System and method for predicting service life of steel structure bridge after fatigue crack repair

Abstract

The invention relates to the technical field of bridge engineering, in particular to a system and method for predicting the service life of a steel structure bridge after fatigue crack repair. The method comprises the steps that load data and stress data in the operation process of a bridge structure after fatigue crack repairing are collected; dynamic response characteristic parameters are extracted in combination with stress distribution characteristics of the fatigue crack repair area; performing multi-dimensional data fusion and preprocessing on the dynamic response characteristic parameters, and constructing a data set for training a deep learning algorithm; combining a deep learning algorithm and a fracture mechanics theory to construct a crack propagation rate prediction model and a residual life prediction model; according to the operation state of the bridge and the performance degradation law of the repairing area, the remaining life of the repaired structure is predicted, and multi-stage early warning is conducted. According to the method provided by the invention, bridge structure performance evaluation, life prediction and potential risk early warning can be realized, and the safety and reliability of bridge operation are ensured.

Description

Technical Field

[0001] The present application relates to the technical field of bridge engineering, and specifically to a system and method for predicting the life of a steel structure bridge after fatigue crack repair. Background Art

[0002] There is a lack of effective means to evaluate the performance of bridge structures after fatigue crack repair. The performance of the repaired structure is affected by the load history, crack repair quality and stress distribution characteristics. The existing methods lack an accurate description of the dynamic changes in the performance of the repaired area, making it difficult to quantify the repair effect.

[0003] Traditional crack growth rate calculations mostly rely on a single fracture mechanics formula. Although it has a certain theoretical basis, it ignores the complexity of the time series characteristics during crack growth and cannot dynamically adapt to changes in bridge operating conditions.

[0004] In summary, the life prediction field of steel structure bridges after fatigue crack repair mainly faces the following problems: lack of accurate evaluation of the dynamic performance of the repaired area, insufficient crack growth rate prediction accuracy, and low reliability of the life prediction model. The existence of these problems not only restricts the scientific evaluation of the safety status of the bridge after repair, but also poses a potential threat to the subsequent maintenance and safe operation of the bridge. Therefore, a scientific, systematic, and accurate life prediction method after fatigue crack repair is urgently needed to comprehensively improve the safety management level of bridge structures.

[0005] In view of this, the present application proposes a system and method for predicting the life of a steel structure bridge after fatigue crack repair. Summary of the invention

[0006] To achieve the above-mentioned purpose, the present application provides a system and method for predicting the life of a steel structure bridge after fatigue crack repair. The specific technical scheme is as follows: A method for predicting the life of a steel structure bridge after fatigue crack repair, comprising:

[0007] Collect load and stress data during operation of bridge structures after fatigue crack repair;

[0008] Based on the finite element analysis method and combined with the stress distribution characteristics of the fatigue crack repair area, the dynamic response characteristic parameters are extracted;

[0009] Perform multi-dimensional data fusion and preprocessing on dynamic response characteristic parameters to construct a data set for training deep learning algorithms;

[0010] Combining deep learning algorithms with fracture mechanics theory, we build crack growth rate prediction models and remaining life prediction models.

[0011] Based on the operating status of the bridge and the performance degradation law of the repaired area, the remaining life of the repaired structure is predicted and multi-level early warning is issued.

[0012] Preferably, a strain sensor and a load sensor are arranged on the bridge structure after the fatigue cracks are repaired, so as to collect strain data and load data of the bridge structure in real time during operation;

[0013] The collected load data and strain data are preprocessed, including denoising and outlier removal; the preprocessed load data and strain data are constructed into time series data.

[0014] Preferably, the load condition of the finite element model is set according to the collected load data; the time history of the load is the same as the collected load time series, and the loading frequency of the load is consistent with the sampling frequency; multiple load conditions are set, corresponding to different load levels and loading methods;

[0015] For each load condition, finite element static analysis is performed to obtain the stress distribution cloud diagram of the fatigue crack repair area;

[0016] Based on the stress distribution characteristics of the fatigue crack repair area, dynamic response characteristic parameters are extracted; the dynamic response characteristic parameters include: stress amplitude, stress ratio, stress gradient and stress triaxiality.

