Real-time monitoring method, device, terminal and storage medium for carbide evolution

Abstract

The present invention provides a method, device, terminal and storage medium for real-time monitoring of carbide evolution, and relates to the technical field of material deformation measurement. The method comprises: obtaining multiple images of different stretching stages of a metal sample during in-situ stretching; wherein the metal sample is surface-modified and fluorescently labeled; identifying the fluorescence intensity gradient in each image; converting the fluorescence intensity gradient of each image into a stress value, and obtaining the stress distribution on the surface of the metal sample in the image, so as to obtain the monitoring result of carbide evolution of the metal sample based on the stress distribution on the surface of the metal sample in each image. The present invention can realize real-time visualization of the carbide evolution process and its influence on the matrix and mechanical properties.

Description

Technical Field

[0001] The present invention relates to the technical field of material deformation measurement, and in particular to a method, device, terminal and storage medium for real-time monitoring of carbide evolution. Background Art

[0002] In-situ stretching is a novel experimental method that reveals the behavior of materials during stretching by applying force and measuring the in-situ strain of the material. This method is widely used in research in materials science, engineering and other related fields. In the in-situ stretching test, not only can the morphological characteristics of the sample at the moment of fracture be recorded and photographed, but also the physical quantities of the sample during the stretching process can be controlled in real time, and the force loaded on the sample surface can be read in real time through the mechanical sensor, and the stress-strain curve can be drawn to explore the sample properties from multiple dimensions.

[0003] However, there is still a lack of intuitive and accurate in-situ observation methods for the deformation, cracking and interaction between carbides and the matrix when the material is subjected to tensile forces. Traditional tensile tests can only obtain macroscopic mechanical properties data, and it is difficult to gain insight into the real-time evolution of carbides at the microscopic level, which limits the in-depth understanding of the failure mechanism of carbide-containing materials. Summary of the invention

[0004] The embodiments of the present invention provide a method, device, terminal and storage medium for real-time monitoring of carbide evolution, so as to solve the problem of real-time monitoring of the evolution of carbides at the microscopic level.

[0005] In a first aspect, an embodiment of the present invention provides a method for real-time monitoring of carbide evolution, comprising:

[0006] Acquire multiple images of a metal sample at different stretching stages during in-situ stretching; wherein the metal sample is surface-modified and fluorescently labeled;

[0007] Identify fluorescence intensity gradients in each image;

[0008] The fluorescence intensity gradient of each image is converted into a stress value to obtain the stress distribution on the surface of the metal sample in the image, so as to obtain the carbide evolution monitoring result of the metal sample based on the stress distribution on the surface of the metal sample in each image.

[0009] In a possible implementation, identifying the fluorescence intensity gradient in each image includes:

[0010] For each image, grayscale processing is performed on the image, and the grayscale value of each pixel in the processed image is used as the fluorescence intensity of the pixel, and the fluorescence intensity gradient in the image is calculated based on the grayscale value of each pixel.

[0011] In a possible implementation, calculating the fluorescence intensity gradient in the image based on the grayscale value of each pixel includes:

[0012] Dividing the image into a plurality of grids of a preset size;

[0013] Calculate the grayscale average of each pixel in each grid as the fluorescence intensity of the grid;

[0014] The fluorescence intensity difference between adjacent grids is calculated as the fluorescence intensity gradient between the two adjacent grids.

[0015] In a possible implementation, the fluorescence intensity gradient of each image is converted into a stress value, including:

[0016] A corresponding numerical conversion model is selected based on the stretching stage where the first image is located, and the fluorescence intensity gradient of each two adjacent grids in the first image is input into the numerical conversion model to obtain the stress value of the two adjacent grids; wherein the first image is any image.

[0017] In a possible implementation, before inputting the fluorescence intensity gradient of every two adjacent grids in the first image into the numerical conversion model, the method further includes:

[0018] Obtain the fluorescence intensity gradient and stress value of multiple different metal samples at multiple moments in the in-situ tensile test;

[0019] For each stretching stage of the in-situ stretching experiment, each fluorescence intensity gradient in the stretching stage is used as a training sample, and the stress value corresponding to the fluorescence intensity gradient is used as a sample label to construct a training data set for each stretching stage;

[0020] The neural network model is trained based on the training data set of each stretching stage to obtain a trained numerical conversion model corresponding to each stretching stage.

[0021] In a possible implementation, after converting the fluorescence intensity gradient of each image into a stress value, the following is further included:

[0022] For each image, the stress values ​​of the two adjacent grids in the image are used to draw a stress distribution cloud map corresponding to the image;

[0023] Based on the stress distribution cloud map corresponding to each image, the abnormal stress area changes on the surface of the metal sample during the in-situ stretching process are identified, and the carbide evolution monitoring results of the metal sample are obtained.

