Damage identification method based on Lamb wave temperature compensation and related device

Abstract

The invention belongs to the technical field of train safety monitoring, and particularly relates to a damage identification method based on Lamb wave temperature compensation and a related device. The damage identification method comprises the following steps: acquiring a Lamb wave signal and an actual environment temperature of a to-be-detected component; inputting the Lamb wave signal and the actual environment temperature into a pre-constructed Lamb wave temperature compensation network model, and outputting the Lamb wave signal after temperature compensation; and performing damage identification on the to-be-detected component based on the Lamb wave signal after temperature compensation to obtain a damage identification result. After temperature compensation, interference related to temperature is removed, so that damage identification based on the compensated Lamb wave signal is more accurate.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of train safety monitoring, and particularly relates to a damage identification method based on Lamb wave temperature compensation and a related device. BACKGROUND

[0002] Structural health monitoring is crucial for maintaining the integrity of engineering structures. In recent years, Lamb waves have received extensive attention due to their application in structural health monitoring. This wave has the advantages of long propagation distance and sensitivity to structural damage, and shows great potential in monitoring the integrity of plate-like structures. Lamb waves themselves have the characteristics of frequency dispersion and multi-mode, and they are a complex combination of boundary echoes, reflection waves caused by damage, and noise, which makes it a challenge to extract information about damage from these signals. In Lamb wave-based structural health monitoring technology, researchers mainly focus on the differences in signals under healthy and damaged states, and analyze these differences to extract the scattering characteristics of damage, and then realize the detection and evaluation of structural damage.

[0003] In engineering practice, the performance of Lamb waves is easily affected by environmental temperature fluctuations. Previous studies have shown that even a small change of 0.5℃ in temperature can cause a significant change in the amplitude of Lamb waves. The fluctuation of temperature not only changes the piezoelectric coefficient of the piezoelectric sensor, but also affects the mechanical properties of the material, such as the elastic modulus of aluminum alloy, so that the amplitude of Lamb waves decreases when the temperature rises, and the propagation time of the wave also lengthens accordingly. Given the extensive coverage of high-speed railway networks and the vast geographical span, the structure of the train will inevitably encounter changing weather conditions. Such changing environmental conditions not only pose a challenge to the structural safety of high-speed trains, but also can cause serious interference to the Lamb wave monitoring signal, making it difficult to accurately identify the damage signal and increasing the risk of misjudgment.

[0004] In summary, the traditional damage monitoring method can only be used to monitor damage in a single temperature environment, and does not take into account the influence of temperature on Lamb wave signals, making it unsuitable for monitoring damage in a variable temperature environment. SUMMARY

[0005] The purpose of the present application is to provide a damage identification method based on Lamb wave temperature compensation and a related device, which solves the problem that the existing Lamb wave damage detection method is not suitable for monitoring damage in a variable temperature environment.

[0006] The present application is achieved by the following technical solutions: A damage identification method based on Lamb wave temperature compensation, comprising the following processes: Obtaining the Lamb wave signal of the component to be tested and the actual environmental temperature; inputting the Lamb wave signal and an actual environment temperature into a pre-constructed Lamb wave temperature compensation network model, and outputting a temperature-compensated Lamb wave signal; performing damage identification on the target structure of the component to be measured based on the temperature-compensated Lamb wave signal, and obtaining a damage identification result.

[0007] Further, the construction process of the pre-constructed Lamb wave temperature compensation model is as follows: processing a crack on the target structure of the component to be measured according to a preset crack size; disposing a sensor at the target structure, and acquiring, by the sensor, Lamb wave signals of different damage sizes on the component to be measured at a plurality of different environment temperatures as original data; the plurality of different environment temperatures include normal temperature and abnormal temperature environment temperatures; truncating the original data, acquiring a basic symmetric mode of the Lamb wave signal, and performing normalization processing to obtain sample data; assigning a temperature label to the sample data according to the environment temperature to obtain a sample set; and dividing the sample set into a training set and a test set; constructing a Lamb wave temperature compensation network based on a conditional generative adversarial network; the Lamb wave temperature compensation network comprises a generator and a discriminator; inputting the training set into the Lamb wave temperature compensation network to perform adversarial training until the generator and the discriminator reach a Nash equilibrium; inputting the test set into the trained Lamb wave temperature compensation network to output a temperature-compensated Lamb wave signal until the test is qualified, and obtaining a constructed Lamb wave temperature compensation network model.

