Metal powder sintering composite layer hard alloy friction vibration behavior time-frequency characterization method

Abstract

The invention relates to a friction vibration behavior time-frequency characterization method, in particular to a metal powder sintering composite layer hard alloy friction vibration behavior time-frequency characterization method. The invention aims to solve the problems that when an existing time-frequency method is used for analysis, the extraction of image feature textures is not deep enough, the extracted features are not wide enough and not highly persuasive, time-domain and frequency-domain analysis needs to be combined, and an improved time-frequency analysis method often needs data adaptation and lacks universality. The method comprises the following steps: preprocessing a collected friction vibration data file; carrying out noise reduction processing on the collected vibration signals; acquiring a vibration signal time-frequency analysis image; performing graying processing on the time-frequency image; performing feature extraction on the vibration signal time-frequency image; carrying out dimension reduction screening on the vibration behavior characteristics based on an MIC maximum mutual information coefficient-BP neural network method; and fitting a stability characterization curve of the test system based on the vibration behavior. The invention belongs to the technical field of friction vibration behavior characterization.

Description

Technical Field

[0001] The invention relates to a friction vibration behavior time-frequency characterization method, belonging to the technical field of friction vibration behavior characterization. Background Art

[0002] With the continuous application of difficult-to-process materials such as titanium alloys in the engineering field, the wear resistance and service life of cemented carbide tools used for cutting them have been greatly challenged. The means of strengthening the surface of cemented carbide are also constantly improving. Among them, sintered metal powder composite layers are widely used in the field of alloy surface strengthening. The strengthening effects of composite layers made of different metal powders and different sintering parameters are very different. Each adjustment of parameters, elements and ratios requires friction and wear testing of surface wear resistance and stability. A large amount of test data will be generated during the testing process, among which vibration behavior data and friction data are two important performance test evaluation indicators. Generally speaking, vibration behavior data is cheaper to obtain than friction data.

[0003] At present, the vibration behavior analysis of friction and wear testing mostly stays in the analysis of time domain, frequency domain and single time and frequency domain. Time domain and frequency domain analysis need to process the original data, the calculation space required is large and the anti-interference ability is poor; the existing time-frequency method does not extract the image feature texture deeply enough during analysis, and the extracted features are not extensive and convincing enough, so it is necessary to combine time domain and frequency domain analysis; and the improved time-frequency analysis method often requires data adaptation and lacks versatility.

[0004] Therefore, improving the persuasiveness of time-frequency analysis feature extraction, improving the versatility of analysis methods, reducing the amount of calculation, and shrinking the feature storage space are of great importance for analyzing multi-batch, multi-feature vibration behavior data generated by the strengthening performance detection of sintered metal powder composite layers, and improving the stability of the detection system and data prediction and characterization capabilities. Summary of the invention

[0005] In order to solve the problems that the existing time-frequency analysis methods have insufficient depth in extracting image feature textures, and the extracted features are not extensive enough and not convincing enough, it is necessary to combine time domain and frequency domain analysis. However, the improved time-frequency analysis methods often require data adaptation and lack of versatility. Therefore, a time-frequency characterization method for the friction vibration behavior of a metal powder sintered composite layer cemented carbide is proposed.

[0006] The technical solution adopted by the present invention to solve the above-mentioned problem is: the steps of the present invention include:

[0007] Step 1: preprocessing the collected friction vibration data file;

[0008] Step 2: performing noise reduction processing on the collected vibration signal;

[0009] Step 3, obtaining a time-frequency analysis image of the vibration signal;

[0010] Step 4: grayscale the time-frequency image;

[0011] Step 5, extracting features from the vibration signal time-frequency image;

[0012] Step 6: Perform dimension reduction screening on vibration behavior characteristics based on MIC maximum mutual information coefficient-BP neural network method;

[0013] Step 7: Verify the prediction accuracy of feature selection regression based on random forest RF;

[0014] Step 8: Fit the stability characterization curve of the test system based on the vibration behavior.

[0015] Furthermore, the preprocessing of the collected friction vibration data file in step 1 refers to converting other types of collected files into an Excel spreadsheet's '.xlsx' format file using LabVIEW.

[0016] Furthermore, in step 2, VMD variational mode decomposition is used to remove the obvious interference part in the vibration and perform noise reduction.

[0017] Furthermore, in step 3, short-time Fourier transform and continuous wavelet transform are used to perform time-frequency analysis on the friction signal to obtain a time-frequency image.

