Key Smelting Data Processing Method, Device and Storage Medium for Inclusions in Steel

Abstract

The present application relates to a method, device and storage medium for processing key smelting data of inclusions in steel. The method comprises: collecting steelmaking process data, inclusion detection result data and first target smelting data of a target steel grade; if the second target smelting data satisfies a normality test, extracting a data concentration area, determining the step length of the data in the concentration area, and grouping the target data in the concentration area according to the step length; fitting the number of groups for each smelting data item, and performing a preset variance analysis verification on the inclusion detection result, importing the inclusion detection result data, target smelting data and number of groups that meet the verification into a preset data processing software to form visual data, extracting the numerical range where the molten steel impurities exceed the preset value based on the visual data, and optimizing the steelmaking process. This scheme can find the factor that has the greatest impact on the cleanliness of the molten steel in a large amount of smelting data, and then quantitatively optimize the existing process in a targeted manner, thereby improving the purity of the steelmaking process.

Description

Technical Field

[0001] The present application relates to a method, device and storage medium for processing key smelting data of inclusions in steel, and belongs to the technical field of data processing in the steelmaking process of a steel furnace. Background Art

[0002] Inclusions formed during the steelmaking process due to deoxidation and alloying, refractory erosion, slag inclusion, and secondary oxidation of molten steel will affect the cleanliness of the molten steel. Their quantity, size, type, and distribution will have varying degrees of impact on the quality of steel products. Therefore, effectively controlling inclusions in steel is the key to improving the quality of steel products.

[0003] Controlling inclusions in steel has always been a major challenge in the metallurgical industry. Current methods for removing inclusions include ladle bottom-blowing gas agitation, ladle electromagnetic agitation, tundish air curtain walls, mold electromagnetic braking, slag washing, and filters. This multitude of inclusion control measures only highlights the difficulty of inclusion removal. Ladle bottom-blowing gas agitation is the simplest and most effective inclusion control measure in the industry. It plays a key role in inclusion removal and is recognized as the optimal solution. Almost all steel mills are equipped with ladle bottom-blowing equipment.

[0004] However, changes in the steelmaking process and equipment can easily lead to ripple effects, creating numerous process loopholes. To ensure smooth production, factories are reluctant to change their process flows and equipment. Therefore, analyzing the relationship between steelmaking process parameters and inclusions, and then adjusting the process flows and equipment conditions based on the relationships between the analyzed data, is extremely valuable. The steelmaking process is subject to various constraints, and excessive inclusions can be caused by dozens or even hundreds of process parameters. These process parameters, in turn, influence each other and have varying degrees of impact on the final result. Finding the key factors that have the greatest impact on molten steel cleanliness within this vast amount of smelting data, and then targeted quantitative optimization of existing processes, has become a pressing issue. Summary of the Invention

[0005] The present application provides a method, device and storage medium for processing key smelting data of inclusions in steel, in order to solve the problem in the prior art of "being unable to find the key factors that have the greatest impact on the cleanliness of molten steel in these massive amounts of smelting data, and then to quantitatively optimize the existing process in a targeted manner."

[0006] To solve the above technical problems, this application provides the following technical solutions:

[0007] In a first aspect, an embodiment of the present application provides a method for processing key smelting data of inclusions in steel, wherein the method comprises:

[0008] Collect steelmaking process data, inclusion detection result data and first target smelting data (key smelting data) of the target steel grade within a preset time period;

[0009] If the second target smelting data that meets the target screening condition meets the preset normality test, extracting the second target smelting data concentrated area, determining the step length of the second target data in the second target smelting data concentrated area, and grouping the second target data in the target smelting data concentrated area according to the step length;

[0010] After performing Gauss Amp fitting for each number of groups of second target smelting data and performing preset variance analysis verification on the inclusion detection results, the inclusion detection result data, target smelting data, and number of groups that meet the preset variance analysis verification are stored in a preset format and imported into a preset data processing software to form visual data. The numerical range where the steel melt impurities exceed the preset value is then extracted based on the visual data to optimize the steelmaking process;

[0011] The first target smelting data includes the second target smelting data.

[0012] In one embodiment, the steelmaking process data includes molten iron data, converter data, refining data, continuous casting data, ladle data, and team data;

[0013] The inclusion detection result data includes the maximum inclusion size, the number of inclusions per unit area, and the calculated inclusion deduction points.

[0014] In one embodiment, the method further includes:

[0015] Screening the first target smelting data to determine the second target smelting data that meets the target screening condition;

[0016] Among them, the first target smelting data includes the average refining bottom blowing soft stirring pressure P and the interval time T from the start of soft stirring to the start of continuous casting; the second target smelting data is the average refining bottom blowing soft stirring pressure P with a numerical range of 1.0 to 6.0 bar and the interval time T from the start of soft stirring to the start of continuous casting with a numerical range of 10 to 60 minutes; and the amount of data included in the second target smelting data and the amount of data included in the interval time T from the start of soft stirring to the start of continuous casting are both greater than 300.

