A method and system for CFRP post-impact integrity analysis

Abstract

The application provides a CFRP post-lightning integrity analysis method and system, and belongs to the technical field of electric digital data processing.The application combines portable ATR-FTIR testing and artificial neural networks, proposes a lightning post-integrity analysis nondestructive testing method for CFRP, and can accurately predict the residual strength of the composite material after lightning strike.The method can directly, accurately and efficiently evaluate the post-lightning integrity of the CFRP, can avoid the damage and high cost of the traditional method, can overcome the defects of the complex structure and error accumulation of the existing neural network model, and realizes more simple and accurate post-lightning CFRP integrity evaluation.

Description

Technical Field

[0001] This invention belongs to the field of electrical digital data processing technology, and relates to the use of machine learning in computer-aided design, and to neural networks. Specifically, it relates to a method and system for CFRP integrity analysis after lightning strike. Background Technology

[0002] Carbon fiber reinforced polymer (CFRP) composites have been widely used in aerospace, transportation, and other fields. However, when CFRP is used in the main load-bearing structures of modern engineering structures such as aircraft, its insufficient electrical conductivity makes it difficult to quickly conduct high-amplitude lightning currents, leaving it vulnerable to lightning strikes. Although modern aircraft employ lightning protection (LSP) systems such as metal mesh / sheets, improperly designed LSP systems can still cause thermal damage to the CFRP surface after a lightning strike. This thermal damage can sometimes be detected visually, manifesting as blistering or delamination. In the long term, such thermal damage can have catastrophic consequences. Therefore, accurately assessing the structural integrity of CFRP after a lightning strike is crucial for improving flight safety, reducing maintenance costs, and extending the service life of modern composite aircraft.

[0003] Traditional methods for analyzing the integrity of composite materials include in-plane compression tests, drop hammer impact tests, or quasi-static indentation tests. Liu Xiaoming's team evaluated the residual strength of carbon fiber composite laminates after lightning strikes through axial compression tests (Liu Xiaoming, Yu Xiaosang, Wang Jiu, et al. Axial compression tests of carbon fiber composite laminates with different protective features after lightning strikes [J]. Journal of Materials Science and Engineering, 2016(3):4.DOI:10.14136 / j.cnki.issn1673-2812.2016.03.008.). However, this test requires sampling from engineering structures and testing in a laboratory. In aerospace applications, the cost of preparing samples from aircraft is extremely high; therefore, in-situ nondestructive testing (NDI) may be an effective way to solve this problem. However, existing ultrasonic testing can only detect the location and area of ​​damage, and cannot quantitatively assess the degree of damage and residual mechanical properties.

[0004] The prior art Chinese invention patent CN118571381A discloses a method for assessing lightning damage to composite materials based on neural networks. It predicts the damage area, damage depth, and residual strength by cascading three neural network models. This prior art has the following disadvantages: (1) The process is cumbersome, and the cascaded network structure leads to the accumulation of errors at each level, which reduces the prediction accuracy; (2) The macroscopic mechanical properties of materials are actually determined by their microscopic chemical structure (determined by the state of chemical bonds and molecular structure). Optical images can only reflect the damage morphology and cannot directly reveal the relationship between structure and performance. However, this prior art relies on visual images as input, which can only capture macroscopic morphological features and cannot reflect changes in the microscopic chemical structure of materials.

[0005] The existing Chinese invention patent CN119164911A discloses a method and system for determining the aging degree of silicone rubber surface, which uses FTIR data to train a neural network model to achieve aging degree classification. This existing technology has the following disadvantages: (1) Its application is limited to silicone rubber materials, and the evaluation target is the aging degree rather than the remaining strength. It fails to establish a direct quantitative relationship (mapping relationship) between chemical structure and mechanical properties; (2) The method does not consider the material response characteristics under extreme thermal effects such as lightning strikes, and its model structure and feature extraction method are not suitable for lightning strike damage assessment of CFRP. Summary of the Invention

[0006] This invention is made to solve the above-mentioned problems, and aims to provide a method and system for CFRP integrity analysis after lightning strike.

