Non-oriented silicon steel grain distribution characteristic characterization method and system

Abstract

The invention provides a non-oriented silicon steel grain distribution characteristic characterization method and system, and relates to the technical field of metal materials, the non-oriented silicon steel grain distribution characteristic characterization method comprises the following steps: preparing a plurality of samples to be detected by using non-oriented silicon steel plates produced by a plurality of production processes; analyzing the plurality of to-be-detected samples to obtain a plurality of grain size distribution data sets corresponding to the to-be-detected samples; based on the grain size distribution density, grain grade density is obtained; inputting the grain grade density and the grain distribution parameter into a preset Rayleigh model to obtain a grain size distribution fitting image corresponding to the to-be-detected sample; on the basis of the grain size distribution fitting image, feature fitting parameters are obtained to represent grain distribution features of the to-be-detected sample. According to the non-oriented silicon steel grain distribution characteristic characterization method, the technical problem that the influence of different process parameters on grain distribution is difficult to accurately quantify in related technologies is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of metal materials, and particularly to a method and system for characterizing the grain distribution characteristics of non-oriented silicon steel. Background Art

[0002] With the development of energy-saving and high-efficiency motor technologies, higher requirements are put forward for the low iron loss and high magnetic permeability of silicon steel. Non-oriented silicon steel is an important electrical engineering material with isotropic magnetic properties and is mainly used in the core components of motors and transformers.

[0003] The grain size has a significant impact on the magnetic properties of silicon steel, such as magnetic permeability, iron loss, and coercivity. The effects of grain size on hysteresis loss and eddy current loss are contradictory, and at the same time, the uniformity of grain distribution has a significant impact on coercivity and magnetic permeability. Therefore, how to further produce non-oriented electrical steel with the minimum total core loss Wt by optimizing the process to control the grain size distribution is the focus of current research.

[0004] In the prior art, the measurement of grain size distribution by microstructure image analysis technology mostly stays at the histogram level. There is no unified mathematical model to summarize the grain size, and it is impossible to accurately describe the distribution law of grains, nor is it easy to accurately quantify the influence of different process parameters on grain distribution, such as annealing time, cooling rate, etc. It is impossible to establish a clear mathematical correlation model with material properties (such as magnetic properties), which limits the efficiency of process optimization. Therefore, it is necessary to develop a model that can combine the distribution law of microstructure with macroscopic properties to guide the selection and optimization of process parameters. Establish a quantitative relationship between grain distribution and performance, form a general optimization tool, and improve the efficiency of process design. In recent years, the prior art has tried to introduce Gaussian distribution or lognormal distribution, but due to the random growth of metal grain size in all directions, these models perform poorly in describing the distribution law after grain refinement.

[0005] Therefore, the prior art needs to be further developed. Summary of the Invention

[0006] The purpose of the present invention is to overcome the above technical deficiencies and provide a method and system for characterizing the grain distribution characteristics of non-oriented silicon steel to solve the technical problem in the related art that it is difficult to accurately quantify the influence of different process parameters on grain distribution.

[0007] To achieve the above technical objectives, the present invention adopts the following technical solutions: A method for characterizing the grain distribution characteristics of non-oriented silicon steel is provided, including: preparing a plurality of samples to be detected from non-oriented silicon steel sheets produced by various production processes; analyzing the plurality of samples to be detected to obtain a grain size distribution data set corresponding to each sample to be detected; obtaining the grain grade density based on the grain size distribution parameters; inputting the grain grade density and the grain distribution parameters into a preset Rayleigh model to obtain a grain size distribution fitting image corresponding to the sample to be detected; and obtaining characteristic fitting parameters based on the grain size distribution fitting image to characterize the grain distribution characteristics of the sample to be detected.

[0008] Further, the non-oriented silicon steel produced by various production processes includes: normalized non-oriented silicon steel, double cold rolled non-oriented silicon steel, continuously annealed non-oriented silicon steel, batch annealed non-oriented silicon steel, thin slab continuous casting and rolling non-oriented silicon steel, and laser scribed non-oriented silicon steel.

[0009] Further, the grain orientations of the non-oriented silicon steel produced by various production processes are different; the grain distributions of the non-oriented silicon steel produced by various production processes are different.

[0010] Further, the method for preparing a plurality of samples to be detected includes selecting sampling positions in different directions on the non-oriented silicon steel sheet and cutting the corresponding samples to be detected from the plurality of sampling positions.

[0011] Further, the method for analyzing the plurality of samples to be detected includes: collecting data of the samples to be detected through an electron backscatter diffraction system to obtain data to be analyzed, where the data to be analyzed includes the grain size and the corresponding proportion of the grain size; and obtaining the grain size distribution data set based on the data to be analyzed.

[0012] Further, the method for obtaining the grain grade density includes: obtaining the grain size interval based on the grain size; evenly dividing the grain size interval into n grade intervals, where n ∈ Z; and classifying the grains according to the grade interval corresponding to the grain size to obtain the grain grade density.

[0013] Further, the method for characterizing the grain distribution characteristics of non-oriented silicon steel further includes: sequentially detecting the samples to be detected supported by the non-oriented silicon steel produced by various production processes to obtain the corresponding characteristic fitting parameters; and establishing a process-grain distribution characteristic correlation model based on the production process corresponding to the sample to be detected and the characteristic fitting parameters of the sample to be detected.

