CN117745308B - Special steel smelting process traceability optimization method based on product performance analysis - Google Patents
Special steel smelting process traceability optimization method based on product performance analysis Download PDFInfo
- Publication number
- CN117745308B CN117745308B CN202410169950.0A CN202410169950A CN117745308B CN 117745308 B CN117745308 B CN 117745308B CN 202410169950 A CN202410169950 A CN 202410169950A CN 117745308 B CN117745308 B CN 117745308B
- Authority
- CN
- China
- Prior art keywords
- blowing
- stirring
- optimization
- sample
- iteration
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000004458 analytical method Methods 0.000 title claims abstract description 294
- 238000005457 optimization Methods 0.000 title claims abstract description 293
- 238000000034 method Methods 0.000 title claims abstract description 151
- 229910000831 Steel Inorganic materials 0.000 title claims abstract description 131
- 239000010959 steel Substances 0.000 title claims abstract description 131
- 238000003723 Smelting Methods 0.000 title claims abstract description 108
- 238000007664 blowing Methods 0.000 claims abstract description 429
- 238000003756 stirring Methods 0.000 claims abstract description 213
- 230000001174 ascending effect Effects 0.000 claims abstract description 18
- 230000002159 abnormal effect Effects 0.000 claims description 45
- 230000000875 corresponding effect Effects 0.000 claims description 38
- 238000001514 detection method Methods 0.000 claims description 22
- 238000007689 inspection Methods 0.000 claims description 15
- 238000010276 construction Methods 0.000 claims description 14
- 238000012545 processing Methods 0.000 claims description 13
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 claims description 10
- 229910052760 oxygen Inorganic materials 0.000 claims description 10
- 239000001301 oxygen Substances 0.000 claims description 10
- 238000007418 data mining Methods 0.000 claims description 9
- 238000013527 convolutional neural network Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 5
- 230000002596 correlated effect Effects 0.000 claims description 4
- 238000003066 decision tree Methods 0.000 claims description 4
- 238000013386 optimize process Methods 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 13
- 230000006870 function Effects 0.000 description 31
- 230000000630 rising effect Effects 0.000 description 7
- 238000009628 steelmaking Methods 0.000 description 7
- 238000004519 manufacturing process Methods 0.000 description 6
- 150000001247 metal acetylides Chemical class 0.000 description 6
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 description 4
- 229910052751 metal Inorganic materials 0.000 description 4
- 239000002184 metal Substances 0.000 description 4
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 3
- 229910052799 carbon Inorganic materials 0.000 description 3
- 239000013078 crystal Substances 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000005255 carburizing Methods 0.000 description 2
- 238000005520 cutting process Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 229910052742 iron Inorganic materials 0.000 description 2
- 238000012804 iterative process Methods 0.000 description 2
- 238000005498 polishing Methods 0.000 description 2
- OAICVXFJPJFONN-UHFFFAOYSA-N Phosphorus Chemical compound [P] OAICVXFJPJFONN-UHFFFAOYSA-N 0.000 description 1
- 241001062472 Stokellia anisodon Species 0.000 description 1
- UCKMPCXJQFINFW-UHFFFAOYSA-N Sulphide Chemical compound [S-2] UCKMPCXJQFINFW-UHFFFAOYSA-N 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000001816 cooling Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- PCHJSUWPFVWCPO-UHFFFAOYSA-N gold Chemical compound [Au] PCHJSUWPFVWCPO-UHFFFAOYSA-N 0.000 description 1
- 239000010931 gold Substances 0.000 description 1
- 229910052737 gold Inorganic materials 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- WPBNNNQJVZRUHP-UHFFFAOYSA-L manganese(2+);methyl n-[[2-(methoxycarbonylcarbamothioylamino)phenyl]carbamothioyl]carbamate;n-[2-(sulfidocarbothioylamino)ethyl]carbamodithioate Chemical compound [Mn+2].[S-]C(=S)NCCNC([S-])=S.COC(=O)NC(=S)NC1=CC=CC=C1NC(=S)NC(=O)OC WPBNNNQJVZRUHP-UHFFFAOYSA-L 0.000 description 1
- 238000002844 melting Methods 0.000 description 1
- 230000008018 melting Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000003647 oxidation Effects 0.000 description 1
- 238000007254 oxidation reaction Methods 0.000 description 1
- 229910052698 phosphorus Inorganic materials 0.000 description 1
- 239000011574 phosphorus Substances 0.000 description 1
- 229910052710 silicon Inorganic materials 0.000 description 1
- 239000010703 silicon Substances 0.000 description 1
- 239000002893 slag Substances 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Investigating And Analyzing Materials By Characteristic Methods (AREA)
Abstract
The invention provides a special steel smelting process traceability optimization method based on product performance analysis, which relates to the technical field of special steel smelting, and comprises the following steps: obtaining inclusion content, coarse carbide content, inclusion uniformity coefficient and carbide uniformity coefficient; constructing a traceability analysis channel; obtaining a top-blowing traceability analysis result, a stirring traceability analysis result and a bottom-blowing traceability analysis result; obtaining a top-blowing optimization step length, a stirring optimization step length and a bottom-blowing optimization step length; constructing a gradient function; performing tracing iteration through gradient ascending iteration optimization; the method comprises the steps of carrying out multiple rounds of tracing iteration and optimizing iteration to obtain optimized top-blowing parameters, optimized stirring parameters and optimized bottom-blowing parameters, solving the technical problems of unreasonable setting of optimization rules of process parameters, and further causing poor purity and quality of special steel in the prior art, and achieving the technical effects of improving the optimization accuracy of the process parameters and improving the purity smelting quality of the special steel.
Description
Technical Field
The invention relates to the technical field of special steel smelting, in particular to a special steel smelting process traceability optimization method based on product performance analysis.
Background
The smelting of the special steel comprises a plurality of processes of feeding, slagging, deslagging and the like, wherein in the smelting process of the smelting furnace, the processes of top blowing, bottom blowing and stirring of the smelting furnace are included, and the processes of top blowing, bottom blowing and stirring of the smelting furnace have important influence on the purity and smelting quality of the special steel, so that the processes of top blowing, bottom blowing and stirring of the smelting furnace in the smelting process of the smelting furnace are optimized, and the quality of the special steel can be effectively improved.
At present, the technical problems of poor purity and quality of special steel caused by unreasonable setting of optimization rules of technological parameters and poor parameter optimization results exist in the prior art.
Disclosure of Invention
The invention provides a special steel smelting process traceability optimization method based on product performance analysis, which is used for solving the technical problems of poor purity and quality of special steel caused by unreasonable process parameter optimization rule setting and poor parameter optimization result in the prior art.