[0017] In one embodiment, the second target smelting data concentration area includes:

[0018] The data concentration area of the average pressure of refining bottom blowing soft stirring is (μ p -k p σ p , μ p +k p σp ), and the data concentration area of the interval from the start of soft stirring to the start of continuous casting is (μ t -k t σ t , μ t +k t σ t );

[0019] Among them, μ p is the average value of P, σ p is the standard deviation of P, k p =0~5;μ t is the average value of T, σ t is the standard deviation of T, k t =0~5;

[0020] In one embodiment, the method further includes:

[0021] Determine whether there is a preset correspondence between the steelmaking process data of the target steel grade within a preset time period and the inclusion detection result data;

[0022] If so, a preset change is made to the steelmaking process information represented by the steelmaking process data of the target steel grade within the preset time period;

[0023] Otherwise, a preset data fitting scheme is adopted to fit the steelmaking process data and inclusion detection result data of the target steel grade within a preset time period in a preset manner, determine the matching degree of the steelmaking process data and the inclusion detection result data, and sort the steelmaking process data and inclusion detection result data according to the matching degree.

[0024] In one embodiment, the preset variance analysis verification includes:

[0025] The sum of squares SSA, degrees of freedom d1, mean square, F value, and significance value sig of component data; the sum of squares SSE, degrees of freedom d2, and mean square MSE of within-group data; the sum of squares SST and degrees of freedom d3 of total data.

[0026] In a second aspect, according to an embodiment of the present application, a device for processing key smelting data of inclusions in steel is provided, comprising:

[0027] Collect steelmaking process data, inclusion detection result data and first target smelting data of the target steel grade within a preset time period;

[0028] If the second target smelting data that meets the target screening condition meets the preset normality test, extracting the second target smelting data concentrated area, determining the step length of the second target data in the second target smelting data concentrated area, and grouping the second target data in the target smelting data concentrated area according to the step length;

[0029] After performing Gauss Amp fitting for each number of groups of second target smelting data and performing preset variance analysis verification on the inclusion detection results, the inclusion detection result data, target smelting data, and number of groups (data volume) that meet the preset variance analysis verification are stored in a preset format and imported into a preset data processing software to form visual data. The numerical range where the steel melt impurities exceed the preset value is then extracted based on the visual data to optimize the steelmaking process;

[0030] The first target smelting data includes a larger amount of data than the second target smelting data.

[0031] In one embodiment, the apparatus further comprises:

[0032] A screening module, configured to screen the first target smelting data and determine the second target smelting data that meets the target screening conditions;

[0033] Among them, the first target smelting data includes the average refining bottom blowing soft stirring pressure P and the interval time T from the start of soft stirring to the start of continuous casting; the second target smelting data is the average refining bottom blowing soft stirring pressure P with a numerical range of 1.0 to 6.0 bar and the interval time T from the start of soft stirring to the start of continuous casting with a numerical range of 10 to 60 minutes; and the amount of data included in the second target smelting data and the amount of data included in the interval time T from the start of soft stirring to the start of continuous casting are both greater than 300.

[0034] In a third aspect, according to an embodiment of the present application, a device for processing key smelting data of inclusions in steel is provided, the device comprising a processor, a memory, and a computer program stored in the memory and runnable on the processor, wherein the computer program is loaded and executed by the processor to implement the steps of any of the above-mentioned methods for processing key smelting data of inclusions in steel.

[0035] In a fourth aspect, according to an embodiment of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, wherein when the computer program is executed by a processor, it is used to implement the steps of the key metallurgical data processing method for inclusions in steel described in any one of the above items.

[0036] The beneficial effects of this application are:

[0037] The embodiment of the present application provides a method for processing key smelting data of inclusions in steel. The method collects steelmaking process data, inclusion detection result data, and first target smelting data of a target steel grade within a preset time period. Then, if the second target smelting data that meets the target screening condition meets the preset normality test, a central area of the second target smelting data is extracted, a step size of the second target data in the central area of the second target smelting data is determined, and the second target data in the central area of the target smelting data are grouped according to the step size. Finally, Gauss Amp fitting is performed on the number of groups of each type of second target smelting data, and a preset variance analysis is performed on the inclusion detection results. The inclusion detection result data, target smelting data, and the number of groups (data volume) that meet the preset variance analysis verification are stored in a preset format and imported into preset data processing software to form visual data. The numerical range in which the molten steel impurities exceed the preset value is extracted based on the visual data to optimize the steelmaking process. The number of first target smelting data is not less than the number of second target smelting data. This solution can identify the key factors that have the greatest impact on the cleanliness of molten steel from massive amounts of smelting data, and then quantitatively optimize existing processes in a targeted manner, effectively improving the purity of the steelmaking process.