[0014] Further, the method for characterizing the grain distribution characteristics of non-oriented silicon steel further includes: obtaining the characteristic fitting parameters corresponding to the performance according to the required performance of the material; and obtaining the production process corresponding to the performance according to the process-grain distribution characteristic correlation model.

[0015] A grain distribution characteristic characterization system, the grain distribution characteristic characterization system includes: a sample preparation unit for preparing a plurality of samples to be detected from materials produced by a variety of production processes; a data acquisition unit for analyzing the plurality of samples to be detected to obtain grain size distribution parameters corresponding to the plurality of samples to be detected; a data processing unit for obtaining a grain grade density based on the grain size distribution parameters; a fitting unit for inputting the grain grade density and the grain size distribution parameters into a preset fitting model to obtain a grain size distribution fitting image corresponding to the sample to be detected; and a characterization unit for obtaining characteristic fitting parameters based on the grain size distribution fitting image to characterize the grain distribution characteristics of the sample to be detected.

[0016] A computer-readable storage medium stores computer-readable instructions thereon, and when the computer-readable instructions are executed by a processor, each step of the method for characterizing the grain distribution characteristics of non-oriented silicon steel as described in any one of the above is implemented.

[0017] Beneficial effects:

[0018] 1. The method for characterizing the grain distribution characteristics of non-oriented silicon steel according to the present invention selects a mathematical model (i.e., the Rayleigh distribution model) that conforms to the grain distribution law, classifies the grain sizes into grades, performs mathematical fitting, and extracts the characteristic parameters after fitting, which can effectively quantify the grain distribution law of non-oriented silicon steel under different process conditions, and accurately characterize the grain distribution characteristics of non-oriented silicon steel produced under different process conditions. In the prior art, the ability to distinguish the change in the grain size distribution characteristics caused by the subtle change in the process parameters in the same process is relatively low, and it is difficult to accurately and intuitively characterize the influence of the subtle change in the process parameters in the same process on the grain size distribution, resulting in the existing mathematical models being unsuitable for guiding the actual production of non-oriented silicon steel. However, the present invention samples and detects from different directions of non-oriented silicon steel to accurately quantify the subtle changes in the grain size distribution at different angles of non-oriented silicon steel. Therefore, while the present application can satisfy the quantification of the grain distribution characteristics of non-oriented silicon steel produced by different production processes, it can also accurately quantify the change in the grain distribution characteristics of non-oriented silicon steel caused by only individual production parameters being different under the same production process. Therefore, the present application can be used to guide actual production, design and adjust the production process, so as to improve the stability and efficiency of the preparation process.

[0019] 2. The method for characterizing the grain distribution characteristics of non-oriented silicon steel of the present invention performs data fitting and modeling through software such as MATLAB, extracts the grain size distribution characteristics, can effectively guide the control of the uniformity of grain size, and further improve the performance stability and consistency of the material. It realizes the prediction of grain distribution based on data-driven, enables the grain size distribution to be directly related to the performance of the material, further guides the process optimization, and then constructs the correlation modeling of process - grain distribution characteristics - performance, improving the flexibility and accuracy of process design and adjustment.

[0020] 3. The method for characterizing the grain distribution characteristics of non-oriented silicon steel of the present invention statistically analyzes the grain size distribution of non-oriented silicon steel under various processes, establishes the corresponding relationship between the process and the grain size, and then constructs the correlation model of process - grain distribution characteristics - performance, realizing the technical effect of reverse inferring the material production process according to the required material performance, which is beneficial to improving the stability and efficiency of the preparation process.

[0021] 4. The method for characterizing the grain distribution characteristics of non-oriented silicon steel of the present invention provides a new method for grain size distribution modeling and optimization. This method has good universality. When the growth of the grain size of the metal is random in all directions, when the modulus length in the two-dimensional plane form is statistically analyzed, its probability density can be fitted by the Rayleigh distribution. Therefore, the Rayleigh distribution model adopted by the present invention can be widely applied to the grain distribution analysis of non-oriented silicon steel produced by different production processes, and the error is controlled within 10%, and even in some cases, the accuracy can reach an error of no more than 2%, providing an effective tool for predicting the material performance under different processes and having high versatility.

[0022] 5. The method for characterizing the grain distribution characteristics of non-oriented silicon steel of the present invention selects the Rayleigh distribution to fit the grain size grade and distribution. When the grain growth of the material is dominated by multi-dimensional random factors, the modulus length statistics of its two-dimensional projection size is more in line with the Rayleigh distribution. Therefore, the Rayleigh distribution is selected to fit the grain size grade and distribution of non-oriented silicon steel produced under different process conditions, and the fitting error is controlled within 10%, and even in some cases, the accuracy can reach an error of no more than 2%. It optimizes the performance of the mathematical model in describing the distribution law after grain refinement of non-oriented silicon steel produced under different process conditions. At the same time, after using the Rayleigh distribution to fit the grain size grade and distribution, the fitting parameters are easier to extract. The mean σ is used to describe the density peak of the grain size distribution, and the variance Var(R) is used to describe the degree of non-uniformity of the distribution, which is convenient for quantifying the grain distribution characteristics. Combined with the correlation model of process - grain distribution characteristics - performance, it effectively solves the technical problem in the related art that it is difficult to accurately quantify the influence of different process parameters on the grain distribution. Description of the Drawings

