Cable bearing bridge structure damage identification method and device, storage medium and product

Abstract

The invention discloses a cable bearing bridge structure damage identification method and device, a storage medium and a product, and the method comprises the steps: constructing a sample data set, and enabling the input quantity of each sample data to be an image obtained through the splicing of time-frequency domain images after the acceleration response signals of a main beam and an inhaul cable are converted into the time-frequency domain images; a damage identification model is constructed, the damage identification model comprises an input layer, an initial convolution layer, a maximum pooling layer, a first combination layer, a second combination layer, a third combination layer, a fourth combination layer, a first global average pooling layer, a first full connection layer and an output layer which are connected in sequence, and the initial convolution layer is provided with six input channels; each combination layer comprises a convolution block, an SE attention mechanism and a residual block which are connected in sequence; training and verifying the damage identification model by using the sample data set to obtain a target damage identification model; and performing damage identification on the actual spliced image by using the target damage identification model. According to the invention, the damage identification precision is obviously improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of bridge health monitoring, and in particular relates to a method, equipment, storage medium and product for identifying damage of a cable-bearing bridge structure. Background Art

[0002] Traditional bridge health monitoring methods mainly rely on the measurement of physical quantities such as structural stress, strain, and displacement. Although these monitoring methods can provide certain structural status information, they have the following shortcomings:

[0003] (1) Manual inspection is costly and inefficient: Manual inspection requires a lot of manpower and time, and the inspection results are easily affected by subjective factors, making it difficult to ensure the comprehensiveness and accuracy of the inspection.

[0004] (2) Limitations of static testing methods: The health status of a structure is mainly assessed by measuring its static responses such as displacement and strain. However, these static response signals are often difficult to reflect the dynamic characteristics and potential damage of the structure during actual service.

[0005] (3) Limited detection accuracy and sensitivity: Traditional sensors have limited detection accuracy and sensitivity, making it difficult to capture tiny structural changes, resulting in limited detection accuracy.

[0006] (4) Complex data processing: Traditional methods generate large amounts of data, and processing and analyzing this data requires a lot of time and human resources.

[0007] (5) Poor real-time performance: Some monitoring methods cannot achieve real-time monitoring and cannot detect and handle sudden structural problems in a timely manner.

[0008] In recent years, with the development of sensor technology and data processing technology, bridge health monitoring methods based on vibration signals have gradually attracted attention. Vibration signals can reflect the dynamic response of bridges under load. By analyzing vibration signals, the dynamic characteristics of the structure can be obtained, thereby judging the health status of the structure. However, the existing bridge health monitoring methods based on vibration signals still face the following challenges in practical applications:

[0009] (1) Difficulty in feature extraction: Bridge vibration signals contain a large amount of noise and complex nonlinear characteristics. Traditional vibration feature extraction methods are difficult to effectively extract useful information.

[0010] (2) Inaccurate damage identification: Existing methods have low accuracy in identifying different types of structural damage, especially for initial minor damage.

[0011] (3) Acceleration sensors installed at key locations of the bridge can collect vibration signals of the bridge under external loads in real time, which can more comprehensively reflect the dynamic characteristics and damage of the bridge structure. However, the analysis method of a single sensor signal has certain limitations and it is difficult to fully utilize the advantages of multimodal data.

[0012] (4) Multi-sensor data fusion problem: Bridge health monitoring usually requires the deployment of multiple sensors. The data collected by sensors at different locations have differences in time domain and frequency domain characteristics. How to effectively fuse these multimodal data is a difficult problem. Summary of the invention

[0013] The purpose of the present invention is to provide a cable-bearing bridge structure damage identification method, equipment, storage medium and product to solve the problems that traditional methods are difficult to effectively extract useful information, have poor results in identifying minor damages, and have difficulty in fully utilizing multimodal data.

[0014] The present invention solves the above technical problems through the following technical solutions: A method for identifying damage to a cable-bearing bridge structure, comprising:

[0015] Constructing a sample data set; wherein each sample data in the sample data set includes an input quantity and an output quantity, the input quantity is a spliced ​​image, and the output quantity is a damage condition of the spliced ​​image; the spliced ​​image is obtained by splicing the acceleration response signals of the main beam and the acceleration response signals of the cable after they are converted into time-frequency domain images respectively;

[0016] Constructing a damage recognition model; wherein the damage recognition model includes an input layer, an initial convolution layer, a maximum pooling layer, a first combination layer, a second combination layer, a third combination layer, a fourth combination layer, a first global average pooling layer, a first fully connected layer and an output layer connected in sequence; the initial convolution layer has 6 input channels; each combination layer includes a convolution block, an SE attention mechanism and a residual block connected in sequence;

[0017] Using the sample data set to train and verify the damage identification model to obtain a target damage identification model;

[0018] Obtain actual acceleration response signals of the main beam and cables;

[0019] The actual acceleration response signals of the main beam and the cable are converted into time-frequency domain images respectively, and then the time-frequency domain images of the main beam and the cable are spliced ​​to obtain an actual spliced ​​image;

[0020] The target damage identification model is used to perform damage identification on the actual spliced ​​image.

[0021] Furthermore, a sample data set is constructed through field experiments, including:

[0022] Building a cable-supported bridge test model, setting a first sensor on the main beam of the cable-supported bridge test model, and setting a second sensor on the cable of the cable-supported bridge test model;

[0023] Simulating different damage conditions on the cable-bearing bridge test model, using a first sensor to collect acceleration response signals of the main beam under different damage conditions, and using a second sensor to collect acceleration response signals of the cable under different damage conditions;

[0024] Preprocess the acceleration response signals of the main beam and cables under different damage conditions;

[0025] Convert the pre-processed acceleration response signals of the main beam and cables into time-frequency domain images;

[0026] The time-frequency domain images of the main beam and the cable under the same damage condition are spliced ​​to obtain a spliced ​​image;

[0027] The sample data set is constructed according to the spliced ​​images and the damage conditions to which they belong.