[0008] Further, the test set is inputted into the trained Lamb wave temperature compensation network to output a temperature-compensated Lamb wave signal until the test is qualified, and a constructed Lamb wave temperature compensation network model is obtained, specifically as follows: inputting the test set into the trained Lamb wave temperature compensation network to output a temperature-compensated Lamb wave signal, and performing damage identification on a crack of the target structure of the component to be measured by using the temperature-compensated Lamb wave signal to output a predicted damage size; performing error calculation on the predicted damage size and the preset crack size until the error of the predicted damage size and the preset crack size meets a requirement, and then the Lamb wave temperature compensation network model is constructed.

[0009] Further, the process of inputting the training set into the Lamb wave temperature compensation network to perform adversarial training is as follows: training the discriminator: inputting the original input signal and the temperature label into the generator, and the generator generates a reconstructed signal; The reconstructed signal and the target real signal are input into the discriminator, and the output result is fed back to the discriminator through a loss function to complete the update of the parameters of the discriminator; in this process, the parameters of the generator are frozen; The generator is trained: The parameters of the discriminator are fixed, the original input signal and the temperature label are input into the generator, and the generator generates a reconstructed signal; The reconstructed signal and the target real signal are input into the discriminator, and the similarity is evaluated through a loss function, if the similarity evaluation meets the preset requirement, the training is stopped, if the similarity evaluation does not meet the preset requirement, the similarity loss is back propagated to the generator, so that the parameters of the generator are iteratively updated, until the similarity evaluation meets the preset requirement, then the generator training is completed; The generator and the discriminator are alternately trained until Nash equilibrium is reached; The original input signal is a Lamb wave signal collected in a very warm environment, and the target real signal is a Lamb wave signal collected at room temperature.

[0010] Further, the temperature-compensated Lamb wave signal is used for damage identification of the to-be-tested component, specifically: The temperature-compensated Lamb wave signal is input into a damage detection network, and a predicted damage result is output.

[0011] Further, the damage detection network adopts a convolutional neural network.

[0012] The application discloses a damage identification system based on Lamb wave temperature compensation, which comprises the following processes: A data acquisition module is used to acquire Lamb wave signals and actual environment temperatures of a to-be-tested component; A temperature compensation module is used to input the Lamb wave signals and the actual environment temperatures into a pre-constructed Lamb wave temperature compensation network model, and output temperature-compensated Lamb wave signals; A damage identification module is used to identify damages of the to-be-tested component based on the temperature-compensated Lamb wave signals, and obtain damage identification results.

[0013] The application further discloses a computer device, which comprises a memory, a processor, and a computer program stored in the memory and running on the processor, and the processor implements the steps of the damage identification method based on Lamb wave temperature compensation when executing the computer program.

[0014] The application further discloses a computer readable storage medium, which stores a computer program, and the computer program implements the steps of the damage identification method based on Lamb wave temperature compensation when executed by a processor.

[0015] The application further discloses a computer program product comprising computer programs / instructions, which, when executed by a processor, realize the steps of the damage identification method based on Lamb wave temperature compensation.