[0023] Figure 1It is a flowchart of a method for characterizing the grain distribution characteristics of non-oriented silicon steel according to an embodiment of the present invention;

[0024] Figure 2 It is a schematic structural diagram of a system for characterizing the grain distribution characteristics of non-oriented silicon steel according to an embodiment of the present invention;

[0025] Figure 3 It is a grain size distribution diagram of different positions on the RD-TD plane of a cold-rolled sheet in the second embodiment of the present invention. a is the L1 position, b is the L2 position, c is the L3 position, d is the L4 position, and e is the L5 position;

[0026] Figure 4 It is a grain size distribution diagram of different positions on the RD-ND plane of a cold-rolled sheet in the second embodiment of the present invention. a is the L1 position, b is the L2 position, c is the L3 position, d is the L4 position, and e is the L5 position;

[0027] Figure 5 It is a grain size distribution diagram of different positions on the TD-ND plane of a cold-rolled sheet in the second embodiment of the present invention. a is the L1 position, b is the L2 position, c is the L3 position, d is the L4 position, and e is the L5 position;

[0028] Figure 6 It is a fitting diagram of the Rayleigh distribution of the grain size of a cold-rolled sheet in the second embodiment of the present invention. a is the RD-TD plane, b is the RD-ND plane, and c is the TD-ND plane;

[0029] Figure 7 It is a fitting diagram of the Rayleigh distribution of the grain size of a normalized sheet in the third embodiment of the present invention. a is the RD-TD plane, b is the RD-ND plane, and c is the TD-ND plane. Detailed implementation manners

[0030] In order to enable those skilled in the art to better understand the solution of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0031] According to an embodiment of the present invention, a method for characterizing the grain distribution characteristics of non-oriented silicon steel is provided. Please refer to Figure 1 , including:

[0032] S100 Prepare a plurality of samples to be detected from materials produced by a variety of production processes;

[0033] Specifically, during the preparation process of non-oriented electrical steel, different process parameters (such as hot rolling, cold rolling, annealing, etc.) have different effects on the grain distribution. Therefore, it is necessary to detect the materials produced by different production processes to establish the corresponding relationship between the production process and the grain distribution.

[0034] Specifically, the non-oriented electrical steel produced by multiple production processes includes: normalized non-oriented electrical steel, double cold rolled non-oriented electrical steel, continuous annealing non-oriented electrical steel, batch annealing non-oriented electrical steel, thin slab continuous casting and rolling non-oriented electrical steel, and laser scribed non-oriented electrical steel.

[0035] Among them, for the normalized non-oriented electrical steel, after hot rolling and normalizing (normalizing treatment), the grains are refined to 20 - 50 μm, the size distribution is relatively uniform, the grain orientation is closer to random distribution, related to the rolling direction, and a weak {111} texture may appear in some areas; for the double cold rolled non-oriented electrical steel, intermediate annealing promotes recrystallization, and the final grain size after double cold rolling is smaller, basically in the range of 15 - 40 μm. The two rolling paths (such as cross rolling) destroy the initial texture, the grain orientation has higher randomness, and the residual stress is lower; for the continuous annealing non-oriented electrical steel, due to rapid annealing, fine equiaxed grains are formed, and the grain size is usually in the range of 20 - 60 μm, but there may be unrecrystallized areas locally. At the same time, the annealing time is short, and the grain orientation retains part of the cold rolling deformation texture, such as weak {100} or {111}; for the batch annealing non-oriented electrical steel, slow annealing leads to grain coarsening, and the grain size is usually in the range of 50 - 100 μm, with a relatively wide size distribution. At the same time, the annealing time is long, and the grains tend to the stable orientation with the lowest energy, such as {100} or {110}, forming a weak texture locally; for the thin slab continuous casting and rolling non-oriented electrical steel, due to the relatively large initial grain size of the continuous casting thin slab, usually in the range of 50 - 150 μm, it is partially refined to 30 - 80 μm after hot rolling. And because of the short process path, the columnar crystal residues of the original as-cast structure lead to local orientation segregation, such as segregation along the rolling direction; for the laser scribed non-oriented electrical steel, the matrix grain size is the same as the original process (such as the normalized type or double cold rolled type), there is no significant change in the grains in the surface laser treatment area, and the grain orientation remains unchanged. However, the laser scribing introduces local stress and forcibly divides the magnetic domains (refined to a width of <1 mm).

[0036] In summary, different preparation processes of non-oriented silicon steel result in slight differences in both the grain size and grain randomness of non-oriented silicon steel. The resolution of the mathematical models (such as the logarithmic model) in the prior art is relatively low and cannot accurately describe the above-mentioned slight differences. The Rayleigh distribution model has good resolution and is applicable to non-oriented silicon steel produced by the above-mentioned various processes. It can more accurately describe the grain distribution characteristics, achieving a fitting error of less than 10%, and even in some cases, the error can be no more than 2%. At the same time, in this embodiment, samples to be detected are taken from multiple directions of non-oriented silicon steel, providing multi-dimensional crystal data, further improving the accuracy of the description of the grain size distribution data set.