[0028] Furthermore, the damage condition includes the damage position, vehicle speed, vehicle mass and damage degree; the vehicle speed is 0.5 m / s, 0.75 m / s or 1 m / s, and the vehicle mass is 5 kg, 10 kg or 20 kg; the cables are numbered in order from right to left;

[0029] The first damage condition: no damage, a combination of different vehicle speeds and different vehicle masses;

[0030] The second damage condition: the first cable is damaged, with different combinations of vehicle speeds and vehicle masses, and the damage degree is 10%, 20%, or 50%;

[0031] The third damage condition: the third cable is damaged, with different combinations of vehicle speeds and vehicle masses, and the damage degree is 10%, 20% or 50%;

[0032] The fourth damage condition: the sixth cable is damaged, with different combinations of vehicle speeds and vehicle masses, and the damage degree is 10%, 20%, or 50%;

[0033] The fifth damage condition: the first and fourth cables are damaged, with different combinations of vehicle speeds and vehicle masses, and the damage degree is 10%, 20% or 50%;

[0034] The sixth damage condition: the fifth and sixth cables are damaged, with a combination of different vehicle speeds and different vehicle masses, and the degree of damage is 10%, 20% or 50%.

[0035] Furthermore, the S transform is used to realize the conversion of time-frequency domain images. The specific formula is:

[0036]

[0037] Among them, S(τ,f) represents the time-frequency domain image, x(t) represents the acceleration response signal, ω(τ-t,f) represents the Gaussian window function, t represents time, τ represents the position of the Gaussian window on the time axis t, f represents the frequency of the acceleration response signal, and i represents the sign of the imaginary part.

[0038] Furthermore, the convolution block includes a first convolution layer, a first activation layer, a second convolution layer, a second activation layer, a third convolution layer and a batch normalization layer connected in sequence;

[0039] The number of residual blocks of the first combination layer is 2, the number of residual blocks of the second combination layer is 3, the number of residual blocks of the third combination layer is 5, and the number of residual blocks of the fourth combination layer is 2; each of the residual blocks includes a fourth convolutional layer, a third activation layer, a fifth convolutional layer, a fourth activation layer and a sixth convolutional layer connected in sequence.

[0040] Furthermore, the SE attention mechanism includes a second global average pooling layer, a second fully connected layer, a third fully connected layer and a reweighted layer which are connected in sequence.

[0041] Furthermore, the damage identification model is trained using the sample data set, including:

[0042] Using the initial convolutional layer to extract features from the spliced ​​image received by the input layer to obtain a first feature map;

[0043] Performing maximum pooling on the first feature map using the maximum pooling layer to obtain a second feature map;

[0044] Using the first combined layer to perform feature extraction on the second feature map to obtain a third feature map;

[0045] Using the second combined layer to perform feature extraction on the third feature map to obtain a fourth feature map;

[0046] Using the third combined layer to perform feature extraction on the fourth feature map to obtain a fifth feature map;

[0047] Using the fourth combined layer to perform feature extraction on the fifth feature map to obtain a sixth feature map;

[0048] Using the first global average pooling layer to perform global average pooling on the sixth feature map to obtain a single feature;

[0049] Using the first fully connected layer to flatten the single feature and map it to the classification space to obtain the predicted damage category;

[0050] A loss value is calculated according to the predicted damage category and the damage condition of the stitched image, and the parameters of the damage recognition model are adjusted according to the loss value until the number of training times is reached or the loss value is less than the set loss accuracy.

[0051] Based on the same concept, the present invention also provides an electronic device, including a memory, a processor, and a computer program / instructions stored in the memory, wherein the processor executes the computer program / instructions to implement the cable-supported bridge structure damage identification method as described above.

[0052] Based on the same concept, the present invention also provides a computer-readable storage medium on which a computer program / instruction is stored. When the computer program / instruction is executed by a processor, the cable-supported bridge structure damage identification method as described above is implemented.

[0053] Based on the same concept, the present invention also provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the cable-supported bridge structure damage identification method as described above.

[0054] Beneficial Effects

[0055] Compared with the prior art, the advantages of the present invention are:

[0056] The present invention converts the acceleration response signals of the main beam and the cable into time-frequency domain images, fully retains the time dependence and nonlinear characteristics of the acceleration response signals, and is conducive to the damage identification model to more accurately capture and analyze the slight changes and early damage of the bridge structure; the time-frequency domain images of the main beam and the cable are spliced ​​to form a spliced ​​image with high-level features and low-level details, which can provide richer features, further helping the damage identification model to capture more feature information, and thus help to identify slight damage; the initial convolution layer of the traditional residual neural network model is improved so that the model can reason about the spliced ​​image; the SE attention mechanism is integrated in each combined layer, which improves the learning ability of the damage identification model, alleviates the gradient vanishing problem, and improves the recognition accuracy of bridge structure damage. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] In order to more clearly illustrate the technical solution of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only one embodiment of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0058] Figure 1 is a flow chart of a method for identifying damage to a cable-bearing bridge structure in an embodiment of the present invention;

[0059] Figure 2 is a schematic diagram of cable numbering of a cable-supported bridge test model in an embodiment of the present invention;

[0060] Figure 3 1 is a schematic diagram of the acceleration response signal of the main beam in the embodiment of the present invention converted into a time-frequency domain image; wherein the color code number represents the amplitude of the midpoint of the image;

[0061] Figure 4 It is a schematic diagram of converting the acceleration response signal of the cable in the embodiment of the present invention into a time-frequency domain image;

[0062] Figure 5 is a diagram of the improved residual neural network model architecture in an embodiment of the present invention;

[0063] Figure 6 is an image tensor with 6 channels in the embodiment of the present invention;

[0064] Figure 7 is the output feature map of 64 channels in an embodiment of the present invention;