According to a first aspect of the invention, a special steel smelting process traceability optimization method based on product performance analysis is provided, comprising the following steps: collecting metallographic images of special steel samples produced by smelting, performing convolution characteristic analysis, obtaining the inclusion content and coarse carbide content of a plurality of areas, and calculating to obtain the inclusion content, the coarse carbide content, the inclusion uniformity coefficient and the carbide uniformity coefficient, wherein the special steel is special steel for a shaver; constructing a traceability analysis channel, wherein the traceability analysis channel comprises a top-blowing analysis branch, a stirring analysis branch and a bottom-blowing analysis branch; when any one of the inclusion content, the coarse carbide content, the inclusion uniformity coefficient and the carbide uniformity coefficient is abnormal, respectively inputting the inclusion content, the coarse carbide content, the inclusion uniformity coefficient and the carbide uniformity coefficient into the top-blowing analysis branch, the stirring analysis branch and the bottom-blowing analysis branch to obtain a top-blowing tracing analysis result, a stirring tracing analysis result and a bottom-blowing tracing analysis result which comprise abnormal grades; performing optimization accuracy decision processing according to the top-blowing traceability analysis result, the stirring traceability analysis result and the abnormal grade in the bottom-blowing traceability analysis result to obtain a top-blowing optimization step length, a stirring optimization step length and a bottom-blowing optimization step length; constructing a gradient function according to the inclusion content, coarse carbide content, inclusion uniformity coefficient and carbide uniformity coefficient; sequentially adopting the top-blowing optimization step length, the stirring optimization step length and the bottom-blowing optimization step length, and sequentially carrying out tracing iteration on preset top-blowing parameters, preset stirring parameters and preset bottom-blowing parameters in a preset smelting process through gradient ascending iterative optimization according to the gradient function until a temporary gradient ascending optimization result is obtained and is used as a tracing result; and carrying out optimization iteration for preset times on the corresponding process parameters in the tracing result, and continuing to carry out multi-round tracing iteration and optimization iteration until preset iteration conditions are reached, so as to obtain optimized top-blowing parameters, optimized stirring parameters and optimized bottom-blowing parameters, and generate an optimized smelting process.
According to a second aspect of the invention, there is provided a special steel smelting process traceability optimization system based on product performance analysis, comprising: the convolution characteristic analysis module is used for collecting metallographic images of special steel samples produced by smelting, carrying out convolution characteristic analysis, obtaining the inclusion content and coarse carbide content of a plurality of areas, and calculating to obtain the inclusion content, coarse carbide content, inclusion uniformity coefficient and carbide uniformity coefficient, wherein the special steel is a special steel of the shaver; the traceability analysis channel construction module is used for constructing a traceability analysis channel, and the traceability analysis channel comprises a top-blowing analysis branch, a stirring analysis branch and a bottom-blowing analysis branch; the traceability analysis module is used for respectively inputting the inclusion content, the coarse carbide content, the inclusion uniformity coefficient and the carbide uniformity coefficient into the top-blowing analysis branch, the stirring analysis branch and the bottom-blowing analysis branch when any one of the inclusion content, the coarse carbide content, the inclusion uniformity coefficient and the carbide uniformity coefficient is abnormal, so as to obtain a top-blowing traceability analysis result, a stirring traceability analysis result and a bottom-blowing traceability analysis result which comprise abnormal grades; the optimization precision decision processing module is used for performing optimization precision decision processing according to the top-blowing traceability analysis result, the stirring traceability analysis result and the abnormal grade in the bottom-blowing traceability analysis result to obtain a top-blowing optimization step length, a stirring optimization step length and a bottom-blowing optimization step length; the gradient function construction module is used for constructing a gradient function according to the inclusion content, the coarse carbide content, the inclusion uniformity coefficient and the carbide uniformity coefficient; the tracing iteration module is used for sequentially adopting the top-blowing optimization step length, the stirring optimization step length and the bottom-blowing optimization step length, sequentially tracing iteration is carried out on preset top-blowing parameters, preset stirring parameters and preset bottom-blowing parameters in a preset smelting process through gradient ascending iteration optimization according to the gradient function until a temporary gradient ascending optimization result is obtained and is used as a tracing result; and the optimized smelting process generation module is used for carrying out optimization iteration for preset times on the corresponding process parameters in the tracing result, and continuing to carry out multiple rounds of tracing iteration and optimization iteration until the preset iteration conditions are reached, so as to obtain optimized top-blowing parameters, optimized stirring parameters and optimized bottom-blowing parameters, and generate the optimized smelting process.
According to the special steel smelting process traceability optimization method based on product performance analysis, the beneficial effects are as follows:
1. The method comprises the steps of collecting metallographic images of special steel samples produced through smelting, carrying out convolution characteristic analysis to obtain inclusion content, coarse carbide content, inclusion uniformity coefficient and carbide uniformity coefficient, carrying out optimization iteration sequentially on preset top-blowing parameters, preset stirring parameters and preset bottom-blowing parameters in a preset smelting process according to gradient functions by carrying out trace iteration on the preset top-blowing parameters, the preset stirring parameters and the preset bottom-blowing parameters in a gradient rising mode until a temporary optimization result of gradient rising is obtained, carrying out optimization iteration of preset times on corresponding process parameters in the trace result as the trace result, and continuing to carry out multi-round trace iteration and optimization iteration until preset iteration conditions are reached, obtaining optimized top-blowing parameters, optimized stirring parameters and optimized bottom-blowing parameters, generating optimization smelting process parameters, and improving the optimization accuracy of the process parameters, and improving the technical effect of special steel purity smelting quality.
2. Sequentially adopting a top-blowing optimization step length, a stirring optimization step length and a bottom-blowing optimization step length, sequentially carrying out tracing iteration on preset top-blowing parameters, preset stirring parameters and preset bottom-blowing parameters in a preset smelting process according to a gradient function through gradient ascending iteration optimization until a temporary gradient ascending optimization result is obtained, and realizing staged optimization of the top-blowing parameters, the stirring parameters and the bottom-blowing parameters as a tracing result, thereby providing a basis for subsequent refined optimization and further improving the smelting quality of special steel in the converter steelmaking process and the technical effect of the purity of the special steel.
3. Carrying out optimization iteration for preset times on corresponding process parameters in the tracing result, and continuing to carry out multi-round tracing iteration and optimization iteration until preset iteration conditions are reached, obtaining optimized top-blowing parameters, optimized stirring parameters and optimized bottom-blowing parameters, generating an optimized smelting process, realizing fine optimization iteration for the tracing result, ensuring the process parameter optimization efficiency, preventing the optimization result from being in local optimum, and further improving the technical effect of optimizing accuracy of the process parameters
It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the invention or to delineate the scope of the invention. Other features of the present invention will become apparent from the description that follows.
Drawings
In order to more clearly illustrate the invention or the technical solutions of the prior art, the following brief description will be given of the drawings used in the description of the embodiments or the prior art, it being obvious that the drawings in the description below are only exemplary and that other drawings can be obtained from the drawings provided without the inventive effort for a person skilled in the art.
Fig. 1 is a schematic flow chart of a tracing optimization method of a special steel smelting process based on product performance analysis provided by the embodiment of the invention;
FIG. 2 is a schematic flow chart of acquiring a tracing result in the embodiment of the invention;
FIG. 3 is a schematic flow chart of obtaining optimized top-blowing parameters, optimized stirring parameters, and optimized bottom-blowing parameters in an embodiment of the present invention;
fig. 4 is a schematic structural diagram of a tracing optimization system for a special steel smelting process based on product performance analysis according to an embodiment of the present invention.