[0037] It can be understood that the grain orientations of non-oriented silicon steel corresponding to various production processes are different, and the grain distributions of non-oriented silicon steel corresponding to various production processes are different. The Rayleigh distribution model is applicable to non-oriented silicon steel produced by various processes, and can achieve a fitting effect with an error of less than 10%, and even less than 2%.

[0038] In this embodiment, the method for preparing multiple samples to be detected includes selecting sampling positions in multiple different directions on the non-oriented silicon steel sheet and cutting the corresponding samples to be detected from the multiple sampling positions.

[0039] In specific practice, the method for preparing samples to be detected includes:

[0040] The observation surface of the non-oriented silicon steel material is polished successively with 400# - 2000# sandpaper, and after polishing, mechanical polishing and vibratory polishing are used for polishing respectively, where the vibratory time is one and a half hours.

[0041] Specifically, the preparation of metallographic samples should include three surfaces: RD-TD, RD-ND, and ND-TD. To ensure the statistical effect, 5 (or more) samples need to be taken from each surface. In this way, sampling from multiple angles can provide multi-dimensional crystallographic information and avoid the one-sidedness of single-section analysis.

[0042] S200 Analyze the multiple samples to be detected to obtain the grain size distribution parameters corresponding to the multiple samples to be detected;

[0043] In specific practice, the method for analyzing the multiple samples to be detected includes:

[0044] S210 Collect data of the samples to be detected through an electron backscatter diffraction system to obtain data to be analyzed, where the data to be analyzed includes the grain size and the corresponding proportion of the grain size;

[0045] S220 Obtain the grain size distribution data set based on the data to be analyzed.

[0046] Specifically, the prepared specimen is placed in the specimen chamber of the EBSD system of the field emission scanning electron microscope to obtain EBSD experimental data, which is imported into the Oxford Crystal software for analysis to obtain the grain size distribution data set.

[0047] Through the above experiments, the grain size distribution of the materials produced by different processes can be quantitatively detected in sequence, and then the correlation model between the process and the grain distribution characteristics can be established.

[0048] S300 obtains the grain grade density based on the grain size distribution parameters;

[0049] Specifically, based on the grain size distribution parameters, the grain size i and the grain size interval i ∈ [i min , i max are obtained, where i min is the minimum grain size and i max is the maximum grain size;

[0050] Specifically, the equivalent circle diameter quantitative statistical method is selected to statistically analyze the grain size;

[0051] The grain size interval is evenly divided into n grade intervals, where n ∈ Z; the grains are classified according to the grade intervals corresponding to the grain size to obtain the grain grade density.

[0052] It should be noted that the Rayleigh distribution can be fitted as long as x is non - negative. The reason for evenly dividing the grain size interval into n is that the grain size distribution within this interval is random, that is, grains of each size exist. Only by statistically counting the number of grains distributed in one interval can quantification be better achieved;

[0053] In specific practice, the percentage of the grain size ratio needs to be converted into a decimal for input into the next fitting model.

[0054] It should be noted that since the proportion of large - sized grains is extremely small, grains with a size greater than 180 μm are ignored in the grain size grade classification.

[0055] Example 1:

[0056] In this example, the grain grade classification method includes:

[0057] The grain size i range is divided into 10 grade intervals, which are: i < 20μm is denoted as 1, 20 ≤ i < 40μm is denoted as 2, 40 ≤ i < 60μm is denoted as 3, 60 ≤ i < 80μm is denoted as 4, 80 ≤ i < 100μm is denoted as 5, 100 ≤ i < 120μm is denoted as 6, 120 ≤ i < 140μm is denoted as 7, 140 ≤ i < 160μm is denoted as 8, 160 ≤ i < 180μm is denoted as 9, and i > 180μm is denoted as 10.

[0058] S400 inputs the grain grade density and the grain distribution parameter into a preset Rayleigh model to obtain a grain size distribution fitting image corresponding to the sample to be detected;

[0059] In specific practice, the preset Rayleigh model is a MATLAB mathematical analysis model established based on the Rayleigh distribution. The fitting error of the preset Rayleigh model does not exceed 10%, and in some cases, the accuracy can even reach an error not exceeding 2%.

[0060] Specifically, the grain size distribution of plates obtained by different processes such as rolling and heat treatment is mathematically analyzed through the preset Rayleigh model, and a fitting result graph of the two is obtained for the grain grade and the proportion (decimal), so that the error is not greater than 10%, and in some cases, the accuracy can even reach an error not exceeding 2%.

[0061] It should be noted that using the Rayleigh distribution for mathematical fitting is not the only option. Other easy-to-describe functions that can represent the grain distribution characteristics can be used for fitting. However, in this embodiment, the non-oriented silicon steel mentioned has the grain size closest to the Rayleigh distribution after processes such as hot rolling, cold rolling, and heat treatment. Therefore, choosing the Rayleigh distribution to describe the grain distribution is more accurate and the characteristic parameter σ is easy to extract, solving the technical problem that the existing mathematical models in the prior art perform poorly in describing the distribution law after grain refinement and it is difficult to accurately quantify the influence of different process parameters on the grain distribution.

