obtained to be 0.6 and 0.3 respectively, indicating that vehicle weight is the main controlling factor affecting corrosion aging.
[0282] In the dynamically updated corrosion risk prediction model, the vehicle weight distribution is used as an important input parameter. When the corrosion area ratio exceeds 15%, a warning is issued in time, and the weight limit standard is adjusted from 30t to 25t. At the same time, combined with the corrosion development trend and traffic characteristics, the best maintenance time window is set to within 2 years, and the maintenance plan is optimized. By replacing corroded steel bars and adding cathodic protection and other measures, the remaining service life of the bridge is extended by more than 10 years.
[0283] Furthermore, in this embodiment, step S8 specifically includes:
[0284] According to the stress characteristics of the bridge structure, select the appropriate finite element analysis software, establish the three-dimensional solid model of the bridge structure, and divide the grid reasonably; the stress characteristics of the bridge structure include beam type, arch type, and rigid frame type; the grid size should not be too large or too small, generally 1 / 10 to 1 / 8 of the beam height;
[0285] In the potential corrosion area of the bridge, the area is divided into several sub-areas according to the distribution of steel bar corrosion. The degree of corrosion in each sub-area is approximately uniform. For each sub-area, the equivalent corroded steel bar material parameters are generated by reducing the cross-sectional area and equivalent yield strength of the steel bar according to the cross-sectional loss rate of the corroded steel bar. When selecting the key stress points of the bridge, in addition to considering the load size, the corrosion sensitivity should also be considered. The parts with high corrosion sensitivity mainly include the alternating positive and negative bending moment action area, the shear and torque action concentration area, the anchorage area and the overlap area. In the finite element model, the above areas should be paid special attention to, and the calculation accuracy should be improved by means of local mesh encryption and refinement of unit types.
[0286] Using parametric modeling technology, the influence area is preset in the geometric model or mesh model. By changing the distance and size of the influence area, different finite element models are automatically generated to simulate the spatial variability of the corrosion area and evaluate the influence of location factors.
[0287] The modified CEB-FIPModel Code 2010 constitutive relationship is used to study the bond-slip behavior of corroded steel bars and concrete.
[0288] When modifying the bond-slip constitutive model, dimensionless parameters reflecting the degree of corrosion are introduced, and quantitative correlations between various parameters and corrosion parameters are established, and calibration is performed through pull-out test. Dimensionless parameters reflecting the degree of corrosion include volume expansion rate and residual section rate.
[0289] The modified corrosion degradation-bond slip constitutive model is embedded in the finite element model to evaluate the degree of weakening of the overall bond strength of the bridge by different corrosion degrees; the meteorological parameters of the area where the bridge is located, such as temperature and humidity, are input into the finite element model, and the multi-field coupling analysis of environment-material-corrosion degradation is used to explore the interactive influence mechanism of environmental conditions and material properties on the development of corrosion and bond degradation, which needs to consider the influence of temperature on the kinetics of corrosion reaction and the physical and mechanical properties of concrete, and the influence of humidity on the ion migration of the corrosion medium and the pore structure of concrete;
[0290] When analyzing the mechanism of vehicle load on corrosion and adhesion, it is necessary to consider the relationship between the stress amplitude and cycle number of steel bars induced by vehicle dynamic load and the kinetics of corrosion electrochemical reaction, study the evolution law of traffic flow characteristics and corrosion current density, product composition and distribution, and quantitatively evaluate the load-corrosion-adhesion chain effect;
[0291] Collect vehicle traffic data since the bridge was built, including vehicle type, vehicle weight, vehicle speed, and traffic volume. Use the rain flow counting method to calculate the load spectrum. Combined with the finite element analysis results, quantitatively evaluate the impact of repeated vehicle loads on the fatigue performance of corroded steel bars to obtain the predicted values of fatigue life and reliability.