[0024] In a possible implementation, after drawing a stress distribution cloud map corresponding to each image using the stress values ​​of each of the two adjacent grids in the image, the method further includes:

[0025] Identify the interface interaction data, mechanical property curves and crack information of the metal sample surface in each image;

[0026] The stress distribution cloud map, interface interaction data, mechanical property curves and crack information corresponding to each image are input into the strain deviation model to evaluate the degree of local strain concentration caused by carbides in the metal sample.

[0027] In a second aspect, an embodiment of the present invention provides a device for real-time monitoring of carbide evolution, comprising:

[0028] An acquisition module, used to acquire multiple images of the metal sample at different stretching stages during in-situ stretching; wherein the metal sample is surface-modified and fluorescently labeled;

[0029] A recognition module for identifying fluorescence intensity gradients in each image;

[0030] The analysis module is used to convert the fluorescence intensity gradient of each image into a stress value to obtain the stress distribution on the surface of the metal sample in the image, so as to obtain the carbide evolution monitoring result of the metal sample based on the stress distribution on the surface of the metal sample in each image.

[0031] In a third aspect, an embodiment of the present invention provides a terminal, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the method described in the first aspect or any possible implementation manner of the first aspect are implemented.

[0032] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method described in the first aspect or any possible implementation method of the first aspect are implemented.

[0033] The embodiments of the present invention provide a method, device, terminal and storage medium for real-time monitoring of carbide evolution. By utilizing the correlation between fluorescence intensity and stress distribution after fluorescence labeling, the fluorescence intensity distribution of multiple images of a metal sample during in-situ stretching is converted into stress distribution, so as to reversely infer the influence of carbides on the surface of the metal sample on the matrix. Ultimately, real-time visualization of the carbide evolution process and its influence on the matrix and mechanical properties can be achieved, breaking through the limitations of previous post-analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0035] Figure 1 is a flow chart of a method for real-time monitoring of carbide evolution provided by an embodiment of the present invention;

[0036] Figure 2 is a schematic structural diagram of a carbide evolution real-time monitoring device provided by an embodiment of the present invention;

[0037] Figure 3 is a schematic diagram of a terminal provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0038] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present invention. However, it should be clear to those skilled in the art that the present invention may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present invention.

[0039] In order to make the purpose, technical solutions and advantages of the present invention more clear, specific embodiments will be described below in conjunction with the accompanying drawings.

[0040] In-situ stretching is a novel experimental method that reveals the behavior of materials during stretching by applying force and measuring the in-situ strain of the material. This method is widely used in research in materials science, engineering and other related fields. In the in-situ stretching test, not only can the morphological characteristics of the sample at the moment of fracture be recorded and photographed, but also the physical quantities of the sample during the stretching process can be controlled in real time, and the force loaded on the sample surface can be read in real time through the mechanical sensor, and the stress-strain curve can be drawn to explore the sample properties from multiple dimensions.

[0041] Fluorescence probe microscopy is a microscopy technique that uses fluorescent probes to observe and analyze samples. Fluorescent probes are a type of fluorescent molecule that has characteristic fluorescence in the ultraviolet-visible-near infrared region, and whose fluorescence properties (excitation and emission wavelengths, intensity, lifetime, polarization, etc.) can change sensitively with changes in the properties of the environment.

[0042] There is still a lack of intuitive and accurate in-situ observation methods for the deformation, cracking and interaction of carbides with the matrix when the material is subjected to tensile forces. Traditional tensile tests can only obtain macroscopic mechanical properties data, but it is difficult to gain insight into the real-time evolution of carbides at the microscopic level, which limits the in-depth understanding of the failure mechanism of carbide-containing materials.

[0043] In view of the above problems, the present invention includes the design of a fluorescent probe-assisted in-situ stretching machine, and accurately analyzes the dynamic characteristics of carbides during the stretching process. The present invention aims to provide a new analysis method and device, which utilizes the specific labeling and optical signal feedback characteristics of the fluorescent probe, combined with the real-time loading function of the in-situ stretching machine, to achieve dynamic microscopic analysis of carbides throughout the stretching process, accurately capture key changes such as its structure and stress distribution, reduce errors caused by manual identification, and add video recording functions, so as to better evaluate the positive and negative effects of carbides in mechanical properties at different deformation stages and different strains. At the same time, the influence of carbides on the matrix organization can be observed through the matrix fluorescent points. Finally, the model is used to realize the automatic evaluation of the evolution process of carbides and the influence on the matrix and mechanical properties. Finally, the in-situ and real-time visualization of the carbide stretching process is achieved, breaking through the limitations of previous post-analysis.

[0044] See also Figure 1 , which shows a flow chart of the implementation of the method for real-time monitoring of carbide evolution provided by an embodiment of the present invention, and is described in detail as follows:

[0045] Step 101, obtaining multiple images of a metal sample at different stretching stages during an in-situ stretching process; wherein the metal sample has been surface modified and fluorescently labeled.