Reference numerals illustrate: the system comprises a convolution characteristic analysis module 11, a traceability analysis channel construction module 12, a traceability analysis module 13, an optimization accuracy decision processing module 14, a gradient function construction module 15, a traceability iteration module 16 and an optimization smelting process generation module 17.
Detailed Description
Exemplary embodiments of the present invention will now be described with reference to the accompanying drawings, in which various details of the embodiments of the present invention are included to facilitate understanding, and are to be considered merely exemplary. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the invention. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.
Example 1
Fig. 1 is a diagram of a tracing optimization method of a special steel smelting process based on product performance analysis, which is provided by the embodiment of the invention, and the method comprises the following steps:
Step S100: collecting metallographic images of special steel samples produced by smelting, performing convolution characteristic analysis, obtaining the inclusion content and coarse carbide content of a plurality of areas, and calculating to obtain the inclusion content, the coarse carbide content, the inclusion uniformity coefficient and the carbide uniformity coefficient, wherein the special steel is special steel for a shaver;
The step S100 of the embodiment of the present invention further includes:
step S110: carrying out metallographic image acquisition on a plurality of areas of the special steel to obtain a plurality of metallographic images of the areas;
Step S120: carrying out data mining and identification on smelting quality inspection data of the special steel to obtain a sample metallographic image set, a sample inclusion content set and a sample coarse carbide content set;
Step S130: based on a convolutional neural network, constructing a metallographic recognition path according to the sample metallographic image set, the sample inclusion content set and the sample coarse carbide content set;
Step S140: inputting the metallographic images into the metallographic recognition path to obtain inclusion contents of a plurality of areas and coarse carbide contents of the plurality of areas;
step S150: and calculating the mean value and the mean square error according to the inclusion contents of the plurality of regions and the coarse carbide contents of the plurality of regions, and obtaining the inclusion contents, the coarse carbide contents, the inclusion uniformity coefficient and the carbide uniformity coefficient.
Specifically, the special steel smelting process traceability optimization method based on product performance analysis can be applied to optimization of the smelting process of special steel, preferably can be used for debugging of a new production workshop of the special steel, and the special steel is a special steel for a shaver, so that the new production workshop of the special steel is a workshop for producing the special steel for the shaver, and the production quality of the special steel in the new production workshop of the special steel is improved by debugging and optimizing smelting process parameters.
The metallographic image is obtained by cutting, embedding, polishing and corroding a metal sample, exposing the metal to microscopic crystal structures such as crystal grains, crystal boundaries, defects, inclusions and the like, and carrying out microscopic photographing under an optical microscope, but along with the rapid development of microscope technology, the current process of obtaining a gold image is simpler and more convenient, the sample can be directly scanned and obtained through a specific microscope without carrying out the processes such as cutting, embedding, polishing, corroding and the like, and specifically, the produced special steel sample is extracted from a production workshop to be optimized for the special steel smelting process, and the metallographic image of the special steel sample can be directly acquired through the existing SEM (scanning electron microscope), TEM (transmission electron microscope) and the like, so that the metallographic image of the special steel sample is obtained and uploaded, and the special steel is shaver special steel. Further performing convolution characteristic analysis on the metallographic image to obtain the inclusion content and coarse carbide content of the regions, and calculating to obtain the inclusion content, coarse carbide content, inclusion uniformity coefficient and carbide uniformity coefficient, wherein the specific process is as follows:
Specifically, the special steel sample can be placed on a scanning platform of a microscope for acquiring metallographic images, different areas of the special steel sample can be scanned by the microscope through multiple times of adjustment of the placement positions of the special steel sample, and a plurality of metallographic images of the plurality of areas can be acquired. The smelting quality inspection data of the special steel are subjected to data mining and identification, the smelting quality inspection data refer to detection records of quality inspection of the special steel when the special steel is produced in historical time, the smelting quality inspection data comprise various different types of quality inspection data, including metallographic images of detection samples, carbon content detection, detection of nonmetallic inclusions, hardness, flexibility, netlike carbides and other detection results of the special steel, in the embodiment, a sample metallographic image set, a sample inclusion content set and a sample coarse carbide content set are extracted from the smelting quality inspection data, inclusions in the special steel generally refer to nonmetallic inclusions such as sulfide inclusions, and the types and the contents of nonmetallic inclusions in the steel can be identified according to the optical and morphological characteristics of the nonmetallic inclusions by adopting equipment such as an optical microscope, a scanning electron microscope, a spectrometer and the like; the coarse carbide is the netlike carbide formed after the special steel is smelted, the carburizing process is carried out in the smelting engineering of the special steel, and the netlike carbide is formed when carbon is not uniformly diffused to the surface of the part in the carburizing process. The nonmetallic inclusion and the coarse carbide can be detected by the prior art, and a sample metallographic image set, a sample inclusion content set and a sample coarse carbide content set can be directly extracted from smelting quality inspection data.
The convolutional neural network is an artificial deep neural network of a forward propagation type, is widely used for image recognition, in this embodiment, based on the convolutional neural network, a metallographic recognition path is constructed by taking the sample metallographic image set, the sample inclusion content set and the sample coarse carbide content set as construction data sets, the sample metallographic image set, the sample inclusion content set and the data in the sample coarse carbide content set have a corresponding relation, the sample metallographic image in the sample metallographic image set is recognized through the convolutional neural network, and the output result of the metallographic recognition path is supervised and adjusted by the corresponding sample inclusion content and the corresponding sample coarse carbide content in the sample inclusion content set and the sample coarse carbide content set, so that the metallographic recognition path is trained to a convergence state, and then the recognition accuracy of the metallographic recognition path is tested, so that the metallographic recognition path meeting the requirements is obtained.
And further inputting the metallographic images into the metallographic recognition path in sequence, so that the inclusion contents of the multiple regions and the coarse carbide contents of the multiple regions can be obtained. And calculating the mean value and the mean square error according to the inclusion contents of the plurality of regions and the coarse carbide contents of the plurality of regions, wherein the mean value of the inclusion contents of the plurality of regions and the coarse carbide contents of the plurality of regions is used as the inclusion content and the coarse carbide content, and the mean square error of the inclusion contents of the plurality of regions and the coarse carbide content of the plurality of regions is used as the inclusion uniformity coefficient and the carbide uniformity coefficient. The quality analysis of the special steel sample is realized by analyzing the inclusions and coarse carbides of the special steel sample, and basic data is provided for the follow-up process tracing optimization.
Step S200: constructing a traceability analysis channel, wherein the traceability analysis channel comprises a top-blowing analysis branch, a stirring analysis branch and a bottom-blowing analysis branch;
the step S200 of the embodiment of the present invention further includes:
Step S210: performing data mining on smelting quality inspection data of the special steel to obtain a sample inclusion content set, a coarse carbide content set, a sample inclusion uniformity coefficient set and a sample carbide uniformity coefficient set;
Step S220: respectively carrying out abnormal analysis of oxygen top blowing, molten pool stirring and bottom blowing according to the sample data in the sample inclusion content set, the coarse carbide content set, the sample inclusion uniformity coefficient set and the sample carbide uniformity coefficient set to obtain a sample top blowing traceable analysis result set, a sample stirring traceable analysis result set and a sample background traceable analysis result set;
Step S230: the sample inclusion content set, the coarse carbide content set, the sample inclusion uniformity coefficient set and the sample carbide uniformity coefficient set are respectively combined with the sample top-blowing traceable analysis result set, the sample stirring traceable analysis result set and the sample background blowing traceable analysis result set to construct a top-blowing analysis branch, a stirring analysis branch and a bottom-blowing analysis branch;
step S240: and integrating the top-blowing analysis branch, the stirring analysis branch and the bottom-blowing analysis branch to obtain the traceability analysis channel.
