Bridge damage identification method under moving load

Abstract

The application discloses a bridge damage identification method under a moving load, comprising the following steps: 1, substituting bridge damage displacement data obtained through finite element simulation into a residual neural network to obtain a predicted damage reduction coefficient; 2, based on a Newmark-beta method, embedding available physical knowledge of a bridge structure into a feature learning process, and substituting the damage reduction coefficient predicted by the network into a physical equation to obtain predicted damage displacement data; 3, substituting the damage reduction coefficient and the displacement damage data obtained by solving into a loss function to obtain a loss value, and updating network parameters; and 4, predicting bridge damage by using the trained physical-guided residual neural network. The application has high damage identification precision under the condition of considering the uncertainty of vehicle weight and the uncertainty of bridge structure parameters, and can improve the accuracy of damage identification of the network when the measured data is less.

Description

TECHNICAL FIELD

[0001] The application relates to the field of bridge detection, in particular to a bridge damage identification method under mobile load. BACKGROUND

[0002] During the service period of a large number of bridges, diseases and damages will occur due to various factors such as environmental erosion, long-term load effect and material aging, resulting in structural performance degradation and safety reduction. Accurate monitoring of the damage of the bridge is helpful for evaluating the safety performance and use state of the structure and helping the management personnel to operate the bridge in service. In the past decade, people have carried out a large amount of research on vibration-based structural damage identification technology, and have made substantial achievements. These achievements are based on the principle that the change of structural characteristics caused by damage will lead to a detectable change in the vibration characteristics of the structure. Therefore, the current health state of the structure can be determined by extracting damage-sensitive features and patterns from the measured vibration characteristics.

[0003] In practice, physical and deep learning methods are often used for damage identification. The physical-based method directly or indirectly uses the physical law that controls the behavior of the structure in order to extract meaningful information about damage and its evolution from the measured vibration response. The physical-based method needs to establish an accurate bridge finite element model, however, in practical applications, even with the knowledge of the entire model, it is difficult to establish a complete mechanism model through the knowledge, and in addition, the structural uncertainty of the bridge may seriously affect the performance of the structural model updating and damage identification. The deep learning-based method learns abstract features from training data under different damage conditions by using a large amount of training data without an accurate initial structural finite element model, and makes predictions according to the learned relationship. These training data sets can consider the influence of bridge structural parameter uncertainty, and output more accurate damage identification results under the influence of uncertainty. However, training a reliable machine learning model usually requires sufficient data, which cannot be provided by most bridge damage identification engineering tasks. Another problem of the deep learning-based damage identification method is that the deep learning model is only trained on data, and lacks physical interpretability for many engineering problems. SUMMARY

[0004] The application is to solve the problems of the existing damage identification method, and provides a bridge damage identification method under mobile load, so as to have higher damage identification accuracy under the consideration of vehicle weight uncertainty and bridge structural parameter uncertainty, and improve the accuracy of network damage identification when the measured data is less.

[0005] In order to achieve the above purpose, the application adopts the following technical scheme.

[0006] The bridge damage identification method under mobile load has the characteristics that the method comprises the following steps:

[0007] Step 1: Establish the vehicle-axle coupling model:

[0008] A bridge model was constructed using finite element method (FEM) software, and moving loads were applied to the bridge model for simulation to determine the length of the bridge model. The elastic modulus is ;

[0009] A vehicle model was constructed using finite element method software, and the parameters of the vehicle model were determined, including: the weight of the vehicle model. The speed of the vehicle model ;

[0010] The vehicle model moves at a constant speed After using the bridge model, the number of sampling points is calculated as follows: ;in, Sampling frequency, For the vehicle model, the time it takes to cross the bridge is [time]. ;

[0011] Step 2: Set the damage conditions and obtain the displacement response data of the bridge model.

[0012] Step 2.0: Define and initialize the types of operating conditions. ;

[0013] Step 2.1: Define and initialize the index of the current cell point. ;

[0014] Step 2.2: Define the i-th damage reduction factor as... and initialize ;

[0015] Step 2.3: On the bridge model, the first... Add the i-th damage reduction coefficient to each unit point. The first bridge model was obtained. Types of damage conditions ;in, Indicates the first Apply the i-th damage reduction coefficient to each unit point. The first time Types of damage conditions; p is the total number of unit points on the bridge model;

[0016] Step 2.4: In Random initialization within the range of the first Variables Thus, the vehicle model is in the first Uncertainty about vehicle weight Using a bridge model, and employing The method obtains the bridge model in the first... Types of damage conditions Displacement response data ,in, The peak value represents the change.