[0018] Furthermore, the grayscale processing of the time-frequency image in step 4 specifically includes:

[0019] Decolorize the image, simplify the complexity of the image, and shorten the calculation time;

[0020] The grayscale algorithm is used to process the image according to 256 grayscale levels. The characteristic information such as image color and shape changes will be saved in the grayscale image in the form of grayscale pixels, and feature conversion is performed using the pixel distribution and pixel number of each grayscale level as characteristic information.

[0021] Furthermore, the feature extraction of the vibration signal time-frequency image in step 5 specifically includes:

[0022] Compare and adapt the feature value extraction methods based on the image features obtained by time-frequency analysis;

[0023] For STFT image features, select 7 Hu invariant moment features A total of 8 features including grayscale mean are extracted as image feature values;

[0024] For the CWT image features, 8 features including the mean and variance of the energy, contrast, uniformity, and entropy of the CWT image grayscale co-occurrence matrix in the four directions of 0°, 45°, 90°, and 135° were selected as image feature values ​​for extraction; finally, a total of 16 time-frequency features of friction vibration were obtained.

[0025] Furthermore, in step 6, the step of reducing the dimension of the vibration behavior characteristics based on the MIC maximum mutual information coefficient-BP neural network method specifically includes:

[0026] Step 601, obtaining 16 eigenvalues ​​of friction and wear vibration of the metal powder sintered composite layer reinforced cemented carbide through step 5;

[0027] Step 602: Taking the friction force, a parameter representing the stability of the friction and wear test system, as the output of feature screening, and taking the 16 eigenvalues ​​of STFT and CWT as input, feature selection based on the maximum mutual information coefficient of MIC is performed, which is specifically implemented as follows:

[0028] By discretizing the input feature X and the output feature Y in two-dimensional space, the data is divided into grids according to the distribution of the feature values, and the mutual information between the feature and the target variable is calculated in each grid. Finally, all grids are traversed to find the grid with the maximum mutual information, so as to obtain the MIC value of each feature as the contribution value of the feature;

[0029] Based on the experience of feature selection, when the difference between the maximum feature contribution and the minimum feature contribution is large, half of the difference between the maximum MIC value and the minimum MIC value between the features is used as the threshold for preliminary screening of feature importance, that is, the features with importance greater than 0.3 are retained in the preliminary screening;

[0030] Step 602: Use the BP neural network friction prediction RMSE value to perform the next step of feature screening. Each time a feature is cut, 30 predicted RMSE values ​​are obtained to calculate the average.

[0031] Furthermore, the specific steps of fitting the stability characterization curve of the test system based on the vibration behavior in step 8 include:

[0032] Step 801: take the grayscale mean, entropy mean, contrast mean, The entropy variance and energy mean characteristics are numerically based on the vector normalization formula:

[0033]

[0034] In formula (1), c represents the coefficient, which is used to increase the data level and is set to 100; n represents the number of samples; a i Represents the actual measurement data; y iRepresents the normalized data; the data is vectorized and normalized by using the norm(x) function in MATLAB to eliminate the dimension effect;

[0035] Step 802: Based on the normalized data, feature weights are divided. The specific implementation is: based on the importance of the MIC detection contribution value, the total contribution of the screening features is set to C, and the contribution value of each feature is x. i , i represents different features, and the contribution value of each feature accounts for the percentage of the total contribution value of the screening features, that is: where y i That is, the weight assigned to each retained feature;

[0036] Step 803: multiply each normalized feature by its own weight and then sum them up to obtain six-feature fusion parameters, and fit the vibration behavior time-frequency characteristic law curve based on the fusion parameters;

[0037] Step 804: by gray level mean, entropy mean, contrast mean, The vector normalization processing, weight division and weighted fitting of entropy variance and energy mean are performed to obtain the time-frequency characteristic law curve of vibration behavior, thereby characterizing the stability change curve of the sintered metal powder composite layer alloy strengthening performance detection system based on vibration behavior.