[0038] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application and to implement it in accordance with the contents of the specification, the following is a detailed description of the preferred embodiments of the present application in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 、 Figure 2 They are respectively flow charts of a method for processing key smelting data of inclusions in steel in one embodiment of the present application;

[0040] Figure 3 This is a schematic diagram of the normal distribution verification of the average pressure P of the refining bottom blowing soft stirring in one embodiment of the present application;

[0041] Figure 4 This is a schematic diagram of the normal distribution verification of the interval time T between the start of soft stirring and the start of continuous casting in one embodiment of the present application;

[0042] Figure 5 This is a schematic diagram of a Gaussian fitting normality test of the grouped data of the average pressure P of the refining bottom blowing soft stirring in one embodiment of the present application;

[0043] Figure 6 Schematic diagram of Gaussian fitting normality test of grouped data of the interval time T from the start of soft stirring to the start of continuous casting in one embodiment of the present application;

[0044] Figure 7 This is a three-dimensional matrix diagram of the data volume (number of furnaces) distribution in one embodiment of this application.mn Schematic diagram;

[0045] Figure 8 This is a three-dimensional matrix diagram of the distribution of inclusion detection exceeding standard rate in one embodiment of the present application. mn ;

[0046] Figure 9 This is a schematic diagram of a key smelting data processing device for inclusions in steel provided in one embodiment of the present application;

[0047] Figure 10 This is a block diagram of a key smelting data processing device for inclusions in steel provided in one embodiment of the present application. DETAILED DESCRIPTION

[0048] The following examples are used to illustrate the present application but are not intended to limit the scope of the present application.

[0049] Changes in steelmaking processes and equipment can easily lead to numerous process vulnerabilities. Analyzing the relationship between process parameters and inclusions in molten steel is crucial. However, analyzing these complex and lengthy steelmaking process parameters presents a daunting challenge. Therefore, quantitatively determining the impact of a single factor on inclusions in molten steel is difficult in current production. Existing clean steel production technologies are not standardized across different steel grades, operating conditions, and process conditions, making widespread adoption extremely challenging.

[0050] To address the above-mentioned issues, the present invention provides a universal steelmaking process parameter analysis method. Utilizing steelmaking process data from steelmaking production, without changing existing process flow and equipment conditions, the method summarizes, refines, processes and analyzes the target smelting data. By adopting variance analysis and creating a three-dimensional matrix, the method establishes a relationship between the target smelting data and inclusion detection information, thereby quantitatively analyzing the parameter range that is most favorable for inclusion removal and achieving control of excessive inclusions in steel. This solution, based on actual steelmaking data, provides guidance for improving the cleanliness of molten steel.

[0051] This application embodiment provides a method for processing key smelting data of inclusions in steel. Figure 1 As shown, the method includes:

[0052] Step S12: collecting steelmaking process data, inclusion detection result data, and first target smelting data of the target steel grade within a preset time period;

[0053] The technical solution provided in the embodiments of the present application can be used to analyze the steelmaking process of one or more target steel grades.

[0054] In an embodiment of the present application, the steelmaking process data collected for the target steel grade within a preset time period may include but is not limited to molten iron data, converter data, refining data, continuous casting data, ladle data, and team data; and the inclusion detection result data of the target molten steel within the preset time period can be divided into the maximum inclusion size, the number of inclusions per unit area, the calculated inclusion deduction points, etc. according to different statistical methods.

[0055] In the embodiment of the present application, the first target data includes the average refining bottom blowing soft stirring pressure P of the target steel grade within a preset time period and the interval time T from the start of soft stirring to the start of continuous casting.

[0056] Step S14: if the second target smelting data that meets the target screening condition meets the preset normality test, extracting the second target smelting data concentrated area, determining the step length of the second target data in the second target smelting data concentrated area, and grouping the second target data in the target smelting data concentrated area according to the step length;

[0057] In an embodiment of the present application, the second target smelting data is first screened, and then the second target smelting data that meets the target screening conditions is determined, wherein the target screening conditions are the numerical range of the average refining bottom blowing soft stirring pressure P in the first target data and the number (data volume) of the average refining bottom blowing soft stirring pressure P in the target time period, and the numerical range of the interval time T from the start of soft stirring to the start of continuous casting and the number (data volume) of the interval time T from the start of soft stirring to the start of continuous casting in the target time period.

[0058] In an embodiment of the present application, after the second target data is screened out, a normality test is performed on the second target data and the data center is extracted. For example, a normality test can be performed on the values of the average refining bottom blowing soft stirring pressure P and the average refining bottom blowing soft stirring pressure T in the target time period using a PP chart with the help of IBM SPSS Statistics 21.0 software.