[0016] Compared with the prior art, the application has the following beneficial technical effects: The application discloses a damage identification method based on Lamb wave temperature compensation, a temperature compensation model is constructed, and the influence of temperature on a Lamb wave signal is considered through temperature compensation. Because temperature change can change the physical properties of a material, such as the elastic modulus and density, and further affect the propagation characteristics of a Lamb wave, so that the original Lamb wave signal is mixed with interference caused by temperature factors. After temperature compensation, the temperature-related interference is removed, so that damage identification based on the compensated Lamb wave signal is more accurate. In a to-be-tested component with large temperature changes in different seasons or working environments, the damage size can be more reliably detected, and misjudgment or missed judgment caused by temperature changes can be avoided. And by collecting Lamb wave signals at different temperatures on the target structure of the to-be-tested component, the method is suitable for various to-be-tested components with similar structures, has a certain universality, and can be applied to damage detection in different fields.

[0017] Further, when the Lamb wave temperature compensation model is constructed, the original data is truncated to obtain basic symmetric modes as sample data, and a temperature label is given according to the environmental temperature. This processing method effectively extracts key information related to damage and temperature, reduces the redundant components in the data, improves the data quality, and is beneficial to the training and learning of the subsequent model.

[0018] The Lamb wave temperature compensation network is constructed based on a conditional generative adversarial network (CGAN), and the mapping relationship of Lamb wave signals at different temperatures is learned through the adversarial training of the generator and the discriminator. The alternating training of the generator and the discriminator until the Nash equilibrium is reached can make the reconstructed signal generated by the generator as close as possible to the target real signal (Lamb wave signal at normal temperature), so as to realize effective compensation of the Lamb wave signal at the temperature of the very warm environment. This adversarial training mechanism has strong learning ability and adaptability, and can capture the complex influence law of temperature change on the Lamb wave signal. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 FIG. 1 is a flowchart of an adversarial training process of a temperature compensation model of the application; Figure 2 FIG. 2 is an experimental overall arrangement of an aluminum plate 6061 experimental piece in Embodiment 3 of the application; Figure 3Lamb wave signal schematic diagram for embodiment 3 of the present application; Figure 4 Structure diagram of the generator provided by the present application; Figure 5a Lamb wave signal corresponding to the temperature of-30 DEG C and normal temperature under the condition of no damage; Figure 5b Lamb wave signal corresponding to the temperature of-30 DEG C and normal temperature under the condition of damage size of 5mm; Figure 5c Lamb wave signal corresponding to the temperature of-10 DEG C and normal temperature under the condition of damage size of 4.5mm; Figure 5d Lamb wave signal corresponding to the temperature of 70 DEG C and normal temperature under the condition of damage size of 5mm; Figure 5e Lamb wave signal corresponding to the temperature of 50 DEG C and normal temperature under the condition of damage size of 3mm; Figure 5f Lamb wave signal corresponding to the temperature of-10 DEG C and normal temperature under the condition of damage size of 7mm; Figure 6a Performance of Lamb wave signal without temperature compensation in damage prediction and damage diagnosis; Figure 6b Performance of the damage identification method based on Lamb wave temperature compensation in damage diagnosis; Figure 7 Flow chart of the damage identification method based on Lamb wave temperature compensation; Figure 8 Block diagram of the damage identification system based on Lamb wave temperature compensation. DETAILED DESCRIPTION

[0020] In order to make the objectives, technical solutions and advantages of the present application clearer and more comprehensible, the following will be further described in detail in combination with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application, and are not used to limit the present application, that is, the described embodiments are only a part of the embodiments of the present application, but not all the embodiments.

[0021] The components described and shown in the drawings and embodiments of the present application can be arranged and designed in various different configurations, therefore, the detailed description of the embodiments of the present application provided in the following drawings is not intended to limit the scope of the claimed present application, but only represents a selected embodiment of the present application. Based on the drawings and embodiments of the present application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the present application.

[0022] It is to be understood that the terms "comprising," "including," and other variants mean "including but not limited to" so that processes, elements, methods, articles or apparatus that consist of, include, or are described with "comprising", "including", and other variants are not limited to those elements specified after the complementing term.

[0023] The application provides a damage identification method based on Lamb wave temperature compensation, which can detect damage in a complex service environment.