[0046] In this embodiment, the metal surface is modified in the early stage of monitoring, and the metal matrix is ​​marked with a fluorescent probe to highlight the position, shape, distribution, etc. of the carbide, so as to realize intuitive positioning of the carbide, instead of traditional manual macroscopic judgment. The specific implementation process can be as follows:

[0047] 1. Surface modification design

[0048] a) Prepare 4% nitric acid alcohol etching solution and corrode the sample surface for 30 seconds;

[0049] b) The corroded sample is placed in a beaker containing anhydrous ethanol for ultrasonic cleaning. After cleaning, it is taken out and dried for plasma treatment for surface modification. The sample is treated with oxygen plasma for 20 minutes to increase the polarity and hydrophilicity of the surface, which is conducive to the subsequent attachment of 3-aminopropyltriethoxysilane (APTES). The treated sample is immersed in an ethanol solution containing APTES to introduce amino groups for surface functionalization, while no reaction or low reaction occurs on the carbide surface.

[0050] c) Heat to 25°C and keep warm for 20 minutes. Take out the sample, wash it with anhydrous ethanol several times and dry it in a drying oven at 80°C for 30 minutes to ensure that APTES is completely cured.

[0051] d) Add fluorescein isothiocyanate (FITC) solution to a prepared solution of 5 mM 1-(3-methylaminopropyl)-3-ethylcarbodiimide hydrochloride (EDC) and 10 mM N-hydroxysuccinimide (NHS) at a ratio of 1:10, and heat to 25°C in a beaker and stir ultrasonically for 20 minutes for activation. Activated fluorescein allows the isothiocyanate group in FITC to react with the functionalized metal substrate surface with amino groups to form a covalent bond.

[0052] e) Fluorescein labeling was performed using fluorescein isothiocyanate (FITC). The functionalized sample was placed in an activated FITC solution and ultrasonically stirred in a beaker at 25°C for 1 hour for labeling. The sample was then taken out and the surface was cleaned with anhydrous ethanol to remove unreacted fluorescein and dried in a drying oven at 25°C.

[0053] 2. Perform in-situ stretching of metal samples, equipped with an in-situ stretching machine, high-precision force sensors and displacement sensors, to precisely control the stretching rate and load size. High-sensitivity fluorescence microscopes and spectrometers are arranged around the loading platform of the stretching machine. During the stretching process, the fluorescence microscope captures the fluorescence imaging of the fluorescent probe in real time, and the spectrometer synchronously records the peak position and intensity changes of the fluorescence spectrum. The fluorescence microscope and the in-situ stretching machine are connected to a computer to build a platform to collect images during the stretching process.

[0054] 3. Analysis of carbide evolution process

[0055] According to the stretching stage, you can collect images or videos of the evolution of carbides in the initial stage, elastic deformation stage, yield stage, strengthening stage, maximum strength stage, and necking stage, or collect image information under different stresses (stress: 5%, 10%, 15%... until it breaks). Before recognition, you can also perform noise reduction on these images to remove noise and noise caused by ambient light interference, while enhancing image contrast to make the boundaries of the fluorescent area clearer.

[0056] Step 102: Identify the fluorescence intensity gradient in each image.

[0057] In this embodiment, since the presence of carbides will interfere with the stress transfer of the metal matrix, the stress concentration of the metal matrix around the carbides will be reflected in the change of fluorescence intensity. Therefore, the stress distribution on the surface of the metal sample in each image can be effectively described by the fluorescence intensity gradient.

[0058] Specifically, a gradient operator (such as Sobel, Prewitt) may be applied to extract the rate of change of the fluorescence intensity in the image.

[0059] Step 103, converting the fluorescence intensity gradient of each image into a stress value to obtain the stress distribution on the surface of the metal sample in the image, so as to obtain the carbide evolution monitoring result of the metal sample based on the stress distribution on the surface of the metal sample in each image.

[0060] In this embodiment, the physical environment of the fluorescent probe used to mark the metal matrix will change significantly due to stress concentration. The stress concentration area is accompanied by local plastic deformation, which will cause the lattice around the fluorescent probe molecule to distort. The luminescence characteristics of fluorescent substances are closely related to the spatial conformation of the molecules and the intermolecular interactions. Lattice distortion disturbs the energy level structure of the fluorescent molecules, and the energy difference between the originally stable excited state and the ground state will change, which will directly affect the intensity of the fluorescence emission.

[0061] Therefore, the effect of carbide on the matrix can be inferred based on the difference in the spatial distribution of fluorescence intensity. The type of carbide crystal structure distortion can be determined based on the red-shift and blue-shift phenomena of the fluorescence spectrum over the stretching time. Combined with mechanical data, a carbide microstructure-macroscopic mechanical property correlation model can be established. During the stretching process, the stress distribution can be identified in real time through images and correlation models, and the position, shape, and distribution of carbides can be identified.