[0065] Figure 8 is a feature extraction thermal visualization diagram of the last residual block in the second combination layer, the third combination layer, and the fourth combination layer in an embodiment of the present invention;

[0066] Fig. 9 is a recognition accuracy curve of the recognition method of the present invention (stitched image + improved residual neural network model) and the traditional recognition method (main beam time-frequency domain image + traditional residual neural network model) in the embodiment of the present invention;

[0067] Fig.10 : is the loss value curve of the recognition method of the present invention (stitched image + improved residual neural network model) and the traditional recognition method (main beam time-frequency domain image + traditional residual neural network model) in the embodiment of the present invention;

[0068] Fig.11 It is a visualization diagram of damage categories of the main beam time-frequency domain image + traditional residual neural network model in an embodiment of the present invention;

[0069] Fig.12 This is a visualization diagram of the damage category of the cable time-frequency domain image + the traditional residual neural network model in the embodiment of the present invention.

[0070] Fig.13 It is a visualization diagram of damage categories of the spliced ​​image + improved residual neural network model in an embodiment of the present invention;

[0071] Fig.14It is a visualization diagram of the damage recognition accuracy of the recognition method of the present invention and the traditional recognition method in the embodiment of the present invention. DETAILED DESCRIPTION

[0072] The following is a clear and complete description of the technical solutions in the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0073] The technical solution of the present application is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0074] like Figure 1 As shown, a cable-bearing bridge structure damage identification method provided by an embodiment of the present invention comprises the following steps:

[0075] Step 1: Build a sample dataset.

[0076] The sample data set includes multiple sample data, each sample data includes an input quantity and an output quantity, the input quantity is a spliced ​​image, and the output quantity is a damage condition of the spliced ​​image; wherein the spliced ​​image is obtained by converting the acceleration response signal of the main beam and the acceleration response signal of the cable into time-frequency domain images respectively and then splicing them together.

[0077] The sample data set can be constructed by a finite element simulation method or by a field test. In a specific embodiment of the present invention, constructing the sample data set by a field test specifically includes:

[0078] Step 1.1: Build a cable-supported bridge test model, set a first sensor on the main beam of the cable-supported bridge test model, and set a second sensor on the cable of the cable-supported bridge test model.

[0079] like Figure 2 As shown, the cable-bearing bridge test model includes a main beam and cables. The cables are numbered from right to left, and the 8 cables are numbered S1 to S8. A first sensor is set at the middle span of the main beam, and a second sensor is set on any cable. In this embodiment, the second sensor is set on the fifth cable S5. The first sensor collects the acceleration response signal of the main beam, and the second sensor collects the acceleration response signal of the cable. During the test, the setting positions of the first sensor and the second sensor remain unchanged.

[0080] In this embodiment, the first sensor and the second sensor are piezoelectric acceleration sensors with a sensitivity of 100 mV / g, a frequency range of 0.5 to 7000 Hz, and a measurement range of ±50 g.

[0081] The present invention uses a dynamic data acquisition device to obtain the acceleration response signals collected by the first sensor and the second sensor under different damage conditions, and exports data from the dynamic data acquisition device and saves it in an Excel table or other data format to facilitate subsequent data processing and analysis. The dynamic data acquisition device of this embodiment uses a DH3822 strain tester, and a DHDAS dynamic signal acquisition and analysis system is installed on the DH3822 strain tester, with a sampling rate of 500Hz.

[0082] Step 1.2: Simulate different damage conditions on the cable-bearing bridge test model, use the first sensor to collect the acceleration response signal of the main beam under different damage conditions, and use the second sensor to collect the acceleration response signal of the cable under different damage conditions.

[0083] Table 1 Different damage conditions

[0084] Working condition number Damage location Speed Vehicle weight Damage level Condition 1 healthy 0.5m / s,0.75m / s,1m / s 5kg,10kg,20kg / Condition 2 S1 0.5m / s,0.75m / s,1m / s 5kg,10kg,20kg 10％,20％,50％ Condition 3 S3 0.5m / s,0.75m / s,1m / s 5kg,10kg,20kg 10％,20％,50％ Condition 4 S6 0.5m / s,0.75m / s,1m / s 5kg,10kg,20kg 10％,20％,50％ Condition 5 S1,S4 0.5m / s,0.75m / s,1m / s 5kg,10kg,20kg 10％,20％,50％ Condition 6 S5,S6 0.5m / s,0.75m / s,1m / s 5kg,10kg,20kg 10％,20％,50％

[0085] As shown in Table 1, the damage conditions of this embodiment include damage location, vehicle speed, vehicle mass, and damage degree. The moving vehicle is used as an external load of the bridge. The vehicle is driven by a motor, and the vehicle speed is controlled by controlling the speed of the motor. The vehicle speed is 0.5 m / s, 0.75 m / s, or 1 m / s, and the vehicle mass is 5 kg, 10 kg, or 20 kg. The first damage condition: no damage, combination of different vehicle speeds and different vehicle masses; the second damage condition: the first cable S1 is damaged, the combination of different vehicle speeds and different vehicle masses, the damage degree is 10%, 20% or 50%; the third damage condition: the third cable S3 is damaged, the combination of different vehicle speeds and different vehicle masses, the damage degree is 10%, 20% or 50%; the fourth damage condition: the sixth cable S6 is damaged, the combination of different vehicle speeds and different vehicle masses, the damage degree is 10%, 20% or 50%; the fifth damage condition: the first and fourth cables S1 / S4 are damaged, the combination of different vehicle speeds and different vehicle masses, the damage degree is 10%, 20% or 50%; the sixth damage condition: the fifth and sixth cables S5 / S6 are damaged, the combination of different vehicle speeds and different vehicle masses, the damage degree is 10%, 20% or 50%.