[0024] The features and performance of the application are further described in detail below in combination with embodiments.

[0025] Embodiment 1 As shown in Figure 7 The application discloses a damage identification method based on Lamb wave temperature compensation, comprising the following processes: Obtain the Lamb wave signal and the actual environment temperature of the measured component; Input the Lamb wave signal and the actual environment temperature into the pre-constructed Lamb wave temperature compensation network model, and output the temperature-compensated Lamb wave signal; Based on the temperature-compensated Lamb wave signal, the damage of the measured component is identified to obtain the damage identification result.

[0026] Embodiment 2 Based on embodiment 1, please refer to Figure 1 The application provides a construction method of the Lamb wave temperature compensation network model, comprising the following processes: According to the preset size, process a crack on the target structure of the measured component as an artificial damage; N sensors are arranged at the target structure of the measured component; the sensors obtain the Lamb wave signal of different damage sizes on the measured component under multiple different environment temperatures as original data; the multiple different environment temperatures include normal temperature and abnormal temperature environment temperature; Then, using a fixed-size sliding window, the original data is truncated to obtain the basic symmetric mode of the Lamb wave signal, i.e. S0 mode, as sample data; According to the environment temperature, the sample data is given a temperature label to obtain a sample set; the sample set is divided into a training set and a test set; Construct a Lamb wave temperature compensation network based on a conditional generative adversarial network; Input the training set into the Lamb wave temperature compensation network for adversarial training until the generator and the discriminator reach Nash equilibrium; The test set is input into the trained Lamb wave temperature compensation network, which outputs the temperature-compensated Lamb wave signal until the test is passed, thus obtaining the constructed Lamb wave temperature compensation network model.

[0027] Specifically, the original data is truncated using a sliding window of fixed size to obtain the basic symmetrical modes of the Lamb wave signal, and then normalized according to the following formula.

[0028]

[0029]

[0030] in, The signal is after normalization. For signal The average value; Let n be the number of signal sampling points, where n is the nth sampling point.

[0031] Specifically, the Lamb wave temperature compensation network includes a generator and a discriminator. For example... Figure 4 As shown, the designed generator consists of an encoder, a transformer, and a decoder. The encoder and decoder can be viewed as inverse operations, while the transformer mainly acts as a transition to mitigate gradient vanishing.

[0032] The encoder and decoder architecture consists of multiple trainable convolutional layers used to extract features from the input Lamb wave signal. These convolutional layers include convolutionals, normalization instances, and activation functions. At each convolutional layer, all kernels are convolved with the input and a bias is added. The features are then transformed using the activation function, as shown in the following expression.

[0033]

[0034] In the formula, for Layer The output of each convolutional feature map. "Indicates convolution operation, Indicates in The i-th neuron in layer and Layer Weights between feature maps; This is the activation function.

[0035] The converter consists of multiple residual modules, which are designed to connect the encoder and decoder. The main purpose is to alleviate the gradient vanishing and gradient exploding problems that occur during the data mapping process.

[0036] The discriminator and generator use a discriminative loss function formed by their own outputs. Alternately update the parameters of the discriminator and the generator.

[0037] The adversarial training process is divided into two steps: First, train the discriminator D: input the original input signal and the temperature label y into the generator, which generates the reconstructed signal ; the reconstructed signal and the target real signal are input into the discriminator D, which discriminates between the input reconstructed signal and the target real signal, constructs the cross-entropy loss, and optimizes the parameters of the discriminator; in this process, the parameters of the generator G are frozen; The original input signal is the Lamb wave signal collected in a very warm environment, and the real signal is the Lamb wave signal collected at room temperature; Second, the generator should be trained: Fix the parameters of the discriminator D, input the original input signal and the temperature label y into the generator, which generates the reconstructed signal ; the reconstructed signal and the target real signal are evaluated by the loss function, and if the similarity evaluation meets the preset requirements, the training is stopped; if the similarity evaluation does not meet the preset requirements, the similarity loss is backpropagated to the generator, and the parameters of the generator are iteratively updated until the similarity evaluation meets the preset requirements, and the generator training is completed.