[0059] In the embodiment of the present application, if the normality test is passed, the second target smelting data concentration area (μ p -k p σ p , μ p +k p σ p ), determine the step size of the second target data in the second target smelting data concentration area, and group the second target data in the target smelting data concentration area according to the step size; as an optional embodiment, the step size of the average pressure P of the refining bottom blowing soft stirring is ΔP=0.05~0.50bar, the step size of the interval time T from the start of soft stirring to the start of continuous casting is ΔT=1.0~10min, and the number of P groups is defined as m, then

[0060] The number of T groups is defined as n, then The number of groups is related to the selection of interval step length. If the number of groups divided by the average pressure P of refining bottom blowing soft stirring is set to m, then

[0061] Assume that the number of groups of the interval time T from the start of soft stirring to the start of continuous casting is n, then The number of groups is related to the selection of the interval step size, which can be found in Table 1 below:

[0062] Table 1 Schematic diagram of the average pressure P of refining bottom blowing soft stirring and the interval time T from the start of soft stirring to the start of continuous casting

[0063]

[0064]

[0065] Step S16: After performing Gauss Amp fitting on the number of groups of each data item in the second target smelting data and performing a preset variance analysis verification on the inclusion detection results, the inclusion detection result data, the target smelting data, and the number of groups (data volume) that meet the preset variance analysis verification are stored in a preset format and imported into a preset data processing software to form visual data. The numerical range where the molten steel impurities exceed the preset value is then extracted based on the visual data to optimize the steelmaking process.

[0066] The number of the first target smelting data is not less than the number of the second target smelting data, and the set formed by the first target smelting data includes the combination formed by the second target data.

[0067] In an embodiment of the present application, after grouping the average pressure P of the refining bottom blowing soft stirring and the time interval T from the start of soft stirring to the start of continuous casting, Gaussian fitting (Gauss Amp fitting) is performed on the number m of groups of the average pressure P of the refining bottom blowing soft stirring and the number n of groups of the time interval T from the start of soft stirring to the start of continuous casting, and the inclusion detection result data corresponding to the second target data in each group is subjected to a preset variance analysis verification, such as ANOVA variance analysis verification. If the ANOVA variance analysis verification is passed, the key smelting data and the data volume (number of smelting furnaces) and the inclusion detection results can be made into a three-dimensional matrix table. The m and n that have passed the variance test are arranged in order from small to large to form a matrix, and the amount of data corresponding to each point in the matrix is calculated, where m is the number of groups of the average pressure P of the refining bottom blowing soft stirring, and n is the number of groups of the time interval T from the start of soft stirring to the start of continuous casting. Any point in the matrix (m i , n j ) are all regarded as a set of data, where i = 1, 2, 3 ... m; j = 1, 2, 3 ... n; m i Indicates the number of groups i in m groups, nj Indicates the number of groups j in n groups, (m i , n j ) form a two-dimensional point, such as the points in Table 7 and Table 8. Next, calculate the inclusion detection result data of each point in the matrix, which can be the average size, number, penalty of the inclusion detection result, or the proportion exceeding a certain threshold. Form a three-dimensional matrix distribution L of data volume (number of furnaces) respectively mn And the three-dimensional matrix distribution of inclusion detection information I mn The calculated three-dimensional matrix distribution can be imported into data processing software including Origin, Excel, etc. to form visual data. mn with I mn By superimposing and verifying, in the area where the number of furnaces is concentrated, the numerical range of the area where the steel liquid impurities exceed the preset value is determined from a quantitative perspective, and the data is fed back to the second target smelting data to provide direction for process optimization.

[0068] In the embodiment of the present application, the steelmaking process data can be divided into molten iron data, converter data, refining data, continuous casting data, ladle data and team data according to different statistical methods;

[0069] The inclusion detection result data includes the maximum inclusion size, the number of inclusions per unit area, and the calculated inclusion deduction points.

[0070] In the examples of this application, see Figure 2 As shown, the method further includes:

[0071] Step S13: Screening the first target smelting data to determine the second target smelting data that meets the target screening conditions;

[0072] Among them, the first target smelting data includes the average refining bottom blowing soft stirring pressure P and the interval time T from the start of soft stirring to the start of continuous casting; the second target smelting data is the average refining bottom blowing soft stirring pressure P with a numerical range of 1.0 to 6.0 bar and the interval time T from the start of soft stirring to the start of continuous casting with a numerical range of 10 to 60 minutes; and the amount of data included in the second target smelting data and the amount of data included in the interval time T from the start of soft stirring to the start of continuous casting are both greater than 300.

[0073] In the embodiment of the present application, the second target smelting data concentration area includes:

[0074] The data concentration area of the average pressure of refining bottom blowing soft stirring is (μ p -k p σ p , μ p +k p σ p), and the data concentration area of the interval from the start of soft stirring to the start of continuous casting is (μ t -k t σ t , μ t +k t σ t );

[0075] Among them, μ p is the average value of P, σ p is the standard deviation of P, k p =0~5;μ t is the average value of T, σ t is the standard deviation of T, k t =0~5;

[0076] In an embodiment of the present application, the method further includes:

[0077] 1) Determine whether there is a preset corresponding relationship between the steelmaking process data of the target steel grade within a preset time period and the inclusion detection result data;

[0078] 2) If so, performing a preset change to the steelmaking process information represented by the steelmaking process data of the target steel grade within a preset time period;

[0079] 3) Otherwise, a preset data fitting scheme is adopted to fit the steelmaking process data and inclusion detection result data of the target steel grade within a preset time period in a preset manner, determine the matching degree of the steelmaking process data and the inclusion detection result data, and sort the steelmaking process data and inclusion detection result data according to the matching degree.