[0062] Specifically, the probability density function represented by the Rayleigh distribution is: x ≥ 0;

[0063] where x is the random variable of the Rayleigh distribution, σ is the characteristic parameter, the Rayleigh distribution is unimodal, reaches the maximum value at x = σ, and as x increases, the tail of the Rayleigh distribution decays rapidly (exponential decay). Therefore, the probability of large values appearing is relatively low.

[0064] It should be noted that the grain size in the thickness direction of the non-oriented silicon steel material is restricted during the rolling process. Therefore, the distribution on the normal plane basically conforms to the Rayleigh distribution with σ = 1. Subsequently, during the heat treatment process, the recrystallization process in all directions of the grains is affected by various random factors such as internal defects of the material, temperature effect, and stored energy after deformation. The statistical characteristics of the modulus length of the final grain size distribution on the two-dimensional plane tend to be the Rayleigh distribution. Therefore, the Rayleigh distribution is selected as the probability distribution model in this embodiment.

[0065] Specifically, the Rayleigh distribution parameter σ determines the density peak position of the grain size distribution, and the variance quantifies the non-uniformity of the distribution. This law stems from the coupling effect of the isotropic random mechanism of grain growth and the two-dimensional observation method - the rolling process restricts the normal growth, and during the heat treatment process, the multi-dimensional random factors project onto the two-dimensional plane as independent normal components, and the statistical characteristics of their modulus length (single-peak right-skewed, positive value distribution) naturally fit the Rayleigh distribution. Therefore, the corresponding law can be obtained that when the growth of the grain size of the metal is random in all directions, when the modulus length in the form of a two-dimensional plane is statistically analyzed, its probability density can be fitted by the Rayleigh distribution. At the same time, the mean σ can be used to describe the density peak of the grain size distribution, and the variance Var(R) can be used to describe the degree of non-uniformity of the distribution.

[0066] It should be noted that in this embodiment, the special laws of the grain size distribution of the non-oriented silicon steel include the following points:

[0067] 1. Two-dimensional statistical characteristics: The grain size in this embodiment is measured in the rolling direction - transverse two-dimensional plane by the EBSD system, and the equivalent circle diameter is taken as the statistic, which is essentially a two-dimensional modulus length parameter and is different from the three-dimensional real size. When the modulus length in the form of a two-dimensional plane is statistically analyzed, its probability density is more accurately fitted by the Rayleigh distribution.

[0068] 2. Independence of orthogonal components: The size changes of the grains in the orthogonal directions in this embodiment are dominated by random perturbations such as rolling stress and temperature gradient, forming two independent and identically distributed normal random variables, and the statistical quantity of their modulus length naturally satisfies the Rayleigh distribution, as

[0069] 3. Verification of experimental distribution: The measured data of the grain distribution in this embodiment show a single-peak right-skewed characteristic, and the peak position coincides with the theoretical value of the Rayleigh distribution parameter σ = 1, which is in line with the recrystallization characteristics of the grains after being restricted by the rolling in the thickness direction.

[0070] 4. Correlation with process mechanism: The non-oriented silicon steel in this embodiment restricts the normal grain growth during the rolling process, and the random recrystallization during the heat treatment strengthens the two-dimensional plane statistics: the isotropic superposition of random factors in the orthogonal directions leads to the modulus length statistics approaching the Rayleigh distribution.

[0071] Therefore, when the grain growth of the material is dominated by multi-dimensional random factors, the modulus length statistics of its two-dimensional projection size conform to the Rayleigh distribution, where σ determines the peak position of the distribution, and the variance characterizes the grain uniformity. The Rayleigh distribution model can effectively connect the microscopic random mechanism with the macroscopic statistical characteristics. While the fitting error does not exceed 10%, the characteristic parameter σ is easy to extract, achieving the technical effect of accurately quantifying the grain distribution with the characteristic parameter.

[0072] It should be noted that in this embodiment, the grains of non-oriented silicon steel are magnified and observed in a two-dimensional plane in the field emission scanning electron microscope in the rolling direction - transverse direction - normal direction respectively, and the grain size is obtained through the EBSD system. Among them, the equivalent circle diameter size is taken as the grain size of a certain plane to participate in the statistics. Simply put, the grain size is observed as a modulus length characteristic in a two-dimensional space, rather than the three-dimensional true size. The size changes of the grains in two orthogonal directions (such as the rolling direction and the transverse direction) in the plane are approximately independent, following the isotropic growth mechanism, and are affected by a large number of tiny and independent random factors (such as local temperature fluctuations, stress distributions), which are orthogonal components dominated by randomness and each follow a similar normal distribution. The grain size is quantified as the combined modulus length of the sizes in these two directions. The Rayleigh distribution is the modulus length of two independent and identically distributed normal random variables. Therefore, when judging the grain size distribution in a two-dimensional field of view, it is more in line with the characteristics of the Rayleigh distribution.

[0073] At the same time, all values in the overall statistics of the grain size are positive, and the distribution shows a right skew, which is consistent with the single-peak right-skew characteristic of the Rayleigh distribution. The peak of the Rayleigh distribution is located at R = σ, and the peak of the experimental data coincides with the theoretical fit.