[0062] The embodiment of the present invention utilizes the correlation between fluorescence intensity and stress distribution after fluorescence labeling to convert the fluorescence intensity distribution of multiple images of the metal sample during in-situ stretching into stress distribution, thereby realizing the reverse inference of the influence of carbides on the surface of the metal sample on the matrix. Ultimately, real-time visualization of the carbide evolution process and its influence on the matrix and mechanical properties can be achieved, breaking through the limitations of previous post-analysis.

[0063] In a possible implementation, identifying the fluorescence intensity gradient in each image includes:

[0064] For each image, grayscale processing is performed on the image, and the grayscale value of each pixel in the processed image is used as the fluorescence intensity of the pixel, and the fluorescence intensity gradient in the image is calculated based on the grayscale value of each pixel.

[0065] In this embodiment, the collected image can be imported into ImageJ and converted into grayscale mode, and the grayscale value can reflect the fluorescence intensity. Then, select the "Measure" option under the "Analyze" menu, and the software will pop up a result window to display various parameters of the selected area, including the mean value (Mean) and standard deviation (StdDev) of the grayscale value.

[0066] In a possible implementation, calculating the fluorescence intensity gradient in the image based on the grayscale value of each pixel includes:

[0067] Dividing the image into a plurality of grids of a preset size;

[0068] Calculate the grayscale average of each pixel in each grid as the fluorescence intensity of the grid;

[0069] The fluorescence intensity difference between adjacent grids is calculated as the fluorescence intensity gradient between the two adjacent grids.

[0070] In this embodiment, the fluorescence intensity gradient calculation is as follows: the processed image is divided into fine (1mm×1mm) grid areas, and the fluorescence intensity is calculated grid by grid. The fluorescence intensity gradient is calculated by comparing the fluorescence intensity difference between adjacent grids (the fluorescence intensity gradient is often larger in the stress concentration area).

[0071] Adjacent grids refer to grids that are directly adjacent in the horizontal or vertical direction in the divided image grids. For a grid located in the i-th row and j-th column, its adjacent grids are the grids in the i-1-th row and j-th column, the i+1-th row and j-th column, the i-th row and j-1-th column, and the i-th row and j+1-th column (provided that these grids are within the image range). When calculating the fluorescence intensity gradient, the fluorescence intensity difference between these adjacent grids is mainly considered.

[0072] Table 1

[0073]

[0074] Table 1 shows an example of fluorescence intensity of the image after gridding. In this example, each square represents a 1mm×1mm grid area, and the value in the square represents the fluorescence intensity gradient value calculated in the area (the value is assumed here for illustration). It can be seen from the figure that the intensity gradient value varies in different areas. For example, the intensity gradient of some grids in the middle part is relatively large, which suggests that there is stress concentration and carbide influence in these areas.

[0075] For each 1mm×1mm grid area, the fluorescence intensity value is obtained through the above operation. Assuming that Iij represents the fluorescence intensity of the grid in the i-th row and j-th column (where i and j are the row and column indices of the grid in the image), Iij is obtained by selecting pixels in the grid area and calculating their grayscale average.

[0076] In a possible implementation, the fluorescence intensity gradient of each image is converted into a stress value, including:

[0077] A corresponding numerical conversion model is selected based on the stretching stage where the first image is located, and the fluorescence intensity gradient of each two adjacent grids in the first image is input into the numerical conversion model to obtain the stress value of the two adjacent grids; wherein the first image is any image.

[0078] In this embodiment, based on the theory of material mechanics, the fluorescence intensity gradient data is input into a pre-constructed numerical conversion model. The model is trained with images and stress-strain curves of different stages, and combined with the fluorescence-stress coupling law of experimental metal materials, it can convert the fluorescence intensity gradient into the corresponding stress value, and then draw a stress distribution cloud map of the metal matrix at each stretching stage, and the stress anomaly at the location of the carbide is highlighted.

[0079] In a possible implementation, before inputting the fluorescence intensity gradient of every two adjacent grids in the first image into the numerical conversion model, the method further includes:

[0080] Obtain the fluorescence intensity gradient and stress value of multiple different metal samples at multiple moments in the in-situ tensile test;

[0081] For each stretching stage of the in-situ stretching experiment, each fluorescence intensity gradient in the stretching stage is used as a training sample, and the stress value corresponding to the fluorescence intensity gradient is used as a sample label to construct a training data set for each stretching stage;

[0082] The neural network model is trained based on the training data set of each stretching stage to obtain a trained numerical conversion model corresponding to each stretching stage.