[0292] Compare the finite element simulation results with the actual bridge test results, and use parameter identification methods such as partial least squares regression and artificial neural network to correct the material parameters, load parameters and boundary conditions in the finite element model, so that the prediction results after model calibration and the measured values can achieve the best fit and improve the prediction accuracy;
[0293] Based on the structural reliability theory of probability theory, the changing law of the probability of structural failure under different corrosion degradation levels is calculated through random finite element analysis. Taking the failure probability of the 100-year design reference period as a reference, when the adhesion degradation in the rusted area causes the probability of structural failure to increase by one order of magnitude, the corresponding adhesion degradation rate can be used as the key threshold.
[0294] The sensitivity analysis method can be used to further quantify the influence weight of the adhesion degradation rate on the failure probability. Combined with the risk matrix method, the risk scenarios formed by the combination of different adhesion degradation rates and failure consequence levels can be classified and graded, so as to determine the risk levels of different corrosion areas and formulate differentiated inspection and maintenance strategies.
[0295] Here is an example:
[0296] Taking a prestressed concrete continuous rigid frame bridge as an example, the ABAQUS finite element software was used to establish the geometric solid model of the entire bridge, and the C3D8R eight-node linear reduced integration unit was used for meshing, generating a total of about 500,000 units and 600,000 nodes;
[0297] According to on-site inspections, 10% of the bridge deck and 5% of the pier were defined as corrosion risk areas. Based on the detection result that the cross-sectional area of the corroded steel bars was reduced by 20%, the geometric scaling method was used to generate an equivalent elastic modulus of 200 GPa and a yield strength of 300 MPa for the corrosion area. The mid-span, support, and pier bottom were identified as corrosion sensitive areas, the grid size was encrypted to 0.1 m, and the Gaussian point positioning corrosion unit method was introduced to automatically generate finite element models for 10 different corrosion distribution scenarios. The modified CEB-FIP bond-slip constitutive model was used, and the volume expansion rate (with a value of 0.1 to 0.5) was introduced as the characteristic parameter of corrosion degradation. The parameters such as slip strength, residual bond stress, and characteristic slip were associated with it, and calibrated through 18 sets of pull-out test data.
[0298] COMSOL software was used to conduct temperature-humidity-corrosion multi-field coupling analysis, considering the effect of temperature gradient on concrete pore water chemistry, using non-Fick diffusion law to describe the migration of chloride ions in the corroded area, coupled with the adsorption / desorption mechanism driven by potential difference. At the same time, the vehicle-bridge coupled vibration equation was introduced, and transient dynamic analysis was used to simulate the stress-strain response of the steel bars in the corroded area under vehicle load, extract the stress amplitude and cycle number, and substitute them into the fatigue life prediction model based on the Paris formula;
[0299] Traffic flow data of the bridge over the past five years was collected, and it was found that the proportion of fatigue vehicles was about 30%, the average axle load was 80kN, and the daily traffic volume was 10,000 vehicles;
[0300] The partial least squares regression was performed on the finite element analysis results and the on-site test data. The correction values of material parameters (such as concrete elastic modulus) and load parameters (such as dynamic load amplification factor) were determined through leave-one-out cross-validation. The relative error of the overall stiffness prediction of the bridge after calibration was reduced to less than 5%.
[0301] Combining the reliability analysis principle, random finite element simulation was carried out on the key corrosion area. Taking into account random variables such as corrosion current density and concrete strength, the first-order second moment method (FORM) was used to calculate the structural failure probability. When the bond strength of the corrosion area degraded by 20%, the failure probability increased from 3% to 38%, which is close to the threshold of the 100-year design reference period (40%).
[0302] The Sobol sensitivity analysis method shows that the importance weight of the bond strength degradation rate is as high as 0.65, so 20% is determined as the key threshold of bond degradation. The risk levels are divided according to the bond degradation rate ≤10%, 10%-20%, and ≥20%, and corresponding maintenance strategies are formulated.