[0007] This invention provides a method for analyzing the integrity of CFRP after a lightning strike, characterized by the following steps: S10, by conducting high-temperature exposure tests on CFRP at different temperatures and comparing the changes in its infrared spectra, the evolution of chemical bonds in CFRP at different temperatures is determined to identify key chemical structures; S20, after scanning the lightning-struck area of ​​CFRP with an ATR-FTIR spectrometer to obtain its historical infrared spectral data, characteristic peaks of key chemical structures are selected and their intensities are calculated as target input data; S30, historical mechanical property data of CFRP after a lightning strike is obtained through mechanical property testing and used as target output data; S40, a mapping relationship between target input data and target output data is established using an artificial neural network to train a CFRP post-lightning strike integrity prediction model; S50, after scanning the lightning-struck area of ​​the CFRP to be analyzed with an ATR-FTIR spectrometer to obtain its historical infrared spectral data, characteristic peaks of key chemical structures are selected and their intensities are calculated, and then input into the prediction model, correspondingly outputting the mechanical property data of CFRP to achieve CFRP post-lightning strike integrity analysis.

[0008] The CFRP integrity analysis method after lightning strike provided by this invention may also have the following features: In step S10, the CFRP is placed in a muffle furnace and ablated at a constant temperature of 100 ℃ to 300 ℃ in an air atmosphere, and the changes in the infrared spectrum of the CFRP at different temperatures are compared, thereby obtaining the evolution law of chemical bonds of the CFRP at different temperatures to determine the key chemical structures therein.

[0009] The CFRP post-lightning-strike integrity analysis method provided by this invention may also have the following features: in steps S10 to S20, the key chemical structures include the CH stretching vibrations of methyl and methylene groups, the stretching vibrations of carbonyl groups, and the skeletal vibrations of benzene rings.

[0010] The CFRP post-lightning-strike integrity analysis method provided by the present invention may also have the following feature: in step S20, baseline adjustment and smoothing are performed on the historical infrared spectral data.

[0011] The CFRP post-lightning-strike integrity analysis method provided by this invention may also have the following feature: In step S30, the mechanical performance test includes a three-point bending test.

[0012] Historical mechanical property data includes data on flexural strength.

[0013] The CFRP post-lightning strike integrity analysis method provided by the present invention may also have the following features: in step S40, constraints are set for the prediction models under different training conditions, including the setting of the maximum number of iterations and the setting of the minimum training target convergence error, and the artificial neural network is a BP neural network.

[0014] The CFRP post-lightning strike integrity analysis method provided by this invention may also have the following features: In step S40, the prediction model under different training conditions is evaluated, and the accuracy of its prediction results is reflected by calculating the relative error. If the relative error does not meet the evaluation requirements, the prediction accuracy of the prediction model is improved by expanding the training dataset of the target input data and the target output data and optimizing the structure of the artificial neural network until the relative error meets the evaluation requirements.

[0015] The CFRP post-lightning strike integrity analysis method provided by this invention may also have the following features: wherein optimizing the structure of the artificial neural network includes adjusting the number of layers, adjusting the number of hidden layer neurons, and optimizing the activation function.

[0016] The CFRP post-lightning strike integrity analysis method provided by this invention may also have the following feature: wherein, in step S50, the CFRP post-lightning strike integrity analysis further includes quantifying the concentration of key chemical structures.c The relationship between the intensity of characteristic peaks and key chemical structures, and the intensity of characteristic peaks is obtained from absorbance values ​​in historical infrared spectral data. A The concentration of key chemical structures was calculated based on the difference from the baseline. c : In the above formula, θ Angle of incidence n s and n c The refractive indices of CFRP and ATR crystals are respectively. K The molar absorption coefficient, K It depends on the CFRP material and the wavelength of the incident light. λ .