[0017] Preferably, the collected load time series data and strain time series data are fused with the extracted dynamic response characteristic parameters in a multi-dimensional manner to construct a fused data set;

[0018] Preprocess the fused data set, including data normalization, data enhancement, and data partitioning;

[0019] Based on the preprocessed fusion data set, a training deep learning data set for training the deep learning algorithm is constructed, and the training deep learning data set is divided into a training data set, a validation data set, and a test data set.

[0020] Preferably, a crack growth rate prediction model is constructed based on a long short-term memory neural network (LSTM) model, and a network structure of the LSTM model is designed, wherein the network structure of the LSTM model includes an input layer, an LSTM layer, a fully connected layer, and an output layer; the input layer is used to input a training data set; the LSTM layer is composed of a plurality of LSTM units, each of which includes an input gate, a forget gate, an output gate, and a memory unit, and is used to learn and memorize features of time series data; the fully connected layer is used to extract features output by the LSTM layer, and map the features to predicted values ​​of the crack growth rate; the output layer uses a linear activation function to output predicted crack growth rate values;

[0021] The training data set is used to train the LSTM model and optimize the model parameters. The mean square error is used as the loss function. During the LSTM model training process, the validation data set is used to validate the model. The LSTM model that meets the performance requirements is used as the crack growth rate prediction model.

[0022] Preferably, the linear elastic fracture mechanics theory is introduced to establish a relationship between the crack growth rate and the stress intensity factor range; the Paris formula is used to describe the relationship between the crack growth rate and calculate the stress intensity factor range;

[0023] Based on the Paris formula and the calculation formula of the stress intensity factor range, a remaining life prediction model is constructed. The crack growth rate predicted by the LSTM model is substituted into the Paris formula. The cumulative number of crack growth cycles is obtained by integration, and the remaining life prediction value of the repaired structure is obtained.

[0024] Preferably, a multi-level warning is given to the safety status of the repaired structure based on the remaining life prediction value; a warning threshold and an alarm threshold are set, and when the remaining life prediction value is lower than the warning threshold, a warning signal is issued; when the remaining life prediction value is lower than the alarm threshold, an alarm signal is issued.

[0025] A life prediction system for a steel structure bridge after fatigue crack repair, which is used to implement a life prediction method for a steel structure bridge after fatigue crack repair, comprising: a data acquisition module, a feature extraction module, a data fusion module, a remaining life prediction module and a multi-level early warning module;

[0026] The data acquisition module is used to collect load data and stress data during the operation of the bridge structure after the fatigue cracks are repaired;

[0027] The feature extraction module extracts dynamic response feature parameters based on the finite element analysis method and the stress distribution characteristics of the fatigue crack repair area;

[0028] The data fusion module performs multi-dimensional data fusion and preprocessing on the dynamic response characteristic parameters to construct a data set for training the deep learning algorithm;

[0029] The remaining life prediction module combines the deep learning algorithm and fracture mechanics theory to construct a crack growth rate prediction model and a remaining life prediction model;

[0030] The multi-level warning module predicts the remaining life of the repaired structure and issues a multi-level warning based on the operating status of the bridge and the performance degradation law of the repaired area.

[0031] An electronic device comprises: a processor and a memory, wherein the memory stores a computer program that can be called by the processor; the processor executes the method for predicting the life of a steel structure bridge after fatigue crack repair by calling the computer program stored in the memory.

[0032] A computer-readable storage medium stores instructions. When the instructions are run on a computer, the computer executes the method for predicting the life of a steel structure bridge after fatigue crack repair.

[0033] Beneficial effects of this application: The load and stress data collected by this application can accurately reflect the operating status of the bridge, provide basic data support for subsequent analysis, and ensure the reliability and accuracy of the model input.