[0080] After that, if there is a preset corresponding relationship between the steelmaking process data in the detection result data, there is no need to perform the subsequent steps S12, S14, and S16, and only the process needs to be changed. If there is no preset corresponding relationship, the fitting process is executed again, and whether to execute the first target data, the second target data extraction and subsequent processes are determined based on the matching degree. As an optional embodiment, when it is determined that the fitting degree is lower than the set threshold, the steps S12, S14, and S16 are executed.

[0081] It should be pointed out here that the embodiments of the present application are only explained using the average pressure P of refining bottom blowing soft stirring and the interval time T from the start of soft stirring to the start of continuous casting as examples, and it does not limit the key smelting data to the average pressure P of refining bottom blowing soft stirring and the interval time T from the start of soft stirring to the start of continuous casting or one of the two.

[0082] In the embodiment of the present application, the preset variance analysis verification includes:

[0083] The sum of squares of deviations SSA, degrees of freedom d1, mean square, F value, and significance value sig of the inter-group data; the sum of squares of deviations SSE, degrees of freedom d2, and mean square MSE of the intra-group data; the sum of squares of deviations SST and degrees of freedom d3 of the total data.

[0084] The calculation method of the sum of squares of the deviations of the total data is:

[0085]

[0086] The calculation method for the sum of squares of the deviations between groups of data is:

[0087]

[0088] The calculation method for the sum of squares of the deviations of data within a group is:

[0089]

[0090] The degree of freedom d1 of the between-group data is calculated as:

[0091] d1=k-1

[0092] The calculation method of the degree of freedom d2 of the within-group data is:

[0093] d2=nk

[0094] The degree of freedom d3 of the total data is calculated as:

[0095] d3=n-1

[0096] The calculation method of the mean square MSA of the between-group data is:

[0097]

[0098] The calculation method of the mean square MSE of the within-group data is:

[0099]

[0100] The calculation method of the F value of the between-group data is:

[0101]

[0102] A specific embodiment is given below for illustration:

[0103] As a specific solution, we used the clean steel production process data and inclusion detection results of a certain plant for more than one year. The steelmaking process is: converter - LF refining - large square billet continuous casting. The inclusion detection results are counted according to the detected size. In this example, if the maximum size of the detected inclusion exceeds 35μm, the furnace is considered unqualified, and the excess rate is used as the I mn The results are shown.

[0104] After removing abnormal information, the total data volume is shown in Table 2:

[0105] Table 2 Amount of information data on the clean steel process of a certain factory

[0106]

[0107] The soft stirring process pressure data is missing due to system reasons, but as long as the normality test is passed and the randomness of the data is confirmed, the data is considered valid. Next, the process data is matched one by one with the inclusion detection results data. Taking the converter team as an example, the results are shown in Table 3:

[0108] Table 3 Converter team and inclusion exceeding standard

[0109]

[0110] Table 3 shows that the differences in inclusion exceedance rates between the three teams are almost negligible, indicating that the variable "converter team" has little correlation with inclusion detection. Analyzing the various process parameters using this method revealed varying degrees of difference, but generally speaking, no single parameter plays a decisive role in the inclusion detection data.

[0111] Next, the key smelting parameters are summarized and refined.

[0112] First, the data was screened to remove outliers and confirm the data volume. Each heat of steel provides three data points: P, T, and inclusion test results. After removing outliers such as negative values, the remaining data volume is 1,339 heats, meeting the requirement of a data volume greater than 300. In addition, outliers outside the P and T ranges were excluded from the statistics.

[0113] Then, a normality test is performed on the data P and T. The present invention uses IBM SPSS Statistics 21.0 software to perform a normality test on P and T. The PP graph test is selected as the test method. The actual data cumulative probability is used as the X-axis, and the corresponding normal distribution cumulative probability is used as the Y-axis. The PP graph reflects the degree of conformity between the actual cumulative probability of the variable and the theoretical cumulative probability. If the data obeys the normal distribution, the data points should basically coincide with the diagonal. Figure 3 Figure 4 All data showed that they were in accordance with normal distribution.

[0114] After passing the normality test, P and T need to be grouped. First, calculate the mean values of P and T respectively, and use formula (2) to calculate the standard deviation σ and standard deviation coefficient k p Take 1.5, k tTake 2.0, the interval step length ΔP is 0.10 bar, and ΔT is 3.4 minutes. The specific grouping results are shown in Table 4:

[0115] Table 4 P, T group data composition

[0116]

[0117] Since grouping breaks the continuity of the original data, it is necessary to further verify the grouped data. First, verify the number of groups and use the Gauss Amp function to fit the number of groups. The results are shown in the attached figure. Figure 5 、 Figure 6 As shown, the expression of Gauss Amp function is as follows.