[0086] Each damage condition forms multiple sub-conditions through different combinations of vehicle speed, vehicle mass and damage degree. Taking the second damage condition as an example, the sub-conditions formed by the second damage condition include:

[0087] (1) S1 damage, vehicle speed is 0.5 m / s, vehicle weight is 5 kg, and damage degree is 10%;

[0088] (2) S1 damage, vehicle speed is 0.75 m / s, vehicle weight is 5 kg, and damage degree is 10%;

[0089] (3) S1 damage, vehicle speed is 1 m / s, vehicle weight is 5 kg, and damage degree is 10%;

[0090] (4) S1 damage, vehicle speed is 0.5 m / s, vehicle weight is 10 kg, and damage degree is 10%;

[0091] (5) S1 damage, vehicle speed is 0.5 m / s, vehicle weight is 20 kg, and damage degree is 10%;

[0092] (6) S1 damage, vehicle speed is 0.75 m / s, vehicle weight is 10 kg, and damage degree is 10%;

[0093] (7) S1 damage, vehicle speed is 0.75 m / s, vehicle weight is 20 kg, and damage degree is 10%;

[0094] (8) S1 damage, vehicle speed is 1 m / s, vehicle weight is 10 kg, and damage degree is 20%;

[0095] (9) S1 damage, vehicle speed is 1 m / s, vehicle weight is 20 kg, and damage degree is 20%;

[0096] (10) S1 damage, vehicle speed is 0.5 m / s, vehicle weight is 5 kg, and damage degree is 20%;

[0097] (11) S1 damage, vehicle speed is 0.75 m / s, vehicle weight is 5 kg, and damage degree is 20%;

[0098] (12) S1 damage, vehicle speed is 1 m / s, vehicle weight is 5 kg, and damage degree is 20%;

[0099] (13) S1 damage, vehicle speed is 0.5 m / s, vehicle weight is 10 kg, and damage degree is 20%;

[0100] (14) S1 damage, vehicle speed is 0.5 m / s, vehicle weight is 20 kg, and damage degree is 20%;

[0101] (15) S1 damage, vehicle speed is 0.75 m / s, vehicle weight is 10 kg, and damage degree is 20%;

[0102] (16) S1 damage, vehicle speed is 0.75 m / s, vehicle weight is 20 kg, and damage degree is 20%;

[0103] (17) S1 damage, vehicle speed is 1 m / s, vehicle weight is 10 kg, and damage degree is 20%;

[0104] (18) S1 damage, vehicle speed is 1 m / s, vehicle weight is 20 kg, and damage degree is 20%;

[0105] (19) S1 damage, vehicle speed is 0.5 m / s, vehicle weight is 5 kg, and damage degree is 50%;

[0106] (20) S1 damage, vehicle speed is 0.75 m / s, vehicle weight is 5 kg, and damage degree is 50%;

[0107] (21) S1 damage, vehicle speed is 1 m / s, vehicle weight is 5 kg, and damage degree is 50%;

[0108] (22) S1 damage, vehicle speed is 0.5 m / s, vehicle weight is 10 kg, and damage degree is 50%;

[0109] (23) S1 damage, vehicle speed is 0.5 m / s, vehicle weight is 20 kg, and damage degree is 50%;

[0110] (24) S1 damage, vehicle speed is 0.75 m / s, vehicle weight is 10 kg, and damage degree is 50%;

[0111] (25) S1 damage, vehicle speed is 0.75 m / s, vehicle weight is 20 kg, and damage degree is 50%;

[0112] (26) S1 damage, vehicle speed is 1m / s, vehicle weight is 10kg, and damage degree is 50%;

[0113] (27) S1 damage, vehicle speed is 1m / s, vehicle weight is 20kg, and damage degree is 50%.

[0114] Through the combination of different vehicle speeds, vehicle weights and damage degrees, multiple acceleration response signals can be obtained under each damage condition.

[0115] Step 1.3: Preprocess the acceleration response signals of the main beam and cables under different damage conditions.

[0116] In a specific implementation of the present invention, preprocessing includes normalization processing, calculating the maximum and minimum values ​​of the acceleration response signal, and normalizing the amplitude of each acceleration response signal to map it to the [-1, 1] interval to ensure consistency.

[0117] Step 1.4: Convert the preprocessed acceleration response signals of the main beam and the cable into time-frequency domain images to obtain the time-frequency domain images of the main beam and the cable.

[0118] In a specific implementation of the present invention, the pre-processed acceleration response signals of the main beam and the cable are converted into time-frequency domain images using S transformation. The specific formula is:

[0119]

[0120]

[0121] Among them, S(τ,f) represents the time-frequency domain image, x(t) represents the acceleration response signal, ω(τ-t,f) represents the Gaussian window function, t represents time, τ represents the position of the Gaussian window on the time axis t, f represents the frequency of the acceleration response signal, and i represents the sign of the imaginary part. The time-frequency domain images of the main beam and the cable are shown in Figure 3 and Figure 4 shown.

[0122] Each point in the time-frequency domain image corresponds to the intensity value of the acceleration response signal at a specific time and frequency. The S transform converts the one-dimensional acceleration response signal into a two-dimensional image, retaining the time dependence and nonlinear characteristics of the acceleration response signal. While retaining the characteristic information in the original time series, it better displays the time-frequency characteristics of the signal, which is conducive to the damage identification model to more accurately capture and analyze the subtle changes and early damage of the bridge structure, and significantly improve the accuracy and sensitivity of monitoring.

[0123] Step 1.5: Splice the time-frequency domain images of the main beam and the cable under the same damage condition to obtain a spliced ​​image.

[0124] The PLL library is used to load the main beam time-frequency domain images and cable time-frequency domain images under the same damage condition, and their sizes are adjusted to 224×224. The function torchvision.transforms.ToTensor() is used to convert the resized main beam time-frequency domain images and cable time-frequency domain images into tensor format. The tensor size is C×224×224, where C represents the number of image channels, and C is equal to 3. The main beam time-frequency domain images and cable time-frequency domain images in tensor format are spliced ​​in the channel dimension to form a spliced ​​image with 6 channels and in tensor format.