[0303] Furthermore, in this embodiment, step S8 also includes:
[0304] For the hazardous areas that exceed the bond strength threshold, the finite element analysis method is used to conduct stress analysis and hazard assessment, determine its impact on the overall safety of the bridge structure, calculate the maintenance and repair scope, and formulate corresponding maintenance and repair strategies;
[0305] Specifically, according to the bridge design drawings and on-site test data, the specific location, area and shape of the hazardous area exceeding the cohesion threshold are obtained;
[0306] Through these data, determine the relative position of the damaged area in the entire bridge structure and its potential impact on the overall structure; use material science testing methods, such as tensile testing and compression testing, to obtain the material properties of the damaged area, including but not limited to elastic modulus and Poisson's ratio;
[0307] Through these material properties, the mechanical behavior and possible failure modes of the material can be determined. Through structural stress analysis tools and load data, the stress analysis state of the hazardous area can be determined, including the size and distribution of the load;
[0308] After obtaining this data, the structural safety of the hazardous area under actual operating conditions can be evaluated. The connection mode and boundary conditions between the hazardous area and the surrounding structure are obtained, and the stress distribution and potential weaknesses of the structure as a whole are determined by analyzing the stability of the connection and the constraints of the boundary conditions. The material properties and geometric dimensions of the overall structure of the bridge are obtained by reviewing the overall design parameters of the bridge. This information is crucial for evaluating the mechanical properties and durability of the entire structure.
[0309] Use appropriate finite element software tools to select appropriate finite element meshing methods and unit types based on the complexity of the hazardous area, the accuracy requirements of the analysis, and the force analysis; after obtaining the mesh model, more accurate numerical simulation and structural analysis can be performed; by setting the convergence accuracy and the number of iterations, the numerical solution of the finite element analysis can be performed; after obtaining the simulation results, the specific impact of the hazardous area on the overall safety of the bridge structure can be determined, and the scope and methods of maintenance and repair that may be required can be evaluated;
[0310] Here is an example:
[0311] Through the analysis of the bridge design drawings and on-site inspection data, it was determined that the hazardous area is located at the bottom of the third T-beam of the second span of the main span, with an area of about 5 square meters and an irregular ellipse; this area is located at 1 / 4 of the entire T-beam length, which has a certain impact on the stress of the overall structure. Tensile and compression tests were conducted on the materials in the hazardous area, and the elastic modulus of the material was 30GPa and the Poisson's ratio was 2, which indicates that the material belongs to the category of normal concrete and is not prone to brittle fracture under normal loads; and through structural stress analysis, it was determined that the maximum compressive stress in the hazardous area under the combined action of dead load and live load is 10MPa, and the maximum tensile stress is 2MPa, both of which do not exceed the material strength limit. However, long-term load effects may cause crack expansion;
[0312] The hazardous area is connected to the adjacent T-beams by continuous longitudinal reinforcement, and the ends of the T-beams are consolidated with the pier caps, so the overall restraint is good. However, if the cracks expand further, the effect of these connections may be weakened. The overall design parameters of the bridge were reviewed. The T-beams of the bridge are made of C50 concrete with a design elastic modulus of 35 GPa. The mid-span T-beam height is 8 meters, the web thickness is 2 meters, the flange width is 2 meters, and the thickness is 25 meters.
[0313] Hexahedral units were used to mesh the hazardous area and adjacent structures, and the mesh was locally encrypted, with a total number of about 500,000 units. Numerical simulations of stress, displacement, and cracks were performed through solver settings. Finite element analysis results showed that the stress level in the hazardous area was controllable under the current state, but the stress concentration factor at the crack tip was high, requiring surface sealing and crack embedding methods for repair, and continuous monitoring of crack expansion to ensure safe use.