Specifically, a traceable analysis channel is constructed, wherein the traceable analysis channel comprises a top-blowing analysis branch, a stirring analysis branch and a bottom-blowing analysis branch, top blowing is a converter steelmaking method for blowing iron-making water from the top of a converter into steel in the converter steelmaking process, the furnace is an upright crucible-shaped container, and an upright water-cooling oxygen lance is inserted into the furnace from the top to supply oxygen, so that elements such as silicon, manganese, carbon, phosphorus and the like in molten iron are removed by oxidation, and the molten steel reaches the preset chemical composition and temperature; stirring is a technology for supplying energy to a molten metal pool to enable molten metal and slag to move and improve the dynamics of metallurgical reaction; bottom blowing is a converter steelmaking method in which oxygen is blown into a molten pool in a furnace through an oxygen nozzle at the bottom of the converter to smelt molten iron into steel. The three branches are connected in parallel, and the traceability analysis channel is a neural network model in machine learning. Based on a neural network, a network structure of a traceable analysis channel is built, input data of the traceable analysis channel are inclusion content, coarse carbide content, inclusion uniformity coefficient and carbide uniformity coefficient, after the inclusion content, coarse carbide content, inclusion uniformity coefficient and carbide uniformity coefficient receive the input data, a top-blowing analysis branch, a stirring analysis branch and a bottom-blowing analysis branch simultaneously analyze the inclusion content, coarse carbide content, inclusion uniformity coefficient and carbide uniformity coefficient, output of the top-blowing analysis branch is a top-blowing traceable analysis result, output of the stirring analysis branch is a stirring traceable analysis result, and output of the bottom-blowing analysis branch is a bottom-blowing traceable analysis result. The top-blowing traceability analysis result, the stirring traceability analysis result and the bottom-blowing traceability analysis result refer to rationality of the top-blowing parameter, the stirring parameter and the bottom-blowing parameter respectively, and the rationality can be specifically represented by abnormal grades, and the higher the abnormal grade is, the larger the errors of the top-blowing parameter, the stirring parameter and the bottom-blowing parameter are indicated.
Specifically, data mining is carried out on smelting quality inspection data of the special steel, a sample inclusion content set, a coarse carbide content set, a sample inclusion uniformity coefficient set and a sample carbide uniformity coefficient set are extracted, and data in the sample inclusion content set, the coarse carbide content set, the sample inclusion uniformity coefficient set and the sample carbide uniformity coefficient set have a corresponding relation. According to the sample data in the sample inclusion content set, the coarse carbide content set, the sample inclusion uniformity coefficient set and the sample carbide uniformity coefficient set, the oxygen top-blowing, the molten pool stirring and the bottom-blowing anomaly analysis are respectively carried out through the prior art, and the oxygen top-blowing, the molten pool stirring and the bottom-blowing anomaly analysis can be carried out through an expert system, so that a sample top-blowing traceability analysis result set, a sample stirring traceability analysis result set and a sample background traceability analysis result set are obtained, wherein the sample top-blowing traceability analysis result set, the sample stirring traceability analysis result set and the sample background traceability analysis result set respectively comprise a sample top-blowing anomaly grade set, a sample stirring anomaly grade set and a sample bottom-blowing anomaly grade set.
The sample inclusion content set, the coarse carbide content set, the sample inclusion uniformity coefficient set and the sample carbide uniformity coefficient set are adopted to respectively combine the sample top-blowing traceability analysis result set, the sample stirring traceability analysis result set and the sample background traceability analysis result set to construct a top-blowing analysis branch, a stirring analysis branch and a bottom-blowing analysis branch, in short, according to the sample inclusion content set, the coarse carbide content set, the sample inclusion uniformity coefficient set and the sample carbide uniformity coefficient set, each group of sample inclusion content, the coarse carbide content, the sample inclusion uniformity coefficient and the sample carbide uniformity coefficient is input into the top-blowing analysis branch, the stirring analysis branch and the bottom-blowing analysis branch, the output result of the top-blowing analysis branch is supervised and adjusted through the top-blowing traceability analysis result set, the output result of the bottom-blowing analysis branch is supervised and adjusted through the sample background traceability analysis result set, and the top-blowing analysis branch, the stirring analysis branch and the bottom-blowing analysis branch is controlled and converged state are input into the top-blowing analysis branch, and the bottom-blowing analysis branch is controlled to the converged state, and the output accuracy is tested. And finally, integrating the top-blowing analysis branch, the stirring analysis branch and the bottom-blowing analysis branch to obtain the traceability analysis channel, namely, the output data of the traceability analysis channel is formed by combining the output data of the top-blowing analysis branch, the stirring analysis branch and the bottom-blowing analysis branch, so that the traceability analysis channel is constructed, and the technical effect of providing model support for traceability optimization of the smelting process of the special steel is achieved.
Step S300: when any one of the inclusion content, the coarse carbide content, the inclusion uniformity coefficient and the carbide uniformity coefficient is abnormal, respectively inputting the inclusion content, the coarse carbide content, the inclusion uniformity coefficient and the carbide uniformity coefficient into the top-blowing analysis branch, the stirring analysis branch and the bottom-blowing analysis branch to obtain a top-blowing tracing analysis result, a stirring tracing analysis result and a bottom-blowing tracing analysis result which comprise abnormal grades;
Specifically, when any one of the inclusion content, coarse carbide content, inclusion uniformity coefficient and carbide uniformity coefficient is abnormal, the inclusion content, coarse carbide content, inclusion uniformity coefficient and carbide uniformity coefficient are respectively input into the top-blowing analysis branch, the stirring analysis branch and the bottom-blowing analysis branch, respectively, the traceable analysis is carried out to judge whether the top-blowing parameter, the stirring parameter and the bottom-blowing parameter are proper or not, if not, the parameters are abnormal, the abnormal grade is output, the top-blowing parameter, the stirring parameter and the bottom-blowing parameter refer to the technological parameters such as oxygen supply amount, stirring speed and the like when the top-blowing, stirring and bottom-blowing process are carried out, in this process, the inclusion content, the coarse carbide content, the inclusion uniformity coefficient and the carbide uniformity coefficient are required to be abnormal, specifically, according to the type of special steel, according to the smelting quality inspection data of the special steel, based on data mining, extracting inclusion content, coarse carbide content, inclusion uniformity coefficient and carbide uniformity coefficient when the special steel is qualified, analyzing to obtain qualified inclusion content range, coarse carbide content range, inclusion uniformity coefficient range and carbide uniformity coefficient range, further respectively comparing the inclusion content, coarse carbide content, inclusion uniformity coefficient, carbide uniformity coefficient and inclusion content range, coarse carbide content range, inclusion uniformity coefficient range and carbide uniformity coefficient range, and determining whether the inclusion content, coarse carbide content, inclusion uniformity coefficient and carbide uniformity coefficient are all in the inclusion content range, coarse carbide content range, and if any one of the inclusion uniformity coefficient range and the carbide uniformity coefficient range is not in the qualified range, the abnormal condition occurs, and then the inclusion content, the coarse carbide content, the inclusion uniformity coefficient and the carbide uniformity coefficient are respectively input into the top-blowing analysis branch, the stirring analysis branch and the bottom-blowing analysis branch, and the top-blowing traceability analysis result, the stirring traceability analysis result and the bottom-blowing traceability analysis result which comprise abnormal grades are output.