[0125] The specific process of stitching is: through the torch.concat function, two 3×224×224 images (main beam time-frequency domain image and cable time-frequency domain image) are stacked along the channel dimension to form a stitched image with 6 channels. The size of the stitched image is 6×224×224.

[0126] Step 1.6: Construct a sample dataset based on the stitched images and their corresponding damage conditions.

[0127] For each damage condition, the acceleration response signals of 200 main beams and 200 cables are collected. A total of 2400 sample data are generated for the six conditions, that is, the sample data set contains 2400 sample data. The sample data set is divided into a training data set, a verification data set and a test data set according to the ratio of 6:2:2. The damage identification model is trained with the training data set, and the trained damage identification model is verified with the verification data set to obtain the target damage identification model.

[0128] Step 2: Construct damage identification model;

[0129] In a specific embodiment of the present invention, the damage identification model adopts an improved residual neural network model, such as Figure 5 As shown, the improved residual neural network model includes an input layer, an initial convolution layer new_conv1, a maximum pooling layer Maxpool1, a first combination layer Stage1, a second combination layer Stage2, a third combination layer Stage3, a fourth combination layer Stage4, a first global average pooling layer Avgpool, a first fully connected layer FC and an output layer connected in sequence; the initial convolution layer new_conv1 has 6 input channels; each combination layer (i.e., the first combination layer Stage1, the second combination layer Stage2, the third combination layer Stage3, and the fourth combination layer Stage4) includes a convolution block Conv Block, a SE attention mechanism and a residual block Identity Block connected in sequence.

[0130] When constructing the damage identification model, first load a traditional residual neural network model pre-trained on a large-scale dataset and extract the parameters of the initial convolutional layer conv1 of the traditional residual neural network model (i.e., transfer learning TL). The initial convolutional layer conv1 has 3 input channels and outputs 64 feature maps. Since the spliced ​​image obtained by splicing the time-frequency domain images of the main beam and the cable is a tensor with 6 input channels (such as Figure 6As shown), the initial convolution layer conv1 of the residual neural network model cannot be applied to the spliced ​​image. The present invention improves the initial convolution layer conv1, and recreates a new initial convolution layer new_conv1, which has 6 input channels and outputs 64 feature maps. The parameters of the initial convolution layer conv1 are copied to the first 3 channels and the last 3 channels of the initial convolution layer new_conv1, that is, new_conv1.weight[:,:3,:,:]=conv1.weight, new_conv1.weight[:,3:6,:,:]=conv1.weight, weight represents the weight parameter, and : represents the assignment of values ​​to all spatial positions and convolution kernels. The gradient is not calculated during the copying process to ensure that the model parameters are not updated during the weight parameter copying process.

[0131] The parameters of the initial convolutional layer new_conv1 are:

[0132] new_conv1=nn.Conv2d(in_channels=6, out_channels=64, kernel_size=7, stride=2, padding=3) (3)

[0133] Among them, nn.Conv2d is a two-dimensional convolutional layer class defined in PyTorch; n_channels=6 means that the number of input channels is 6, that is, the number of channels of the spliced ​​image is 6; out_channels=64 means that the number of output channels is 64, that is, the number of channels of the feature map obtained after processing by the initial convolution layer new_conv1 is 64; kernel_size=7 means that the size of the convolution kernel is 7×7; stride=2 means that the stride is 2, that is, the step length of each sliding of the convolution kernel on the spliced ​​image is 2, that is, each time it moves 2 pixels; padding=3 means that the padding pixels are 3 pixels to prevent the image size from shrinking too much after the convolution operation.

[0134] Replacing the initial convolution layer conv1 of the residual neural network model with the initial convolution layer new_conv1 can process the spliced ​​image. The initial convolution layer new_conv1 uses 64 convolution kernels of size 7×7, a stride of 2, and a padding of 3. The output feature map size is (64, 112, 112). The 6-channel spliced ​​image is input to the initial convolution layer new_conv1 for forward propagation, and the output 64-channel feature map result (i.e., the first feature map) is obtained, as shown in Figure 7 shown.

[0135] The maximum pooling layer Maxpool1 performs a maximum pooling operation on the first feature map output by the initial convolution layer new_conv1 to obtain the second feature map, the size of which is (64, 56, 56). The pooling kernel size of the maximum pooling layer Maxpool1 is 3×3, the stride is 2, and the border padding is 0.

[0136] Each combination layer of the traditional residual neural network model only includes a convolution block Conv Block and a residual block Identity Block. The present invention improves the combination layer by inserting an SE attention mechanism between the convolution block Conv Block and the residual block Identity Block, that is, each combination layer includes a convolution block Conv Block, an SE attention mechanism and a residual block Identity Block connected in sequence.

[0137] In this embodiment, each convolution block Conv Block includes a first convolution layer, a first activation layer, a second convolution layer, a second activation layer, a third convolution layer and a batch normalization layer connected in sequence. The specific formula of the convolution block Conv Block is:

[0138] Xout = BN(Conv1×1(ReLU(Conv3×3(ReLU(Conv1×1(Xin)))))) (4)

[0139] In formula (4), Xin represents the input feature map of the convolution block Conv Block, Xout represents the output feature map of the convolution block ConvBlock, Conv1×1 represents that the convolution kernel size of the first convolution layer and the third convolution layer is 1×1, Conv3×3 represents that the convolution kernel size of the second convolution layer is 3×3, ReLU represents the ReLU activation function of the first activation layer and the second activation layer, and BN represents the batch normalization layer.