[0314] For non-hazardous areas that do not exceed the bond strength threshold, the long-term benefits after replacement or repair are evaluated and maintenance and repair strategies are formulated based on the structural characteristics of the bridge and monitoring data combined with cost-benefit analysis;
[0315] Specifically, through the Geographic Information System (GIS) and on-site investigation, the specific location and boundary data of non-hazardous areas that do not exceed the cohesion threshold are obtained to obtain accurate regional geographical location information;
[0316] According to the obtained regional geographical location information, use area measurement tools such as GIS or CAD software to calculate the total area of the non-hazardous area to obtain the precise area size of the area, use adhesion testing instruments, such as tensile testing machines, to obtain the specific value of the adhesion in the area, determine the specific gap between the adhesion value and the safety threshold, and obtain the adhesion value and its comparison result with the threshold; obtain environmental monitoring data of the area, including the readings of humidity and temperature sensors, and by analyzing these data, determine the influence of surrounding environmental factors (such as humidity and temperature) on the adhesion of the area, and obtain the impact assessment results of environmental factors;
[0317] According to the design documents and structural analysis reports of the bridge, the role and influence of the area in the overall structure of the bridge are evaluated to obtain the impact of the area on bridge safety. By estimating the material cost and labor cost, the cost data for repairing or replacing the area is obtained, the comprehensive cost is calculated, and the total cost estimate required to repair the area is obtained;
[0318] Based on the adhesion test results before and after the repair, the effectiveness of the repair measures is evaluated, such as whether the adhesion reaches the expected enhancement value, and the repair effect and the expected improvement in service life are obtained. By analyzing the repair effect and the expected remaining service life of the bridge, the long-term economic benefits after the repair are calculated, and the long-term benefit analysis results are obtained;
[0319] Use maintenance history records and expert consultation to develop a daily maintenance plan suitable for the area, including maintenance frequency and measures, to obtain the maintenance plan and the expected effect of extending the service life; comprehensively consider the maintenance plan, cost, repair effect and long-term benefits, determine the best time for repair through optimization models and decision analysis tools, and develop a detailed repair plan and implementation plan to obtain the final repair decision and implementation plan;
[0320] Here is an example:
[0321] Through GIS and on-site investigation, a non-hazardous area with a length of 100 meters and a width of 50 meters was determined. Its specific boundary coordinates are (12234E, 3123N), (12236E, 3123N), (12236E, 3121N), (12234E, 3121N), forming a rectangular area. Using the GIS area measurement tool, the area of the rectangular area was calculated to be 100 meters × 50 meters = 5000 square meters. The average adhesion value in the area was measured by a tensile testing machine to be 2MPa, the safety threshold was 5MPa, and the adhesion value was 3MPa lower than the threshold. By analyzing the environmental monitoring data, it was found that the average humidity around the area was 60% and the average temperature was 20℃. It was assessed that the humidity and temperature were within the normal range and had little effect on the adhesion. According to the design documents and structural analysis, it was assessed that the area was located at the edge of the bridge, had little impact on the overall structure, and had a low risk impact on the safety of the bridge.
[0322] It is estimated that the repair of this area will require 200,000 yuan in material costs, 100,000 yuan in labor costs, and a total cost of 300,000 yuan. The bonding strength before the repair was 2MPa, and after the repair, the bonding strength was measured to increase to 6MPa, reaching the expected enhancement value of 5MPa, and it is expected that the service life of this area can be extended by 10 years; the analysis shows that the expected remaining service life of the bridge is 30 years, and it can be extended for 10 years after the repair. Based on the annual toll income of 1 million yuan, the long-term economic benefits are estimated to be 10 million yuan, and a daily maintenance plan is formulated to conduct inspections once a quarter, repair problems in a timely manner, and conduct a comprehensive inspection once a year, which is expected to extend the service life by 5 years; through optimization model calculations, comprehensive consideration of maintenance costs, repair effects and long-term benefits, it is determined that the best time for repair is 2 years later, and a detailed phased implementation plan is formulated, with an estimated total construction period of 3 months.
[0323] In the description of the present invention, it is necessary to understand that the directions or positional relationships indicated by directional words such as "front, back, up, down, left, right", "lateral, vertical, horizontal" and "top, bottom" are usually based on the directions or positional relationships shown in the drawings. They are only for the convenience of describing the present invention and simplifying the description. Unless otherwise specified, these directional words do not indicate or imply that the device or element referred to must have a specific direction or be constructed and operated in a specific direction. Therefore, they cannot be understood as limiting the scope of protection of the present invention.
[0324] For those skilled in the art, various other corresponding changes and deformations can be made according to the technical solutions and concepts described above, and all of these changes and deformations should fall within the protection scope of the claims of the present invention.