[0038] The beneficial effects of the present invention are:

[0039] 1. The analysis method proposed in the present invention is highly suitable for the vibration behavior analysis of the surface properties of sintered metal powder composite layer cemented carbide and the stability detection of the test system, which can significantly improve the accuracy of the analysis and the accuracy of the prediction of related parameters;

[0040] 2. Compared with traditional vibration signal time domain and frequency domain analysis, the method of the present invention can effectively reduce the amount of data from the root, convert complex original data into images for storage, save storage and computing space, and extract a large number of image features to dilute the interference of impurity signals, thereby improving the anti-interference ability of the analysis method;

[0041] 3. Compared with the traditional single time-frequency analysis method and the improved time-frequency analysis method, the method of the present invention is more comprehensive and complete in extracting time-frequency image features, has stronger versatility, and has a good balance between the accuracy and versatility of the vibration behavior time-frequency processing method;

[0042] 4. The machine learning algorithm used in the present invention performs feature fusion at the feature layer level to achieve feature selection and data prediction with multiple inputs and a single output, and has good robustness during operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1It is a structural schematic diagram of a sintered metal powder composite layer cemented carbide surface performance detection system;

[0044] Figure 2 This is a schematic diagram of the signal data file format conversion program based on LabVIEW;

[0045] Figure 3 This is a schematic diagram of vibration signal noise reduction processing based on VMD;

[0046] Figure 4 This is a schematic diagram of obtaining the time-frequency image of vibration signals based on STFT and CWT;

[0047] Figure 5 This is a schematic diagram of grayscale conversion of STFT and CWT time-frequency images based on grayscale algorithm;

[0048] Figure 6 It is a schematic diagram of the time-frequency image feature value extraction program;

[0049] Figure 7 It is a schematic diagram of the MIC maximum mutual information coefficient feature selection procedure and importance image;

[0050] Figure 8 This is a schematic diagram of the random forest RF prediction verification results;

[0051] Fig. 9 It is a schematic diagram of the stability characterization curve of the test system;

[0052] Fig.10 This is a file format conversion diagram;

[0053] Fig.11 is a schematic diagram of a partial denoised image;

[0054] Fig.12 It is a schematic diagram of STFT and CWT time-frequency processing of two-dimensional, three-dimensional and grayscale images;

[0055] Fig.13 It is a flow chart of the present invention. DETAILED DESCRIPTION

[0056] Specific implementation method 1: Figures 1 to 13 As shown, a time-frequency characterization method for friction vibration behavior of a metal powder sintered composite layer cemented carbide comprises the following specific steps:

[0057] Step 1: preprocessing the collected friction vibration data file;

[0058] Step 2: performing noise reduction processing on the collected vibration signal;

[0059] Step 3, obtaining a time-frequency analysis image of the vibration signal;

[0060] Step 4: grayscale the time-frequency image;

[0061] Step 5, extracting features from the vibration signal time-frequency image;

[0062] Step 6: Perform dimension reduction screening on vibration behavior characteristics based on MIC maximum mutual information coefficient-BP neural network method;

[0063] Step 7: Verify the prediction accuracy of feature selection regression based on random forest RF;

[0064] Step 8: Fit the stability characterization curve of the test system based on the vibration behavior.

[0065] Specific implementation method 2: Figures 1 to 13 As shown, the preprocessing of the collected friction vibration data file in step 1 refers to using LabVIEW to convert other types of collected files into an Excel spreadsheet's '.xlsx' format file.

[0066] Specific implementation method three: Figures 1 to 13 As shown, in step 2, VMD variational mode decomposition is used to remove the obvious interference part in the vibration and perform noise reduction.

[0067] Among them, if the vibration information of the entire detection system is fully considered, noise reduction may not be performed when the interference from external impurity signals is small.

[0068] The image was denoised using the [imf, residual] = vmd(x, 'NumIMF', a) program in MATLAB. A total of 9 layers of decomposition were performed to reconstruct the IMF2-IMF8 components.

[0069] Specific implementation method four: Figures 1 to 13 As shown, in step 3, short-time Fourier transform and continuous wavelet transform are used to perform time-frequency analysis on the friction signal to obtain a time-frequency image.

[0070] Among them, STFT uses Hanning window to set the window length to 127.

[0071] CWT needs to adjust the wavelet function according to the range of the cone curve in the image, so that the fuzzy features outside the cone curve do not affect the overall time-frequency characteristics. According to the signal characteristics, wavelet functions such as 'haar', 'db', 'sym', 'cmor', 'mexh', 'gaus', 'bior', 'morse', 'amor', and 'bump' can be called to reduce the influence of the fuzzy features in the cone curve. The vibration signal obtained in the friction and wear test uses the 'amor' wavelet function to perform CWT time-frequency analysis to meet the needs of feature extraction.

[0072] The MATLAB short-time Fourier transform toolbox and [wt,f] = cwt (x, 'wavelet function', fs) are used to perform STFT and CWT time-frequency analysis of vibration signals. The obtained time-frequency images and influence cone curves are verified as follows: Figure 4 shown.

