Raman spectroscopy-based coating damage evolution analysis method

Abstract

The application is suitable for the technical field of surface analysis and material characterization, and provides a coating damage evolution analysis method based on Raman spectrum, which comprises the following steps: step one, detecting characteristic substances before coating damage; step two, detecting characteristic substances after coating damage; step three, testing electrochemical corrosion performance before and after coating damage; step four, establishing a Raman spectrum database of characteristic substances before and after damage; step five, training a coating damage evolution identification model; step six, designing a coating damage evolution identification system; and step seven, verifying system analysis results.The coating damage evolution analysis method based on Raman spectrum provided by the application takes Raman spectrum technology as the core, combines technologies such as scanning electron microscope / energy spectrometer and electrochemical workstation, adopts models such as principal component analysis-support vector machine and random forest for training comparison, combines the best model with a spectrum pretreatment system, and forms an integrated coating damage evolution analysis system.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of surface analysis and material characterization, and particularly relates to a coating damage evolution analysis method based on Raman spectroscopy. BACKGROUND

[0002] With the wide application of functional coatings in the industries of machinery, aerospace, marine equipment, electronic devices, etc., different proportion composite coatings with characteristics of corrosion resistance, wear resistance, insulation or high temperature protection, etc. are often prepared on the surface of various substrate materials. The quality and reliability of the coating are directly related to the service life and safety of the product, and the corrosion resistance is often one of the important indicators for evaluating the comprehensive performance of the coating. Therefore, it is very important to deeply understand the damage mechanism and evolution process of the coating under different working conditions for material research and development and engineering application.

[0003] At present, the commonly used detection methods for the microstructure, composition of the coating and the phase and chemical changes thereof before and after damage include scanning electron microscopy, energy spectrum analysis, X-ray diffraction and infrared spectroscopy, etc. Although these methods have certain advantages in qualitative analysis and partial semi-quantitative research, they usually require complicated sample preparation steps and long testing time. In addition, some coatings will produce multiple corrosion products or phase change structures under complex environment, and it is difficult to quickly and accurately identify and track the change process using traditional means.

[0004] Raman spectroscopy is increasingly valued in the field of material research due to its fast testing speed, non-destructive nature and insensitivity to moisture. However, there are still some problems in the precise identification of the damage evolution of multi-layer or composite coatings using a single Raman detection method: firstly, the Raman characteristic peaks of different coatings and corrosion products may overlap or be submerged in complex background noise; secondly, there is a lack of systematic standard spectrum database, making it difficult to quickly find and compare the product phase of new coatings or composite coatings, and lacking verification methods for the test results; thirdly, the existing data processing and analysis models are limited, and it is difficult to balance the accuracy and real-time analysis demand of massive spectral data.

[0005] Therefore, there is an urgent need for a coating damage evolution analysis method based on Raman spectroscopy, which combines scanning electron microscopy, electrochemistry and Raman spectroscopy, establishes a unified damage product database and introduces advanced data preprocessing and model training methods, to improve the reliability of data and the efficiency of coating damage monitoring and identification, and finally forms a systematic damage identification and performance evaluation method. The present application aims to solve the above difficulties and deficiencies, and provides a comprehensive analysis process and method, which can provide important technical support for the damage identification and evaluation of coatings in actual service environment, and has a positive influence on the design and maintenance strategy of coatings. SUMMARY

[0006] In view of the deficiencies of the prior art, the purpose of the embodiments of the present application is to provide a coating damage evolution analysis method based on Raman spectrum to solve the problems in the foregoing background.

[0007] To achieve the above-mentioned purpose, the present application provides the following technical solutions.

[0008] The coating damage evolution analysis method based on Raman spectrum comprises the following steps.

[0009] Step one, detecting characteristic substances before coating damage: scanning electron microscope (SEM) and energy dispersive spectrometer (EDS) are used to test the micro-morphology and chemical composition of the coating, combined with Raman spectrum technology, the characteristic peak position of the coating characteristic substance is determined, and the characteristic substance before coating damage is determined according to literature, database and the like.

[0010] Step two, detecting characteristic substances after coating damage: a small knife is used to form a clear "crossed" scratch on the sample surface, a salt spray accelerated test is used to simulate an extreme environment to prepare a damaged coating sample, and step one is repeated combined with SEM / EDS and Raman spectrum technology to determine the characteristic substances after damage.