[0017] This invention also provides a CFRP post-lightning-strike integrity analysis system, characterized by using any of the aforementioned CFRP post-lightning-strike integrity analysis methods, comprising: a data input module for users to input target input data and target output data; a model building and training module, based on an artificial neural network and connected to the data input module, for establishing a mapping relationship between the target input data and the target output data, thereby training a CFRP post-lightning-strike integrity prediction model; a prediction analysis module, connected to the model building and training module, for users to input the intensity of characteristic peaks of key chemical structures in the historical infrared spectral data of the CFRP lightning-strike area to be analyzed, and then outputting the mechanical performance data of the CFRP corresponding to the prediction model trained by the model building and training module; and an output module, connected to the prediction analysis module, for outputting and displaying the mechanical performance data and generating an integrity evaluation report.

[0018] The CFRP post-lightning strike integrity analysis method and system of the present invention have the following beneficial effects:

[0019] (1) This invention proposes a non-destructive testing method for CFRP integrity analysis after lightning strike by combining portable ATR-FTIR testing with artificial neural networks, which can accurately predict the residual strength of composite materials after lightning strike. The handheld ATR-FTIR can complete the on-site chemical structure acquisition without disassembly or sampling; combined with the trained prediction model (BP-ANN model), the residual bending strength is output instantaneously, realizing on-site detection and quantitative evaluation, avoiding the problems of high cost, long cycle and difficulty in quantification in traditional evaluation methods.

[0020] (2) This invention relies on the sensitivity of FTIR to the characteristic functional groups (-C=O, -CH2-, benzene ring skeleton, etc.) of composite materials after lightning strike to establish a nonlinear mapping relationship between the chemical structure concentration of composite materials after lightning strike and the macroscopic residual strength. For the first time, it realizes the direct correlation between matrix chemical degradation and structural load-bearing capacity, and provides data support for revealing the lightning damage mechanism, predicting life and formulating maintenance standards.

[0021] (3) By dynamically optimizing the number of network layers, nodes and activation function, the prediction error of this invention is stably converged to <5%. This method provides a new method for non-destructive field evaluation of aircraft composite material structures, and also provides a general method for evaluating the residual mechanical properties of composite material structures under other conditions.

[0022] (4) The present invention can directly, accurately and efficiently assess the integrity of CFRP after lightning strike. It can avoid the destructiveness and high cost of traditional methods, and overcome the drawbacks of complex structure and error accumulation of existing neural network models, thus achieving a simpler and more accurate assessment of CFRP integrity after lightning strike. Attached Figure Description

[0023] Figure 1 This is a flowchart illustrating the CFRP post-lightning strike integrity analysis method according to an embodiment of the present invention.

[0024] Figure 2 This is a comparison of the infrared spectra of CFRP before and after a lightning strike in the test example of this invention.

[0025] Figure 3 This is a comparison diagram of the mechanical properties of CFRP before and after a lightning strike in the test example of this invention.

[0026] Figure 4 This is a CFRP integrity evaluation diagram after a lightning strike, based on a test example of the present invention.

[0027] Figure 5 This is a neural network design diagram of a test example of the present invention. Detailed Implementation

[0028] To make the technical means, creative features, objectives and effects of the present invention easy to understand, the following embodiments, in conjunction with the accompanying drawings, specifically illustrate a CFRP post-lightning strike integrity analysis method and system of the present invention.

[0029] Example

[0030] Figure 1 This is a flowchart illustrating the CFRP post-lightning strike integrity analysis method according to an embodiment of the present invention.

[0031] like Figure 1 As shown, this embodiment provides a method for CFRP integrity analysis after a lightning strike, including the following steps:

[0032] S10, determine the key chemical structures in CFRP, including the following sub-steps S11~S13:

[0033] S11, High-temperature exposure test: CFRP was placed in a muffle furnace and ablated at a constant temperature of 100 ℃~300 ℃ for 15 min in an air atmosphere. The temperature distribution on the sample surface and inside was similar, and the degree of ablation damage was similar.