[0017] Step 2.5: [The text appears to be incomplete and contains several grammatical errors. A more accurate translation would require the full context.] Assign to Then, return to step 2.4 and execute sequentially until... Until then, thus obtaining the result in the first place. Apply the i-th damage reduction coefficient at each unit point. time Displacement response data of a bridge model under various damage conditions; among which, Indicates the first Apply the i-th damage reduction coefficient at each unit point. The number of uncertain vehicle weight types set at the time;

[0018] Step 2.6: [The text appears to be incomplete and contains several grammatical errors. A more accurate translation would require Assign to Then, return to step 2.3 and execute sequentially until... Until then, thus obtaining the result in the first place. Apply at each unit point When the damage reduction factor is Displacement response data of a bridge model under various damage conditions; among which, This indicates the number of types of damage reduction coefficients set.

[0019] Step 2.7: [The text appears to be incomplete and contains several grammatical errors. A more accurate translation would require Assign to Then, return to step 2.2 and execute sequentially until... Thus, under the uncertain vehicle weight of the vehicle model, the following conditions are obtained: sequentially applying force at unit point p... When the damage reduction factor is Types of damage conditions Displacement response data ; n represents the total number of damage conditions, and ;

[0020] Step 3: Put The input is processed in a residual neural network to obtain the first... Predicted values ​​of various damage conditions ;

[0021] Step 4: Construct a physical-guided residual neural network and use it to obtain the bridge model in the first step. Predicted values ​​of various damage conditions Displacement response prediction of the bridge model ;

[0022] Step 4.1: Obtain the global stiffness matrix of the bridge model under the ith damage condition prediction value by using formula (1)

[0023] (1)

[0024] In formula (1), K represents the global stiffness matrix of the bridge model under the undamaged state;

[0025] Step 4.2: Construct the Newmark-β method structure dynamics equation by using formula (2):

[0026] (2)

[0027] In formula (2), M represents the global mass matrix of the bridge model, C represents the global damping matrix of the bridge model, K represents the global stiffness matrix of the bridge model under the ith damage condition prediction value, a represents the acceleration response data of the bridge model under the ith damage condition prediction value, v represents the velocity response data of the bridge model under the ith damage condition prediction value, d represents the displacement response data of the bridge model under the ith damage condition prediction value, and f represents the moving load of the bridge model under the ith damage condition prediction value.

[0028] Step 4.3: Solve by using the Newmark-β method structure dynamics equation to obtain the bridge displacement response prediction value of the bridge model under the ith damage condition prediction value.

[0029] Step 5: Construct the total loss function by using formula (3):

[0030] (3)

[0031] In formula (3), w represents the weight of adjusting the data loss, and w represents the weight of adjusting the physical loss. ​​​​​​​​​​​​​​​​​​​​​​​​​Represents the data loss function. Let represent the physical loss function, and we have:

[0032] (4)

[0033] (5)

[0034] Step 6: Based on and The Adam optimizer was used to train the physics-guided residual neural network, and the total loss function was calculated. Update network parameters until The model converges until the bridge damage identification model with optimal physical guidance under the uncertain vehicle weight is obtained, which is used to identify damage to a certain part of the bridge.

[0035] The bridge damage identification method under moving load described in this invention is also characterized by:

[0036] Replace the vehicle model uncertainty in step 2.4 with the parameter uncertainty of the bridge model, thereby The magnitude of the moment of inertia of the bridge model is changed, and following steps 2.0-2.7, the unknown moment of inertia of the bridge model is obtained by sequentially applying the following parameters to the p-element points. When the damage reduction factor is Types of damage conditions Displacement response data Following steps 3-6, a bridge damage identification model with optimal physical guidance under uncertain inertial moments is obtained; where, The bridge model under the uncertain moment of inertia represents the first... Various damage conditions, The bridge model under the uncertain moment of inertia represents the first... Types of damage conditions Displacement response data; .

[0037] The present invention provides an electronic device, including a memory and a processor, wherein the memory is used to store a program that supports the processor in executing the bridge damage identification method, and the processor is configured to execute the program stored in the memory.