[0011] Step three, electrochemical corrosion performance test before and after coating damage: the electrochemical corrosion performance in the coating damage evolution process is tested to determine the difference in electrochemical corrosion performance before and after coating damage.

[0012] Step four, establishing a Raman spectrum database of characteristic substances before and after damage: standard substances of characteristic substances appearing in the damage evolution process are purchased, 100 groups of spectrum data of each standard substance are collected by using a Raman spectrometer, the spectrum characteristic information is optimized through data preprocessing and expansion, the amount of Raman spectrum data of each characteristic substance is increased, and a coating damage evolution Raman spectrum database is constructed.

[0013] Step five, training a coating damage evolution identification model: a principal component analysis (PCA)-support vector machine (SVM) model and a random forest model are used to train the Raman spectrum database of coating damage evolution, and the application efficacy of the models is compared to obtain a classification and identification model with the best performance.

[0014] Step six, design of a coating damage evolution identification system: the trained classification model with the best performance is combined with a Raman spectrum preprocessing system to construct an integrated coating damage evolution analysis system.

[0015] Step seven, verification of system analysis results: the coating is subjected to electrochemical corrosion performance test, the analysis results of the coating damage evolution identification system are verified, and a correlation between "damage characterization" and "performance evolution" is established.

[0016] As a further technical solution of the present application, in step one, the qualitative analysis of the coating damage product can be performed by combining the use of X-ray diffractometer, Fourier transform infrared spectroscopy and other structure or composition characterization methods.

[0017] As a further technical solution of the present application, in step two, the cleanliness of the sample surface is strictly controlled before and after scratching, and the method of fixed load or constant depth scratching is used to ensure the repeatability of the scratch depth / width, and the scratch morphology parameters are recorded, and the parameter conditions of the salt spray test are adjusted according to the application scene of the coating, and the coating damage can also be caused by other environmental simulation.

[0018] As a further technical solution of the present application, in step three, the electrochemical impedance spectroscopy and potentiodynamic polarization curve in the coating damage evolution process are tested by using an electrochemical workstation.

[0019] As a further technical solution of the present application, in step four, the Raman spectrum data expansion method is expanded by introducing a data enhancement algorithm, adding spectra with different noise levels, frequency shift perturbation and the like.

[0020] As a further technical solution of the present application, in step five, the PCA-SVM model and the random forest model are both optimized by using the grid search method, the leave-one-out cross-validation is used to reduce the fitting, and the confusion matrix, accuracy, sensitivity, specificity and the like are used to comprehensively compare the model performance.

[0021] As a further technical solution of the present application, in step five, the use steps of the random forest model are as follows:

[0022] Step one, standardizing the coating Raman spectrum data set to obtain Raman spectrum data in a unified format;

[0023] Step two, using the bootstrap method or other multiple subsample sampling methods to randomly extract multiple sample subsets from the training set spectrum data for training multiple decision trees;

[0024] Step three, voting or weighted fusion of the classification results of each decision tree to obtain the final classification result of the random forest model, and identifying the coating components.

[0025] As a further technical solution of the present application, in step six, the coating damage evolution analysis system includes coating Raman spectrum pretreatment and damage classification and identification.

[0026] As a further technical solution of the present application, in step seven, the use steps of the electrochemical corrosion performance test are as follows:

[0027] Step one, using an electrochemical workstation, respectively test the sample before and after the coating damage by electrochemical impedance spectroscopy and potentiodynamic polarization curve test;

[0028] Step two, record the corrosion potential, current density, polarization resistance and other parameters;

[0029] Step three, compare and analyze the above parameters to evaluate the difference of corrosion resistance before and after the coating damage, and cross verify with the Raman spectrum identification result.

[0030] Compared with the prior art, the beneficial effects of the present application are:

[0031] (1) The present application realizes the rapid discrimination and accurate evaluation of the coating damage process, compared with the traditional technical solution relying on a single detection method, which can effectively reduce the detection blind area and improve the detection accuracy;

[0032] (2) The present application incorporates the electrochemical corrosion performance test results into the coating damage evaluation system, and the damage characteristic substances identified by Raman spectrum are mutually verified, which can more comprehensively and accurately reveal the damage evolution and failure mechanism of the coating under actual working conditions.