[0034] S12, compare the changes in the infrared spectrum of CFRP after exposure tests at room temperature and high temperature.

[0035] S13, obtain the evolution of chemical bonds in CFRP at different temperatures, and thus determine that the key chemical structures are methylene, methyl, carbonyl and benzene ring skeleton.

[0036] S20, Obtain the target input data, including the following sub-steps S21~S22:

[0037] S21. The infrared spectral history data of the CFRP was obtained by scanning the lightning strike area using a portable ATR-FTIR spectrometer.

[0038] S22: After baseline adjustment and smoothing of historical infrared spectral data, characteristic peaks of key chemical structures are selected and their intensities are calculated as target input data.

[0039] Among them, the intensity of the characteristic peak is obtained from the absorbance values ​​in the historical infrared spectral data. A Calculate the difference from the baseline.

[0040] S30 uses historical data on the mechanical properties of CFRP after a lightning strike obtained through a three-point bending test as the target output data.

[0041] Among them, the historical mechanical property data is the data on bending strength.

[0042] S40, train the CFRP post-lightning integrity prediction model using a BP neural network, specifically including the following sub-steps S41~S43:

[0043] S41, use a BP neural network to establish a mapping relationship between the target input data obtained in step S22 and the target output data obtained in step S30, thereby training a CFRP post-lightning integrity prediction model.

[0044] S42 sets constraints on the prediction models obtained under different training conditions. The constraints include setting the maximum number of iterations and setting the minimum training objective convergence error.

[0045] S43: Evaluate the prediction model under different training conditions. The accuracy of the prediction result is reflected by calculating the relative error. If the relative error does not meet the evaluation requirements, return to step S41 and improve the prediction accuracy of the prediction model by expanding the training dataset of the target input data and target output data and optimizing the structure of the artificial neural network until the relative error meets the evaluation requirements.

[0046] Optimizing the structure of artificial neural networks includes adjusting the number of layers, adjusting the number of neurons in the hidden layers, and optimizing the activation function.

[0047] S50, Integrity Analysis, includes the following sub-steps S51~S53:

[0048] S51. After scanning the lightning-struck area of ​​CFRP to be analyzed using an ATR-FTIR spectrometer to obtain its historical infrared spectral data, the characteristic peaks of the key chemical structures are selected and their intensities are calculated. The data are then input into the prediction model, and the mechanical property data of CFRP are output accordingly.

[0049] S52, concentration of quantitative key chemical structures c Specifically, it includes the following sub-steps S521~S523:

[0050] S521, Determine the thickness of the CFRP that can be penetrated by infrared light via a portable ATR-FTIR spectrometer / the thickness of the absorption layer:

[0051] (Equation 1)

[0052] In the above formula, θ Angle of incidence n s and n c The refractive indices of CFRP and ATR crystals are respectively. λ This represents the wavelength of the incident light. Among them, d p and λ ,θ, n s and n c Related, when λ When both θ and θ are fixed, if n s If it is a fixed value, then d p It is also a fixed value.

[0053] S522, the absorbance formula is: A = Kd p c (Equation 2)

[0054] In the above formula,A This is the absorbance value. K The molar absorption coefficient, K It depends on the CFRP material and the wavelength of the incident light. λ , c The concentration of key chemical structures in CFRP after a lightning strike.

[0055] S523, combining Equations 1 and 2, determines the concentration of key chemical structures in CFRP after a lightning strike. c :

[0056] .

[0057] S53, combine the mechanical property data output in step S51 and the concentration of key chemical structures output in step S52. c Together they form the CFRP integrity analysis report after a lightning strike.