[0038] The present invention discloses a computer-readable storage medium on which a computer program is stored, wherein the computer program is executed by a processor to perform the steps of the bridge damage identification method.

[0039] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0040] 1、 Compared with the traditional model correction method, the present application does not need an accurate finite element model, the network learns the characteristics through the damage conditions under different working conditions, and predicts according to the learning relationship, and the data can consider the influence of the uncertainty of the vehicle weight and the uncertainty of the bridge structure parameters, so as to realize the accurate prediction of the bridge damage condition in the face of complex situations.

[0041] 2、 Compared with the bridge damage identification method based on deep learning, on the basis of deep learning, the present application adds the structural dynamics equation as physical knowledge into the training process of the network, which realizes a process similar to finite element solution in network training, reduces the gap between input response and predicted response, and realizes further optimization of the bridge damage coefficient. At the same time, under the condition of limited data samples, the embedded physical knowledge realizes the constraint of network input and output results, retains its physical meaning, and this method based on physical guidance can ensure that the network obtains more accurate identification results under limited data amount. BRIEF DESCRIPTION OF DRAWINGS

[0042] Figure 1 is the schematic diagram of the calculation process of the present application;

[0043] Figure 2 is the schematic diagram of the numerical simulation of the equal cross section simply supported beam model of the present application;

[0044] Figure 3 is the schematic diagram of the network structure of the present application;

[0045] Figure 4 is the comparison diagram of the damage reduction coefficient prediction value and the true value of the equal cross section simply supported beam bridge under the excitation of the moving load of the present application;

[0046] Figure 5 is the comparison diagram of the damage reduction coefficient prediction value and the true value of the equal cross section simply supported beam bridge under the condition of small data set of the present application.

[0047] Figure 6 is the comparison diagram of the damage reduction coefficient prediction value and the true value of the equal cross section simply supported beam bridge under the condition of considering the influence of the uncertainty of the bridge structure of the present application. DETAILED DESCRIPTION

[0048] In this embodiment, as shown in the equal cross section simply supported beam, Figure 2 the elastic modulus of the bridge is , the density of the bridge per meter is , the cross section height of the bridge is , and the width is , and the finite element software is used for simulation.

[0049] A bridge damage identification method under moving loads employs a physics-guided residual neural network for damage identification. The dynamic control equations are incorporated into feature learning as physical constraints to guide the network's learning. The Newmark-β method, as available physical knowledge, is formulated into the loss function, providing physical constraints for the network output. This allows for accurate identification of bridge damage even with limited data samples and significant uncertainties in bridge structural parameters, thus improving the accuracy of bridge damage identification. Specifically, the principle of this bridge damage identification method under moving loads utilizes a simply supported beam model as follows: Figure 1 As shown, the method includes the following steps:

[0050] Step 1: Establish the vehicle-axle coupling model:

[0051] A bridge model was constructed using finite element method (FEM) software, and moving loads were applied to the bridge model for simulation to determine the length of the bridge model. The elastic modulus is ;

[0052] The bridge model has p=10 element points, from which points are selected. =3 unit points are used as observation points, and the first one is used as the observation point. The unit point number corresponding to each observation point is denoted as j, and the displacement response data of the bridge model under different damage conditions are recorded.

[0053] A vehicle model was constructed using finite element method (FEM) software, and its parameters were determined, including the weight of the vehicle model. The front wheelbase is The rear wheelbase is The speed of the vehicle model ;

[0054] The vehicle model moves at a constant speed After using the bridge model, the number of sampling points is calculated as follows: The number of sampling points is calculated to be 100; among them, Sampling frequency, For the vehicle model, the time it takes to cross the bridge is [time]. ;

[0055] Step 2: Set the damage conditions and obtain the displacement response data of the bridge model:

[0056] Step 2.0: Define and initialize the types of operating conditions. ;

[0057] Step 2.1: Define and initialize the index of the current cell point. ;

[0058] Step 2.2: Define the i-th damage reduction factor as... and initialize , This represents the damage reduction factor for the bridge model, which is used to reduce the stiffness of the bridge in the finite element model. To simulate bridge damage;

[0059] Step 2.3: On the bridge model, the first... Add the i-th damage reduction coefficient to each unit point. The first bridge model was obtained. Types of damage conditions ;in, Indicates the first Apply the i-th damage reduction coefficient to each unit point. The first time Various damage conditions, In addition to The values ​​for all other locations besides the current location are 0 to ensure that each set of damage conditions has only one damage location.

