Structural damage identification method, device, computer equipment and storage medium

Abstract

This application relates to a structural damage identification method, apparatus, computer device, storage medium, and computer program product. The method comprises: obtaining a vibration response signal from a target device; extracting modal parameters of the target device based on the vibration response signal; and inputting the modal parameters into a pretrained damage identification model to determine the damage level of the target device. This method can improve the efficiency and accuracy of structural damage level identification.

Description

Technical Field

[0001] The present application relates to the technical field of structural damage identification, and in particular to a method, apparatus, computer equipment, and storage medium for equipment damage identification. Background Art

[0002] Structural damage identification is an emerging research direction in recent years. With the vigorous development of the civil industry, more and more buildings and structures such as roads, bridges, and houses are being built. The accompanying problems have also gradually emerged. Accidents and disasters caused by structural damage are common around the world. This has promoted the development of related industries such as structural damage identification. Structural damage identification refers to obtaining structural information through certain methods and then inferring the structural damage situation.

[0003] Traditional structural damage identification methods include visual inspection of concrete surface cracks, concrete crack penetration testing, concrete carbonation depth measurement, concrete chloride ion content determination, steel bar distribution scanners, and battery point test methods for steel bar corrosion. These detection methods require manual judgment on whether the structure is damaged and the extent of the damage, resulting in low identification efficiency and accuracy. Summary of the Invention

[0004] Based on this, it is necessary to provide a structural damage identification method, device, computer equipment and storage medium that can improve the efficiency and accuracy of structural damage identification in response to the above technical problems.

[0005] In a first aspect, the present application provides a method for identifying structural damage, the method comprising:

[0006] Obtaining a vibration response signal of the target device;

[0007] extracting modal parameters of the target device according to the vibration response signal;

[0008] The modal parameters are input into a pre-trained damage identification model to obtain the damage level of the target device.

[0009] In one embodiment, the method further comprises:

[0010] Obtaining a vibration test response signal of the target device;

[0011] Determining modal parameters and damage levels of the target device based on the vibration test response signal;

[0012] The neural network model to be trained is trained according to the modal parameters and the damage level to obtain the pre-trained damage identification model.

[0013] In one embodiment, the vibration test response signal includes: a vibration response signal of the target device during normal operation, and a vibration response signal of the target device at different damage levels.

[0014] In one embodiment, determining the modal parameters and damage level of the target device based on the vibration test response signal includes:

[0015] Extracting modal parameters corresponding to the vibration test response signal according to the vibration test response signal;

[0016] The modal parameters corresponding to the vibration test response signal are compared with the modal parameters of the target device during normal operation to obtain the damage level of the test vibration response signal; wherein the damage level includes no damage, mild damage, moderate damage and severe damage.

[0017] In one embodiment, the training of the neural network model to be trained based on the modal parameters and the damage level to obtain the pre-trained damage identification model includes:

[0018] Pre-training the neural network model according to a predefined self-supervised task based on a vibration test response signal with a damage level of no damage and corresponding modal parameters;

[0019] Based on vibration test response signals with damage levels of mild damage, moderate damage, and severe damage and corresponding modal parameters, the neural network model is fine-tuned to obtain the pre-trained damage identification model.

[0020] In one embodiment, obtaining a vibration response signal of a target device includes:

[0021] Acquire the vibration response signal of the target device collected by the data acquisition system; the data acquisition system includes: an acceleration sensor, a velocity sensor, a stress sensor, an acoustic emission sensor, a laser Doppler vibrometer, a temperature sensor, a humidity sensor and a wind speed sensor.

[0022] In one embodiment, extracting the modal parameters of the target device according to the vibration response signal includes:

[0023] The vibration response signal is transformed to obtain modal parameters of the target device; wherein the transformation processing includes at least one of Fourier transform, short-time Fourier transform and wavelet transform, and the modal parameters include natural frequency, damping ratio, modal shape and modal assurance degree.