Step S400: performing optimization accuracy decision processing according to the top-blowing traceability analysis result, the stirring traceability analysis result and the abnormal grade in the bottom-blowing traceability analysis result to obtain a top-blowing optimization step length, a stirring optimization step length and a bottom-blowing optimization step length;
the step S400 of the embodiment of the present invention further includes:
Step S410: performing optimization precision analysis according to the sample top-blowing traceability analysis result set, the sample stirring traceability analysis result set and the sample background traceability analysis result set to obtain a sample top-blowing optimization step length set, a sample stirring optimization step length set and a sample bottom-blowing optimization step length set, wherein the size of the optimization step length is positively correlated with the size of the abnormal grade;
Step S420: respectively adopting the sample top-blowing traceability analysis result set and the sample top-blowing optimization step length set, the sample stirring traceability analysis result set and the sample stirring optimization step length set, and the sample bottom-blowing traceability analysis result set and the sample bottom-blowing optimization step length set to construct an optimization precision decision channel based on a decision tree, wherein the optimization precision decision channel comprises a top-blowing decision path, a stirring decision path and a bottom-blowing decision path;
Step S430: and inputting the abnormal grades in the top-blowing decision path, the stirring decision path and the bottom-blowing decision path into the top-blowing traceability analysis result, the stirring traceability analysis result and the bottom-blowing traceability analysis result to obtain the top-blowing optimization step length, the stirring optimization step length and the bottom-blowing optimization step length.
Specifically, according to the top-blowing traceability analysis result, the stirring traceability analysis result and the abnormal grade in the bottom-blowing traceability analysis result, performing optimization accuracy decision processing, and obtaining a top-blowing optimization step length, a stirring optimization step length and a bottom-blowing optimization step length, wherein the top-blowing optimization step length, the stirring optimization step length and the bottom-blowing optimization step length refer to single adjustment values when top-blowing parameters, stirring parameters and bottom-blowing parameters are adjusted, in colloquial terms, when process optimization is performed, the process parameters are difficult to be adjusted to the optimal values once, slow and repeated adjustment analysis is needed, so that the optimal process parameters are found to realize process optimization, the larger the abnormal grade is, and the larger the top-blowing optimization step length, the stirring optimization step length and the bottom-blowing optimization step length are, and the specific processes are as follows:
And performing optimization accuracy analysis according to the sample top-blowing traceability analysis result set, the sample stirring traceability analysis result set and the sample background traceability analysis result set to obtain a sample top-blowing optimization step set, a sample stirring optimization step set and a sample bottom-blowing optimization step set, wherein the size of the optimization step is positively correlated with the size of the abnormal grade, i.e. the higher the abnormal grade is, the larger the optimization step is, by way of example, different optimization step sizes can be set based on different abnormal grades, for example, the abnormal grade is 1 grade, the optimization step is a unit, the unit here refers to the minimum adjustment value of the top-blowing parameter, the stirring parameter and the bottom-blowing parameter, and the specific optimization step setting rule can be set by a person skilled in the art by self-combining with the actual situation, and the method is not limited. The method comprises the steps of respectively adopting a sample top-blowing traceability analysis result set and a sample top-blowing optimization step length set, a sample stirring traceability analysis result set and a sample stirring optimization step length set, and a sample bottom-blowing traceability analysis result set and a sample bottom-blowing optimization step length set, constructing an optimization precision decision channel based on a decision tree, wherein the optimization precision decision channel comprises a top-blowing decision path, a stirring decision path and a bottom-blowing decision path, and exemplarily, taking the sample top-blowing traceability analysis result set as a root node, setting a top-blowing abnormal grade value, performing two classifications on data in the sample top-blowing traceability analysis result set, respectively taking data which are larger than and smaller than the top-blowing abnormal grade value as two leaf nodes, respectively setting the abnormal grade value for the two leaf nodes again, performing two classification division on the leaf nodes again, and so on until the division times reach a preset threshold value, and the preset threshold value can be set by combining with historical experience, marking the sample optimization step length in the sample top-blowing optimization step length set to the top of each bottom layer, namely the leaf node, connecting a plurality of decision nodes, so that the top-blowing path can be obtained, and the same stirring path and the bottom-blowing path can be obtained.
And finally, inputting the abnormal grades in the top-blowing traceability analysis result, the stirring traceability analysis result and the bottom-blowing traceability analysis result into the top-blowing decision path, the stirring decision path and the bottom-blowing decision path, and performing traversal comparison to obtain corresponding top-blowing optimization step length, stirring optimization step length and bottom-blowing optimization step length, so that the analysis of the optimization step length is realized, the support is provided for subsequent traceability iteration, and the technical effects of optimizing the top-blowing, melting furnace stirring and bottom-blowing processes of the special steel and improving the production quality of the special steel are achieved. Step S500: constructing a gradient function according to the inclusion content, coarse carbide content, inclusion uniformity coefficient and carbide uniformity coefficient;
specifically, a gradient function is constructed according to the inclusion content, coarse carbide content, inclusion uniformity coefficient and carbide uniformity coefficient, and the following formula is formed:
；
Wherein L is a gradient, 、/>、/>、/>For the weight, the setting of the weight can be based on the content of inclusions, the content of coarse carbides, the uniformity coefficient of inclusions and the degree of correlation between the uniformity coefficient of carbides and the smelting quality of the special steel, and the larger the degree of correlation is, the larger the weight is, the weight can be set by the existing coefficient of variation method, and the coefficient of variation method is a common technical means for the person skilled in the art and is not developed. /(I)For the inclusion content,/>For the coarse carbide content,/>For the inclusion homogeneity coefficient,/>For the carbide homogeneity coefficient,/>For detecting the inclusion content and the coarse carbide content of the special steel sample produced after iteration (after the top blowing parameter, the stirring parameter and the bottom blowing parameter are adjusted according to the top blowing optimizing step length, the stirring optimizing step length and the bottom blowing optimizing step length)For the detection of inclusion content in the ith region,/>For the detection of coarse carbide content in the ith zone,/>To detect the uniformity coefficient of inclusions in the iterated special steel sample,/>, the method comprises the following stepsThe carbide uniformity coefficient is used for detecting the iterated special steel sample.