[0038] Alternately train the generator G and the discriminator D until Nash equilibrium is reached.

[0039] wherein,

[0040]

[0041] is the reconstructed signal, represents maximizing the loss function in the case of fixing the discriminator D, and minimizing the loss function in the case of fixing the generator G; y is the temperature label.

[0042] After obtaining the reconstructed signal through the generator G, a Lamb wave signal similarity measurement method is proposed, which is used as a loss to train and optimize the entire Lamb wave temperature compensation network. The similarity between the reconstructed signal and the target real signal is calculated as follows:

[0043] wherein, n is the number of samples, and respectively, the optimization of this loss function can make the gap between the generated network and the target real signal smaller and smaller, and realize temperature compensation. The whole optimization process updates the parameters in the Lamb wave temperature compensation network through the Adam optimization algorithm. The weights of this network are trained by the data itself, which has the characteristics of self-adaptation. Therefore, it can better adapt to the Lamb wave temperature compensation under variable temperature.

[0044] The test set is input into the trained Lamb wave temperature compensation network, and the temperature compensated Lamb wave signal of the test set is output; then the temperature compensated Lamb wave signal is input into the damage detection network, and the predicted damage size is output; the error calculation is performed on the predicted damage size and the preset crack size until the error of the predicted damage size and the preset crack size meets the requirements, and then the Lamb wave temperature compensation network model is constructed.

[0045] The damage detection network can adopt the existing convolutional neural network, which can effectively capture the local features and spatial information in the Lamb wave signal, and has high accuracy in damage detection.

[0046] Embodiment 3 This embodiment is a specific test verification method, as follows: Twenty-one 2mm thick aluminum plates 6061 are used as experimental pieces, with a size of 400mm*200mm, and the mechanical parameters are shown in Table 1. Different damage sizes are preset on each aluminum plate, and the damage sizes are 0-10mm with an interval of 0.5mm. The specific structure design and sensor arrangement position are the same as those in the above Figure 2 The programmable temperature chamber is used to change the environmental temperature to simulate the service environment of the actual structure.

[0047] Table 1: Mechanical parameters of aluminum plate 6061

[0048] The overall experimental arrangement is shown in Figure 2 The programmable temperature chamber is used to change the environmental temperature to simulate the service environment of the actual structure. The temperature range simulated by the present application is-40℃ to 80℃, and after a certain preset temperature is stably maintained for 10min, one of the sensors is used as an exciter, and the other sensor receives the Lamb wave signal. For each aluminum plate, the operation is repeated, and the Lamb signal waveform corresponding to different environmental temperatures received by the sensor is recorded, as shown in Figure 3 In this way, a large amount of data is obtained as raw data.

[0049] Then the original data is truncated using a fixed-size sliding window to obtain the basic symmetric mode of Lamb wave signal, i.e. S0 mode, and normalized as sample data; According to the environmental temperature, the sample data is given a temperature label to obtain a sample set; the sample set is divided into a training set and a test set; The training set is input into the Lamb wave temperature compensation network for adversarial training until the generator and the discriminator reach Nash equilibrium; The test set is input into the optimized Lamb wave temperature compensation network to output the temperature-compensated Lamb wave signal until the test is qualified, and the constructed Lamb wave temperature compensation network model is obtained.

[0050] As shown in Figure 5a , the original input signal corresponding to the temperature of-30℃ and the undamaged condition is input into the constructed Lamb wave temperature compensation network, and the temperature-compensated Lamb wave signal, i.e. the reconstructed signal, is output. At the same time, the target real signal corresponding to the temperature of 20℃ and the undamaged condition is also input as a control.