[0024] In a second aspect, the present application further provides a structural damage identification device, comprising:

[0025] A signal acquisition module is used to obtain a vibration response signal of a target device;

[0026] A modal parameter extraction module, configured to extract modal parameters of the target device based on the vibration response signal;

[0027] The damage identification module is used to input the modal parameters into a pre-trained damage identification model to obtain the damage level of the target device.

[0028] In a third aspect, the present application further provides a computer device. The computer device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are performed:

[0029] Obtaining a vibration response signal of the target device;

[0030] extracting modal parameters of the target device according to the vibration response signal;

[0031] The modal parameters are input into a pre-trained damage identification model to obtain the damage level of the target device.

[0032] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the following steps:

[0033] Obtaining a vibration response signal of the target device;

[0034] extracting modal parameters of the target device according to the vibration response signal;

[0035] The modal parameters are input into a pre-trained damage identification model to obtain the damage level of the target device.

[0036] The above-mentioned structural damage identification method, apparatus, computer device, and storage medium obtain a vibration response signal from a target device; extract the target device's modal parameters based on the vibration response signal; and input the modal parameters into a pre-trained damage identification model to determine the target device's damage level. Using this method for structural damage identification can improve the efficiency and accuracy of damage identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 is a schematic flow chart of a structural damage identification method in one embodiment;

[0038] Figure 2 A schematic diagram of a process for obtaining a pre-trained damage recognition model in one embodiment;

[0039] Figure 3A schematic diagram of a process for obtaining a pre-trained damage recognition model in one embodiment;

[0040] Figure 4 Schematic diagram of the structure of a wall bushing in a substation according to one embodiment;

[0041] Figure 5 A schematic diagram of a mean square error indicator change trend in one embodiment;

[0042] Figure 6 is a structural block diagram of a structural damage identification device in one embodiment;

[0043] Figure 7 FIG. 1 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0044] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0045] In one embodiment, Figure 1 As shown, a structural damage identification method is provided. This embodiment uses the method applied to a terminal as an example. It can be understood that the method can also be applied to a server, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server.

[0046] In this embodiment, the method includes the following steps:

[0047] Step 102: Acquire a vibration response signal of the target device.

[0048] The vibration response signal represents the vibration output signal of the target device when it is subjected to vibration, including signals such as the velocity response, acceleration response, displacement response, and strain response of the target device.

[0049] Step 104: extracting modal parameters of the target device according to the vibration response signal.

[0050] Among them, the modal parameters of the target device include parameters such as natural frequency, damping ratio and modal shape.

[0051] Specifically, for the vibration response signal, modal parameter extraction methods such as modal parameter identification method and modal damping ratio method are used to extract the modal parameters of the target device.

[0052] Step 106: Input the modal parameters into the pre-trained damage identification model to obtain the damage level of the target device.

[0053] Among them, the pre-trained damage identification model is a model used to determine the damage level of the target equipment, which can be a random forest, a support vector machine, a gradient boosting tree, etc.

[0054] In the above-described structural damage identification method, a vibration response signal of the target device is obtained; modal parameters of the target device are extracted based on the vibration response signal; and the modal parameters are input into a pretrained damage identification model to determine the damage level of the target device. The method of this embodiment, which uses a pretrained damage identification model to identify structural damage, avoids the low recognition efficiency and accuracy associated with manual damage level determination, thereby improving the efficiency and accuracy of damage level identification.

[0055] In one embodiment, Figure 2 As shown, the method further includes:

[0056] Step 202: Acquire a vibration test response signal of the target device.

[0057] Specifically, a vibration test is performed on the target device to obtain a vibration test response signal of the target device. The vibration test includes field testing and finite element simulation testing. The vibration test response signal includes the vibration response signal of the target device during normal operation and the vibration response signal of the target device at different damage levels, including no damage, mild damage, moderate damage, and severe damage.

[0058] Step 204 : determining the modal parameters and damage level of the target device based on the vibration test response signal.