Step S600: sequentially adopting the top-blowing optimization step length, the stirring optimization step length and the bottom-blowing optimization step length, and sequentially carrying out tracing iteration on preset top-blowing parameters, preset stirring parameters and preset bottom-blowing parameters in a preset smelting process through gradient ascending iterative optimization according to the gradient function until a temporary gradient ascending optimization result is obtained and is used as a tracing result;
as shown in fig. 2, step S600 of the embodiment of the present invention further includes:
Step S610: sequentially adopting the top-blowing optimization step length, the stirring optimization step length and the bottom-blowing optimization step length, and adjusting and tracing iteration on the preset top-blowing parameter, the preset stirring parameter and the preset bottom-blowing parameter to obtain an iteration top-blowing parameter, an iteration stirring parameter and an iteration bottom-blowing parameter;
Step S620: sequentially adopting the iterative top-blowing parameter, the iterative stirring parameter and the iterative bottom-blowing parameter to perform trial smelting and sample detection on the special steel, and obtaining a first top-blowing gradient, a first stirring gradient and a first bottom-blowing gradient according to the gradient function detection calculation;
Step S630: judging whether the first top-blowing gradient, the first stirring gradient or the first bottom-blowing gradient is larger than 1, if not, continuing to adjust the tracing iteration, and if so, taking the corresponding iteration top-blowing parameter, iteration stirring parameter or iteration bottom-blowing parameter as the tracing result.
Specifically, the top blowing optimization step length, the stirring optimization step length and the bottom blowing optimization step length are adopted sequentially, and according to the gradient function, iterative optimization is carried out through gradient ascending, in short, the lower the inclusion content and the coarse carbide content in the special steel are, the better the inclusion content and the coarse carbide content are, the better the purity and the smelting quality of the special steel are, the inclusion uniformity coefficient and the carbide uniformity coefficient are the variances of the inclusion content and the coarse carbide content in each area of M, the smaller the variances are, the better the smelting result of the special steel is, the smelting quality is, therefore, the ratio of the inclusion content which is not subjected to iterative optimization to the detected inclusion content, the ratio of the coarse carbide content to the detected coarse carbide content, the ratio of the inclusion uniformity coefficient to the detected inclusion uniformity coefficient, and the ratio of the carbide uniformity coefficient to the detected carbide uniformity coefficient are calculated respectively, then weighting the obtained ratio according to the weight, wherein the weighted result is a gradient, the higher the gradient is, the content of inclusions, the content of coarse carbides, the uniformity coefficient of inclusions and the uniformity coefficient of carbides after iterative optimization are improved, the purity and smelting quality of special steel are improved, therefore, the preset top-blowing parameter, the preset stirring parameter and the preset bottom-blowing parameter in the preset smelting process are sequentially traced and iterated according to the top-blowing optimization step length, the stirring optimization step length and the bottom-blowing optimization step length through gradient rising iterative optimization until a temporary optimization result of gradient rising is obtained, and the traced result is the top-blowing parameter, the stirring parameter and the bottom-blowing parameter corresponding to the temporary optimization result as a tracing result, wherein the preset top-blowing parameter, the preset stirring parameters and the preset bottom blowing parameters can be simply understood as initial top blowing parameters, initial stirring parameters and initial bottom blowing parameters set in the smelting furnace steelmaking process of the special steel, and can be directly read based on actual conditions.
The specific process of acquiring the tracing result is as follows: sequentially adopting the top-blowing optimizing step length, the stirring optimizing step length and the bottom-blowing optimizing step length, adjusting and tracing iteration is carried out on the preset top-blowing parameter, the preset stirring parameter and the preset bottom-blowing parameter to obtain an iteration top-blowing parameter, an iteration stirring parameter and an iteration bottom-blowing parameter, sequentially adopting the iteration top-blowing parameter, the iteration stirring parameter and the iteration bottom-blowing parameter, carrying out trial smelting of the special steel based on the prior art, carrying out sample detection on the special steel obtained by trial smelting to obtain a detected inclusion content, a detected coarse carbide content, a detected inclusion uniformity coefficient and a detected carbide uniformity coefficient, according to the gradient function detection calculation, a first top-blowing gradient, a first stirring gradient and a first bottom-blowing gradient are obtained, that is, only one process parameter of a preset top-blowing parameter, a preset stirring parameter and a preset bottom-blowing parameter is subjected to iterative adjustment in one-time trial smelting, for example, the iterative top-blowing parameter, the preset stirring parameter and the preset bottom-blowing parameter are adopted to perform trial smelting of the special steel, sample detection is performed on the special steel obtained by trial smelting, a detection result is brought into the gradient function, the obtained function value is the first top-blowing gradient, and the like, and the preset stirring parameter and the preset bottom-blowing parameter are respectively adjusted to obtain the corresponding first stirring gradient and the first bottom-blowing gradient.
By adjusting a single process parameter type at a time, whether the adjustment of different types of process parameters is effective can be more clearly determined, so that repeated iterative updating is conveniently performed on the basis of adjusting the effective process parameters, namely, whether the first top-blowing gradient, the first stirring gradient or the first bottom-blowing gradient is larger than 1 is judged, in general, if the gradient is equal to 1, the iterative adjustment of the description parameters has no influence on the purity and quality of the special steel, and if the gradient is smaller than 1, the iterative adjustment of the description parameters reduces the purity and quality of the special steel, therefore, if the first top-blowing gradient, the first stirring gradient or the first bottom-blowing gradient is smaller than or equal to 1, the corresponding iterative top-blowing parameter, the iterative stirring parameter or the iterative bottom-blowing parameter does not reach any optimization effect, and if the first top-blowing gradient, the first stirring gradient or the first bottom-blowing gradient is larger than 1, the iterative top-blowing parameter or the iterative bottom-blowing parameter is continuously adjusted and traced according to a top-blowing optimization step length, a stirring optimization step length or a bottom-blowing optimization step length, and if the first top-blowing gradient or the first bottom-blowing gradient is larger than 1, the corresponding iterative top-blowing parameter or the iterative top-blowing parameter is traced to be used as an iterative result. For example, the first top-blowing gradient is greater than 1, the first stirring gradient and the first bottom-blowing gradient are less than or equal to 1, then the iteration top-blowing parameter corresponding to the first top-blowing gradient is added to the tracing result, the tracing iteration is adjusted again according to the stirring optimization step length and the bottom-blowing optimization step length, the iteration stirring parameter and the iteration bottom-blowing parameter are adjusted again until the corresponding gradient is greater than 1, and the iteration stirring parameter and the iteration bottom-blowing parameter corresponding to the gradient greater than 1 are added to the tracing result. Therefore, the optimization of top blowing parameters, stirring parameters and bottom blowing parameters is realized, and the technical effects of improving the smelting quality of special steel in the converter steelmaking process and improving the purity of the special steel are achieved. Step S700: and carrying out optimization iteration for preset times on the corresponding process parameters in the tracing result, and continuing to carry out multi-round tracing iteration and optimization iteration until preset iteration conditions are reached, so as to obtain optimized top-blowing parameters, optimized stirring parameters and optimized bottom-blowing parameters, and generate an optimized smelting process.