[0118]

[0119] It can be seen that the expressions of Gauss Amp function and normal function are basically the same, the only difference is that the coefficients in front are inconsistent. Figure 7 、 Figure 8 It can be seen that the number of groups of P and T is basically consistent with the normal distribution. It should be noted that the normal distribution requires Such coefficients are to ensure that the integral in the interval (-∞, +∞) is 1. In addition, it can be seen that The integral in the interval (-∞, +∞) is

[0120] After the quantity distribution of each group of P and T meets the conditions, the next step is to perform ANOVA analysis of variance on the actual meaning represented by the group, that is, the maximum size of the inclusions, to further confirm whether the means of each group are consistent. In layman's terms, it is to verify whether the grouping is sufficient to distinguish the differences in the data. The F value in the variance analysis can be obtained through a simple calculation. The next step is to check the F critical value table and compare the size of F and the critical value Fα at the two significance levels of α=0.05 and ɑ=0.01. If F>Fα, it is considered that there is a significant difference in the means between the groups, which is verified by variance. In addition, another way to process the F value is to directly calculate the corresponding sig value. The calculation of the sig value can be done with the help of the FDIST function in EXCEL (returning the degrees of freedom of the right-tailed F probability distribution of two sets of data). X is the F value, Deg_freedom1=d1, Deg_freedom2=d2. That is:

[0121] sig=FDIST(X,Deg freedom1 , Deg freedom1 )

[0122] The variance verification results are shown in Table 5 and Table 6:

[0123] Table 5ANOVA analysis of variance table of inclusion detection results of P

[0124]

[0125] Table 6 ANOVA analysis of variance table of T inclusion detection results

[0126]

[0127] It can be seen that the calculated significance sig<0.05, which means that the variance analysis at the α=0.05 level has passed. In addition, by querying the F critical value table, the same conclusion can be drawn, that is, the variance test has passed.

[0128] After passing the variance test, it is necessary to make a three-dimensional data matrix L for P and T respectively mn and I mn In this example, I mn The exceedance rate of each point in the matrix is used as the display result. Similarly, the average size of each point can also be used. The results are shown in Table 7 and Table 8:

[0129] Table 7 Data volume (number of furnaces) distribution three-dimensional matrix L mn

[0130]

[0131] Table 8 Three-dimensional matrix I of the inclusion detection exceeding standard rate distribution mn

[0132]

[0133]

[0134] Import the 3D matrix data into Origin to visualize the data. The final result is as follows Figure 7 、 Figure 8 As shown by Figure 7 It can be seen that as the P value decreases and the T value decreases, the overall inclusion exceeding rate increases, forming a clear distinction between the lower left and upper right areas, and in the heat with P < 2.27 bar and T < 34.8 min, the overall exceeding rate is significantly higher than that in other areas. Figure 7 、 Figure 8 As can be seen, within the data center, the areas with P > 2.72 bar, P < 2.27 bar, and T < 34.8 min exhibited a high overall exceedance rate. This provides a threshold reference for on-site process execution, indicating that the optimal range for ladle refining soft stirring and bottom blowing pressure is 2.27 to 2.72 bar, and the interval between ladle refining soft stirring and pouring must not be less than 35 minutes. This data can then be used to improve process data to ensure fewer inclusions in the steel and, therefore, the purity of the steelmaking process.

[0135] In summary, the key smelting data processing method for inclusions in steel provided in the embodiment of the present application collects steelmaking process data, inclusion detection result data and first target smelting data of the target steel grade within a preset time period; then, if the second target smelting data that meets the target screening condition meets the preset normality test, the second target smelting data central area is extracted, the step size of the second target data in the second target smelting data central area is determined, and the second target data in the target smelting data central area is grouped according to the step size; finally, Gauss Amp fitting is performed on the number of groups of each second target smelting data, and a preset variance analysis is performed on the inclusion detection result. After that, the inclusion detection result data, target smelting data and the number of groups (data volume) that meet the preset variance analysis verification are stored in a preset format and imported into a preset data processing software to form visual data, and then the numerical range where the steel liquid impurities exceed the preset value is extracted according to the visual data for optimization of the steelmaking process; wherein the number of the first target smelting data is not less than the number of the second target smelting data. This solution can identify the key factors that have the greatest impact on the cleanliness of molten steel from massive amounts of smelting data, and then quantitatively optimize existing processes in a targeted manner, effectively improving the purity of the steelmaking process.

[0136] Example 2

[0137] The present application also provides a device for processing key smelting data of inclusions in steel. Figure 9 As shown, including:

[0138] The collection module 91 is used to collect steelmaking process data, inclusion detection result data and first target smelting data of the target steel grade within a preset time period;

[0139] a grouping module 92 for extracting a second target smelting data concentrated area, determining a step length of the second target data in the second target smelting data concentrated area, and grouping the second target data in the target smelting data concentrated area according to the step length, if the second target smelting data that meets the target screening condition meets a preset normality test;

[0140] The exceeding-standard data determination module 93 is configured to perform Gauss Amp fitting for each number of groups of second target smelting data, and perform a preset variance analysis verification on the inclusion detection results. The module then stores the inclusion detection result data, target smelting data, and number of groups that meet the preset variance analysis verification in a preset format, imports the data into a preset data processing software to form visual data, and then extracts the numerical range where the molten steel impurities exceed the preset value based on the visual data, so as to optimize the steelmaking process.