[0034] This application extracts dynamic response characteristic parameters through finite element analysis, which can accurately characterize the stress distribution characteristics of the repair area and provide key physical property descriptions for crack propagation prediction.

[0035] This application improves the consistency and validity of data through data fusion and preprocessing, builds a high-quality data set, provides high-quality training samples for deep learning algorithms, and improves model accuracy.

[0036] This application uses deep learning to mine time series features and combines it with fracture mechanics theory to build a model that can dynamically predict crack propagation behavior and significantly improve the accuracy of remaining life prediction.

[0037] This application predicts the remaining life through performance degradation laws, and combined with a multi-level early warning mechanism, it can promptly detect potential hidden dangers, provide scientific guidance for bridge maintenance, and improve operational safety.

[0038] The method of this application comprehensively improves the accuracy of performance evaluation and life prediction of repaired bridge structures by combining finite element analysis, deep learning algorithms and fracture mechanics theory. It can provide real-time warning of potential risks, optimize bridge maintenance and management strategies, and ensure the safety and reliability of bridge operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 A flow chart of a method for predicting the life of a steel structure bridge after fatigue crack repair provided in this application;

[0040] Figure 2 A flow chart for constructing a crack growth rate prediction model for a method for predicting the life of a steel structure bridge after fatigue crack repair provided in this application;

[0041] Figure 3 A remaining life prediction and early warning flow chart of a life prediction method for a steel structure bridge after fatigue crack repair provided in this application;

[0042] Figure 4 A structural diagram of a life prediction system for a steel structure bridge after fatigue crack repair provided in this application. DETAILED DESCRIPTION

[0043] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are described in detail below in conjunction with the drawings in the specification.

[0044] In the following description, many specific details are set forth to facilitate a full understanding of the present application, but the present application can also be implemented in other ways different from those described herein, and those skilled in the art can make similar generalizations without violating the connotation of the present application. Therefore, the present application is not limited to the specific embodiments disclosed below.

[0045] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present application. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0046] Example 1

[0047] Reference Figures 1 to 3 The first embodiment of the present application provides a method for predicting the life of a steel structure bridge after fatigue crack repair.

[0048] Step 1: Collect load data and stress data during the operation of the bridge structure after the fatigue cracks are repaired.

[0049] Arrange strain sensors and load sensors on the bridge structure after fatigue crack repair to collect strain data and load data of the bridge in real time during operation;

[0050] Exemplarily, according to the stress distribution characteristics of the fatigue crack repair area, appropriate strain sensors and load sensors are selected; the strain sensor uses a resistive strain gauge, model BE120-05AA, with a range of ±5000 microstrain, a sensitivity coefficient of 2.0±0.1, and a resistance value of 120±0.5Ω; the load sensor uses a pressure load sensor, model CYB-YB-50KN, with a range of 050kN, a comprehensive accuracy of 0.02%FS, and an output signal of 05V.

[0051] Exemplarily, strain gauges are arranged at key positions in and near the fatigue crack repair area, including the crack tip, crack propagation path, stress concentration area, etc.; at least three strain gauges are arranged at each key position, respectively arranged along the longitudinal, transverse and 45° directions of the bridge to measure the strain response in different directions; load sensors are arranged at key sections of the bridge to measure the load acting on the bridge.

[0052] Exemplarily, a wireless sensor network system is built, consisting of wireless strain nodes, wireless load nodes, wireless relay nodes and wireless base stations; the wireless strain nodes and wireless load nodes are respectively connected to strain gauges and load sensors to collect strain data and load data, and transmit the data to wireless relay nodes through wireless communication; the wireless relay nodes transmit the data to the wireless base stations, and the wireless base stations are connected to a host computer to realize data storage and processing.

[0053] Exemplarily, the data acquisition frequency of the wireless sensor network is set to 200 Hz to ensure that the dynamic change characteristics of load and strain can be accurately captured; the data acquisition time T is determined according to actual needs.