[0060] Step 2.4: In Random initialization within the range of the first Variables Thus, the vehicle model is in the first Uncertainty about vehicle weight Using a bridge model, and employing The method obtains the bridge model in the first... Types of damage conditions Displacement response data of the bridge model below ,in, The peak value represents the change.

[0061] Step 2.5: [The text appears to be incomplete and contains several grammatical errors. A more accurate translation would require Assign to Then, return to step 2.4 and execute sequentially until... Until then, thus obtaining the result in the first... Apply the i-th damage reduction coefficient to each unit point. time Displacement response data under various damage conditions; among them, Indicates the first Apply the i-th damage reduction coefficient to each unit point. time The number of vehicle weight types with uncertainties under various damage conditions. .

[0062] Step 2.6: [The text appears to be incomplete and contains several grammatical errors. A more accurate translation would require Assign to Then, return to step 2.3 and execute sequentially until... Until then, thus obtaining the result in the first... Apply at each unit point When the damage reduction coefficient is Displacement response data under various damage conditions; among them, This indicates the number of damage reduction coefficient types set in this method. In this embodiment, , There are 4 different possible values. .

[0063] Step 2.7: [The text appears to be incomplete and contains several grammatical errors. A more accurate translation would require the full context.] Assign to Then, return to step 2.2 and execute sequentially until... Thus, under the uncertain vehicle weight of the vehicle model, the following conditions are obtained: sequentially applying force at unit point p... When the damage reduction coefficient is Types of damage conditions Displacement response data n=2200 represents the total number of damage conditions, and .

[0064] Step 3: Construct a residual neural network, including: convolutional layers, residual blocks, pooling layers, global average pooling layers, fully connected layers, and the ReLU activation function;

[0065] Will In a residual neural network, the input layer is passed first, followed by successive passes through a series of... The last convolutional layer has a kernel size of 1.5. The number of convolution kernels is Each convolutional layer is followed by an activation layer and a normalization layer. After the convolutional layers, the data enters... The structure of a combination of residual blocks, each residual block consisting of... The system consists of several convolutional layers, followed by activation and normalization layers. After passing through residual blocks, the data passes through an adaptive average pooling layer, and finally through a fully connected layer to obtain the first convolutional layer of the bridge model. Predicted values ​​of various damage conditions .

[0066] Step 4: As Figure 3 As shown, a physical-guided residual neural network is constructed and used to obtain the bridge model in the first step. Predicted values ​​of various damage conditions Bridge displacement response prediction values ;

[0067] Step 4.1: Use equation (1) to obtain the bridge model in the first step. Predicted values ​​of various damage conditions global stiffness matrix ;

[0068] (1)

[0069] In formula (1), represents the overall stiffness matrix of the bridge model in the undamaged state.

[0070] Step 4.2: Construct the structural dynamics equation using formula (2):

[0071] (2)

[0072] In formula (2), M represents the mass matrix of the bridge model, C represents the damping matrix of the bridge model, represents the stiffness matrix of the bridge model in the predicted value of the first kind of damage working condition, represents the acceleration of the bridge model in the predicted value of the first kind of damage working condition, represents the speed of the bridge model in the predicted value of the first kind of damage working condition, represents the displacement response data of the bridge model in the predicted value of the first kind of damage working condition, represents the moving load of the bridge model in the predicted value of the first kind of damage working condition.

[0073] Step 4.3: Solve the structural dynamics equation using the Newmark-β method to obtain the bridge displacement response prediction value of the bridge model in the predicted value of the first kind of damage working condition .

[0074] Step 5: Construct the total loss function using formula (3):

[0075] (3)

[0076] In formula (3), is the weight of adjusting data loss, is the weight of adjusting physical loss; represents the data loss function, represents the physical loss function, and has:

[0077] (4)

[0078] (5)​​​​​​

[0079] Step 6: Based on and , the residual neural network of the physical guidance is trained using the Adam optimizer, and the total loss function is calculated , the network parameters are updated until convergence is reached, thereby obtaining the optimal physical guidance bridge damage identification model under the uncertainty of the vehicle model's weight, which is used to realize damage identification of a certain part of the bridge.

[0080] In this embodiment, an electronic device includes a memory for storing a program supporting a processor to execute the above method, and the processor is configured to execute the program stored in the memory.