[0073] Specific implementation method five: Figures 1 to 13 As shown, the grayscale processing of the time-frequency image in step 4 specifically includes:

[0074] Decolorize the image, simplify the complexity of the image, and shorten the calculation time;

[0075] The grayscale algorithm is used to process the image according to 256 grayscale levels. The characteristic information such as image color and shape changes will be saved in the grayscale image in the form of grayscale pixels, and feature conversion is performed using the pixel distribution and pixel number of each grayscale level as characteristic information.

[0076] Specific implementation method six: Figures 1 to 13 As shown, the feature extraction of the vibration signal time-frequency image in step 5 specifically includes:

[0077] Compare and adapt the feature value extraction methods based on the image features obtained by time-frequency analysis;

[0078] For STFT image features, select 7 Hu invariant moment features A total of 8 features including grayscale mean are extracted as image feature values;

[0079] For the CWT image features, 8 features including the mean and variance of the energy, contrast, uniformity, and entropy of the CWT image grayscale co-occurrence matrix in the four directions of 0°, 45°, 90°, and 135° were selected as image feature values ​​for extraction; finally, a total of 16 time-frequency features of friction vibration were obtained.

[0080] Among them, Figure 7 The ranking in is gradually removed to observe the change of RSME value until the value is minimum. The specific implementation is: first, the BP neural network is used to obtain the predicted RMSE value of the system friction after retaining all input features and pre-screening the input features, and the result of the feature pre-screening is verified; then, based on the 10 features pre-screened, one feature with the smallest contribution is reduced each time according to the feature MIC contribution value ranking, and a prediction calculation of the output feature friction is performed to detect the change of RMSE value until the RMSE value of the system friction predicted by the retained features is minimized. This realizes the advanced screening of input features and obtains the final features.

[0081] The prediction calculation is shown in Table 1. The final retained features are: grayscale mean, entropy mean, contrast mean, Entropy variance, energy mean.

[0082] Table 1 RMSE value of friction force predicted by BP neural network

[0083]

[0084] Specific implementation method seven: Figures 1 to 13 As shown, the steps of reducing the dimension of the vibration behavior characteristics based on the MIC maximum mutual information coefficient-BP neural network method in step 6 specifically include:

[0085] Step 601, obtaining 16 eigenvalues ​​of friction and wear vibration of the metal powder sintered composite layer reinforced cemented carbide through step 5;

[0086] Step 602: Taking the friction force, a parameter representing the stability of the friction and wear test system, as the output of feature screening, and taking the 16 eigenvalues ​​of STFT and CWT as input, feature selection based on the maximum mutual information coefficient of MIC is performed, which is specifically implemented as follows:

[0087] By discretizing the input feature X and the output feature Y in two-dimensional space, the data is divided into grids according to the distribution of the feature values, and the mutual information between the feature and the target variable is calculated in each grid. Finally, all grids are traversed to find the grid with the maximum mutual information, so as to obtain the MIC value of each feature as the contribution value of the feature;

[0088] Based on the experience of feature selection, when the difference between the maximum feature contribution and the minimum feature contribution is large, half of the difference between the maximum MIC value and the minimum MIC value between the features is used as the threshold for preliminary screening of feature importance, that is, the features with importance greater than 0.3 are retained in the preliminary screening;

[0089] Step 602: Use the BP neural network friction prediction RMSE value to perform the next step of feature screening. Each time a feature is cut, 30 predicted RMSE values ​​are obtained to calculate the average.

[0090] Specific implementation method eight: Figures 1 to 13 As shown, the specific steps of fitting the stability characterization curve of the test system based on the vibration behavior in step 8 include:

[0091] Step 801: take the grayscale mean, entropy mean, contrast mean, The entropy variance and energy mean characteristics are numerically based on the vector normalization formula:

[0092]

[0093] In formula (1), c represents the coefficient, which is used to increase the data level and is set to 100; n represents the number of samples; a i Represents the actual measurement data; y iRepresents the normalized data; the data is vectorized and normalized by using the norm(x) function in MATLAB to eliminate the dimension effect;

[0094] Step 802: Based on the normalized data, feature weights are divided. The specific implementation is: based on the importance of the MIC detection contribution value, the total contribution of the screening features is set to C, and the contribution value of each feature is x. i , i represents different features, and the contribution value of each feature accounts for the percentage of the total contribution value of the screening features, that is: where y i That is, the weight assigned to each retained feature;

[0095] Step 803: multiply each normalized feature by its own weight and then sum them up to obtain six-feature fusion parameters, and fit the vibration behavior time-frequency characteristic law curve based on the fusion parameters;

[0096] Step 804: by gray level mean, entropy mean, contrast mean, The vector normalization processing, weight division and weighted fitting of entropy variance and energy mean are performed to obtain the time-frequency characteristic law curve of vibration behavior, thereby characterizing the stability change curve of the sintered metal powder composite layer alloy strengthening performance detection system based on vibration behavior.