[0054] The collected load data P(t) and strain data Preprocessing is performed, including denoising and outlier removal. The original data is denoised using wavelet transform, with sym5 as the wavelet basis function and 5 decomposition levels. Outlier detection and removal are performed on the denoised data.

[0055] The preprocessed load data and strain data are constructed into time series data, represented as P(t)|t=1,2,...,m and n0=1,2,...,n, where m is the number of sampling points, m=f s T,f s is the sampling frequency, n is the number of strain sensors;

[0056] In this step, strain sensors and load sensors are arranged to collect strain data of the fatigue crack repair area and load data of the bridge in real time, providing reliable data support for subsequent fatigue fracture analysis and life prediction; the collected data are preprocessed, including denoising and outlier removal, to improve the data quality and lay the foundation for subsequent data analysis; time series data that characterizes the strain response of the fatigue crack repair area and the load action of the bridge are formed, providing data input for subsequent feature parameter extraction, fatigue crack growth rate analysis and life prediction.

[0057] Step 2: Based on the finite element analysis method and combined with the stress distribution characteristics of the fatigue crack repair area, the dynamic response characteristic parameters are extracted.

[0058] According to the bridge design drawings and fatigue crack repair plan, the finite element model of the bridge structure was established using the finite element software ANSYS. The geometric dimensions, material properties, boundary conditions and characteristics of the fatigue crack repair area of ​​the bridge were considered in the model. The eight-node solid element SOLID185 was used to simulate the bridge structure, and the six-node triangular element PLANE183 was used to simulate the fatigue crack repair area. The mesh was encrypted in the fatigue crack repair area and its vicinity to improve the calculation accuracy.

[0059] According to the load data P(t)|t=1,2,...,m collected in step 1, the load condition of the finite element model is set; the load is applied to the key section of the bridge, and the size and direction are consistent with the measured load data; the time history of the load is the same as the collected load time series, and the loading frequency of the load is consistent with the sampling frequency; set multiple load conditions, corresponding to different load levels and loading methods, to simulate the load effect during the actual bridge operation.

[0060] For each load condition, finite element static analysis is performed to obtain the stress distribution cloud map of the fatigue crack repair area. The stress distribution cloud map reflects the stress distribution characteristics of the fatigue crack repair area, including stress concentration and stress gradient. The stress tensor of the fatigue crack repair area is σ ij =[σ xx ,σ yy ,σ zz ,τ xy ,τ yz ,τ zx ] ij , where σ xx ,σ yy ,σ zz is the normal stress component, τ xy ,τ yz ,τ zx is the shear stress component.

[0061] Based on the stress distribution characteristics of the fatigue crack repair area, dynamic response characteristic parameters are extracted; the dynamic response characteristic parameters include: stress amplitude, stress ratio, stress gradient and stress triaxiality.

[0062] Stress amplitude: At the peak and valley moments of the load, the maximum principal stress σ1 and the minimum principal stress σ3 in the fatigue crack repair area are extracted, and the stress amplitude Δσ=σ1-σ3 is calculated.

[0063] Stress ratio: At the peak load moment, the maximum principal stress σ1 and the minimum principal stress σ3 in the fatigue crack repair area are extracted, and the stress ratio R = σ3 / σ1 is calculated.

[0064] Stress gradient: At the peak load moment, extract the stress tensor σ of the adjacent unit in the fatigue crack repair area ij and σi'j' , calculate the stress gradient G = |σ ij -σ i'j' | / d, where d is the distance between the centers of the two units.

[0065] Stress triaxiality: Extract the stress tensor σ of the fatigue crack repair area element at the peak load moment ij , calculate stress triaxiality Y T =σ m / σ eq , where σ m is the average normal stress, σ m The calculation formula is: m =(σ xx +σ yy +σ zz ) / 3;

[0066] σ eq is the equivalent stress, σ eq The calculation formula is:

[0067]

[0068] Assuming that l dynamic response characteristic parameters are extracted, the dynamic response characteristic parameters under the i-th load condition can be expressed as X i =(x i1 ,x i2 ,...,x il ).