[0081] In this embodiment, a computer readable storage medium has a computer program stored thereon, and the computer program is executed by a processor to perform the steps of the above method.

[0082] Figure 2 In this embodiment 1, an equal cross-section simply supported beam bridge is shown as , the density of the bridge per meter is , the cross-section height of the bridge is , and the width is . A finite element model of the bridge is established using finite element software. Considering the uncertainty of the vehicle weight, the data of are selected as the training set, the data of are selected as the validation set, and the data of are selected as the test set. The true values and the prediction results of the damage conditions are shown in Figure 4 and Table 1.

[0083] In this embodiment 2, the basic parameters of the bridge remain unchanged. Considering the uncertainty of the vehicle weight, the data of are selected as the training set, the data of are selected as the validation set, and the data of are selected as the test set. The true values and the prediction results of the damage conditions are shown in Figure 5 and Table 1.

[0084] In this embodiment, the uncertainty of the vehicle model in step 2.4 is replaced by the parameter uncertainty of the bridge model, so that the size of the moment of inertia of the bridge model is changed, and the displacement response data under damage reduction factors are obtained by following the process of steps 2.0-2.7 under the uncertainty of the moment of inertia of the bridge model. ​​Following steps 3-6, a bridge damage identification model with optimal physical guidance under uncertain inertial moments is obtained; where, The bridge model under the uncertain moment of inertia represents the first... Various damage conditions, The bridge model under the uncertain moment of inertia represents the first... Types of damage conditions Displacement response data; n=2200.

[0085] In this embodiment 3, the basic parameters of the bridge remain unchanged. Considering the uncertainties in the bridge model structure, the following parameters are selected: Data as training set Data as a validation set The data was used as the test set. The actual values ​​and predicted results for the damage condition section are shown in [reference needed]. Figure 6 See Table 1.

[0086] Mean squared error (MSE) and regression value (R1) are defined to evaluate the network's structural damage recognition performance in Examples 1, 2, and 3. MSE measures the distance between the ground truth value and the predicted output of the trained model. The regression value R1, ranging from 0 to 1, is also used to evaluate the quality of the trained model. Specifically, the R1 value represents the correlation coefficient, which quantifies the linear correlation between the predicted output and the ground truth value. Table 1 shows a comparison of the mean squared error (MSE) and regression value (R1) for Examples 1, 2, and 3.

[0087] (6)

[0088] (7)

[0089] In the formula, n represents the total number of damage conditions. This indicates the set damage condition parameters. This represents the damage condition parameters predicted by the residual neural network. , Represents the total number of partitioned units The total number of damage reduction coefficient types The product of. Represents the j-th unit The true damage index under a damage reduction factor Represents the Jth unit The average damage index predicted under each damage reduction factor.

[0090] Table 1

[0091]

[0092] From Table 1, by constructing the residual neural network model with physical information, the damage condition of the bridge model under the action of the moving vehicle is identified, and relatively accurate prediction results are obtained under embodiments 1, 2 and 3. In conclusion, the application can better predict the bridge damage condition under the condition of insufficient data, considering structural uncertainty and vehicle weight uncertainty.