[0033] (3) The present application provides perfect data preprocessing and intelligent analysis method, which overcomes the drawbacks of traditional Raman analysis, such as being easily disturbed by background and lacking systematic data processing, and significantly improves the identification accuracy and reliability;

[0034] (4) The present application constructs a Raman spectrum database suitable for various environmental and material systems by standardizing the test and data expansion of typical substances before and after the coating damage, realizes the rapid identification of new coating materials or more complex corrosion products, and has good scalability;

[0035] (5) The present application integrates the optimized spectrum preprocessing method with the trained best classification model, and builds an integrated coating damage evolution analysis system, which greatly improves the detection efficiency and application flexibility;

[0036] In order to more clearly illustrate the structural features and effects of the present application, the present application will be described in detail below in combination with the drawings and specific embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0037] Figure 1 The working flow chart of the coating damage evolution analysis method based on Raman spectrum provided by the embodiment of the present application;

[0038] Figure 2 The Raman spectrum, morphology and element distribution of the coating surface provided by the embodiment of the present application;

[0039] Figure 3The principal component analysis result and the confusion matrix diagram of the support vector machine model provided by the embodiment of the present application are shown in the following table:

[0040] Figure 4 The confusion matrix diagram of the random forest model provided by the embodiment of the present application is shown in the following table:

[0041] Figure 5 The Raman spectrum pretreatment system operation interface provided by the embodiment of the present application is shown in the following table:

[0042] Figure 6 The Raman spectrum classification and identification system operation interface provided by the embodiment of the present application is shown in the following table: DETAILED DESCRIPTION

[0043] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of the present application.

[0044] The specific implementation of the present application will be described in detail below in combination with specific embodiments.

[0045] Embodiment 1

[0046] As shown in the following table, the Raman spectrum-based coating damage evolution analysis method provided by an embodiment of the present application includes the following steps: Figure 1

[0047] Step one, detecting the characteristic substance before coating damage: test the microstructure and chemical composition of the coating by using a scanning electron microscope (SEM) and an energy-dispersive spectrometer (EDS), combine with Raman spectrum technology, determine the characteristic peak position of the characteristic substance of the coating, and determine the characteristic substance before coating damage according to literature, database, etc.

[0048] Step two, detecting the characteristic substance after coating damage: use a knife to form a clear "crossed" scratch on the sample surface, prepare the damaged coating sample by using a salt spray accelerated test to simulate an extreme environment, combine with SEM / EDS and Raman spectrum technology to repeat step one, and determine the characteristic substance after damage.

[0049] Step three, electrochemical corrosion performance test before and after coating damage: test the electrochemical corrosion performance in the coating damage evolution process, and determine the difference in electrochemical corrosion performance before and after coating damage.

[0050] ​Step four, establish the Raman spectrum database of characteristic substances before and after damage: purchase standard substances of characteristic substances appearing in the damage evolution process, collect 100 groups of spectrum data for each standard substance by using Raman spectrometer, optimize the spectrum characteristic information through data preprocessing and expansion, increase the Raman spectrum data of each characteristic substance, and construct the Raman spectrum database of coating damage evolution.

[0051] Step five, train the coating damage evolution identification model: train the Raman spectrum database of coating damage evolution by using principal component analysis (PCA)-support vector machine (SVM) model and random forest model, compare the application efficacy of the models, and obtain the classification and identification model with the best performance.

[0052] Step six, design the coating damage evolution identification system: combine the classification model with the best performance after training and the Raman spectrum preprocessing system to construct an integrated coating damage evolution analysis system.

[0053] Step seven, verify the analysis results of the system: perform the electrochemical corrosion performance test on the coating, verify the analysis results of the coating damage evolution identification system, and establish the correlation between "damage characterization" and "performance evolution".

[0054] Figure 2 The results of detecting the coating by using SEM / EDS and Raman spectrum are shown:

[0055] The microstructure of the corrosion-resistant coating presents a uniform distribution of granular structure and small pores. This structure may be formed by the combination of corrosion-resistant oxide particles (such as chromium oxide or aluminum oxide) and the organic coating matrix. The dense distribution between the particles can effectively block the penetration of corrosive media, enhancing the barrier effect of the coating. In addition, the uniformity of this microstructure and the small pores help the coating to exhibit excellent corrosion resistance in marine or acidic environments;