[0058] This embodiment also provides a CFRP post-lightning strike integrity analysis system, which uses the CFRP post-lightning strike integrity analysis method in this embodiment, including a data input module, a model building and training module, a prediction analysis module, and an output module.

[0059] The data input module is used by the user to input the target input data obtained in step S22 and the target output data obtained in step S30.

[0060] The model building and training module, based on an artificial neural network and connected to the data input module, is used to establish a mapping relationship between the target input data and the target output data according to the method in step S40, thereby training a CFRP post-lightning integrity prediction model.

[0061] The predictive analysis module, connected to the model building and training module, is used to, according to steps S51 and S52, allow users to input the intensity of the characteristic peaks of the key chemical structures in the CFRP lightning-struck area to be analyzed, and then output the mechanical property data of the CFRP and the concentration of the key chemical structures accordingly. c .

[0062] The output module is used to output and display the mechanical property data and the concentration of key chemical structures output by the predictive analysis module according to the method in step S53. c And generate an integrity evaluation report.

[0063] Test case

[0064] This test case uses the CFRP post-lightning strike integrity analysis system in the embodiments and performs the test according to the CFRP post-lightning strike integrity analysis method in the embodiments.

[0065] The CFRP integrity analysis method after a lightning strike in this test case is largely similar to that in the previous examples and will not be repeated here. The specific data / method selections for this test case are as follows:

[0066] In step S21, the scanning position of the portable ATR-FTIR spectrometer is selected using an array method, expanding outward from the center of the CFRP lightning injection point. One infrared spectral data is collected every 1.5 cm, and all collected infrared spectral data are used as historical infrared spectral data.

[0067] In step S22: (1) the number of smoothing points is set to 15; (2) the wavenumbers of the characteristic peaks of the selected key chemical structures are: 2924 cm⁻¹ for the symmetric stretching vibrations of —CH₃ and —CH₂— -1 2870 cm corresponding to asymmetric stretching vibration -1 1663 cm⁻¹ corresponding to carbonyl stretching vibration -1 1590 cm corresponding to the vibration of the benzene ring skeleton -1 and 1512 cm -1 , specifically Figure 2 As shown.

[0068] In step S30, a three-point bending test strip is cut from the CFRP every 1.5 cm. The bending strength of the CFRP is obtained through the three-point bending experiment as its target output data, as detailed below. Figure 3 As shown. Among them, Figure 3 In comparison with the mechanical properties before the lightning strike, the mechanical properties of CFRP after the lightning strike decreased significantly.

[0069] In step S41, the number of training datasets is 40.

[0070] In step S42, the maximum number of iterations for the network is set to 500, and the minimum training objective convergence error is set to 10. -5 .

[0071] Figure 4 This is a CFRP integrity evaluation diagram after a lightning strike, based on a test example of the present invention.

[0072] like Figure 4 As shown, the relative error of the residual intensity predicted by the artificial neural network in this test case is less than 5%.

[0073] Figure 5 This is a neural network design diagram of a test example of the present invention.

[0074] like Figure 5 As shown, in this test case, the number of layers and the number of hidden layer neurons in the optimized neural network are set to 3 and 12, respectively.

[0075] Those skilled in the art should understand that this invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to this invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. A method for CFRP integrity analysis after a lightning strike, characterized in that, Includes the following steps: S10, by conducting high-temperature exposure tests on CFRP at different temperatures and comparing the changes in its infrared spectrum, the evolution of chemical bonds in CFRP at different temperatures can be determined to identify the key chemical structures within it; S20. After obtaining the infrared spectral history data of the lightning-struck area of ​​CFRP by scanning the ATR-FTIR spectrometer, the characteristic peaks of the key chemical structure are selected and their intensities are calculated as target input data. S30 obtains historical data on the mechanical properties of CFRP after a lightning strike through mechanical performance testing, and uses it as the target output data. S40, an artificial neural network is used to establish a mapping relationship between the target input data and the target output data, thereby training a CFRP post-lightning integrity prediction model; S50. After scanning the lightning-struck area of ​​CFRP to be analyzed using an ATR-FTIR spectrometer to obtain its historical infrared spectral data, the characteristic peaks of the key chemical structures are selected and their intensities are calculated. These peaks are then input into the prediction model, and the mechanical property data of CFRP are output accordingly, thereby realizing the integrity analysis of CFRP after lightning strike.