[0083] In this embodiment, because the interaction mechanism between carbides and the matrix and the stress-strain distribution law are different in different stages of material stretching, such as elastic deformation, yielding, strengthening, and necking, the relationship between fluorescence intensity and stress is not exactly the same. In the elastic stage, the material deformation is mainly elastic, and the fluorescence intensity is more sensitive to stress changes and shows a linear correlation trend; in the necking stage, the local deformation of the material is concentrated, the stress distribution is complex, and the relationship between fluorescence intensity and stress becomes complicated. Constructing models separately can better analyze the characteristics of different stages and improve accuracy.

[0084] The neural network model can effectively handle the complex nonlinear relationship between fluorescence intensity and stress values. During the stretching process, factors that affect the fluorescence intensity include the morphology, distribution, and crystal structure changes of carbides, as well as the deformation and stress concentration of the matrix, which makes the relationship between fluorescence intensity and stress complex. The neural network constructs a complex network structure through a large number of neurons and can automatically learn complex patterns and rules in the data.

[0085] In a possible implementation, after converting the fluorescence intensity gradient of each image into a stress value, the following is further included:

[0086] For each image, the stress values ​​of the two adjacent grids in the image are used to draw a stress distribution cloud map corresponding to the image;

[0087] Based on the stress distribution cloud map corresponding to each image, the abnormal stress area changes on the surface of the metal sample during the in-situ stretching process are identified, and the carbide evolution monitoring results of the metal sample are obtained.

[0088] In this embodiment, data from a large number of different material samples in in-situ tensile experiments are collected. This covers a variety of carbide-containing materials, such as alloy steels of different compositions, nonferrous metal alloys, etc. Each material sample collects multiple sets of data during the tensile process, including different tensile stages (elasticity, yielding, strengthening, etc.), fluorescence microscope image data (fluorescence intensity information) at different stress levels, spectrometer data (fluorescence spectrum characteristics), and mechanical sensor data (stress, strain values).

[0089] Obtaining sample labels:

[0090] The high-precision force sensor measures the force applied to the sample in real time, and combines the original size information of the sample to obtain the stress value of the sample at each moment. These stress values ​​are matched with the fluorescence intensity data collected at the same moment to obtain the sample label.

[0091] At each moment of the stretching experiment, there are corresponding fluorescence intensity data and stress data. The stress value at the same moment is associated with the fluorescence intensity data, and this stress value becomes the sample label of the corresponding fluorescence intensity data. By collecting a large amount of such paired data at different stretching stages and different stress levels, a dataset with sample labels for model training is constructed.

[0092] In a possible implementation, after drawing a stress distribution cloud map corresponding to each image using the stress values ​​of each of the two adjacent grids in the image, the method further includes:

[0093] Identify the interface interaction data, mechanical property curves and crack information of the metal sample surface in each image;

[0094] The stress distribution cloud map, interface interaction data, mechanical property curves and crack information corresponding to each image are input into the strain deviation model to evaluate the degree of local strain concentration caused by carbides in the metal sample.

[0095] In this embodiment, during the stretching process of the material, the stress distribution cloud map, interface interaction data, mechanical property curve and crack information are closely related: stress distribution affects interface interaction, and the interface bonding state changes the stress transfer path. The two jointly determine the mechanical properties, and the generation and development of cracks are the result of the combined effect of these factors, reflecting the failure trend of the material. The strain deviation model integrates this information, constructs a comprehensive evaluation system, analyzes the influence mechanism and weight of each factor, and comprehensively evaluates the evolution of carbides and their impact on material properties.

[0096] 1. Stress distribution cloud map: The stress concentration area is the weak point of the material, which is easy to cause cracks and affect the strength and life of the material. The greater the stress concentration coefficient, the higher the risk of material failure. By analyzing the characteristics of the stress distribution cloud map, the internal stress state of the material and the potential failure area can be determined.

[0097] 2. Interface interaction data: Good interface bonding allows for uniform stress transfer and enhances material performance; interface debonding or slippage leads to stress concentration and reduces material strength. Interface bonding strength and debonding degree affect material mechanical properties and crack propagation. Measuring interface bonding strength and observing debonding can assess the impact of interface interaction on material performance.

[0098] 3. Stress-strain curve: The elastic modulus reflects the material's ability to resist elastic deformation, the yield strength indicates the stress at which the material begins to plastically deform, and the tensile strength reflects the maximum stress the material can withstand. Obtain key parameters from the mechanical property curve to intuitively understand the material's macroscopic mechanical properties and deformation behavior.

[0099] 4. Crack information: Crack initiation and expansion are important processes in material failure. Crack length, expansion rate and direction reflect the degree of material damage and failure trend. The longer the crack length and the faster the expansion rate, the shorter the remaining life of the material. Monitoring crack information can evaluate the material damage state and predict the failure time.

[0100] The stress distribution cloud map, interface interaction data, mechanical property curves and crack information can be integrated to measure the uniformity of the overall deformation of the metal matrix in areas with or without carbides, aggregated or sparse, and of different sizes. This information is input into the final model to automatically evaluate the strain deviation in different areas, and then evaluate the local strain concentration caused by carbides, and determine whether carbides promote or hinder the coordinated deformation of the metal matrix.