[0059] Specifically, based on the vibration test response signal, the modal parameters of the target device during the vibration test are extracted. The modal parameters of the target device during the vibration test are then compared with the modal parameters of the target device in its undamaged state to obtain the damage level of the target device.

[0060] For example, the target device is a bridge, whose natural frequency in the undamaged state is 10 Hz. Based on its modal parameters, the following damage levels can be defined:

[0061] 1. Mild damage: The natural frequency changes from -0.1Hz to -0.5Hz (i.e., from 10Hz to 9.9Hz to 9.5Hz). This level of damage is generally caused by small cracks or localized corrosion.

[0062] 2. Moderate damage: The natural frequency changes from -0.5 Hz to -1 Hz (i.e., from 10 Hz to 9.5 Hz to 9 Hz). This level of damage is generally large cracks or loosening of some structural elements.

[0063] 3. Severe damage: The natural frequency range is greater than -1 Hz (i.e., less than 9 Hz). This level of damage generally involves structural integrity, such as large cracks, corrosion, or complete failure of key components.

[0064] In some embodiments, multiple modal parameters of the target device (such as natural frequency, modal shape, and damping ratio) may be integrated to determine the damage level of the target device, thereby improving the accuracy of the damage level determination.

[0065] Step 206 : training the neural network model to be trained according to the modal parameters and the damage level to obtain the pre-trained damage identification model.

[0066] In this embodiment, a vibration test response signal of the target device is first obtained. Based on the vibration test response signal, the modal parameters and damage level of the target device are determined. A pretrained damage identification model is obtained by training a neural network model based on the modal parameters and damage level. Determining the modal parameters and damage level corresponding to the vibration test response signal and then training the model to be trained facilitates the production of a more accurate damage identification model, thereby improving the accuracy of subsequent determination of the target device's damage level.

[0067] In one embodiment, Figure 3 As shown in FIG, according to the modal parameters and damage levels, the neural network model to be trained is trained to obtain a pre-trained damage identification model, including:

[0068] Step 302 : Pre-training a neural network model according to a predefined self-supervision task based on the vibration test response signal with a damage level of no damage and the corresponding modal parameters.

[0069] The self-supervised task can be time series prediction, and the neural network model can be an LSTM (Long Short-Term Memory) network or a GRU (Gated Recurrent Unit) network. By pre-training the neural network model, it can learn the vibration response signal of the target device in an undamaged state, as well as the modal parameters corresponding to the vibration response signal.

[0070] Step 304 : Based on the vibration test response signals of the damage levels of mild damage, moderate damage, and severe damage and the corresponding modal parameters, the neural network model is fine-tuned to obtain a pre-trained damage identification model.

[0071] Specifically, the pre-trained neural network model takes the vibration test response signal and the corresponding modal parameters as input, and the damage level of the target equipment, including mild damage, moderate damage, and severe damage, as output, and is trained based on a supervised learning framework.

[0072] In this embodiment, a neural network model is pre-trained based on vibration test response signals and corresponding modal parameters for damage levels of no damage, according to a predefined self-supervised task. The neural network model is then fine-tuned based on vibration test response signals and corresponding modal parameters for damage levels of mild, moderate, and severe damage, resulting in a pre-trained damage identification model. Using the method of this embodiment for model training can further improve the accuracy of the damage identification model, thereby facilitating the subsequent accuracy of determining the damage level of the target device.

[0073] In one embodiment, obtaining a vibration response signal of a target device includes obtaining the vibration response signal of the target device from a data acquisition system. The data acquisition system includes an acceleration sensor, a velocity sensor, a stress sensor, an acoustic emission sensor, a laser Doppler vibrometer, a temperature sensor, a humidity sensor, and a wind speed sensor.

[0074] For example, the structures to be inspected can include transformers, circuit breakers, disconnectors, voltage transformers, current transformers, capacitors, reactors, lightning arresters, busbars, and the like. Sensors are typically placed at vulnerable locations within the structure, often determined empirically, such as protruding areas of irregular structures, the ends of slender structures, and joints (flanges). Each sensor can store data and transmit it to a data acquisition system via a data cable or Bluetooth.