[0140] Each residual block Identity Block includes the fourth convolution layer, the third activation layer, the fifth convolution layer, the fourth activation layer and the sixth convolution layer connected in sequence. The specific formula of the residual block Identity Block is:

[0141] Xout = Conv1×1(ReLU(Conv3×3(ReLU(Conv1×1(Xin))))) (5)

[0142] In formula (5), Xin represents the input feature map of the residual block Identity Block, Xout represents the output feature map of the residual block Identity Block, Conv1×1 represents that the convolution kernel size of the fourth convolution layer and the sixth convolution layer is 1×1, Conv3×3 represents that the convolution kernel size of the fifth convolution layer is 3×3, and ReLU represents the ReLU activation function of the third activation layer and the fourth activation layer.

[0143] SE attention mechanism, namely Squeeze-and-Excitation attention mechanism, is an attention mechanism widely used in deep learning, which aims to improve the performance and accuracy of the model. SE attention mechanism includes the second global average pooling layer, the second fully connected layer, the third fully connected layer and the reweighted layer connected in sequence. The second global average pooling layer is used to perform global average pooling on the feature map input to the SE attention mechanism, compressing the feature map from (C, H, W) to (C, 1, 1). The specific formula is:

[0144]

[0145] Among them, C represents the number of channels of the feature map, H represents the height of the feature map, W represents the width of the feature map, and x c,i,j represents the feature value of the cth channel at position (i, j) in the feature map input to the SE attention mechanism, z c represents the compressed features of the c-th channel, c = 1, 2, …, C.

[0146] The feature map output by the second global average pooling layer is processed using the second fully connected layer (with ReLU activation function). The specific formula is:

[0147] e c =ReLU(W1·z c ) (7)

[0148] Among them, W1 represents the weight matrix of the second fully connected layer, e c Represents the ReLU activation feature of the cth channel.

[0149] The third fully connected layer (with Sigmoid activation function) is used to process the feature map output by the second fully connected layer. The specific formula is:

[0150] s c =Sigmoid(W2·e c ) (8)

[0151] Among them, W2 represents the weight matrix of the third fully connected layer, s c Represents the Sigmoid activation feature of the c-th channel.

[0152] Use the reweighted layer to activate the Sigmoid feature s of each channel c Multiply it with the feature map input to the SE attention mechanism to obtain the weighted feature. The specific formula is:

[0153]

[0154] in, Represents the weighted feature of the cth channel.

[0155] In this embodiment, the number of residual blocks Identity Blocks in the first combination layer is 2, the number of residual blocks Identity Blocks in the second combination layer is 3, the number of residual blocks Identity Blocks in the third combination layer is 5, and the number of residual blocks Identity Blocks in the fourth combination layer is 2. Each combination layer includes a convolution block Conv Block and multiple residual blocks Identity Block. The output feature map of each combination layer is used as the input feature map of the next combination layer, and is downsampled according to a specific step size to achieve layer-by-layer feature extraction. The specific formulas for each combination layer are:

[0156] Xstage1=IdentityBlock(IdentityBlock(SEModule(ConvBlock(Xin)))(10)

[0157] Xstage2=IdentityBlock(IdentityBlock(IdentityBlock(SEModule(ConvBlock(Xstage1))))) (11)

[0158] Xstage3=IdentityBlock(IdentityBlock(IdentityBlock(IdentityBlock(IdentityBlock(SEModule(ConvB ock(Xstage2))))))(12)

[0159] Xstage4=IdentityBlock(IdentityBlock(SEModule(ConvBlock(Xstage3)))(13)

[0160] Among them, Xin represents the feature map input to the first combination layer stage1, Xstage1 represents the output feature map of the first combination layer stage1, Xstage2 represents the output feature map of the second combination layer stage2, Xstage3 represents the output feature map of the third combination layer stage3, and Xstage4 represents the output feature map of the fourth combination layer stage4 (i.e., the sixth feature map). Figure 8 The feature extraction heat map of the last residual block IdentityBlock in the second combination layer, the third combination layer, and the fourth combination layer is shown, which can intuitively show the areas that the model focuses on.

[0161] The first global average pooling layer is used to perform global average pooling on the output feature map of the fourth combination layer to obtain a single feature; the single feature is flattened using the first fully connected layer and mapped to the classification space to obtain the predicted damage category.

[0162] Step 3: Use the sample data set to train and verify the damage identification model to obtain the target damage identification model.

[0163] In a specific embodiment of the present invention, the damage identification model (i.e., the improved residual neural network model) is trained using a sample data set, including:

[0164] Step 3.1: Use the initial convolutional layer new_conv1 to extract features from the spliced ​​image received by the input layer to obtain the first feature map;

[0165] Step 3.2: Use the maximum pooling layer Maxpool1 to perform maximum pooling on the first feature map to obtain the second feature map;

[0166] Step 3.3: Use the first combination layer Stage1 to extract features from the second feature map to obtain a third feature map;

[0167] Step 3.4: Use the second combination layer Stage2 to extract features from the third feature map to obtain a fourth feature map;

[0168] Step 3.5: Use the third combination layer Stage3 to extract features from the fourth feature map to obtain a fifth feature map;

[0169] Step 3.6: Use the fourth combination layer Stage4 to extract features from the fifth feature map to obtain a sixth feature map;

[0170] Step 3.7: Use the first global average pooling layer Avgpool to perform global average pooling on the sixth feature map to obtain a single feature;

[0171] Step 3.8: Use the first fully connected layer FC to flatten the single feature and map it to the classification space to obtain the predicted damage category;

[0172] Step 3.9: Calculate the loss value based on the predicted damage category and the damage condition of the stitched image, and adjust the parameters of the damage recognition model according to the loss value until the number of training times is reached or the loss value is less than the set loss accuracy.