[0069] In this step, finite element analysis is used to obtain the stress distribution characteristics of the fatigue crack repair area under the actual load, which provides a basis for the extraction of dynamic response characteristic parameters; dynamic response characteristic parameters characterizing the fatigue crack growth characteristics are extracted, including stress amplitude, stress ratio, stress gradient and stress triaxiality, etc. These characteristic parameters are closely related to the fatigue crack growth rate and are important influencing factors for fatigue life prediction; the dynamic response characteristic parameters extracted under different load conditions are constructed into a characteristic parameter set, which provides data support for subsequent data fusion, preprocessing and deep learning model training.

[0070] Step 3: Perform multi-dimensional data fusion and preprocessing on the dynamic response characteristic parameters to construct a data set for training the deep learning algorithm.

[0071] The load time series data P(t)|t=1,2,...,m and strain time series data collected in step 1 are n0=1,2,...,n and the dynamic response characteristic parameter X extracted in step 2 i Perform multi-dimensional data fusion and construct a fused data set D = {(P j ,E j ,Xj )|j=1,2,...,N};P j =(P(t1),P(t2),...,P(t k )) is the load sequence of the jth sample; E j =(ε1(t1),...,ε1(t k ),ε2(t1),,...,ε2(t k ),...,ε(t1),...,ε n (t k )) is the strain sequence of the jth sample, X j =(X i1 ,X i2 ,...,X il ) is the dynamic response characteristic parameter corresponding to the jth sample, N is the total number of samples, and k is the time step of each sample. Through multi-dimensional data fusion, the load, strain and dynamic response characteristic parameters are integrated into a high-dimensional data set, which improves the information density and relevance of the data.

[0072] The fused dataset D is preprocessed, including data normalization, data enhancement and data partitioning.

[0073] The data normalization includes: using the maximum and minimum value normalization method to normalize the load sequence P j , strain sequence E j and dynamic response characteristic parameter X j Normalized to the interval [0,1] respectively, that is:

[0074]

[0075] in, and are the normalized load, strain and dynamic response characteristic parameters, respectively, iq is the unnormalized dynamic response characteristic parameter, min(·) and max(·) represent the minimum and maximum values ​​respectively; normalization can eliminate the influence of different dimensions and orders of magnitude and improve the comparability of data.

[0076] The data enhancement includes: using data expansion technology to expand the fusion data set D to generate new training samples; the data expansion method includes: j and strain sequence E j Add random noise and dynamic response characteristic parameter X j Perform random perturbations, random cropping and splicing of samples; data augmentation can effectively increase the number and diversity of training samples, and improve the generalization ability and robustness of deep learning models.

[0077] The data division includes: dividing the preprocessed fusion data set D into a training set and a validation set, wherein the training set is used to train the deep learning model, and the validation set is used to adjust the model hyperparameters and prevent overfitting.

[0078] Based on the preprocessed fusion data set, a training deep learning data set for training the deep learning algorithm is constructed; each sample is represented as (S j ,y j ), where is the input feature of the jth sample, and are the normalized load sequence, strain sequence and dynamic response characteristic parameter sequence, y j is the crack growth rate or remaining life corresponding to the jth sample; the crack growth rate and remaining life are determined according to the actual monitoring data or theoretical calculation formula and used as the training label of the deep learning model to obtain the training data set D train ={(S j ,y j )|j=1,2,...,N1} and the validation dataset D val ={(Sj,y j )|j=N1+1,N1+2,...,N2}, where N1 and N2 are the number of samples in the training set and the validation set respectively.