[0051] As shown in Figure 5b , the original input signal corresponding to the temperature of-30℃ and the damage size of 5mm is input into the constructed Lamb wave temperature compensation network, and the temperature-compensated Lamb wave signal, i.e. the reconstructed signal, is output. At the same time, the target real signal corresponding to the temperature of 20℃ and the damage size of 5mm is also input as a control.

[0052] As shown in Figure 5c , the original input signal corresponding to the temperature of-10℃ and the damage size of 4.5mm is input into the constructed Lamb wave temperature compensation network, and the temperature-compensated Lamb wave signal, i.e. the reconstructed signal, is output. At the same time, the target real signal corresponding to the temperature of 20℃ and the damage size of 4.5mm is also input as a control.

[0053] As shown in Figure 5d , the original input signal corresponding to the temperature of 70℃ and the damage size of 5mm is input into the constructed Lamb wave temperature compensation network, and the temperature-compensated Lamb wave signal, i.e. the reconstructed signal, is output. At the same time, the target real signal corresponding to the temperature of 20℃ and the damage size of 5mm is also input as a control.

[0054] As shown in Figure 5e , the original input signal corresponding to the temperature of 50℃ and the damage size of 3mm is input into the constructed Lamb wave temperature compensation network, and the temperature-compensated Lamb wave signal, i.e. the reconstructed signal, is output. At the same time, the target real signal corresponding to the temperature of 20℃ and the damage size of 3mm is also input as a control.

[0055] As shown in Figure 5fThe temperature is -10°C, the damage size is 7mm, the corresponding original input signal is input into the built Lamb wave temperature compensation network, and the output temperature compensated Lamb wave signal is the reconstructed signal. Meanwhile, the corresponding target real signal under the condition of temperature 20°C and damage size 7mm is also input as a control.

[0056] Figure 5a to Figure 5f Among them, the three curves respectively represent the original input signal, the target real signal and the reconstructed signal, and the target real signal and the reconstructed signal are very close, proving that the temperature compensation model of the application can realize temperature compensation.

[0057] The Lamb wave signals with and without temperature compensation are input into the damage detection network to predict the damage size.

[0058] The root mean square error (RMSE) is used to evaluate the temperature compensation performance of the temperature compensation network method, and multiple error indicators are used to evaluate the damage diagnosis performance. MAE represents the average absolute error between the predicted damage size and the actual measured damage size, and its formula is as follows:

[0059] RMSE is the square root of the average square of the error between the actual measured damage size and the predicted damage size:

[0060] STD can reflect the dispersion degree between individuals in the data set, and its calculation formula is as follows:

[0061] represents the predicted damage size; represents the actual measured damage size; m represents the sample number.

[0062] The three test indicators are shown in Table 2. The root mean square error of the Lamb wave signal output by the temperature compensation network model of the application for damage identification of the to-be-tested component is less than 0.5mm, while the root mean square error of the Lamb wave signal without temperature compensation directly input into the damage detection network for identification is greater than 0.5mm, indicating that the damage identification method of the application has acceptable performance in predicting the damage size.

[0063] Table 2: Model performance

[0064] In order to more intuitively determine the fluctuation of the data, the error and the interquartile range of the predicted damage size on all sample points are shown in Figure 6a and Figure 6b ​The box plot shown. The middle of the box is a red line, indicating the median of the prediction error. There is a blue line indicating the average number of prediction errors. The top and bottom of the box are the upper and lower quartiles of the data. The values between the upper and lower quartiles are described as the middle 50%. Overall, the diagnostic method without temperature compensation predicts the damage size to cause a significantly higher quartile range of prediction errors than the method proposed in the present application.

[0065] Embodiment 4 As Figure 8 The present application discloses a Lamb wave temperature compensation based damage identification system, comprising the following processes: A data acquisition module for acquiring Lamb wave signals and actual environmental temperature of the component to be measured; A temperature compensation module for inputting the Lamb wave signals and the actual environmental temperature into a pre-constructed Lamb wave temperature compensation network model, and outputting the temperature-compensated Lamb wave signals; A damage identification module for identifying the damage of the component to be measured based on the temperature-compensated Lamb wave signals, and obtaining the damage identification result.