As shown in fig. 3, step S700 of the embodiment of the present invention further includes:
Step S710: carrying out optimization iteration for preset times on iteration process parameters in the tracing result by adopting an optimization step length corresponding to the tracing result, and calculating to obtain a preset number of optimization gradients according to the gradient function, wherein in the optimization iteration, the size of the optimization step length is reduced along with the increase of the iteration times;
step S720: selecting the optimized process parameters corresponding to the maximum gradient in the preset number of optimized gradients as an optimized result and as a basic process parameter of the next round of traceability iteration;
Step S730: and continuing to perform multiple rounds of tracing iteration and optimization iteration until the global preset iteration times are reached, and outputting final top-blowing parameters, stirring parameters and bottom-blowing parameters to obtain the optimized top-blowing parameters, optimized stirring parameters and optimized bottom-blowing parameters.
Specifically, the optimization iteration of the preset times is performed on the corresponding process parameters in the tracing result, the multi-round tracing iteration and the optimization iteration are continued until the preset iteration condition is reached, the optimized top-blowing parameter, the optimized stirring parameter and the optimized bottom-blowing parameter are obtained, and the optimized smelting process is generated, that is, the corresponding process parameters in the tracing result are only temporary optimization results with gradient rising, namely, the process optimization effect can be achieved, but are not optimal optimization results, therefore, the parameters in the tracing result need to be subjected to multi-round optimization iteration to obtain the type of parameters with the largest gradient, then the tracing iteration is performed again, and the multi-round operation is repeated to obtain the optimized top-blowing parameter, the optimized stirring parameter and the optimized bottom-blowing parameter, and the specific process is as follows:
And carrying out optimization iteration of preset times on iteration process parameters in the tracing result by adopting an optimization step length corresponding to the tracing result, carrying out trial smelting and sample detection on the special steel by using the iterative process parameters after each optimization iteration, then calculating according to the gradient function to obtain an optimization gradient corresponding to the iterative process parameters, wherein in the optimization iteration, the size of the optimization step length is reduced along with the increase of the iteration number, preferably, the optimization step length of the first optimization iteration can be reduced by one tenth by using a top-blowing optimization step length, a stirring optimization step length or a bottom-blowing optimization step length as the optimization step length of the first optimization iteration, and each optimization iteration is carried out sequentially to obtain the optimization gradient corresponding to the process parameters after each iteration according to the gradient function calculation until the optimization iteration number reaches the preset number, wherein the preset number can be set by self according to actual conditions, and the maximum optimization process parameter corresponding to the preset number of the optimization gradient is selected as the optimization result, and the process parameter in the optimization result is taken as the basic process parameter of the tracing source of the next round.
And (3) repeating the methods of the step S500 and the step S600 by taking the top-blowing parameter, the stirring parameter and the bottom-blowing parameter in the basic technological parameters as references, continuing to perform multiple rounds of tracing iteration and optimization iteration until reaching the global preset iteration times, wherein the global preset iteration times refer to the total times of tracing iteration and optimization iteration, 1 tracing iteration and optimization iteration are 1 global iteration times, the global preset iteration times can be set by a person skilled in the art, and finally, outputting the final top-blowing parameter, stirring parameter and bottom-blowing parameter as the optimized top-blowing parameter, optimized stirring parameter and optimized bottom-blowing parameter, and the optimized top-blowing parameter, optimized stirring parameter and optimized bottom-blowing parameter form an optimized smelting process.
Therefore, rough tracing iteration is carried out on the process parameters according to the top blowing optimization step length, the stirring optimization step length and the bottom blowing optimization step length, and then the tracing result is subjected to fine optimization iteration by gradually reducing the optimization step length, so that the technical effects of ensuring the process parameter optimization efficiency, preventing the optimization result from sinking into local optimization and further improving the optimization accuracy of the process parameters are achieved.
Based on the analysis, the invention provides a special steel smelting process traceability optimization method based on product performance analysis, which has the following beneficial effects:
1. The method comprises the steps of collecting metallographic images of special steel samples produced through smelting, carrying out convolution characteristic analysis to obtain inclusion content, coarse carbide content, inclusion uniformity coefficient and carbide uniformity coefficient, carrying out optimization iteration sequentially on preset top-blowing parameters, preset stirring parameters and preset bottom-blowing parameters in a preset smelting process according to gradient functions by carrying out trace iteration on the preset top-blowing parameters, the preset stirring parameters and the preset bottom-blowing parameters in a gradient rising mode until a temporary optimization result of gradient rising is obtained, carrying out optimization iteration of preset times on corresponding process parameters in the trace result as the trace result, and continuing to carry out multi-round trace iteration and optimization iteration until preset iteration conditions are reached, obtaining optimized top-blowing parameters, optimized stirring parameters and optimized bottom-blowing parameters, generating optimization smelting process parameters, and improving the optimization accuracy of the process parameters, and improving the technical effect of special steel purity smelting quality.
2. Sequentially adopting a top-blowing optimization step length, a stirring optimization step length and a bottom-blowing optimization step length, sequentially carrying out tracing iteration on preset top-blowing parameters, preset stirring parameters and preset bottom-blowing parameters in a preset smelting process according to a gradient function through gradient ascending iteration optimization until a temporary gradient ascending optimization result is obtained, and realizing staged optimization of the top-blowing parameters, the stirring parameters and the bottom-blowing parameters as a tracing result, thereby providing a basis for subsequent refined optimization and further improving the smelting quality of special steel in the converter steelmaking process and the technical effect of the purity of the special steel.
3. And carrying out optimization iteration for the corresponding process parameters in the tracing result for a preset number of times, and continuing to carry out multi-round tracing iteration and optimization iteration until a preset iteration condition is reached, so as to obtain an optimized top-blowing parameter, an optimized stirring parameter and an optimized bottom-blowing parameter, generate an optimized smelting process, realize the fine optimization iteration for the tracing result, and prevent the optimized result from falling into local optimum while ensuring the process parameter optimization efficiency, thereby improving the technical effect of optimizing the process parameter accuracy.