[0079] In this step, multi-dimensional data fusion is used to integrate the characteristic parameters of load, strain and dynamic response into a high-dimensional data set, which makes full use of various monitoring data and finite element analysis results, improves the information density and correlation of the data, and provides rich characteristic information for the training of deep learning models. Preprocessing techniques such as data normalization, data enhancement and data partitioning are used to standardize, expand and optimize the fused data set, improve the quality and applicability of the data, and enhance the generalization ability and robustness of the deep learning model. A data set for training deep learning algorithms is constructed, and the input features are mapped with crack growth rate or remaining life, which provides direct data support for the training and prediction of deep learning models and lays a data foundation for fatigue life prediction.

[0080] Step 4: Combine deep learning algorithm and fracture mechanics theory to construct crack growth rate prediction model and remaining life prediction model.

[0081] The long short-term memory neural network (LSTM) was selected as the basic algorithm of the crack growth rate prediction model; LSTM is a special recurrent neural network (RNN) that can effectively process time series data and capture the long-term dependencies of data; LSTM overcomes the gradient vanishing and gradient exploding problems of traditional RNN by introducing a gating mechanism, and has stronger modeling capabilities and a more stable training process.

[0082] Design the network structure of the LSTM model, including the input layer, LSTM layer, fully connected layer, and output layer; the input layer receives the training data set D constructed in step 3 train ={(S j ,y j )|j=1,2,...,N1}, where is the input feature of the jth sample, y j is the crack growth rate label of the jth sample; the LSTM layer consists of multiple LSTM units, each of which contains an input gate, a forget gate, an output gate, and a memory unit, which can adaptively learn and memorize the features of time series data; the fully connected layer is used to extract the high-level features output by the LSTM layer and map them to the predicted value of the crack growth rate; the output layer uses a linear activation function to output the predicted crack growth rate value.

[0083] Using the training data set D train Train the LSTM model and optimize the model parameters; use the mean square error (MSE) as the loss function, that is:

[0084]

[0085] Among them, y j is the crack growth rate label of the jth sample, The crack growth rate label predicted by the LSTM model; the Adam optimization algorithm is used to minimize the loss function and adaptively adjust the weight and bias of the model; during the training process, the validation data set D val The model is verified and its generalization performance is monitored to avoid overfitting. When the performance on the validation set reaches the optimal level, the model parameters are saved as the final crack growth rate prediction model.

[0086] The linear elastic fracture mechanics (LEFM) theory is introduced to establish a relationship between the crack growth rate and the stress intensity factor range ΔK; the Paris formula is used to describe the crack growth rate da / dN z The relationship with ΔK is: Where a is the crack size, N z is the number of loading cycles, C and is the material constant.

[0087] The stress intensity factor range ΔK is related to the stress amplitude Δσ and the crack size a and can be expressed as: Where Y is the correction factor related to crack shape and crack size.

[0088] Based on the Paris formula and the calculation formula of the stress intensity factor range, a remaining life prediction model is constructed; the crack growth rate da / dN is predicted using the LSTM model z ; Substituting the predicted crack growth rate into the Paris formula, we get: Integrate the above formula to get the cumulative number of crack propagation cycles N f ,Right now: Where a0 is the initial crack size, a c is the critical crack size;

[0089] Remaining life N r It can be expressed as: N r =N f -N0; where N0 is the current number of loading cycles; by solving the above integral equation, the predicted value of the remaining life of the repaired structure is obtained.

[0090] In this step, the LSTM neural network is used to construct a crack growth rate prediction model, which makes full use of the characteristics of time series data, captures the dynamic change law of the crack growth process, and improves the accuracy of crack growth rate prediction; the fracture mechanics theory is introduced to establish a physical relationship between the crack growth rate and the stress intensity factor range, which enhances the interpretability and rationality of the remaining life prediction model; combining deep learning algorithms and fracture mechanics theory, a remaining life prediction model that comprehensively considers load, strain and dynamic response characteristics is constructed, which improves the reliability and applicability of life prediction; the remaining life is solved by integral equations, and the quantitative life prediction results are obtained, which provides a direct basis for subsequent early warning and decision-making.

[0091] Step 5: Based on the operating status of the bridge and the performance degradation law of the repaired area, predict the remaining life of the repaired structure and issue a multi-level early warning.