[0141] The first target smelting data includes a larger amount of data than the second target smelting data.

[0142] In one embodiment, the device for processing key smelting data of inclusions in steel further includes:

[0143] A screening module, configured to screen the first target smelting data and determine the second target smelting data that meets the target screening conditions;

[0144] Among them, the first target smelting data includes the average refining bottom blowing soft stirring pressure P and the interval time T from the start of soft stirring to the start of continuous casting; the second target smelting data is the average refining bottom blowing soft stirring pressure P with a numerical range of 1.0 to 6.0 bar and the interval time T from the start of soft stirring to the start of continuous casting with a numerical range of 10 to 60 minutes; and the amount of data included in the second target smelting data and the amount of data included in the interval time T from the start of soft stirring to the start of continuous casting are both greater than 300.

[0145] Figure 10 This is a block diagram of a device for processing key smelting data of steel inclusions provided in one embodiment of the present application. The device for processing key smelting data of steel inclusions described in this embodiment can be a computing device such as a desktop computer, a laptop computer, a PDA, or a cloud server. The device may include, but is not limited to, a processor and a memory. The device for processing key smelting data of steel inclusions described in this embodiment includes at least a processor and a memory. The memory stores a computer program. The computer program can be run on the processor. When the processor executes the computer program, the steps of the above-mentioned method for processing key smelting data of steel inclusions are implemented, for example, Figure 1 Or the steps of the method for processing key smelting data of steel inclusions shown in 2. Alternatively, when the processor executes the computer program, the functions of each module in the embodiment of the device for processing key smelting data of steel inclusions are realized.

[0146] Exemplarily, the computer program can be divided into one or more modules, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules can be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program in the key smelting data processing device for steel inclusions. For example, the computer program can be divided into a collection module, a grouping module, and an over-standard data determination module. The specific functions of each module are as follows:

[0147] A collection module is used to collect steelmaking process data, inclusion detection result data and first target smelting data of the target steel grade within a preset time period;

[0148] a grouping module, configured to extract a second target smelting data concentrated area if the second target smelting data that meets the target screening condition meets a preset normality test, determine a step length of the second target data in the second target smelting data concentrated area, and group the second target data in the target smelting data concentrated area according to the step length;

[0149] The module for determining excess data is used to perform Gauss Amp fitting for each number of groups of second target smelting data, and to perform preset variance analysis verification on the inclusion detection results. The module then stores the inclusion detection result data, target smelting data, and number of groups that meet the preset variance analysis verification in a preset format and imports them into preset data processing software to form visual data. The module then extracts the numerical range where the molten steel impurities exceed the preset value based on the visual data to optimize the steelmaking process.

[0150] The first target smelting data includes a larger amount of data than the second target smelting data.

[0151] The processor may include one or more processing cores, such as a 4-core processor, a 6-core processor, etc. The processor may be implemented in at least one hardware form of DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), or PLA (Programmable Logic Array). The processor may also include a main processor and a coprocessor. The main processor is a processor for processing data in the awake state, also known as a CPU (Central Processing Unit); the coprocessor is a low-power processor for processing data in the standby state. In some embodiments, the processor may also include an AI (Artificial Intelligence) processor, which is used to process computing operations related to machine learning. The processor is the control center of the key smelting data processing device for inclusions in steel, and uses various interfaces and lines to connect various parts of the key smelting data processing device for inclusions in steel.

[0152] The memory can be used to store the computer programs and / or modules. The processor implements various functions of the key smelting data processing device for inclusions in steel by running or executing the computer programs and / or modules stored in the memory and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required for a function (such as a sound playback function, an image playback function), etc.; the data storage area can store data created based on the use of the mobile phone (such as audio data, a phone book, etc.). In addition, the memory can include a high-speed random access memory and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, at least one disk storage device, a memory device, or other volatile solid-state storage device.

[0153] Those skilled in the art will appreciate that the device described in this embodiment is merely an example of a device for processing critical data on steel inclusions, and does not constitute a limitation on such a device. Other embodiments may include more or fewer components, or combinations of certain components, or different components. For example, a device for processing critical data on steel inclusions may also include input and output devices, network access devices, buses, and the like. The processor, memory, and peripheral device interfaces may be connected via buses or signal lines. Each peripheral device may be connected to the peripheral device interface via a bus, signal lines, or circuit boards. Illustratively, peripheral devices include, but are not limited to, radio frequency circuits, touch screen displays, audio circuits, and power supplies.

[0154] Of course, the key smelting data processing device for inclusions in steel may also include fewer or more components, which is not limited in this embodiment.

[0155] Optionally, the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it is used to implement the steps of the above-mentioned method for processing key smelting data of inclusions in steel.

[0156] Optionally, the present application also provides a computer product, which includes a computer-readable storage medium, in which a program is stored. The program is loaded and executed by a processor to implement the steps of an embodiment of a method for processing key smelting data of inclusions in steel.

[0157] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0158] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art could make various modifications and improvements without departing from the spirit of the present application, all of which fall within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be determined by the appended claims.