[0074] In summary, due to the isotropic characteristics of the grains of non-oriented silicon steel and the observation method that strengthens the two-dimensional plane statistical characteristics, the grain size distribution characteristics of non-oriented silicon steel conform to the Rayleigh distribution. Therefore, it is concluded that the probability density of non-oriented silicon steel can be fitted by the Rayleigh distribution. At the same time, the mean value σ can be used to describe the density peak of the grain size distribution, and the variance Var(R) can be used to describe the degree of non-uniformity of the distribution. Specifically, Var(R) = σ 2 (4 - π) / 2, achieving the technical effect of accurately quantifying the influence of different processes on the grain distribution of non-oriented silicon steel with parameters.

[0075] S500 Based on the grain size distribution fitting image, obtain characteristic fitting parameters to characterize the grain distribution characteristics of the to-be-detected sample.

[0076] Extract the fitting parameter σ from the grain size distribution fitting image obtained in step S400. See Figures 6 to 7, the fitting parameter σ can be directly extracted from the fitting image of the grain size distribution. At the same time, the fitting parameter σ can intuitively reflect the influence of different processes on the grain size distribution. After repeatedly fitting and extracting the fitting parameters for materials produced by various different processes, a correlation model between the process and the grain distribution characteristics can be established, thereby preparing for subsequent data-driven analysis.

[0077] In this embodiment, the method for characterizing the grain distribution characteristics of non-oriented silicon steel further includes: obtaining the characteristic fitting parameters corresponding to the performance according to the required performance of the material; and obtaining the production process corresponding to the performance according to the characteristic fitting parameters.

[0078] Specifically, the grain size distribution of electrical steel under various processes is statistically analyzed to establish a correlation model between the process and the grain distribution characteristics. At the same time, the grain size is inversely deduced based on the required material properties (such as magnetic properties). According to the relationship of the correlation model between the process and the grain distribution characteristics, the optimal process parameters are obtained, thereby improving the stability and efficiency of the preparation process.

[0079] Example Two:

[0080] In this embodiment, non-oriented silicon steel sheets of 0.2 mm and 0.15 mm produced by various processes are selected, and the chemical compositions are shown in Table 1.

[0081] Table 1 Chemical Compositions of Test Samples (wt.%)

[0082] element C Si Mn P S Als Fe content 0.0017 3.047 0.288 0.011 0.0018 0.8613 balance

[0083] Step 1: Samples are taken from the edge positions near the operating side, the quarter positions, the edge positions near the drive side, the quarter positions, and the middle position (denoted as L1 to L5 respectively) of each non-oriented silicon steel sheet as the test objects. The EBSD observation planes are the RD-TD plane, the RD-ND plane, and the TD-ND plane.

[0084] In specific practice, ultra-thin non-oriented silicon steel needs to go through six processes including hot rolling, normalizing, two cold rolling processes, and two annealing processes. This embodiment takes the single cold-rolled sheet as an example.

[0085] Specifically, the sample size is 12 mm (RD) × 10 mm (TD) × 0.62 mm (ND). Three samples are cut from each steel sheet at different positions. The observation surfaces are polished successively with 400# to 2000# sandpapers, and mechanical polishing and vibratory polishing are used respectively. Among them, the polishing liquid is 0.06 μm Eposal alumina suspension, the frequency of the polishing machine is 85 Hz, the amplitude is 80%, and the vibration time is one and a half hours.

[0086] Step 2: Place the prepared specimen into the specimen chamber of the electron backscatter diffraction system to obtain EBSD experimental data. Import the data into Oxford Crystal software for analysis. Select the equivalent circle diameter for quantitative statistical analysis of the grain size distribution. Refer to Figures 3 to 5 , where Figure 3 a is the grain size distribution at position L1 on the RD-TD plane, Figure 3 b is the grain size distribution at position L2 on the RD-TD plane, Figure 3 c is the grain size distribution at position L3 on the RD-TD plane, Figure 3 d is the grain size distribution at position L4 on the RD-TD plane, Figure 3 e is the grain size distribution at position L5 on the RD-TD plane, Figure 4 a is the grain size distribution at position L1 on the RD-ND plane, Figure 4 b is the grain size distribution at position L2 on the RD-TD plane, Figure 4 c is the grain size distribution at position L3 on the RD-TD plane, Figure 4 d is the grain size distribution at position L4 on the RD-TD plane, Figure 4 e is the grain size distribution at position L5 on the RD-TD plane, Figure 5 a is the grain size distribution at position L1 on the TD-ND plane, Figure 5 b is the grain size distribution at position L2 on the RD-TD plane, Figure 5 c is the grain size distribution at position L3 on the RD-TD plane, Figure 5 d is the grain size distribution at position L4 on the RD-TD plane, Figure 5 e is the grain size distribution at position L5 on the RD-TD plane.

[0087] Step 3: Based on the statistically obtained grain size distribution data, obtain the grain size range and divide the grain size range into 10 intervals. Among them, the grain size i < 20μm is denoted as 1, 20 ≤ i < 40μm is denoted as 2, 40 ≤ i < 60μm is denoted as 3, 60 ≤ i < 80μm is denoted as 4, 80 ≤ i < 100μm is denoted as 5, 100 ≤ i < 120μm is denoted as 6, 120 ≤ i < 140μm is denoted as 7, 140 ≤ i < 160μm is denoted as 8, 160 ≤ i < 180μm is denoted as 9, and i > 180μm is denoted as 10. Then convert the percentage of the grain size in the grain size distribution map into a decimal.