Claims
1. A method for identifying hidden damage of a bridge structure, characterized in that: The following steps are involved: S1. Monitor and collect corrosion image data, structural deformation data, dynamic response data and environmental condition data of bridge structures; S2, analyzing the collected steel bar surface corrosion images to identify the type, distribution characteristics and corresponding current corrosion degree of the corrosion; S3. Construct a corrosion prediction model based on the corrosion degree under multiple time series; S4. Conducting simulation tests on the bonding strength between steel bars and concrete with different preset corrosion degrees, obtaining multiple sets of correlation data between the preset corrosion degrees and the bonding strength, and constructing a correlation analysis model between the steel bar surface corrosion degree and the bonding strength; S5, comparing and analyzing the corrosion prediction model with current monitoring data of different bridge structures, evaluating the accuracy of the corrosion prediction model under current environmental conditions, and adjusting the corrosion prediction model; S6. Compare and analyze the correlation analysis model with current monitoring data of different bridge structures, evaluate the accuracy of the correlation analysis model under the current corrosion degree, and adjust the corrosion prediction model; S7. Based on the corrosion type, distribution characteristics and degree of corrosion, as well as the comparative analysis results of the corrosion prediction model and the current monitoring data, identify the potential risk areas affected by corrosion, and analyze the impact of passing vehicles on these potential risk areas; S8. Evaluate the effect of corrosion on adhesion in potential risk areas based on the stress characteristics of the bridge structure. Determine the degree of weakening of adhesion caused by different degrees of corrosion based on the distance from the area to the stress point, material properties and environmental conditions. Analyze historical data and the characteristics of passing vehicles, set a threshold for the impact of adhesion, and classify the affected areas based on this threshold.
2. The bridge structure hidden damage identification method according to claim 1 is characterized in that: The step S1 specifically includes: Use image sensors to obtain high-definition image data of the bridge structure surface, segment and extract features from the image, automatically identify and locate the damaged area, and calculate the corrosion area ratio and crack length as the evaluation basis for corrosion monitoring; Use strain sensors and accelerometers to collect the strain and vibration acceleration of the bridge structure in real time. According to the change trend and amplitude of the strain and acceleration data, combined with the preset judgment threshold, determine whether the bridge structure has deformed or vibrated abnormally, and obtain monitoring data on the deformation and dynamic response of the bridge structure; Temperature, humidity and corrosion sensors are used to monitor the temperature, humidity and corrosive gas concentration of the bridge environment in real time. According to the correlation between environmental parameters and the characteristics of bridge structure materials, a quantitative evaluation model is established to determine the impact of current environmental conditions on the durability of the bridge structure and obtain environmental condition monitoring results.
3. The bridge structure hidden damage identification method according to claim 1 is characterized in that: The step S2 specifically includes: Based on the high-definition image data collected from the steel bar surface, image preprocessing technology is used to denoise, enhance and standardize the image to improve the image quality and the recognizability of the steel bar surface corrosion features; A semantic segmentation algorithm based on deep learning is used to perform pixel-level segmentation on the rust image, accurately extract the location and boundary information of the rust area, and obtain a binary mask map of the rust area; According to the mask map of the corrosion area, the ratio of the corrosion area to the total surface area of the steel bar is calculated as an important indicator to measure the degree of corrosion; By extracting the grayscale histogram features of the rusted area, the grayscale distribution of the rusted area is obtained; By calculating the local binary pattern features of the rust area, the microstructure and regularity of the rust texture are captured, and these feature parameters are combined into a multidimensional feature vector as the input for subsequent rust type discrimination and degree assessment.
4. The bridge structure hidden damage identification method according to claim 1 is characterized in that: The step S3 specifically includes: Obtain historical inspection data of steel structure, extract the corrosion area and maximum corrosion depth at different time nodes, and construct a time series data set of corrosion degree indexed by time; Perform a stationarity test on the time series of the degree of corrosion. If it is not stationary, perform differential processing until it passes the stationarity test. Adopting the STL decomposition method and based on LOESS fitting, the time series is decomposed into trend term, seasonal term and random term. According to the decomposition results, linear regression or polynomial regression modeling is adopted for trend term, Fourier series or periodic function modeling is adopted for seasonal term, and ARMA modeling is adopted for random term. Select the time series prediction model, and determine the hyperparameters of the model based on the stationarity, autocorrelation and trend characteristics of the corrosion degree time series, combined with the ACF and PACF graphs; The goodness of fit and predictive performance of different models were evaluated by cross-validation and information criteria, and the optimal model was selected; Environmental factors are introduced as covariates and fused with the corrosion degree time series to construct a corrosion prediction model for multivariate time series.