[0101] The final strain deviation model is implemented by:

[0102] 1. Data processing: Convert the stress distribution cloud map into a numerical matrix to extract the coordinates of the stress concentration area, stress amplitude and other features; normalize the interface interaction data to unify the dimension and numerical range; digitize the mechanical property curve to extract characteristic parameters such as elastic modulus and yield strength; quantify crack information, such as converting crack length and extension angle into numerical values.

[0103] 2. Build a neural network model: the input layer nodes correspond to the preprocessed data features, the number of layers and nodes in the hidden layer is determined according to the data complexity, and the output layer is the comprehensive evaluation result.

[0104] 3. Model training and optimization: Collect a large amount of multi-source data under different materials and stretching conditions, annotate the corresponding evaluation results, and form a training data set. Use optimization algorithms such as stochastic gradient descent and Adam to adjust model parameters through back propagation to minimize the error between the predicted results and the actual evaluation results, and improve the accuracy and generalization ability of the model. Use cross-validation technology to divide the training data into multiple subsets, train and verify the model in turn, evaluate the model performance, and select the optimal model parameters.

[0105] 4. Model evaluation and application: Use the test data set to evaluate the trained model, calculate the accuracy, recall rate, mean square error and other indicators, and judge the model performance. If the model performance does not meet the standard, adjust the model structure, parameters or data preprocessing methods, retrain and evaluate. Apply the verified model to the actual material tensile data evaluation, output the evaluation results such as the influence of carbides on the mechanical properties of materials, the risk level of material failure, etc., to provide a decision-making basis for material optimization design and performance improvement.

[0106] Specifically, the step of identifying the interface interaction data may include:

[0107] During the stretching process, the interface area between the carbide and the metal matrix will produce relative displacement and microcrack initiation due to the force, which will cause special changes in the fluorescence signal marked on the metal matrix side. The image recognition algorithm can be used to accurately lock the interface contour, extract the curve of the fluorescence intensity change over time near the interface, identify the interface interaction state between the carbide and the metal matrix through the change characteristics of the fluorescence intensity, and realize the monitoring of the dynamic evolution of the interface interaction affected by the carbide during the stretching process.

[0108] During normal stretching, the interface is well bonded and the fluorescence intensity changes steadily; if the carbide and the matrix debond, the fluorescence intensity at the interface will drop sharply or fluctuate more sharply; when tiny cracks or local slip begin to appear at the interface, the rate of change of the fluorescence intensity will accelerate.

[0109] The changes in the interface fluorescence intensity of the same material sample at different stretching stages are compared. In the elastic deformation stage, the fluorescence intensity changes are usually small and regular; when entering the plastic deformation stage, especially when approaching the yield point of the material, the interface interaction may change, and the fluctuation of the fluorescence intensity will increase significantly. Through this comparison, the dynamic evolution of the interface interaction affected by carbides during the stretching process can be clearly observed.

[0110] By comparing the interfacial fluorescence curves at different stretching times and in different regions, the amplitude, rate of change and fluctuation frequency of fluorescence intensity can be selected as the determining factors of the interfacial interaction state.

[0111] The interface line between carbide and metal matrix is ​​identified in each image by using the Canny edge detection algorithm, and then the interface area can be determined by extending the interface line by 5-10 pixels to both sides. The extended distance can be adjusted according to the actual situation.

[0112] Based on the fluorescence intensity of the interface area in each image and the shooting time, a curve of fluorescence intensity changing with time is drawn. Based on the curve of fluorescence intensity changing with time, the fluorescence intensity change amplitude, fluorescence change rate and fluorescence fluctuation frequency of the target time period are calculated, and weighted summation is performed with a weight of 1:1:1 to obtain a numerical value that can comprehensively reflect the state of interface interaction, that is, the interface interaction state index of the metal sample in the target time period.

[0113] By calibrating this value with a large number of experiments under different material systems and tensile conditions, the range of this value under different interface interaction states is determined (for example, 0-0.3 indicates good bonding, 0.3-0.7 indicates partial debonding, and 0.7-1 indicates severe debonding), thereby quantifying the weakening or strengthening effect of carbides on the interface bonding strength.

[0114] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0115] The following is an embodiment of the device of the present invention. For details not described in detail therein, reference may be made to the corresponding method embodiment described above.

[0116] Figure 2 The structural schematic diagram of the real-time monitoring device for carbide evolution provided by an embodiment of the present invention is shown. For the convenience of explanation, only the part related to the embodiment of the present invention is shown, which is described in detail as follows:

[0117] like Figure 2 As shown, the carbide evolution real-time monitoring device 2 comprises:

[0118] An acquisition module 21 is used to acquire multiple images of the metal sample at different stretching stages during in-situ stretching; wherein the metal sample is surface-modified and fluorescently labeled;

[0119] An identification module 22, used to identify the fluorescence intensity gradient in each image;

[0120] The analysis module 23 is used to convert the fluorescence intensity gradient of each image into a stress value to obtain the stress distribution on the surface of the metal sample in the image, so as to obtain the carbide evolution monitoring result of the metal sample based on the stress distribution on the surface of the metal sample in each image.