[0092] According to the predicted value N of the remaining life of the repaired structure obtained in step 4 r , to provide multi-level warnings for the safety status of the structure; set the warning threshold N warn And the alarm threshold N alarm , when the remaining life prediction value N r When the remaining life prediction value N is lower than the warning threshold, a warning signal is issued, indicating the need to strengthen monitoring and testing and evaluate the safety performance of the structure; r When it falls below the alarm threshold, an alarm signal is issued, indicating the need to take load limiting, reinforcement and other measures to ensure the safe use of the structure.

[0093] Warning threshold N warn And the alarm threshold N alarmIt can be determined based on design documents, evaluation standards or expert experience, and is generally taken as a certain percentage of the design life, such as 80% and 90%; the specific threshold setting can be adjusted according to actual needs.

[0094] This step achieves real-time monitoring of the safety status of the repaired structure and risk warnings through multi-level early warning, providing a scientific basis for operation management and maintenance decisions; this helps to improve the safety and reliability of bridge operation, extend the service life of the bridge, and reduce the cost of the entire life cycle.

[0095] Example 2

[0096] Reference Figure 4 The second embodiment of the present application provides a life prediction system for a steel structure bridge after fatigue crack repair.

[0097] The system comprises: a data acquisition module, a feature extraction module, a data fusion module, a remaining life prediction module and a multi-level early warning module.

[0098] The data acquisition module is used to collect load data and stress data during the operation of the bridge structure after the fatigue cracks are repaired.

[0099] The feature extraction module extracts dynamic response feature parameters based on the finite element analysis method and in combination with the stress distribution characteristics of the fatigue crack repair area.

[0100] The data fusion module performs multi-dimensional data fusion and preprocessing on the dynamic response characteristic parameters to construct a data set for training a deep learning algorithm.

[0101] The remaining life prediction module combines deep learning algorithms and fracture mechanics theory to construct a crack growth rate prediction model and a remaining life prediction model.

[0102] The multi-level warning module predicts the remaining life of the repaired structure and issues a multi-level warning based on the operating status of the bridge and the performance degradation law of the repaired area.

[0103] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also change, modify, replace and modify the above-mentioned embodiments without departing from the scope of protection of the purpose of the present application and the claims, which are all within the protection of the present application.

Claims

1. A method for predicting the life of a steel structure bridge after fatigue crack repair, characterized in that: include: Collect load and stress data during operation of bridge structures after fatigue crack repair; Based on the finite element analysis method and combined with the stress distribution characteristics of the fatigue crack repair area, the dynamic response characteristic parameters are extracted; Perform multi-dimensional data fusion and preprocessing on dynamic response characteristic parameters to construct a data set for training deep learning algorithms; Combining deep learning algorithms with fracture mechanics theory, we build crack growth rate prediction models and remaining life prediction models. Based on the operating status of the bridge and the performance degradation law of the repaired area, the remaining life of the repaired structure is predicted and multi-level early warning is issued.

2. A method for predicting the life of a steel structure bridge after fatigue crack repair according to claim 1, characterized in that: Arrange strain sensors and load sensors on the bridge structure after fatigue crack repair to collect strain data and load data of the bridge structure in real time during operation; Preprocess the collected load data and strain data, including denoising and outlier removal; The preprocessed load data and strain data are constructed into time series data.

3. A method for predicting the life of a steel structure bridge after fatigue crack repair according to claim 2, characterized in that: According to the collected load data, the load condition of the finite element model is set; the time history of the load is the same as the collected load time series, and the loading frequency of the load is consistent with the sampling frequency; Set up multiple load conditions, corresponding to different load levels and loading methods; For each load condition, finite element static analysis is performed to obtain the stress distribution cloud diagram of the fatigue crack repair area; Extract dynamic response characteristic parameters based on stress distribution characteristics of fatigue crack repair area; The dynamic response characteristic parameters include: stress amplitude, stress ratio, stress gradient and stress triaxiality.