2. The CFRP post-lightning strike integrity analysis method according to claim 1, characterized in that: in, In step S10, CFRP is placed in a muffle furnace and ablated at a constant temperature of 100 ℃ to 300 ℃ in an air atmosphere. The changes in the infrared spectra of CFRP at different temperatures are compared to obtain the evolution of chemical bonds in CFRP at different temperatures and determine the key chemical structures therein.

3. The CFRP post-lightning strike integrity analysis method according to claim 1, characterized in that: in, In steps S10 to S20, the key chemical structures include the CH stretching vibrations of methyl and methylene groups, the stretching vibrations of carbonyl groups, and the skeletal vibrations of the benzene ring.

4. The CFRP post-lightning strike integrity analysis method according to claim 1, characterized in that: in, In step S20, baseline adjustment and smoothing are also performed on the infrared spectral historical data.

5. The CFRP post-lightning strike integrity analysis method according to claim 1, characterized in that: in, In step S30, the mechanical property test includes a three-point bending test. The historical mechanical property data includes data on bending strength.

6. The CFRP post-lightning strike integrity analysis method according to claim 1, characterized in that: in, In step S40, constraints are also set for the prediction models obtained under different training conditions. These constraints include setting the maximum number of iterations and setting the minimum training objective convergence error. The artificial neural network is a BP neural network.

7. The CFRP post-lightning strike integrity analysis method according to claim 1, characterized in that: in, In step S40, the prediction models under different training conditions are also evaluated, and the accuracy of their prediction results is reflected by calculating the relative error. If the relative error does not meet the evaluation requirements, the prediction accuracy of the prediction model is improved by expanding the training dataset of the target input data and the target output data and optimizing the structure of the artificial neural network until the relative error meets the evaluation requirements.

8. The CFRP post-lightning strike integrity analysis method according to claim 7, characterized in that: in, Optimizing the structure of the artificial neural network includes adjusting the number of layers, adjusting the number of neurons in the hidden layers, and optimizing the activation function.

9. The CFRP post-lightning strike integrity analysis method according to any one of claims 1 to 8, characterized in that: in, In step S50, the CFRP post-lightning-strike integrity analysis further includes quantifying the concentration of the key chemical structures. c The relationship between the characteristic peak intensity and the key chemical structure. The intensity of the characteristic peak is obtained from the absorbance value in the historical infrared spectral data. A Calculation of the difference from the baseline Concentration of the key chemical structure c : ， In the above formula, θ Angle of incidence n s and n c The refractive indices of CFRP and ATR crystals are respectively. K The molar absorption coefficient, K It depends on the CFRP material and the wavelength of the incident light. λ .

10. A CFRP post-lightning strike integrity analysis system, characterized in that, The CFRP post-lightning strike integrity analysis method according to any one of claims 1 to 9 is used, including: The data input module is used for users to input the target input data and the target output data into it; The model building and training module, based on an artificial neural network and connected to the data input module, is used to establish a mapping relationship between the target input data and the target output data, thereby training a CFRP post-lightning integrity prediction model. The predictive analysis module, connected to the model building and training module, allows users to input the intensity of characteristic peaks of key chemical structures from historical infrared spectral data of the CFRP lightning-affected area to be analyzed. The predictive model, pre-trained by the model building and training module, then outputs the mechanical property data of the CFRP. The output module, connected to the predictive analysis module, is used to output and display the mechanical performance data and generate an integrity evaluation report.

Images