5. The bridge structure hidden damage identification method according to claim 1 is characterized in that: The step S4 specifically includes: According to the preset corrosion degree level, the steel bar specimens are artificially corroded by using the electrochemical accelerated corrosion method; By controlling the corrosion current density and corrosion time, the degree of corrosion can be regulated; During the corrosion process, the mass loss rate and volume expansion rate of the steel bars are regularly monitored to ensure that the preset corrosion degree is reached; Simulate the uneven corrosion in the actual environment, prepare steel bar specimens with different corrosion degrees, measure the corrosion area ratio and corrosion depth on the surface, and quantitatively characterize the corrosion degree; The steel bar specimens after corrosion treatment were cast with concrete into standard specimens, and the concrete mix ratio and curing conditions similar to the actual structure were adopted to ensure the representativeness of the specimens; A control group of uncorroded steel bars and concrete specimens was set up to evaluate the effect of corrosion on the bond strength; The pull-out test was carried out on reinforced concrete specimens with different corrosion degrees. The load-displacement curve and steel bar strain distribution during the pull-out process were continuously recorded by displacement sensors and strain gauge testing equipment to obtain the mechanical parameters reflecting the bonding force. According to the mechanical parameters measured by the pull-out test, the bond force distribution of the reinforced concrete interface under different corrosion degrees is calculated; Combined with the volume expansion effect, loose porous characteristics and cross-sectional loss of the corrosion products, the influence of the corrosion products on the bond transfer mechanism is analyzed to identify the key factors that lead to bond degradation due to corrosion. These characteristics of the corrosion products were quantitatively evaluated through cross-sectional analysis and microscopic morphology characterization of the drawn specimens; Statistical analysis was performed on the bond strength and interfacial stress distribution under different corrosion levels to extract quantitative indicators of bond strength; The correlation analysis model between the degree of steel bar surface corrosion and bonding strength is established using multiple linear regression or nonlinear regression methods.
6. The bridge structure hidden damage identification method according to claim 1 is characterized in that: The step S5 specifically includes: Extract real-time monitoring data of different bridges, record the area type, traffic flow and load type information of the bridge, and build a bridge corrosion monitoring database; The current corrosion state parameters of each bridge in the monitoring database are input into the constructed corrosion prediction model to obtain the predicted value of the corrosion degree of each bridge at the current moment, and compared with the corrosion degree estimated by electrochemical parameters to calculate the prediction error and relative error; The prediction errors of each bridge were statistically analyzed, the mean and standard deviation of the prediction errors were calculated, and the distribution histogram and box plot of the prediction errors were drawn to intuitively evaluate the overall prediction accuracy and stability of the corrosion prediction model on different bridges.
7. The bridge structure hidden damage identification method according to claim 1 is characterized in that: The step S6 specifically includes: Extract real-time bond monitoring data, including the ultimate bond strength between steel bars and concrete obtained by the pull-out method and the steel bar strain distribution obtained by the steel bar strain method, and invert the bond stress distribution between steel bars and concrete based on the strain distribution; For each monitoring section, at least three measuring points are arranged, and at least three specimens are tested in parallel at each measuring point, and the average value is obtained as the representative value of the section; The monitoring frequency is determined based on the importance and deterioration rate of the bridge, and the monitoring data are normalized to form a bond data set across bridges and structure types; The current bonding force monitoring data is input into the established correlation analysis model between the steel bar surface corrosion degree and bonding force. The model predicts the corresponding theoretical bonding force value according to the current corrosion area ratio and corrosion depth parameters as the bonding force prediction value under the current corrosion degree. For each bridge structure, the predicted bond strength value is compared with the measured value, and the difference between the two is calculated as the prediction deviation. The magnitude of the prediction deviation is quantified using indicators such as absolute value and relative percentage. Evaluate the accuracy of the correlation analysis model under the current corrosion level and determine whether the prediction accuracy of the model meets the actual application requirements.