Claims

1. A method for processing key smelting data of inclusions in steel, characterized by: The method comprises: Collect steelmaking process data, inclusion detection result data and first target smelting data of the target steel grade within a preset time period; Screening the first target smelting data to determine the second target smelting data that meets the target screening condition; The first target smelting data includes the average refining bottom blowing soft stirring pressure P and the interval time T from the start of soft stirring to the start of continuous casting; the second target smelting data includes the average refining bottom blowing soft stirring pressure P in the range of 1.0 to 6.0 bar and the interval time T from the start of soft stirring to the start of continuous casting in the range of 10 to 60 min; and the amount of data including the average refining bottom blowing soft stirring pressure P and the interval time T from the start of soft stirring to the start of continuous casting in the second target smelting data is greater than 300; If the second target smelting data that meets the target screening condition in the first target smelting data meets a preset normality test, extracting the second target smelting data concentrated area, determining the step length of the second target data in the second target smelting data concentrated area, and grouping the second target data in the target smelting data concentrated area according to the step length; After performing Gauss Amp fitting on the number of groups of each smelting data in the second target smelting data and performing a preset variance analysis verification on the inclusion detection results, the inclusion detection result data, target smelting data, and the number of groups that meet the preset variance analysis verification are stored in a preset format and imported into a preset data processing software to form visual data. The numerical range where the steel melt impurities exceed the preset value is then extracted based on the visual data to optimize the steelmaking process; The quantity of the first target smelting data is not less than the quantity of the second target smelting data.

2. The method according to claim 1, characterized in that The steelmaking process data includes molten iron data, converter data, refining data, continuous casting data, ladle data and team data; The inclusion detection result data includes the maximum inclusion size, the number of inclusions per unit area, and the calculated inclusion deduction points.

3. The method according to claim 1 or 2, characterized in that The second target smelting data concentration area includes: The data concentration area of the average pressure of refining bottom blowing soft stirring is (µ p -k p σ p , µ p +k p σ p ), and the data concentration area of the interval from the start of soft stirring to the start of continuous casting is (µ t -k t σ t , µ t +k t σ t ); Where, µ p is the average value of P, σ p is the standard deviation of P, k p =0~5;µ t is the average value of T, σ t is the standard deviation of T, k t =0~5.

4. The method according to claim 1 or 2, characterized in that The method further comprises: Determine whether there is a preset correspondence between the steelmaking process data of the target steel grade within a preset time period and the inclusion detection result data; If so, a preset change is made to the steelmaking process information represented by the steelmaking process data of the target steel grade within the preset time period; Otherwise, a preset data fitting scheme is adopted to fit the steelmaking process data and inclusion detection result data of the target steel grade within a preset time period in a preset manner, determine the matching degree of the steelmaking process data and the inclusion detection result data, and sort the steelmaking process data and inclusion detection result data according to the matching degree.

5. The method according to claim 1, wherein The preset variance analysis verification includes: The sum of squares SSA, degrees of freedom d1, mean square, F value, and significance value sig of component data; the sum of squares SSE, degrees of freedom d2, and mean square MSE of within-group data; the sum of squares SST and degrees of freedom d3 of total data.

6. A device for processing key smelting data of inclusions in steel, characterized in that: include: A collection module is used to collect steelmaking process data, inclusion detection result data and first target smelting data of the target steel grade within a preset time period; A screening module, configured to screen the first target smelting data and determine the second target smelting data that meets the target screening conditions; The first target smelting data includes the average refining bottom blowing soft stirring pressure P and the interval time T from the start of soft stirring to the start of continuous casting; the second target smelting data includes the average refining bottom blowing soft stirring pressure P in the range of 1.0 to 6.0 bar and the interval time T from the start of soft stirring to the start of continuous casting in the range of 10 to 60 min; and the amount of data including the average refining bottom blowing soft stirring pressure P and the interval time T from the start of soft stirring to the start of continuous casting in the second target smelting data is greater than 300; a grouping module, configured to extract a second target smelting data concentrated area if the second target smelting data that meets the target screening condition meets a preset normality test, determine a step length of the second target data in the second target smelting data concentrated area, and group the second target data in the target smelting data concentrated area according to the step length; The module for determining excess data is used to perform Gauss Amp fitting for each number of groups of second target smelting data, and to perform preset variance analysis verification on the inclusion detection results. The module then stores the inclusion detection result data, target smelting data, and number of groups that meet the preset variance analysis verification in a preset format and imports them into preset data processing software to form visual data. The module then extracts the numerical range where the molten steel impurities exceed the preset value based on the visual data to optimize the steelmaking process. The first target smelting data includes a larger amount of data than the second target smelting data.

7. A device for processing key smelting data of inclusions in steel, comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, characterized in that: The computer program is loaded and executed by the processor to implement the steps of the method for processing key metallurgical data of inclusions in steel as claimed in any one of claims 1 to 5.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, it is used to implement the steps of the method for processing key metallurgical data of inclusions in steel as claimed in any one of claims 1 to 5.

Images