[0056] The EDS analysis of the coating shows that its elemental composition is complex and diverse, including strontium (31.42wt%), titanium (16.35wt%), chromium (15.83wt%), silicon (12.50wt%), magnesium (4.82wt%), and nickel (4.34wt%). Strontium compounds may exist in the form of strontium chromate, providing cathodic protection function and effectively preventing the penetration of corrosive media. Titanium and chromium further enhance the oxidation resistance of the coating, while the presence of silicon and magnesium improves the weather resistance and mechanical stability of the matrix. The introduction of nickel strengthens the durability of the coating in marine environments;

[0057] From Figure 2 It can be seen that the Raman spectrum of the coating has peaks at 867cm -1 and 893cm -1The obvious peaks appeared, combined with literature data and Raman spectrum database information, these peaks correspond to the characteristic peaks of strontium chromate (SrCr04). SrCr04 is a common anticorrosive material, and its Raman active characteristics are consistent with the detection results, indicating that the coating has excellent corrosion resistance.

[0058] In step one, the qualitative analysis of coating damage products can also be combined with other structure or composition characterization methods such as X-ray diffractometer and Fourier transform infrared spectroscopy.

[0059] In step two, the cleanliness of the sample surface before and after scratching is strictly controlled, and the method of fixed load or constant depth scratching is used to ensure the repeatability of the scratch depth / width, and the scratch morphology parameters are recorded, and the parameter conditions of the salt spray test are adjusted according to the application scene of the coating, and the coating can also be damaged by other environmental simulation.

[0060] In step three, the electrochemical impedance spectroscopy and potentiodynamic polarization curve in the evolution process of coating damage are tested by using an electrochemical workstation.

[0061] In step four, the Raman spectrum data expansion method is carried out by introducing data enhancement algorithm, adding spectra with different noise levels, frequency shift perturbation, etc.

[0062] Example 2

[0063] As a preferred embodiment of the present application, in step five, the PCA-SVM model and the random forest model are both optimized by grid search, and the leave-one-out cross-validation is used to reduce fitting, and the confusion matrix, accuracy, sensitivity, specificity, robustness and other indicators are used to compare the model performance.

[0064] The PCA-SVM model is trained using a radial basis function kernel, and the best parameter combination is set to C=10 and γ=0.1 by grid search, which realizes accurate fitting of complex decision boundary and can effectively distinguish multiple data categories. The proportion of test set and training set is 8:2;

[0065] Figure 3The results of using PCA to reduce the dimension of the Raman spectrum data set and the confusion matrix after importing the reduced data into the SVM model for training are shown. The principal component analysis result shows that the Raman spectrum information accounts for 84.9%, of which the explained variance of principal component 1 is 66.33%, and the explained variance of principal component 2 is 18.57%. As can be seen from the figure, different compounds show a relatively dispersed distribution on the principal component 1 and principal component 2 plane, indicating that the Raman spectrum characteristics of each Cr2O3, Fe, Fe2O3, Mg2SiO4, SiO2, SrCrO4, TiO2(rutile type) and TiO2(anatase type) are significantly different, which is sufficient to show clear clustering partition after dimension reduction. The values of Cr2O3 corresponding to the row and column on the diagonal line are much larger than those on the non-diagonal line, indicating that the recognition effect of SVM on Cr2O3 samples is good; for other categories, such as Fe2O3, Mg2SiO4, etc., also have a high accuracy. Fe and Fe2O3 may have some characteristic peaks close in the spectrum, resulting in a few misjudgments of samples, but the overall accuracy is still considerable; the test set result shows that the PCA-SVM model reaches 100% accuracy, recall rate and F1 score in the classification task of 8 kinds of substances.

[0066] Since the random forest model and the PCA-SVM model are the same in accuracy, sensitivity and specificity, the robustness of the two models is compared to select a more suitable model. The coating Raman spectrum data is expanded to 100 groups of Raman spectrum data by adding noise, and is respectively imported into the above two models. The results show that the accuracy of the random forest model is 97%, while the accuracy of the PCA-SVM model is 99%, so the PCA-SVM model is selected for subsequent coating damage evolution classification and recognition system design.

[0067] Embodiment 3

[0068] As a preferred embodiment of the application, in step five, the use steps of the random forest model are as follows:

[0069] Step one, standardizing the coating Raman spectrum data set to obtain Raman spectrum data in a unified format;

[0070] Step two, using the bootstrap method or other multiple subsample sampling methods to randomly extract multiple sample subsets from the training set spectrum data for training multiple decision trees;

[0071] Step three, voting or weighted fusion of the classification results of each decision tree to obtain the final classification result of the random forest model, and identifying the coating components.