[0088] Step 4: Select a suitable mathematical model. In this embodiment, the Rayleigh distribution model is selected. Extract the five-position data of each EBSD observation plane respectively, and use Matlab software to perform mathematical analysis on the grain size distribution of the 0.62mm cold-rolled plate to make the fitting error not exceed 2%, so as to obtain the grain size distribution fitting diagram of the 0.62mm cold-rolled plate. Refer to Figure 6 .

[0089] From Figure 6 it can be found that the grain size distribution characteristics of the three surfaces of the 0.62 mm cold-rolled sheet are approximately Rayleigh distribution with σ = 1. Therefore, the characteristic parameter σ = 1 of the grain distribution under this process is obtained.

[0090] Example 3:

[0091] Material preparation: Select the non-oriented silicon steel sheet after normalization treatment after hot rolling in the production process of the above non-oriented silicon steel, with a thickness of 2.3 mm. The composition is the same as that of the non-oriented silicon steel sheet in Example 1, as shown in Table 1.

[0092] Step 1: The sample size in this example is 12 mm (RD) × 10 mm (TD) × 2.3 mm (ND). Three samples are cut from each group of steel sheets at different positions. The observation surfaces are polished successively with 400# - 2000# sandpaper, and mechanical polishing and vibration polishing are used respectively. The polishing liquid is selected as Etosi l anhydrous alcohol-based, the frequency of the polishing machine is 85 Hz, the amplitude is 80%, and the vibration time is one and a half hours.

[0093] Put the prepared sample into the sample chamber of the electron backscatter diffraction system of the field emission scanning electron microscope to obtain the EBSD experimental data, and import it into the Oxford Crystal software for analysis. Select the equivalent circle diameter to quantitatively count the grain size distribution.

[0094] Step 2: Divide the grain size range into 10 intervals. Among them, the grain size i < 20 μm is recorded as 1, 20 ≤ i < 40 μm is recorded as 2, 40 ≤ i < 60 μm is recorded as 3, 60 ≤ i < 80 μm is recorded as 4, 80 ≤ i < 100 μm is recorded as 5, 100 ≤ i < 120 μm is recorded as 6, 120 ≤ i < 140 μm is recorded as 7, 140 ≤ i < 160 μm is recorded as 8, 160 ≤ i < 180 μm is recorded as 9, and i > 180 μm is recorded as 10. At the same time, convert the percentage of the grain size in Step 1 into a decimal.

[0095] Step 3: Extract the five-position data of each surface respectively, and use Matlab software to perform mathematical analysis on the grain size distribution of the samples in this example. Select a suitable mathematical model to make its fitting error not greater than 10%, so as to obtain the fitting diagram of the grain size distribution of the normalized plate. See Figure 7 , and the Rayleigh distribution model is selected in this example.

[0096] Step 4: From Figure 7 it can be found that the grain size distribution characteristics of the three surfaces of the 2.3 mm normalized plate are approximately Rayleigh distribution with σ = 3. Therefore, the characteristic parameter σ = 3 of the grain distribution under this process is obtained.

[0097] This embodiment provides a grain distribution characteristic characterization system, and the grain distribution characteristic characterization system includes:

[0098] A sample preparation unit, which is used to prepare a plurality of samples to be detected from materials produced by a variety of production processes;

[0099] A data acquisition unit, which is used to analyze a plurality of the samples to be detected and obtain grain size distribution parameters corresponding to the plurality of samples to be detected;

[0100] A data processing unit, which is used to obtain a grain grade density based on the grain size distribution parameters;

[0101] A fitting unit, which is used to input the grain grade density and the grain distribution parameters into a preset fitting model to obtain a grain size distribution fitting image corresponding to the sample to be detected;

[0102] A characterization unit, which is used to obtain characteristic fitting parameters based on the grain size distribution fitting image to characterize the grain distribution characteristics of the sample to be detected.

[0103] With the above settings, by selecting a mathematical model that conforms to the grain distribution law, classifying the grain size into grades, performing mathematical fitting, and extracting characteristic parameters, it is possible to effectively quantify the grain distribution law under different process conditions, characterize the grain distribution characteristics. At the same time, by statistically analyzing the grain size distribution of electrical steel under multiple processes, a corresponding relationship between the process and the grain size can be established, and then an association model of process - grain distribution characteristic - performance can be constructed, achieving the technical effect of inferring the material production process from the required material performance, which is beneficial to improving the stability and efficiency of the preparation process. The grain distribution characteristic characterization system in this embodiment has good universality and can be widely applied to the grain distribution analysis of different types of metal materials, providing an effective tool for predicting the material performance under different processes, having high generality, and effectively solving the technical problem in the related art that it is difficult to accurately quantify the influence of different process parameters on the grain distribution.

[0104] This embodiment provides a computer-readable storage medium, on which computer-readable instructions are stored. The computer-readable instructions, when executed by a processor, implement each step of the non-oriented silicon steel grain distribution characteristic characterization method as described above.