[0066] Embodiment 5 The present application also discloses a computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the Lamb wave temperature compensation based damage identification method when executing the computer program, and the memory can include a memory such as a high-speed random access memory, and can also include a non-volatile memory such as at least one disk memory; the processor, network interface and memory are connected to each other through an internal bus, which can be an industry standard architecture bus, a peripheral component interconnect standard bus, an extended industry standard structure bus, etc., and the bus can be divided into an address bus, a data bus and a control bus. The memory is used to store programs, specifically, the programs can include program codes, and the program codes include computer operation instructions. The memory can include a memory and a non-volatile memory, and provide instructions and data to the processor.

[0067] Embodiment 6 The present application also discloses a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the steps of the Lamb wave temperature compensation based damage identification method. Specifically, the computer readable storage medium includes but is not limited to, for example, a volatile memory and / or a non-volatile memory. The volatile memory can include a random access memory and / or a cache memory, etc. The non-volatile memory can include a read-only memory, a hard disk, a flash memory, an optical disk, a magnetic disk, etc.

[0068] Those skilled in the art will appreciate that embodiments of the application can be readily used as a method, apparatus, or computer program product. Accordingly, the application can take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the application can take the form of a computer program product on one or more computer readable storage media (including, but not limited to, disk memory, optical storage etc.) embodying computer readable program code.

[0069] The application is described in reference to the flowchart and / or block diagrams of the method, apparatus (system) and computer program product according to embodiments of the application. It will be understood that each block of the flowchart and / or block diagrams, and combinations of blocks in the flowchart and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart and / or block diagram block or blocks. Figure One one or more functions specified in the flowchart and / or block diagram block or blocks. Figure One means for carrying out each of the one or more functions specified in the flowchart and / or block diagram block or blocks.

[0070] These computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture including instructions which implement the flowchart and / or block diagram block or blocks. Figure One one or more functions specified in the flowchart and / or block diagram block or blocks. Figure One means for carrying out each of the one or more functions specified in the flowchart and / or block diagram block or blocks.

[0071] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the flowchart and / or block diagram block or blocks. Figure One one or more functions specified in the flowchart and / or block diagram block or blocks. Figure One means for carrying out each of the one or more functions specified in the flowchart and / or block diagram block or blocks.

[0072] Finally, it should be noted that the above-mentioned embodiments are merely intended for describing and illustrating, not limiting, the technical solution of the present application. Although the present application has been described in detail with reference to the above-mentioned embodiments, those skilled in the art should understand that the specific embodiments of the present application can be modified or replaced by equivalents without departing from the spirit and scope of the present application, and any modification or equivalent replacement without departing from the spirit and scope of the present application should be covered in the protection scope of the claims of the present application.

Claims

1. A Lamb wave temperature-compensated damage identification method, characterized in that, The method comprises the following steps: obtaining a Lamb wave signal of a to-be-tested component and an actual ambient temperature; inputting the Lamb wave signal and the actual ambient temperature into a pre-constructed Lamb wave temperature compensation network model to output a temperature-compensated Lamb wave signal; based on the temperature-compensated Lamb wave signal, identifying damage of the to-be-tested component to obtain a damage identification result.

2. The damage identification method based on Lamb wave temperature compensation according to claim 1, characterized in that, The construction process of the pre-constructed Lamb wave temperature compensation model is as follows: processing a crack on a target structure of the to-be-tested component according to a preset crack size; setting a sensor at the target structure, and the sensor obtains Lamb wave signals of different damage sizes on the to-be-tested component under multiple different ambient temperatures as original data; the multiple different ambient temperatures include normal temperature and abnormal temperature ambient temperatures; cutting the original data to obtain basic symmetric modes of the Lamb wave signals, and performing normalization processing to obtain sample data; assigning temperature labels to the sample data according to the ambient temperatures to obtain a sample set; and dividing the sample set into a training set and a test set; based on a conditional generative adversarial network, constructing a Lamb wave temperature compensation network; the Lamb wave temperature compensation network comprises a generator and a discriminator; inputting the training set into the Lamb wave temperature compensation network to perform adversarial training until the generator and the discriminator reach a Nash equilibrium; inputting the test set into the trained Lamb wave temperature compensation network to output a temperature-compensated Lamb wave signal until the test is qualified, and obtaining a constructed Lamb wave temperature compensation network model.