[0072] Figure 4The confusion matrix of the random forest model is shown. From the figure, it can be seen that all the classified samples are correctly predicted, the diagonal elements are the maximum, and the non-diagonal elements are 0, indicating that the RF model does not have any confusion between classes. Compared with other models, the RF model can effectively deal with the nonlinear characteristics and potential noise of Raman spectrum data due to its characteristics of ensemble learning and diversity based on decision tree. The random forest model shows high accuracy and stability in classifying Raman spectrum data, and the classification results of the test set also reach 100% precision, recall rate and F1 score. This result shows the excellent performance of the RF model in multi-class classification tasks, which can effectively capture the characteristics of different substances in Raman spectrum.

[0073] For the eight standard substance categories (Cr2O3, Fe, Fe3O4, MgSiO4, SiO2, SrCrO4, TiO2(rutile type), TiO2(anatase type)), the classification index is 100%. This shows that there is no false positive in the classification prediction of all substance categories, and the model successfully identifies all test samples without missing classification, achieving perfect balance between precision and recall rate in the classification task, and demonstrating the reliability of the model.

[0074] Example 4

[0075] As a preferred embodiment of the present application, in step six, the coating damage evolution analysis system includes coating Raman spectrum pretreatment and damage classification identification. Before performing coating damage classification identification, the Raman spectrum data is pretreated, and then the pretreated Raman spectrum data is imported into the damage classification identification system. The system automatically performs PCA dimensionality reduction and standardization processing on the Raman spectrum data, and then uses the PCA-SVM model with the best comprehensive performance for classification identification.

[0076] Figure 5 The Raman spectrum pretreatment system operation interface provided for the embodiments of the present application. The system includes spectrum range selection, cosmic ray elimination, baseline calibration and smoothing processing, and other Raman spectrum pretreatment functions. The user inputs the minimum and maximum values of the Raman shift according to personal needs, clicks the "update (range)" button after inputting, and can view the Raman spectrum data graph within the selected range in the right view area; select the appropriate "threshold value" according to the maximum value of the cosmic ray presented in the original Raman spectrum data, which can eliminate the interference of cosmic rays; input the specific value that the user wants to set in the "polynomial order" part, and similarly select the specific value in the lower "regular coefficient" part, and click the right "update (baseline)" to realize the baseline calibration of the Raman spectrum; the user can input the specific value set reasonably in the right "window length" part of the operation interface, and similarly select the specific value in the lower "order" part, and click the right "update (S-G)" to realize the smoothing processing of the Raman spectrum.

[0077] Figure 6 The Raman spectrum classification and identification system operation interface provided by the embodiment of the present application. The user only needs to click the "select file" button in the middle of the software interface, select the Raman spectrum file (xlsx) to be classified, and the classification can be realized. The classification situation is demonstrated in the figure, and the "classification result is as follows: iron oxide (Fe2O3) - after damage Absorbing coating composition; prompt: the result is for reference only, please further analyze combined with experimental data".

[0078] Embodiment 5

[0079] As a preferred embodiment of the present application, in step seven, the use steps of the electrochemical corrosion performance test are as follows:

[0080] Step one, using an electrochemical workstation, respectively test the samples before and after the coating damage by electrochemical impedance spectroscopy and potentiodynamic polarization curve test;

[0081] Step two, record the corrosion potential, current density, polarization resistance and other parameters;

[0082] Step three, compare and analyze the above parameters to evaluate the difference of corrosion resistance before and after the coating damage, and cross verify with the Raman spectrum identification result.

[0083] The Raman spectrum classification and identification result of the coating damage is compared with the electrochemical corrosion performance result of the coating damage. If the Raman spectrum classification and identification result of the coating damage shows that new substances generated after the coating damage appear in the coating, and at the same time the electrochemical corrosion performance of the coating damage decreases, it indicates that the coating has been damaged and needs to be repaired in time.