[0105] Embodiments of the present invention may be implemented in the form of a computer program product implemented on one or more storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing program code. Computer-readable storage media include permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. The information may be computer-readable instructions, data structures, program modules, or other data. Examples of computer-readable storage media include but are not limited to: new types of memories such as phase change memory / resistive random access memory / magnetic random access memory / ferroelectric random access memory (PRAM / RRAM / MRAM / FeRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette tapes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible by a computing device.

[0106] It should be noted that the terms "first", "second", etc. in the specification, claims and above-mentioned drawings of this application are used to distinguish similar objects and do not necessarily have to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present application described here can be implemented in an order other than those illustrated or described here. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0107] Optionally, the specific examples in this embodiment may refer to the examples described in the above embodiments, and will not be elaborated here.

[0108] The serial numbers of the above embodiments of the present application are only for description and do not represent the advantages or disadvantages of the embodiments.

[0109] In the above embodiments of the present application, the descriptions of the various embodiments have their own emphases. For the parts not detailed in a certain embodiment, reference may be made to the relevant descriptions of other embodiments.

[0110] The above is only the preferred embodiment of the present application. It should be pointed out that for those of ordinary skill in the art, without departing from the principle of the present application, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present application.

Claims

1. A method for characterizing the grain distribution characteristics of non-oriented silicon steel, characterized in that, Including: Taking non-oriented silicon steel produced by multiple production processes to prepare multiple samples to be detected; Analyzing the multiple samples to be detected to obtain a grain size distribution data set corresponding to the multiple samples to be detected; Based on the grain size distribution data set, obtaining the grain grade density; Inputting the grain grade density and the grain size distribution data set into a preset Rayleigh model to obtain a grain size distribution fitting image corresponding to the sample to be detected; Based on the grain size distribution fitting image, obtaining characteristic fitting parameters to characterize the grain distribution characteristics of the sample to be detected.

2. The method for characterizing the grain distribution characteristics of non-oriented silicon steel according to claim 1, characterized in that, The non-oriented silicon steel includes normalized non-oriented silicon steel, double cold-rolled non-oriented silicon steel, continuously annealed non-oriented silicon steel, batch annealed non-oriented silicon steel, thin slab continuous casting and rolling non-oriented silicon steel, and laser scribed non-oriented silicon steel.

3. The method for characterizing the grain distribution characteristics of non-oriented silicon steel according to claim 1, wherein, The grain sizes of the multiple non-oriented silicon steels are different; and / or The grain orientations of the multiple non-oriented silicon steels are different.

4. The method for characterizing the grain distribution characteristics of non-oriented silicon steel according to claim 1, wherein The method for preparing the multiple samples to be detected includes selecting sampling positions in multiple different directions on the non-oriented silicon steel sheet and cutting corresponding samples to be detected from the multiple sampling positions.

5. The method for characterizing the grain distribution characteristics of non-oriented silicon steel according to claim 1, wherein, The method for analyzing the multiple samples to be detected includes: Collecting data of the sample to be detected through an electron backscatter diffraction system to obtain data to be analyzed, where the data to be analyzed includes the grain size and the corresponding proportion of the grain size; Based on the data to be analyzed, obtaining the grain size distribution data set.

6. The method for characterizing the grain distribution characteristics of non-oriented silicon steel according to claim 5, wherein The method for obtaining the grain grade density includes: Based on the grain size, obtaining a grain size interval; Dividing the grain size interval evenly into n grade intervals, where n ∈ Z; Classifying the grains according to the grade interval corresponding to the grain size to obtain the grain grade density.

7. The method for characterizing the grain distribution characteristics of non-oriented silicon steel according to claim 1, characterized in that The method for characterizing the grain distribution characteristics of the non-oriented silicon steel further includes: Sequentially detecting the samples to be detected supported by the non-oriented silicon steel produced by the multiple production processes to obtain the corresponding characteristic fitting parameters; Based on the production process corresponding to the sample to be detected and the characteristic fitting parameters of the sample to be detected, establishing a process-grain distribution characteristic correlation model.

8. The method for characterizing the grain distribution characteristics of non-oriented silicon steel according to claim 7, characterized in that The method for characterizing the grain distribution characteristics of the non-oriented silicon steel further includes: According to the required performance of the metal material, obtaining the characteristic fitting parameters corresponding to the performance of the metal material; According to the process-grain distribution characteristic correlation model, obtaining the production process corresponding to the performance of the metal material.

9. An unoriented silicon steel grain distribution characteristic characterization system, characterized in that, The grain distribution characteristic characterization system includes: A sample preparation unit for preparing multiple samples to be detected from non-oriented silicon steel sheets produced by multiple production processes; A data collection unit for analyzing the multiple samples to be detected to obtain a grain size distribution data set corresponding to the multiple samples to be detected; A data processing unit for obtaining the grain grade density based on the grain size distribution parameters; A fitting unit for inputting the grain grade density and the grain distribution parameters into a preset Rayleigh model to obtain a grain size distribution fitting image corresponding to the sample to be detected; A characterization unit, which is used to fit an image based on the grain size distribution to obtain characteristic fitting parameters to characterize the grain distribution characteristics of the sample to be detected.

10. A computer-readable storage medium, on which computer-readable instructions are stored, characterized in that, When the computer-readable instructions are executed by a processor, each step of the method for characterizing the grain distribution characteristics of non-oriented silicon steel as described in any one of claims 1-8 is implemented.

Images