[0075] The data acquisition system can collect the target device's vibration response signals through various sensors, facilitating analysis and extracting modal parameters. To facilitate modal parameter extraction, environmental data and basic information about the target device are often also referenced. Environmental data includes external environmental factors such as temperature, humidity, wind speed, and direction, which may affect the target device's vibration characteristics. Basic information includes the target device's material properties, dimensions, and load conditions.

[0076] In one embodiment, extracting the modal parameters of the target device according to the vibration response signal includes: performing transformation processing on the vibration response signal to obtain the modal parameters of the target device.

[0077] The transformation processing includes at least one of Fourier transform, short-time Fourier transform and wavelet transform, and the modal parameters include natural frequency, damping ratio, modal shape and modal assurance degree.

[0078] In one embodiment, the target device is Figure 4 The wall bushing in the substation shown. A wall bushing, also known as a waterproof bushing or pre-buried wall pipe, is a high-voltage insulator primarily used in power systems to pass current-carrying conductors through grounded structures such as walls, providing insulation and support. Its structural features include an outdoor bushing and an indoor bushing connected by a metal sleeve. The bushing's internal core is made of a single core of glue-impregnated paper. To improve voltage distribution, grading rings are installed at both ends. During installation, the metal sleeve is connected to the valve hall wall at a 10° angle. The fiberglass sleeve is glued to the metal flange. A metal strip connector is installed at the top of the bushing to ensure electric field balance and stability during earthquakes. Accelerometers are installed at both ends of the outer sleeve and the end of the metal flange.

[0079] For this wall bushing, vibration response signals can be obtained through finite element simulation. The finite element software used for this test is Abaqus. The outer flange of the wall bushing should be fixed to the wall. The modeling assumes a rigid connection to the ground, ignoring the actual angle between the bushing and the wall. Based on the characteristics of the structural system, all components are modeled using solid elements. The finite element calculation can be simplified as follows:

[0080] (1) The conductive rod is simplified to a cylindrical component; the capacitor core is simplified to a composite component with cones at both ends and a cylinder in the middle; the outer sleeve is simplified to a hollow circular tube; and the flange is also simplified appropriately. The purpose of simplification is to simplify the modeling and reduce the amount of calculation. The principle of simplification is to ensure that the overall dimensions and weight of the component are consistent with the design requirements, so the material density is adjusted during the actual modeling.

[0081] (2) All materials are assumed to be homogeneous and all connection surfaces are assumed to be ideal rigid connections.

[0082] (3) The pressure equalizing ring is simplified to a connecting plate with concentrated mass applied at both ends, with a mass of 300 kg on each side.

[0083] (4) The actual boundary conditions of the wall bushing are that the bushing is welded to the flange, the flange is connected to the mounting plate by bolts, and the mounting plate is fixed to the wall by embedded parts. The boundary conditions of the mounting plate are assumed to be fixed constraints during modeling and calculation. When the earthquake is input, the recorded acceleration is input along the four sides of the mounting plate.

[0084] (5) The finite element model of the wall bushing is assumed to be a linear elastic component for seismic analysis. This is because the structural damage is considered to be minor and the structure has not entered the plastic deformation stage.

[0085] (6) The actual connection of the wall bushing is that the components are fixed by bolts or welding, and there is no relative movement between the connected components. The connection relationship between the components is "Tie". "Tie" means that the two contact surfaces are bound together, and tension and pressure can be transmitted between the two contact surfaces. There is no relative displacement or relative rotation between the two contact surfaces.

[0086] After damage and cracks occur in actual components, the stiffness of the damaged area decreases, but the mass remains largely unchanged. Therefore, stiffness reduction at different locations is used to simulate damage. In the Abaqus finite element software, seismic waves are input, and measurement points are set at the ends of the outer sleeve and the end of the metal flange to measure the acceleration response signal of the wall bushing.