8. The bridge structure hidden damage identification method according to claim 1 is characterized in that: The step S7 specifically includes: According to the corrosion type identification results, the corrosion of bridge reinforcement is divided into three types: uniform corrosion, local corrosion and pitting corrosion. Combining the morphological characteristics, spatial distribution characteristics and causal mechanism of rust, a set of rust type classification standards with multi-index comprehensive evaluation is formulated. The weight of each factor is determined by the hierarchical analysis method. The quantitative index of the rust type is obtained by weighted summation, and the rust type is then divided according to the size of the index value. Combined with the analysis of corrosion distribution characteristics, the distribution probability and spatial correlation of various types of corrosion in different parts of the bridge are determined to form a probability model of bridge corrosion type and distribution; The corrosion degree assessment method was used to calculate the average corrosion area ratio and maximum corrosion depth of the steel bars in various parts of the bridge. Combined with the corrosion type and distribution probability model, a comprehensive assessment of the overall and local corrosion degree of the bridge was conducted to obtain the spatial distribution map of the bridge corrosion degree.
9. The bridge structure hidden damage identification method according to claim 1 is characterized in that: The step S8 specifically includes: According to the stress characteristics of the bridge structure, select the appropriate finite element analysis software, establish a three-dimensional solid model of the bridge structure, and divide the grid reasonably; In the bridge corrosion risk area, the area is divided into several sub-areas according to the distribution of steel bar corrosion. The degree of corrosion in each sub-area is approximately uniform. For each sub-area, the equivalent corroded steel bar material parameters are generated by reducing the cross-sectional area and equivalent yield strength of the steel bar according to the cross-sectional loss rate of the corroded steel bar. Using parametric modeling technology, the influence area is preset in the geometric model or mesh model. By changing the distance and size of the influence area, different finite element models are automatically generated to simulate the spatial variability of the corrosion area and evaluate the influence of location factors. The modified CEB-FIPModelCode2010 constitutive relationship is used to study the bond-slip behavior of corroded steel bars and concrete; When modifying the bond-slip constitutive model, dimensionless parameters reflecting the degree of corrosion are introduced, and quantitative correlations between each parameter and the corrosion parameter are established, and calibration is performed through pull-out test. The modified corrosion degradation-bond-slip constitutive model was embedded in the finite element model to evaluate the degree of weakening of the overall bond strength of the bridge due to different corrosion degrees.
10. The bridge structure hidden damage identification method according to claim 9, characterized in that: The step S8 further comprises: For the hazardous areas that exceed the bond strength threshold, the finite element analysis method is used to conduct stress analysis and hazard assessment, determine its impact on the overall safety of the bridge structure, calculate the maintenance and repair scope, and formulate corresponding maintenance and repair strategies; For non-hazardous areas that do not exceed the bond strength threshold, the long-term benefits after replacement or repair are evaluated and maintenance and repair strategies are formulated based on the structural characteristics of the bridge and monitoring data, combined with cost-benefit analysis.
Citation Information
Patent Citations
Testing method and system for mechanical property of miniature steel bar with rusting
CN109211763A
Rusted beam model correction and damage identification method
CN113283144A
Prediction method for corrosion rate of steel bridge bolt
CN116563626A
Intelligent evaluation method for aging characteristics of coating
CN117764933A
Steel structure corrosion identification method and device, medium and equipment
CN117934901A
Cited By
Bridge disease diagnosis and evaluation method based on monitoring perception technology
CN119245979A
Radar signal processing system and method based on bridge array
CN119310544A
Bridge construction, management and maintenance integrated data management platform
CN119399010A
An integrated data management platform for bridge construction, management and maintenance
CN119399010B
BMS-based battery health state evaluation method and system
CN119414283A