[0173] In this embodiment, cross entropy is used as the loss function to measure the difference between the predicted damage category and the damage condition of the spliced ​​image. In this embodiment of the present invention, there are 6 damage categories. For each sample data, the damage recognition model generates a prediction vector To represent the predicted probability of the six damage categories. The damage condition of the spliced ​​image is represented by the actual vector y. In the actual vector y, only the elements corresponding to the damage condition of the spliced ​​image are 1, and the rest are 0. The specific formula of the loss function is:

[0174]

[0175] Among them, L represents the cross entropy loss, y i represents the actual label value (0 or 1) of the i-th category, Represents the predicted probability of the i-th class, and N represents the number of damage classes. Calculate the cross entropy loss for each sample data in each batch, and then calculate the derivative of the cross entropy loss with respect to the model parameters. The cross entropy loss function quantifies the error of the model by calculating the gap between the actual label and the predicted probability. The smaller the loss value, the closer the model's predicted probability is to the actual label value.

[0176] Step 4: Obtain the actual acceleration response signals of the main beam and cables.

[0177] Step 5: Convert the actual acceleration response signals of the main beam and the cable into time-frequency domain images respectively, and then splice the time-frequency domain images of the main beam and the cable to obtain the actual spliced ​​image.

[0178] Step 5 is similar to steps 1.4 and 1.5.

[0179] Step 6: Use the target damage recognition model to perform damage recognition on the actual stitching image.

[0180] In order to verify the accuracy of the target damage identification model of the present invention, the damage identification accuracy of the method of the present invention is compared with the damage identification accuracy of the traditional residual neural network model. The comparison results are shown in Table 2 and Fig. 9 , Fig.10 As shown in Table 2, Fig. 9 and Fig.10 It can be seen that the present invention has high damage identification accuracy and low loss value.

[0181] Table 2 Comparison of damage recognition accuracy of different input images

[0182]

[0183] In order to visualize the damage categories of the time-frequency domain images, t-distributed stochastic neighbor embedding (t-SNE) is used to visualize the damage categories of the time-frequency domain images of the main beam, the time-frequency domain images of the cables, and the spliced ​​images, as shown in Figure 2. Fig.11 , Fig.12 , Fig.13 As shown. t-Distributed Stochastic Neighbor Embedding (t-SNE) is a nonlinear dimensionality reduction technique used to map high-dimensional data to a two-dimensional space for easy visualization. The horizontal axis (X-axis) represents the first new coordinate axis after dimensionality reduction, and the vertical axis (Y-axis) represents the second new coordinate axis after dimensionality reduction. Figure 11 to Figure 13 The solid dots in the figure represent the damage categories corresponding to the damage conditions. That is, the horizontal axis (X-axis) represents the position of the data in the first main direction in the low-dimensional embedding space, which can also be interpreted as the distribution of the data on the first significant feature; the vertical axis (Y-axis) represents the position of the data in the second main direction in the low-dimensional embedding space, which can also be interpreted as the distribution of the data on another significant feature. The horizontal and vertical axes themselves have no specific physical meaning or units. These coordinates are generated when the high-dimensional data points are embedded into the low-dimensional space through the t-SNE algorithm.

[0184] Depend on Fig.13 It can be seen that although the solid dots of the damage categories overlap slightly, the method of the present invention can correctly identify the clusters of the same damage category, indicating that the method of the present invention has good damage identification ability. In contrast, there are different degrees of confusion in the damage categories of the main beam time-frequency domain image and the cable time-frequency domain image, the cluster boundaries are fuzzy, and the data of different damage categories are close or even overlapped. Therefore, the damage identification ability of the traditional recognition method is poor. The traditional recognition method is prone to damage category errors, thereby reducing the accuracy of damage identification. Damage category visualization provides an intuitive result of the distribution of damage identification results. The results show that the method of the present invention has superior damage identification performance and can effectively distinguish various types of data.

[0185] Another way to express Table 2 is as follows Fig.14 As shown by Fig.14 The high damage recognition accuracy of the method of the present invention can be intuitively seen. Compared with the traditional recognition method, the present invention significantly improves the damage recognition accuracy, indicating the effectiveness of introducing the SE attention mechanism in each combination layer and splicing the time-frequency domain images of the main beam and the cable.

[0186] Example 2

[0187] An embodiment of the present invention also provides an electronic device, which includes: a memory, a processor, and a computer program / instructions stored in the memory, and the processor executes the computer program / instructions to implement the cable-bearing bridge structure damage identification method in the embodiment of the present application.

[0188] Although not shown, the electronic device includes a processor, which can perform various appropriate operations and processes according to the program and / or data stored in the read-only memory (ROM) or the program and / or data loaded from the storage portion into the random access memory (RAM). The processor can be a multi-core processor, or it can include multiple processors. In some embodiments, the processor can include a general main processor and one or more special coprocessors, such as a central processing unit, a graphics processing unit (GPU), a neural network processor (NPU), a digital signal processor (DSP), etc. In RAM, various programs and data required for device operation are also stored. The processor, ROM and RAM are connected to each other via a bus. The input / output (I / O) interface is also connected to the bus.

[0189] The processor and memory are used together to execute the program / instructions stored in the memory. When the program / instructions are executed by the computer, the methods, steps or functions described in the above embodiments can be implemented.

[0190] Although not shown, an embodiment of the present invention further provides a computer-readable storage medium having a computer program / instruction stored thereon, which, when executed by a processor, implements the cable-bearing bridge structure damage identification method in the embodiment of the present application.

[0191] Storage media in embodiments of the present invention include permanent and non-permanent, removable and non-removable items that can be implemented by any method or technology to store information. Examples of storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission medium that can be used to store information that can be accessed by a computing device.

[0192] Readable storage media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include temporary computer-readable media (transitory media), such as modulated data signals and carrier waves.

[0193] Although not shown, an embodiment of the present invention further provides a computer program product, including: a computer program / instruction, which, when executed by a processor, implements the cable-bearing bridge structure damage identification method in the embodiment of the present application.

[0194] What is disclosed above is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or modifications within the technical scope disclosed in the present invention, which should be covered within the protection scope of the present invention.