[0087] The following is an example of the structural damage identification method provided by this application. This example uses the method applied to a terminal as an example. The main steps include:

[0088] Step 1: Data collection and preprocessing.

[0089] Vibration response signals of the target device are collected from the undamaged and damaged state. The variance of the acceleration is calculated from each data segment of the vibration response signal as feature data. The feature data is then standardized or normalized to be within a normal distribution range.

[0090] Step 2: feature extraction.

[0091] CNN (Convolutional Neural Networks) is used to extract feature data.

[0092] Step three: model design and training.

[0093] Design a deep neural network model consisting of multiple convolutional, pooling, and fully connected layers. Train the model using feature data labeled with damage levels. Optimize damage level prediction using a loss function such as MSE (mean squared error). Use optimizers such as Adam (adaptive moment estimation) or SGD (stochastic gradient descent) to minimize the training loss and perform training.

[0094] Step 4: Verification and testing.

[0095] Use the validation set to validate the neural network model and adjust hyperparameters to improve the generalization ability of the neural network model. Evaluate the performance of the neural network model on an independent test set to ensure that the neural network model is not overfitting.

[0096] Step 5: Damage detection and location.

[0097] For new unlabeled damage level feature data, the trained neural network model is used to make predictions. Based on the prediction results, the damage level of the target device is determined.

[0098] Step six, post-processing.

[0099] The prediction results of the neural network model are analyzed and combined with other sensors or detection technologies to provide more accurate damage identification.

[0100] Step 7: Continuous monitoring and updating.

[0101] The neural network model is updated and optimized regularly to adapt the neural network model to new damage patterns.

[0102] Figure 5 is the change trend of the mean square error index of the neural network model in this embodiment during the training process. Figure 5 As shown in the figure, as training progresses, the degree of fit of the model, that is, the degree of fit between the modal parameters (acceleration variance) and the damage level, becomes increasingly better. The closer the mean square error is to 0, the better the fit, while the closer the mean square error is to 1, the worse the fit.

[0103] The steps in this embodiment utilize big data analysis and machine learning technology to identify the damage level of the target device, which can improve the accuracy of damage level identification.

[0104] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0105] Based on the same inventive concept, embodiments of the present application also provide a structural damage identification device for implementing the aforementioned structural damage identification method. The solution provided by this device is similar to the solution described in the aforementioned method. Therefore, the specific limitations of one or more structural damage identification device embodiments provided below can be found in the aforementioned limitations of the structural damage identification method and will not be further elaborated here.

[0106] In one embodiment, Figure 6 As shown, a structural damage identification device 600 is provided, comprising: a signal acquisition module 601, a modal parameter extraction module 602 and a damage identification module 603, wherein:

[0107] The signal acquisition module 601 is used to acquire a vibration response signal of a target device.

[0108] The modal parameter extraction module 602 is used to extract the modal parameters of the target device according to the vibration response signal.

[0109] The damage identification module 603 is used to input the modal parameters into the pre-trained damage identification model to obtain the damage level of the target equipment.

[0110] The structural damage identification device 600 of this embodiment is used to identify structural damage. Since the structural damage is identified through a pre-trained damage identification model, the problems of low recognition efficiency and accuracy caused by manual judgment of damage levels are avoided, thereby improving the efficiency and accuracy of damage level identification.

[0111] In one embodiment, the damage identification device is also used to obtain a vibration test response signal of the target device; based on the vibration test response signal, determine the modal parameters and damage level of the target device; and train the neural network model to be trained according to the modal parameters and damage level to obtain a pre-trained damage identification model.

[0112] In one embodiment, the vibration test response signal includes: a vibration response signal of the target device during normal operation, and a vibration response signal of the target device at different damage levels.

[0113] In one embodiment, the damage identification device is also used to determine the modal parameters and damage level of the target device based on the vibration test response signal, including: extracting the modal parameters corresponding to the vibration test response signal according to the vibration test response signal; comparing the modal parameters corresponding to the vibration test response signal with the modal parameters of the target device during normal operation to obtain the damage level of the test vibration response signal.