3. The damage identification method based on Lamb wave temperature compensation according to claim 2, characterized in that, inputting the test set into the trained Lamb wave temperature compensation network to output a temperature-compensated Lamb wave signal until the test is qualified, and obtaining a constructed Lamb wave temperature compensation network model, specifically: inputting the test set into the trained Lamb wave temperature compensation network to output a temperature-compensated Lamb wave signal, and using the temperature-compensated Lamb wave signal to identify damage of a target structure crack of the to-be-tested component to output a predicted damage size; performing error calculation on the predicted damage size and the preset crack size until the error of the predicted damage size and the preset crack size meets the requirements, and then the Lamb wave temperature compensation network model is constructed.

4. The damage identification method based on Lamb wave temperature compensation according to claim 2, characterized in that, The process of inputting the training set into the Lamb wave temperature compensation network to perform adversarial training is as follows: training the discriminator: inputting the original input signal and the temperature label into the generator to generate a reconstructed signal; inputting the reconstructed signal and the target real signal into the discriminator, and feeding back the output result to the discriminator through a loss function to complete updating of parameters of the discriminator; in this process, the parameters of the generator are frozen; training the generator: fixing the parameters of the discriminator, inputting the original input signal and the temperature label into the generator, and the generator generates a reconstructed signal; The reconstructed signal and the target real signal are input into the discriminator, similarity evaluation is performed through a loss function, if the similarity evaluation reaches a preset requirement, the training is stopped; if the similarity evaluation does not reach the preset requirement, the similarity loss is back-propagated to the generator, the parameters of the generator are iteratively updated until the similarity evaluation reaches the preset requirement, and then the generator training is completed; The generator and the discriminator are alternately trained until a Nash equilibrium is reached. The original input signal is a Lamb wave signal collected in a very warm environment, and the target real signal is a Lamb wave signal collected at normal temperature.

5. The Lamb wave temperature-compensated damage identification method of claim 1, wherein, The damage identification method based on the temperature-compensated Lamb wave signal comprises the following steps: The temperature-compensated Lamb wave signal is input into the damage detection network, and a predicted damage result is output.

6. The damage identification method based on Lamb wave temperature compensation according to claim 5, characterized in that, The damage detection network adopts a convolutional neural network.

7. A Lamb wave temperature-compensated damage identification system, characterized by, The method comprises the following steps: A data acquisition module is configured to acquire a Lamb wave signal and an actual environment temperature of a to-be-tested component; A temperature compensation module is configured to input the Lamb wave signal and the actual environment temperature into a pre-constructed Lamb wave temperature compensation network model, and output a temperature-compensated Lamb wave signal; A damage identification module is configured to perform damage identification on the to-be-tested component based on the temperature-compensated Lamb wave signal, and obtain a damage identification result.

8. A computer device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, The processor executes the computer program to implement the steps of the damage identification method based on Lamb wave temperature compensation according to any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, the computer program comprising instructions that, when executed by a computer, cause the computer to perform the method of any one of claims 1 to 8. The computer program is executed by the processor to implement the steps of the damage identification method based on Lamb wave temperature compensation according to any one of claims 1 to 6.

10. A computer program product comprising computer programs / instructions, characterized in that, The computer program / instruction is executed by the processor to implement the steps of the damage identification method based on Lamb wave temperature compensation according to any one of claims 1 to 6.

Images