[0097] Specific implementation method nine: Figures 1 to 13 As shown in Figure 1, the feature selection regression prediction accuracy verification based on random forest RF in step 7 specifically includes:

[0098] The 4th, 5th, and 6th features selected step by step in Table 1 in step 6 are used as input respectively.

[0099] Taking friction as the output, feature selection and prediction accuracy verification are performed. In order to adapt to the data, prevent prediction overfitting, and improve generalization ability, the number of decision trees is selected as 150 and the number of leaves is 1.

[0100] The random forest prediction verification results are as follows Figure 8 As shown, the selected grayscale mean, entropy mean, contrast mean, The entropy variance and energy mean are the best features. In this way, the vibration behavior characteristics with the best description and the greatest prediction contribution to the stability of the sintered metal powder composite layer strengthening performance detection system are obtained.

[0101] Example

[0102] A time-frequency characterization method for friction vibration behavior of metal powder sintered composite layer cemented carbide, specifically comprising:

[0103] Step 1: Use LabVIEW to convert data files in other formats into '.xlsx' format files;

[0104] Step 2: Perform VMD noise reduction processing on the original signal data according to the signal characteristics and component preservation needs;

[0105] Step 3: Perform STFT and CWT time-frequency analysis on the vibration signal data after noise reduction to obtain a time-frequency image, and perform grayscale processing on the time-frequency image using a grayscale algorithm;

[0106] Step 4: Extract features based on the features of the time-frequency image. Select the Hu invariant moment for the STFT image. Grayscale mean; CWT image selects GLCM grayscale co-occurrence matrix: mean and variance of energy, contrast, uniformity, and entropy in four directions of 0°, 45°, 90°, and 135°. A total of 16 features are extracted;

[0107] Step 5: Taking the system stability friction as input and 16 features as output, the MIC maximum mutual information coefficient method is used to preliminarily select features, and exclude features with output contribution levels lower than 0.3;

[0108] Step 6: Use BP neural network with MIC maximum mutual information coefficient method to gradually screen the features;

[0109] Step 7: Use the random forest RF regression prediction algorithm to verify the selected feature information and predict the system stability friction again;

[0110] Step 8. Perform curve fitting on the obtained screening features; use vector normalization method to normalize the features; divide the weights of the features to be fitted based on the feature contribution values ​​obtained by MIC maximum mutual information coefficient method, and finally fit the stability change curve of the detection system based on vibration behavior according to the weights.

[0111] The above is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technician familiar with this profession can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent replacement and improvement made to the above embodiments without departing from the content of the technical solution of the present invention, based on the technical essence of the present invention, within the spirit and principles of the present invention, still fall within the protection scope of the technical solution of the present invention.

Claims

1. A time-frequency characterization method for friction vibration behavior of metal powder sintered composite layer cemented carbide, characterized in that: The specific steps include: Step 1: preprocessing the collected friction vibration data file; Step 2: performing noise reduction processing on the collected vibration signal; Step 3, obtaining a time-frequency analysis image of the vibration signal; Step 4: grayscale the time-frequency image; Step 5, extracting features from the vibration signal time-frequency image; Step 6: Perform dimension reduction screening on vibration behavior characteristics based on MIC maximum mutual information coefficient-BP neural network method; Step 7: Verify the prediction accuracy of feature selection regression based on random forest RF; Step 8: Fit the stability characterization curve of the test system based on the vibration behavior.

2. The time-frequency characterization method for friction vibration behavior of metal powder sintered composite layer cemented carbide according to claim 1 is characterized in that: The preprocessing of the collected friction vibration data file in step 1 refers to using LabVIEW to convert other types of collected files into an Excel spreadsheet's '.xlsx' format file.

3. The method for time-frequency characterization of friction vibration behavior of metal powder sintered composite layer cemented carbide according to claim 1, characterized in that: In step 2, VMD variational mode decomposition is used to remove the obvious interference parts in the vibration and perform noise reduction.