[0114] Among them, the injury levels include no injury, mild injury, moderate injury and severe injury.

[0115] In one embodiment, the damage identification device is also used to pre-train the neural network model based on the vibration test response signal with a damage level of no damage and the corresponding modal parameters according to a predefined self-supervision task; and fine-tune the neural network model based on the vibration test response signal with damage levels of mild damage, moderate damage and severe damage and the corresponding modal parameters to obtain a pre-trained damage identification model.

[0116] In one embodiment, the signal acquisition module 601 is further configured to acquire a vibration response signal of a target device acquired by a data acquisition system.

[0117] The data acquisition system includes: acceleration sensor, velocity sensor, stress sensor, acoustic emission sensor, laser Doppler vibrometer, temperature sensor, humidity sensor and wind speed sensor.

[0118] In one embodiment, the modal parameter extraction module 602 is further configured to perform transformation processing on the vibration response signal to obtain the modal parameters of the target device.

[0119] The transformation processing includes at least one of Fourier transform, short-time Fourier transform and wavelet transform, and the modal parameters include natural frequency, damping ratio, modal shape and modal assurance degree.

[0120] Each module in the structural damage identification device 600 may be implemented in whole or in part through software, hardware, or a combination thereof. Each module may be embedded in or independent of a processor in a computer device in the form of hardware, or may be stored in a memory in the computer device in the form of software, so that the processor can call and execute the corresponding operations of each module.

[0121] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 7As shown. The computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a structural damage identification method is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad provided on the computer device housing, or an external keyboard, touchpad or mouse.

[0122] Those skilled in the art will understand that Figure 7 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0123] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0124] Get the vibration response signal of the target device.

[0125] Extract the modal parameters of the target device based on the vibration response signal.

[0126] The modal parameters are input into the pre-trained damage identification model to obtain the damage level of the target equipment.

[0127] In one embodiment, when the processor executes the computer program, it further implements the following steps: obtaining a vibration test response signal of the target device; determining the modal parameters and damage level of the target device based on the vibration test response signal; and training the neural network model to be trained according to the modal parameters and damage level to obtain a pre-trained damage identification model.

[0128] In one embodiment, the vibration test response signal includes: a vibration response signal of the target device during normal operation, and a vibration response signal of the target device at different damage levels.

[0129] In one embodiment, when the processor executes the computer program, it further implements the following steps: determining the damage level of the target device based on the vibration test response signal includes: extracting the modal parameters corresponding to the vibration test response signal according to the vibration test response signal; comparing the modal parameters corresponding to the vibration test response signal with the modal parameters of the target device during normal operation to obtain the damage level of the test vibration response signal.

[0130] Among them, the injury levels include no injury, mild injury, moderate injury and severe injury.

[0131] In one embodiment, when the processor executes the computer program, the following steps are further implemented: based on the vibration test response signal with a damage level of no damage and the corresponding modal parameters, the neural network model is pre-trained according to a predefined self-supervision task; based on the vibration test response signal with damage levels of mild damage, moderate damage, and severe damage and the corresponding modal parameters, the neural network model is fine-tuned to obtain a pre-trained damage identification model.

[0132] In one embodiment, when the processor executes the computer program, the processor further implements the following steps: acquiring a vibration response signal of the target device collected by the data collection system.

[0133] The data acquisition system includes: acceleration sensor, velocity sensor, stress sensor, acoustic emission sensor, laser Doppler vibrometer, temperature sensor, humidity sensor and wind speed sensor.

[0134] In one embodiment, when the processor executes the computer program, the following steps are further implemented: the modal parameter extraction module is further used to transform the vibration response signal to obtain the modal parameters of the target device.

[0135] The transformation processing includes at least one of Fourier transform, short-time Fourier transform and wavelet transform, and the modal parameters include natural frequency, damping ratio, modal shape and modal assurance degree.

[0136] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0137] Get the vibration response signal of the target device.

[0138] Extract the modal parameters of the target device based on the vibration response signal.