4. A method for predicting the life of a steel structure bridge after fatigue crack repair according to claim 3, characterized in that: The collected load time series data and strain time series data are fused with the extracted dynamic response characteristic parameters in multiple dimensions to construct a fused data set; Preprocess the fused data set, including data normalization, data enhancement, and data partitioning; Based on the preprocessed fusion data set, a training deep learning data set for training the deep learning algorithm is constructed, and the training deep learning data set is divided into a training data set, a validation data set, and a test data set.

5. A method for predicting the life of a steel structure bridge after fatigue crack repair according to claim 4, characterized in that: A crack growth rate prediction model is constructed based on a long short-term memory neural network (LSTM) model, and a network structure of the LSTM model is designed. The network structure of the LSTM model includes an input layer, an LSTM layer, a fully connected layer, and an output layer. The input layer is used to input a training data set. The LSTM layer is composed of a plurality of LSTM units, each of which includes an input gate, a forget gate, an output gate, and a memory unit, and is used to learn and memorize the features of time series data. The fully connected layer is used to extract the features of the output of the LSTM layer and map the features to the predicted value of the crack growth rate. The output layer uses a linear activation function to output the predicted crack growth rate value. The training data set is used to train the LSTM model and optimize the model parameters. The mean square error is used as the loss function. During the LSTM model training process, the validation data set is used to validate the model. The LSTM model that meets the performance requirements is used as the crack growth rate prediction model.

6. A method for predicting the life of a steel structure bridge after fatigue crack repair according to claim 5, characterized in that: The linear elastic fracture mechanics theory is introduced to establish the relationship between the crack growth rate and the range of stress intensity factors; the Paris formula is used to describe the relationship between the crack growth rate and calculate the range of stress intensity factors; Based on the Paris formula and the calculation formula of the stress intensity factor range, a remaining life prediction model is constructed. The crack growth rate predicted by the LSTM model is substituted into the Paris formula. The cumulative number of crack growth cycles is obtained by integration, and the remaining life prediction value of the repaired structure is obtained.

7. A method for predicting the life of a steel structure bridge after fatigue crack repair according to claim 6, characterized in that: Based on the predicted value of remaining life, a multi-level early warning is provided for the safety status of the repaired structure; Set the warning threshold and alarm threshold. When the remaining life prediction value is lower than the warning threshold, a warning signal is issued; when the remaining life prediction value is lower than the alarm threshold, an alarm signal is issued.

8. A life prediction system for a steel structure bridge after fatigue crack repair, which is used to implement a life prediction method for a steel structure bridge after fatigue crack repair as claimed in any one of claims 1 to 7, characterized in that: include: Data acquisition module, feature extraction module, data fusion module, remaining life prediction module and multi-level warning module; The data acquisition module is used to collect load data and stress data during the operation of the bridge structure after the fatigue cracks are repaired; The feature extraction module extracts dynamic response feature parameters based on the finite element analysis method and the stress distribution characteristics of the fatigue crack repair area; The data fusion module performs multi-dimensional data fusion and preprocessing on the dynamic response characteristic parameters to construct a data set for training the deep learning algorithm; The remaining life prediction module combines the deep learning algorithm and fracture mechanics theory to construct a crack growth rate prediction model and a remaining life prediction model; The multi-level warning module predicts the remaining life of the repaired structure and issues a multi-level warning based on the operating status of the bridge and the performance degradation law of the repaired area.

9. An electronic device, characterized in that: include: A processor and a memory, wherein the memory stores a computer program that can be called by the processor; the processor executes the method for predicting the life of a steel structure bridge after fatigue crack repair as described in any one of claims 1 to 7 by calling the computer program stored in the memory.

10. A computer-readable storage medium, characterized in that: Instructions are stored, and when the instructions are run on a computer, the computer is caused to execute a method for predicting the life of a steel structure bridge after fatigue crack repair as described in any one of claims 1 to 7.

Images