[0121] In a possible implementation, the identification module 22 is specifically configured to:

[0122] For each image, grayscale processing is performed on the image, and the grayscale value of each pixel in the processed image is used as the fluorescence intensity of the pixel, and the fluorescence intensity gradient in the image is calculated based on the grayscale value of each pixel.

[0123] In a possible implementation, the identification module 22 is specifically configured to:

[0124] Dividing the image into a plurality of grids of a preset size;

[0125] Calculate the grayscale average of each pixel in each grid as the fluorescence intensity of the grid;

[0126] The fluorescence intensity difference between adjacent grids is calculated as the fluorescence intensity gradient between the two adjacent grids.

[0127] In a possible implementation, the analysis module 23 is specifically used for:

[0128] A corresponding numerical conversion model is selected based on the stretching stage where the first image is located, and the fluorescence intensity gradient of each two adjacent grids in the first image is input into the numerical conversion model to obtain the stress value of the two adjacent grids; wherein the first image is any image.

[0129] In a possible implementation, the analysis module 23 is further configured to:

[0130] Before inputting the fluorescence intensity gradient of each two adjacent grids in the first image into the numerical conversion model, obtaining the fluorescence intensity gradient and stress value of multiple different metal samples at multiple moments in the in-situ tensile test;

[0131] For each stretching stage of the in-situ stretching experiment, each fluorescence intensity gradient in the stretching stage is used as a training sample, and the stress value corresponding to the fluorescence intensity gradient is used as a sample label to construct a training data set for each stretching stage;

[0132] The neural network model is trained based on the training data set of each stretching stage to obtain a trained numerical conversion model corresponding to each stretching stage.

[0133] In a possible implementation, the analysis module 23 is further configured to:

[0134] After converting the fluorescence intensity gradient of each image into a stress value, for each image, the stress values ​​of each of the two adjacent grids in the image are used to draw a stress distribution cloud map corresponding to the image;

[0135] Based on the stress distribution cloud map corresponding to each image, the abnormal stress area changes on the surface of the metal sample during the in-situ stretching process are identified, and the carbide evolution monitoring results of the metal sample are obtained.

[0136] In a possible implementation, the analysis module 23 is further configured to:

[0137] For each image, after drawing a stress distribution cloud map corresponding to the image using the stress values ​​of each of the two adjacent grids in the image, the interface interaction data, mechanical property curves and crack information on the surface of the metal sample in each image are identified;

[0138] The stress distribution cloud map, interface interaction data, mechanical property curves and crack information corresponding to each image are input into the strain deviation model to evaluate the degree of local strain concentration caused by carbides in the metal sample.

[0139] The embodiment of the present invention utilizes the correlation between fluorescence intensity and stress distribution after fluorescence labeling to convert the fluorescence intensity distribution of multiple images of the metal sample during in-situ stretching into stress distribution, thereby realizing the reverse inference of the influence of carbides on the surface of the metal sample on the matrix. Ultimately, real-time visualization of the carbide evolution process and its influence on the matrix and mechanical properties can be achieved, breaking through the limitations of previous post-analysis.

[0140] Figure 3 is a schematic diagram of a terminal provided by an embodiment of the present invention. Figure 3 As shown, the terminal 3 of this embodiment includes: a processor 30, a memory 31, and a computer program 32 stored in the memory 31 and executable on the processor 30. When the processor 30 executes the computer program 32, the steps in the above-mentioned embodiments of the method for real-time monitoring of carbide evolution are implemented, for example Figure 1 Alternatively, when the processor 30 executes the computer program 32, the functions of each module / unit in the above-mentioned device embodiments are realized, for example, Figure 2 Functions of modules / units 21 to 23 are shown.

[0141] Exemplarily, the computer program 32 may be divided into one or more modules / units, which are stored in the memory 31 and executed by the processor 30 to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, which are used to describe the execution process of the computer program 32 in the terminal 3. For example, the computer program 32 may be divided into Figure 2 Modules / units 21 to 23 are shown.

[0142] The terminal 3 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The terminal 3 may include, but is not limited to, a processor 30 and a memory 31. Those skilled in the art will appreciate that Figure 3 It is only an example of terminal 3 and does not constitute a limitation on terminal 3. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the terminal may also include input and output devices, network access devices, buses, etc.