[0139] The modal parameters are input into the pre-trained damage identification model to obtain the damage level of the target equipment.

[0140] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: obtaining a vibration test response signal of the target device; determining the modal parameters and damage level of the target device based on the vibration test response signal; and training the neural network model to be trained according to the modal parameters and damage level to obtain a pre-trained damage identification model.

[0141] In one embodiment, the vibration test response signal includes: a vibration response signal of the target device during normal operation, and a vibration response signal of the target device at different damage levels.

[0142] In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: determining the damage level of the target device based on the vibration test response signal includes: extracting the modal parameters corresponding to the vibration test response signal according to the vibration test response signal; comparing the modal parameters corresponding to the vibration test response signal with the modal parameters of the target device during normal operation to obtain the damage level of the test vibration response signal.

[0143] Among them, the injury levels include no injury, mild injury, moderate injury and severe injury.

[0144] In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: based on the vibration test response signal with a damage level of no damage and the corresponding modal parameters, the neural network model is pre-trained according to a predefined self-supervision task; based on the vibration test response signal with damage levels of mild damage, moderate damage and severe damage and the corresponding modal parameters, the neural network model is fine-tuned to obtain a pre-trained damage identification model.

[0145] In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: obtaining a vibration response signal of the target device collected by the data collection system.

[0146] The data acquisition system includes: acceleration sensor, velocity sensor, stress sensor, acoustic emission sensor, laser Doppler vibrometer, temperature sensor, humidity sensor and wind speed sensor.

[0147] In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: the modal parameter extraction module is further used to transform the vibration response signal to obtain the modal parameters of the target device.

[0148] The transformation processing includes at least one of Fourier transform, short-time Fourier transform and wavelet transform, and the modal parameters include natural frequency, damping ratio, modal shape and modal assurance degree.

[0149] In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the following steps:

[0150] Get the vibration response signal of the target device.

[0151] Extract the modal parameters of the target device based on the vibration response signal.

[0152] The modal parameters are input into the pre-trained damage identification model to obtain the damage level of the target equipment.

[0153] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: obtaining a vibration test response signal of the target device; determining the modal parameters and damage level of the target device based on the vibration test response signal; and training the neural network model to be trained according to the modal parameters and damage level to obtain a pre-trained damage identification model.

[0154] In one embodiment, the vibration test response signal includes: a vibration response signal of the target device during normal operation, and a vibration response signal of the target device at different damage levels.

[0155] In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: determining the damage level of the target device based on the vibration test response signal includes: extracting the modal parameters corresponding to the vibration test response signal according to the vibration test response signal; comparing the modal parameters corresponding to the vibration test response signal with the modal parameters of the target device during normal operation to obtain the damage level of the test vibration response signal.

[0156] Among them, the injury levels include no injury, mild injury, moderate injury and severe injury.

[0157] In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: based on the vibration test response signal with a damage level of no damage and the corresponding modal parameters, the neural network model is pre-trained according to a predefined self-supervision task; based on the vibration test response signal with damage levels of mild damage, moderate damage and severe damage and the corresponding modal parameters, the neural network model is fine-tuned to obtain a pre-trained damage identification model.

[0158] In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: obtaining a vibration response signal of the target device collected by the data collection system.

[0159] The data acquisition system includes: acceleration sensor, velocity sensor, stress sensor, acoustic emission sensor, laser Doppler vibrometer, temperature sensor, humidity sensor and wind speed sensor.

[0160] In one embodiment, when the computer program is executed by the processor, the following steps are further implemented: the modal parameter extraction module is further used to transform the vibration response signal to obtain the modal parameters of the target device.

[0161] The transformation processing includes at least one of Fourier transform, short-time Fourier transform and wavelet transform, and the modal parameters include natural frequency, damping ratio, modal shape and modal assurance degree.

[0162] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0163] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, database or other media used in the embodiments provided in this application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processor involved in the various embodiments provided herein may be, but are not limited to, a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic unit, a data processing logic unit based on quantum computing, and the like.