Claims

1. A method for identifying damage to a cable-supported bridge structure, characterized in that: The identification method comprises: Constructing a sample data set; wherein each sample data in the sample data set includes an input quantity and an output quantity, the input quantity is a spliced ​​image, and the output quantity is a damage condition of the spliced ​​image; the spliced ​​image is obtained by splicing the acceleration response signals of the main beam and the acceleration response signals of the cable after they are converted into time-frequency domain images respectively; Constructing a damage recognition model; wherein the damage recognition model includes an input layer, an initial convolution layer, a maximum pooling layer, a first combination layer, a second combination layer, a third combination layer, a fourth combination layer, a first global average pooling layer, a first fully connected layer and an output layer connected in sequence; the initial convolution layer has 6 input channels; each combination layer includes a convolution block, an SE attention mechanism and a residual block connected in sequence; Using the sample data set to train and verify the damage identification model to obtain a target damage identification model; Obtain actual acceleration response signals of the main beam and cables; The actual acceleration response signals of the main beam and the cable are converted into time-frequency domain images respectively, and then the time-frequency domain images of the main beam and the cable are spliced ​​to obtain an actual spliced ​​image; The target damage identification model is used to perform damage identification on the actual spliced ​​image.

2. The cable-supported bridge structure damage identification method according to claim 1, characterized in that: A sample data set was constructed through field experiments, including: Building a cable-supported bridge test model, setting a first sensor on the main beam of the cable-supported bridge test model, and setting a second sensor on the cable of the cable-supported bridge test model; Simulating different damage conditions on the cable-bearing bridge test model, using a first sensor to collect acceleration response signals of the main beam under different damage conditions, and using a second sensor to collect acceleration response signals of the cable under different damage conditions; Preprocess the acceleration response signals of the main beam and cables under different damage conditions; Convert the pre-processed acceleration response signals of the main beam and cables into time-frequency domain images; The time-frequency domain images of the main beam and the cable under the same damage condition are spliced ​​to obtain a spliced ​​image; The sample data set is constructed according to the spliced ​​images and the damage conditions to which they belong.

3. The cable-supported bridge structure damage identification method according to claim 1, characterized in that: The damage condition includes the damage position, vehicle speed, vehicle mass and damage degree; the vehicle speed is 0.5m / s, 0.75m / s or 1m / s, and the vehicle mass is 5kg, 10kg or 20kg; the cables are numbered from right to left; The first damage condition: no damage, a combination of different vehicle speeds and different vehicle masses; The second damage condition: the first cable is damaged, with different combinations of vehicle speeds and vehicle masses, and the damage degree is 10%, 20%, or 50%; The third damage condition: the third cable is damaged, with different combinations of vehicle speeds and vehicle masses, and the damage degree is 10%, 20% or 50%; The fourth damage condition: the sixth cable is damaged, with different combinations of vehicle speeds and vehicle masses, and the damage degree is 10%, 20%, or 50%; The fifth damage condition: the first and fourth cables are damaged, with different combinations of vehicle speeds and vehicle masses, and the damage degree is 10%, 20% or 50%; The sixth damage condition: the fifth and sixth cables are damaged, with a combination of different vehicle speeds and different vehicle masses, and the degree of damage is 10%, 20% or 50%.

4. The cable-supported bridge structure damage identification method according to claim 1 or 2, characterized in that: The S transform is used to realize the conversion of time-frequency domain images. The specific formula is: Among them, S(τ,f) represents the time-frequency domain image, x(t) represents the acceleration response signal, ω(τ-t,f) represents the Gaussian window function, t represents time, τ represents the position of the Gaussian window on the time axis t, f represents the frequency of the acceleration response signal, and i represents the sign of the imaginary part.

5. The cable-supported bridge structure damage identification method according to claim 1, characterized in that: The convolution block includes a first convolution layer, a first activation layer, a second convolution layer, a second activation layer, a third convolution layer and a batch normalization layer connected in sequence; The number of residual blocks of the first combination layer is 2, the number of residual blocks of the second combination layer is 3, the number of residual blocks of the third combination layer is 5, and the number of residual blocks of the fourth combination layer is 2; each of the residual blocks includes a fourth convolutional layer, a third activation layer, a fifth convolutional layer, a fourth activation layer and a sixth convolutional layer connected in sequence.

6. The cable-supported bridge structure damage identification method according to claim 1, characterized in that: The SE attention mechanism includes a second global average pooling layer, a second fully connected layer, a third fully connected layer and a reweighted layer connected in sequence.

7. The cable-supported bridge structure damage identification method according to claim 1, characterized in that: Using the sample data set to train the damage identification model includes: Using the initial convolutional layer to extract features from the spliced ​​image received by the input layer to obtain a first feature map; Performing maximum pooling on the first feature map using the maximum pooling layer to obtain a second feature map; Using the first combined layer to perform feature extraction on the second feature map to obtain a third feature map; Using the second combined layer to perform feature extraction on the third feature map to obtain a fourth feature map; Using the third combined layer to perform feature extraction on the fourth feature map to obtain a fifth feature map; Using the fourth combined layer to perform feature extraction on the fifth feature map to obtain a sixth feature map; Using the first global average pooling layer to perform global average pooling on the sixth feature map to obtain a single feature; Using the first fully connected layer to flatten the single feature and map it to the classification space to obtain the predicted damage category; A loss value is calculated according to the predicted damage category and the damage condition of the stitched image, and the parameters of the damage recognition model are adjusted according to the loss value until the number of training times is reached or the loss value is less than the set loss accuracy.

8. An electronic device comprising a memory, a processor, and a computer program / instruction stored in the memory, characterized in that: The processor executes the computer program / instructions to implement the cable-supported bridge structure damage identification method according to any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program / instruction stored thereon, characterized in that: When the computer program / instruction is executed by a processor, the cable-supported bridge structure damage identification method as described in any one of claims 1 to 7 is implemented.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instruction is executed by a processor, the cable-supported bridge structure damage identification method as described in any one of claims 1 to 7 is implemented.

Images