[0143] The processor 30 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0144] The memory 31 may be an internal storage unit of the terminal 3, such as a hard disk or memory of the terminal 3. The memory 31 may also be an external storage device of the terminal 3, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the terminal 3. Further, the memory 31 may also include both an internal storage unit and an external storage device of the terminal 3. The memory 31 is used to store the computer program and other programs and data required by the terminal. The memory 31 may also be used to temporarily store data that has been output or is to be output.

[0145] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0146] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0147] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0148] In the embodiments provided by the present invention, it should be understood that the disclosed devices / terminals and methods can be implemented in other ways. For example, the device / terminal embodiments described above are only schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0149] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0150] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0151] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned various carbide evolution real-time monitoring method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device that can carry the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practices in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practices, computer-readable media does not include electrical carrier signals and telecommunication signals.

[0152] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A method for real-time monitoring of carbide evolution, characterized in that: include: Acquiring multiple images of a metal sample at different stretching stages during in-situ stretching; wherein the metal sample is surface-modified and fluorescently labeled; Identify fluorescence intensity gradients in each image; Converting the fluorescence intensity gradient of each image into a stress value to obtain the stress distribution on the surface of the metal sample in the image, so as to obtain the carbide evolution monitoring result of the metal sample based on the stress distribution on the surface of the metal sample in each image; The step of identifying the fluorescence intensity gradient in each image comprises: For each image, grayscale processing is performed on the image, and the grayscale value of each pixel in the processed image is used as the fluorescence intensity of the pixel, and the fluorescence intensity gradient in the image is calculated based on the grayscale value of each pixel; The step of calculating the fluorescence intensity gradient in the image based on the grayscale value of each pixel comprises: Dividing the image into a plurality of grids of a preset size; Calculate the grayscale average of each pixel in each grid as the fluorescence intensity of the grid; The fluorescence intensity difference between adjacent grids is calculated as the fluorescence intensity gradient between the two adjacent grids.

2. The method for real-time monitoring of carbide evolution according to claim 1, characterized in that: The step of converting the fluorescence intensity gradient of each image into a stress value comprises: A corresponding numerical conversion model is selected based on the stretching stage of the first image, and the fluorescence intensity gradient of each two adjacent grids in the first image is input into the numerical conversion model to obtain the stress value of the two adjacent grids; wherein the first image is any image.

3. The method for real-time monitoring of carbide evolution according to claim 2, characterized in that: Before inputting the fluorescence intensity gradient of every two adjacent grids in the first image into the numerical conversion model, the method further includes: Obtain the fluorescence intensity gradient and stress value of multiple different metal samples at multiple moments in the in-situ tensile test; For each stretching stage of the in-situ stretching experiment, each fluorescence intensity gradient in the stretching stage is used as a training sample, and the stress value corresponding to the fluorescence intensity gradient is used as a sample label to construct a training data set for each stretching stage; The neural network model is trained based on the training data set of each stretching stage to obtain a trained numerical conversion model corresponding to each stretching stage.

4. The method for real-time monitoring of carbide evolution according to claim 2, characterized in that: After converting the fluorescence intensity gradient of each image into a stress value, the method further includes: For each image, the stress values ​​of the two adjacent grids in the image are used to draw a stress distribution cloud map corresponding to the image; Based on the stress distribution cloud map corresponding to each image, the abnormal stress area changes on the surface of the metal sample during the in-situ stretching process are identified to obtain the carbide evolution monitoring results of the metal sample.

5. The method for real-time monitoring of carbide evolution according to claim 4, characterized in that: After drawing the stress distribution cloud map corresponding to each image based on the stress values ​​of the two adjacent grids in the image, the method further includes: Identify interface interaction data, mechanical property curves and crack information of the metal sample surface in each image; The stress distribution cloud map, interface interaction data, mechanical property curve and crack information corresponding to each image are input into the strain deviation model to evaluate the local strain concentration degree caused by carbides on the metal sample.

6. A real-time monitoring device for carbide evolution, characterized in that: include: An acquisition module, used to acquire multiple images of the metal sample at different stretching stages during in-situ stretching; wherein the metal sample is surface-modified and fluorescently labeled; A recognition module for identifying fluorescence intensity gradients in each image; An analysis module, used to convert the fluorescence intensity gradient of each image into a stress value, and obtain the stress distribution on the surface of the metal sample in the image, so as to obtain the carbide evolution monitoring result of the metal sample based on the stress distribution on the surface of the metal sample in each image; The identification module is specifically used for: For each image, grayscale processing is performed on the image, and the grayscale value of each pixel in the processed image is used as the fluorescence intensity of the pixel, and the fluorescence intensity gradient in the image is calculated based on the grayscale value of each pixel; The identification module is specifically used for: Dividing the image into a plurality of grids of a preset size; Calculate the grayscale average of each pixel in each grid as the fluorescence intensity of the grid; The fluorescence intensity difference between adjacent grids is calculated as the fluorescence intensity gradient between the two adjacent grids.

7. A terminal comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

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

Images