[0164] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0165] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. A structural damage identification method, characterized in that: The method comprises: Obtaining a vibration response signal of a target device; the target device includes a transformer, a circuit breaker, a disconnector, a voltage transformer, a current transformer, a capacitor, a reactor, a lightning arrester, and a busbar; Extracting modal parameters of the target device according to the vibration response signal; the modal parameters include natural frequency, damping ratio, modal shape and modal assurance degree; Inputting the modal parameters into a pre-trained damage identification model to obtain a damage level of the target device; The obtaining of the vibration response signal of the target device includes: Acquiring a vibration response signal of a target device collected by a data acquisition system; the data acquisition system includes: an acceleration sensor, a velocity sensor, a stress sensor, an acoustic emission sensor, a laser Doppler vibrometer, a temperature sensor, a humidity sensor, and a wind speed sensor; The method further comprises: Obtaining a vibration test response signal of the target device; Extracting modal parameters corresponding to the vibration test response signal based on the vibration test response signal; the vibration test response signal includes: a vibration response signal of the target device during normal operation, and a vibration response signal of the target device at different damage levels; the vibration response signal represents a vibration output signal of the target device when subjected to vibration, including signals such as a velocity response, an acceleration response, a displacement response, and a strain response of the target device; Compare the modal parameters corresponding to the vibration test response signal with the modal parameters of the target device during normal operation to obtain a damage level of the vibration test response signal; wherein the damage level includes no damage, mild damage, moderate damage, and severe damage: Based on the vibration test response signal with no damage level and the corresponding modal parameters, the neural network model to be trained is pre-trained according to the predefined self-supervised task; Based on vibration test response signals with damage levels of mild damage, moderate damage, and severe damage and corresponding modal parameters, the neural network model is fine-tuned to obtain the pre-trained damage identification model.

2. The method according to claim 1, characterized in that Extracting the modal parameters of the target device according to the vibration response signal includes: The vibration response signal is transformed to obtain modal parameters of the target device; wherein the transformation processing includes at least one of Fourier transform, short-time Fourier transform and wavelet transform.

3. A structural damage identification device, characterized in that: The device comprises: A signal acquisition module is used to acquire vibration response signals of target devices; the target devices include transformers, circuit breakers, disconnectors, voltage transformers, current transformers, capacitors, reactors, lightning arresters, and busbars; The obtaining of the vibration response signal of the target device includes: Acquiring a vibration response signal of a target device collected by a data acquisition system; the data acquisition system includes: an acceleration sensor, a velocity sensor, a stress sensor, an acoustic emission sensor, a laser Doppler vibrometer, a temperature sensor, a humidity sensor, and a wind speed sensor; A modal parameter extraction module, configured to extract modal parameters of the target device based on the vibration response signal; the modal parameters include natural frequency, damping ratio, modal shape, and modal assurance degree; a damage identification module, configured to input the modal parameters into a pre-trained damage identification model to obtain a damage level of the target device; The damage identification device is also used to obtain the vibration test response signal of the target equipment; Extracting modal parameters corresponding to the vibration test response signal based on the vibration test response signal; the vibration test response signal includes: a vibration response signal of the target device during normal operation, and a vibration response signal of the target device at different damage levels; the vibration response signal represents a vibration output signal of the target device when subjected to vibration, including signals such as a velocity response, an acceleration response, a displacement response, and a strain response of the target device; Compare the modal parameters corresponding to the vibration test response signal with the modal parameters of the target device during normal operation to obtain a damage level of the vibration test response signal; wherein the damage level includes no damage, mild damage, moderate damage, and severe damage: Based on the vibration test response signal with no damage level and the corresponding modal parameters, the neural network model to be trained is pre-trained according to the predefined self-supervised task; Based on vibration test response signals with damage levels of mild damage, moderate damage, and severe damage and corresponding modal parameters, the neural network model is fine-tuned to obtain the pre-trained damage identification model.

4. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 2 are implemented.

5. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 2 are implemented.

Images