[0084] The above only describes the preferred embodiments of the present application and does not limit the present application. Any modification, equivalent replacement and improvement made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method of coating damage evolution analysis based on Raman spectroscopy, characterized in that, Comprising the following steps: Step one, detecting characteristic substances before coating damage: Scanning electron microscopy and energy dispersive spectrometer are used to test the micro-morphology and chemical composition of the coating, combined with Raman spectroscopy technology, to determine the characteristic peak position of the characteristic substances of the coating, and according to the literature and database, to determine the characteristic substances before coating damage; Step two, detecting characteristic substances after coating damage: A small knife is used to form a clear "cross" scratch on the surface of the sample, and a salt spray accelerated test is used to simulate extreme environment to prepare the damaged coating sample, combined with scanning electron microscopy, energy dispersive spectrometer and Raman spectroscopy technology, to repeat step one and determine the characteristic substances after damage; Step three, electrochemical corrosion performance test before and after coating damage: The electrochemical corrosion performance in the coating damage evolution process is tested to determine the difference in electrochemical corrosion performance before and after coating damage; Step four, establishing Raman spectrum database of characteristic substances before and after damage: The standard substances of characteristic substances appearing in the damage evolution process are purchased, and 100 groups of spectral data of each standard substance are collected by using Raman spectrometer, the spectral characteristic information is optimized through data preprocessing and expansion, the amount of Raman spectrum data of each characteristic substance is increased, and the Raman spectrum database of coating damage evolution is constructed; Step five, training coating damage evolution recognition model: The principal component analysis-support vector machine model and the random forest model are used to train the Raman spectrum database of coating damage evolution, and the application effectiveness of the models is compared to obtain the best classification and recognition model; Step six, design of coating damage evolution recognition system: The best classification model is combined with the Raman spectrum preprocessing system to construct an integrated coating damage evolution analysis system; Step seven, verification of system analysis results: The coating is tested for electrochemical corrosion performance to verify the analysis results of the coating damage evolution recognition system, and the correlation between "damage characterization" and "performance evolution" is established.

2. The Raman spectroscopy-based coating damage evolution analysis method of claim 1, wherein, In step two, the qualitative analysis of coating damage products can be performed by combining X-ray diffractometer, Fourier transform infrared spectroscopy and other structure or composition characterization methods.

3. The Raman spectroscopy-based coating damage evolution analysis method of claim 1, wherein, In step two, the sample surface cleanliness is strictly controlled before and after scratching, and a fixed load or constant depth scratching method is used to ensure the repeatability of the scratch depth / width, and the scratch morphology parameters are recorded. The parameters of the salt spray test are adjusted according to the application scenario of the coating, and the coating damage can also be simulated by other environmental simulation methods.

4. The Raman spectroscopy-based coating damage evolution analysis method of claim 1, wherein, In step three, the electrochemical impedance spectroscopy and potentiodynamic polarization curve in the coating damage evolution process are tested by using an electrochemical workstation.

5. The Raman spectroscopy-based coating damage evolution analysis method of claim 1, wherein, In step four, the Raman spectrum data expansion method is performed by introducing data enhancement algorithm to add different noise level spectra and frequency shift perturbation.

6. The Raman spectroscopy-based coating damage evolution analysis method of claim 1, wherein, In step five, the principal component analysis-support vector machine model and the random forest model are optimized by using grid search method, cross-validation is used to reduce fitting, and the model performance is compared by using confusion matrix, accuracy, sensitivity and specificity indicators.

7. The Raman spectroscopy-based coating damage evolution analysis method of claim 1, wherein, In step five, the steps of using the random forest model are as follows: Step one, standardize the coating Raman spectrum data set to obtain uniform format Raman spectrum data; Step two, randomly extract multiple sample subsets from the training set of spectral data using the bootstrap method or other multiple subsample sampling methods, respectively for training multiple decision trees; Step three, vote or weighted fusion of the classification results of each decision tree to obtain the final classification result of the random forest model, and identify the coating components.

8. The Raman spectroscopy-based coating damage evolution analysis method of claim 1, wherein, In step six, the coating damage evolution analysis system includes coating Raman spectrum pretreatment and damage classification and identification.

9. The Raman spectroscopy-based coating damage evolution analysis method of claim 1, wherein, In step seven, the use steps of the electrochemical corrosion performance test are as follows: Step one, use an electrochemical workstation to test the electrochemical impedance spectroscopy and potentiodynamic polarization curve of the sample before and after the coating damage respectively; Step two, record the corrosion potential, current density and polarization resistance parameters; Step three, compare and analyze the above parameters to evaluate the difference in corrosion resistance before and after the coating damage, and cross verify with the Raman spectrum identification result.

Images