Claims

1. A method for identifying damage of a bridge under moving load, characterized by, The method comprises the following steps: Step 1: establishing a vehicle-bridge coupling model The finite element software is used to construct a bridge model, and a moving load is applied to the bridge model for simulation, so that the length of the bridge model is determined as , and the elastic modulus is ; The vehicle model is constructed by using a finite element software, and parameters of the vehicle model are determined, including: a weight of the vehicle model is , and a driving speed of the vehicle model is . The vehicle model is at a constant speed After passing the bridge model, the number of sampling points is calculated as ; wherein is the sampling frequency, is the time of the vehicle model to pass the bridge, and ; Step 2: setting a damage working condition and obtaining displacement response data of the bridge model Step 2.0: Define and initialize the kind of case ; Step 2.1 : Define and initialize the sequence number of the current cell point ; Step 2.2: Define the i-th impairment reduction factor as and initialize ; Step 2.3: On the bridge model, the first... Add the i-th damage reduction factor to each unit point. The first bridge model was obtained. Types of damage conditions ;in, Indicates the first Apply the i-th damage reduction coefficient at each unit point. The first time Types of damage conditions; p is the total number of unit points on the bridge model; Step 2.4: In Random initialization within the range of the first Variables Thus, the vehicle model is in the first Uncertainty about vehicle weight Using a bridge model, and employing The method obtains the bridge model in the first... Types of damage conditions Displacement response data ,in, The peak value represents the change. Step 2.5: [The text appears to be incomplete and contains several grammatical errors. A more accurate translation would require Assign to Then, return to step 2.4 and execute sequentially until... Until then, thus obtaining the result in the first... Apply the i-th damage reduction coefficient at each unit point. time Displacement response data of a bridge model under various damage conditions; among which, Indicates the first Apply the i-th damage reduction coefficient at each unit point. The number of uncertain vehicle weight types set at the time; Step 2.6: Assigning values to , and then returning to step 2.3 for sequential execution until , thereby obtaining the displacement response data of the bridge model under the damage working condition of the th damage reduction coefficient applied at the th unit point; wherein represents the number of damage reduction coefficient types set; and represents the number of damage reduction coefficients set. ​ Step 2.7: [The text appears to be incomplete and contains several grammatical errors. A more accurate translation would require Assign to Then, return to step 2.2 and execute sequentially until... Thus, under the uncertain vehicle weight of the vehicle model, the following conditions are obtained: sequentially applying force at unit point p... When the damage reduction factor is Types of damage conditions Displacement response data ; n represents the total number of damage conditions, and ; Step 3: The input residual neural network is processed to obtain the bridge model prediction value under the first damage working condition ;​​ Step 4: Constructing the residual neural network under physical guidance and used to obtain the displacement response prediction value of the bridge model under the prediction value of the bridge model under the damage working condition ;​​ Step 4.1: Obtain the bridge model using formula (1) at the predicted value of the overall stiffness matrix under the damage working condition ; (1) In formula (1), represents the overall stiffness matrix of the bridge model in the undamaged state; Step 4.2: constructing a Newmark-β method structural dynamics equation by using formula (2) (2) In formula (2), M represents the overall mass matrix of the bridge model, C represents the overall damping matrix of the bridge model, represents the overall stiffness matrix of the bridge model under the predicted value of the first kind of damage working condition, represents the acceleration response data of the bridge model under the predicted value of the first kind of damage working condition, represents the velocity response data of the bridge model under the predicted value of the first kind of damage working condition, represents the displacement response data of the bridge model under the predicted value of the first kind of damage working condition, represents the moving load of the bridge model under the predicted value of the first kind of damage working condition, represents the moving load of the bridge model under the predicted value of the first kind of damage working condition, represents the moving load of the bridge model under the predicted value of the first kind of damage working condition, represents the moving load of the bridge model under the predicted value of the first Step 4.3: The bridge displacement response prediction value under the damage working condition prediction value is obtained by solving the structure dynamics equation by using the Newmark-β method ;​​ Step 5: Constructing the total loss function with formula (3) : (3) in formula (3), is a weight that adjusts data loss, is a weight that adjusts physical loss; represents a data loss function, represents a physical loss function, and has: (4) (5) Step 6: Based on and , the residual neural network of the physical guidance is trained using the Adam optimizer, and the total loss function is calculated The network parameters are updated until convergence is reached, thereby obtaining the optimal physical guidance bridge damage identification model under the uncertainty of the vehicle model's weight, which is used to achieve damage identification of a certain part of the bridge.

2. The bridge damage identification method under a moving load according to claim 1, characterized in that: Replace the vehicle model uncertainty in step 2.4 with the parameter uncertainty of the bridge model, thereby The magnitude of the moment of inertia of the bridge model is changed, and following steps 2.0-2.7, the unknown moment of inertia of the bridge model is obtained by sequentially applying the following parameters to the p-element points. When the damage reduction factor is Types of damage conditions Displacement response data Following steps 3-6, a bridge damage identification model with optimal physical guidance under uncertain inertial moments is obtained; whereby, The bridge model under uncertain moments of inertia represents the first... Various damage conditions, The bridge model under uncertain moments of inertia represents the first... Types of damage conditions Displacement response data; .

3. An electronic device comprising a memory and a processor, characterized in that The memory is configured to store a program supporting the processor to execute the bridge damage identification method according to claim 1 or 2, and the processor is configured to execute the program stored in the memory.

4. A computer-readable storage medium having stored thereon a computer program, characterized in that The computer program is configured to execute the steps of the bridge damage identification method according to claim 1 or 2 when the processor runs the computer program.

Images