4. The method for time-frequency characterization of friction vibration behavior of metal powder sintered composite layer cemented carbide according to claim 1, characterized in that: In step 3, short-time Fourier transform and continuous wavelet transform are used to perform time-frequency analysis on the friction signal to obtain a time-frequency image.

5. The time-frequency characterization method for friction vibration behavior of metal powder sintered composite layer cemented carbide according to claim 1, characterized in that: The grayscale processing of the time-frequency image in step 4 specifically includes: Decolorize the image, simplify the complexity of the image, and shorten the calculation time; The grayscale algorithm is used to process the image according to 256 grayscale levels. The characteristic information such as image color and shape changes will be saved in the grayscale image in the form of grayscale pixels, and feature conversion is performed using the pixel distribution and pixel number of each grayscale level as characteristic information.

6. The method for time-frequency characterization of friction vibration behavior of metal powder sintered composite layer cemented carbide according to claim 1, characterized in that: The feature extraction of the vibration signal time-frequency image in step 5 specifically includes: Compare and adapt the feature value extraction methods based on the image features obtained by time-frequency analysis; For STFT image features, select 7 Hu invariant moment features A total of 8 features including grayscale mean are extracted as image feature values; For the CWT image features, 8 features including the mean and variance of the energy, contrast, uniformity, and entropy of the CWT image grayscale co-occurrence matrix in the four directions of 0°, 45°, 90°, and 135° were selected as image feature values ​​for extraction; finally, a total of 16 time-frequency features of friction vibration were obtained.

7. The time-frequency characterization method for friction vibration behavior of metal powder sintered composite layer cemented carbide according to claim 1, characterized in that: In step 6, the steps of reducing the dimension of the vibration behavior characteristics based on the MIC maximum mutual information coefficient-BP neural network method specifically include: Step 601, obtaining 16 eigenvalues ​​of friction and wear vibration of the metal powder sintered composite layer reinforced cemented carbide through step 5; Step 602: Taking the friction force, a parameter representing the stability of the friction and wear test system, as the output of feature screening, and taking the 16 eigenvalues ​​of STFT and CWT as input, feature selection based on the maximum mutual information coefficient of MIC is performed, which is specifically implemented as follows: By discretizing the input feature X and the output feature Y in two-dimensional space, the data is divided into grids according to the distribution of the feature values, and the mutual information between the feature and the target variable is calculated in each grid. Finally, all grids are traversed to find the grid with the maximum mutual information, so as to obtain the MIC value of each feature as the contribution value of the feature; Based on the experience of feature selection, when the difference between the maximum feature contribution and the minimum feature contribution is large, half of the difference between the maximum MIC value and the minimum MIC value between the features is used as the threshold for preliminary screening of feature importance, that is, the features with importance greater than 0.3 are retained in the preliminary screening; Step 602: Use the BP neural network friction prediction RMSE value to perform the next step of feature screening. Each time a feature is cut, 30 predicted RMSE values ​​are obtained to calculate the average.

8. The time-frequency characterization method for friction vibration behavior of metal powder sintered composite layer cemented carbide according to claim 1, characterized in that: The specific steps of fitting the stability characterization curve of the test system based on vibration behavior in step 8 include: Step 801: take the grayscale mean, entropy mean, contrast mean, The entropy variance and energy mean characteristics are numerically based on the vector normalization formula: In formula (1), c represents the coefficient, which is used to increase the data level and is set to 100; n represents the number of samples; a i Represents the actual measurement data; y i Represents the normalized data; the data is vector-normalized by using the norm(x) function in MATLAB to achieve normalization and eliminate the dimension effect; Step 802: Based on the normalized data, feature weights are divided. The specific implementation is: based on the importance of the MIC detection contribution value, the total contribution of the screening features is set to C, and the contribution value of each feature is x. i , i represents different features, and the contribution value of each feature accounts for the percentage of the total contribution value of the screening features, that is: where y i That is, the weight assigned to each retained feature; Step 803: multiply each normalized feature by its own weight and then sum them up to obtain six-feature fusion parameters, and fit the vibration behavior time-frequency characteristic law curve based on the fusion parameters; Step 804: by gray level mean, entropy mean, contrast mean, The vector normalization processing, weight division and weighted fitting of entropy variance and energy mean are performed to obtain the time-frequency characteristic law curve of vibration behavior, thereby characterizing the stability change curve of the sintered metal powder composite layer alloy strengthening performance detection system based on vibration behavior.

Images