Example 2
Based on the same inventive concept as the special steel smelting process traceability optimization method based on product performance analysis in the foregoing embodiment, as shown in fig. 4, the invention further provides a special steel smelting process traceability optimization system based on product performance analysis, where the system includes:
The convolution characteristic analysis module 11 is used for collecting metallographic images of special steel samples produced by smelting, carrying out convolution characteristic analysis, obtaining the inclusion content and coarse carbide content of a plurality of areas, and calculating to obtain the inclusion content, coarse carbide content, inclusion uniformity coefficient and carbide uniformity coefficient, wherein the special steel is razor special steel;
The traceability analysis channel construction module 12 is used for constructing a traceability analysis channel, and the traceability analysis channel construction module 12 comprises a top-blowing analysis branch, a stirring analysis branch and a bottom-blowing analysis branch;
The traceability analysis module 13 is configured to, when any one of the inclusion content, the coarse carbide content, the inclusion uniformity coefficient and the carbide uniformity coefficient is abnormal, input the inclusion content, the coarse carbide content, the inclusion uniformity coefficient and the carbide uniformity coefficient into the top-blowing analysis branch, the stirring analysis branch and the bottom-blowing analysis branch respectively to obtain a top-blowing traceability analysis result, a stirring traceability analysis result and a bottom-blowing traceability analysis result including abnormal grades;
the optimizing precision decision processing module 14 is configured to perform optimizing precision decision processing according to the top-blowing traceability analysis result, the stirring traceability analysis result and the abnormal level in the bottom-blowing traceability analysis result, so as to obtain a top-blowing optimizing step length, a stirring optimizing step length and a bottom-blowing optimizing step length;
A gradient function construction module 15, wherein the gradient function construction module 15 is used for constructing a gradient function according to the inclusion content, the coarse carbide content, the inclusion uniformity coefficient and the carbide uniformity coefficient;
The tracing iteration module 16 is configured to sequentially perform tracing iteration on a preset top-blowing parameter, a preset stirring parameter and a preset bottom-blowing parameter in a preset smelting process by sequentially adopting the top-blowing optimization step length, the stirring optimization step length and the bottom-blowing optimization step length according to the gradient function and performing gradient ascending iteration optimization until a temporary optimization result of gradient ascending is obtained as a tracing result;
The optimized smelting process generating module 17 is configured to perform optimization iteration for a preset number of times on the corresponding process parameters in the tracing result, and continue performing multiple rounds of tracing iteration and optimization iteration until a preset iteration condition is reached, obtain an optimized top-blowing parameter, an optimized stirring parameter and an optimized bottom-blowing parameter, and generate an optimized smelting process. Further, the convolution characteristic analysis module 11 is further configured to:
carrying out metallographic image acquisition on a plurality of areas of the special steel to obtain a plurality of metallographic images of the areas;
carrying out data mining and identification on smelting quality inspection data of the special steel to obtain a sample metallographic image set, a sample inclusion content set and a sample coarse carbide content set;
Based on a convolutional neural network, constructing a metallographic recognition path according to the sample metallographic image set, the sample inclusion content set and the sample coarse carbide content set;
inputting the metallographic images into the metallographic recognition path to obtain inclusion contents of a plurality of areas and coarse carbide contents of the plurality of areas;
And calculating the mean value and the mean square error according to the inclusion contents of the plurality of regions and the coarse carbide contents of the plurality of regions, and obtaining the inclusion contents, the coarse carbide contents, the inclusion uniformity coefficient and the carbide uniformity coefficient.
Further, the traceability analysis channel construction module 12 is further configured to:
Performing data mining on smelting quality inspection data of the special steel to obtain a sample inclusion content set, a coarse carbide content set, a sample inclusion uniformity coefficient set and a sample carbide uniformity coefficient set;
Respectively carrying out abnormal analysis of oxygen top blowing, molten pool stirring and bottom blowing according to the sample data in the sample inclusion content set, the coarse carbide content set, the sample inclusion uniformity coefficient set and the sample carbide uniformity coefficient set to obtain a sample top blowing traceable analysis result set, a sample stirring traceable analysis result set and a sample background traceable analysis result set;
The sample inclusion content set, the coarse carbide content set, the sample inclusion uniformity coefficient set and the sample carbide uniformity coefficient set are respectively combined with the sample top-blowing traceable analysis result set, the sample stirring traceable analysis result set and the sample background blowing traceable analysis result set to construct a top-blowing analysis branch, a stirring analysis branch and a bottom-blowing analysis branch;
and integrating the top-blowing analysis branch, the stirring analysis branch and the bottom-blowing analysis branch to obtain the traceability analysis channel.
Further, the optimization accuracy decision processing module 14 is further configured to:
Performing optimization precision analysis according to the sample top-blowing traceability analysis result set, the sample stirring traceability analysis result set and the sample background traceability analysis result set to obtain a sample top-blowing optimization step length set, a sample stirring optimization step length set and a sample bottom-blowing optimization step length set, wherein the size of the optimization step length is positively correlated with the size of the abnormal grade;
Respectively adopting the sample top-blowing traceability analysis result set and the sample top-blowing optimization step length set, the sample stirring traceability analysis result set and the sample stirring optimization step length set, and the sample bottom-blowing traceability analysis result set and the sample bottom-blowing optimization step length set to construct an optimization precision decision channel based on a decision tree, wherein the optimization precision decision channel comprises a top-blowing decision path, a stirring decision path and a bottom-blowing decision path;
and inputting the abnormal grades in the top-blowing decision path, the stirring decision path and the bottom-blowing decision path into the top-blowing traceability analysis result, the stirring traceability analysis result and the bottom-blowing traceability analysis result to obtain the top-blowing optimization step length, the stirring optimization step length and the bottom-blowing optimization step length.
Further, the gradient function construction module 15 further includes: and constructing a gradient function according to the inclusion content, the coarse carbide content, the inclusion uniformity coefficient and the carbide uniformity coefficient, wherein the gradient function comprises the following formula:
；
Wherein L is a gradient, 、/>、/>、/>Is weight,/>For the inclusion content,/>For the coarse carbide content,/>For the inclusion homogeneity coefficient,/>For the carbide homogeneity coefficient,/>For the number of regions for detecting the inclusion content and the coarse carbide content of the special steel sample produced after the iteration,/>For the detection of inclusion content in the ith region,/>For the detection of coarse carbide content in the ith zone,/>To detect the uniformity coefficient of inclusions in the iterated special steel sample,/>, the method comprises the following stepsThe carbide uniformity coefficient is used for detecting the iterated special steel sample.
Further, the traceability iteration module 16 is further configured to:
Sequentially adopting the top-blowing optimization step length, the stirring optimization step length and the bottom-blowing optimization step length, and adjusting and tracing iteration on the preset top-blowing parameter, the preset stirring parameter and the preset bottom-blowing parameter to obtain an iteration top-blowing parameter, an iteration stirring parameter and an iteration bottom-blowing parameter;
Sequentially adopting the iterative top-blowing parameter, the iterative stirring parameter and the iterative bottom-blowing parameter to perform trial smelting and sample detection on the special steel, and obtaining a first top-blowing gradient, a first stirring gradient and a first bottom-blowing gradient according to the gradient function detection calculation;
Judging whether the first top-blowing gradient, the first stirring gradient or the first bottom-blowing gradient is larger than 1, if not, continuing to adjust the tracing iteration, and if so, taking the corresponding iteration top-blowing parameter, iteration stirring parameter or iteration bottom-blowing parameter as the tracing result.
Further, the optimized smelting process generation module 17 is further configured to:
carrying out optimization iteration for preset times on iteration process parameters in the tracing result by adopting an optimization step length corresponding to the tracing result, and calculating to obtain a preset number of optimization gradients according to the gradient function, wherein in the optimization iteration, the size of the optimization step length is reduced along with the increase of the iteration times;
Selecting the optimized process parameters corresponding to the maximum gradient in the preset number of optimized gradients as an optimized result and as a basic process parameter of the next round of traceability iteration;
and continuing to perform multiple rounds of tracing iteration and optimization iteration until the global preset iteration times are reached, and outputting final top-blowing parameters, stirring parameters and bottom-blowing parameters to obtain the optimized top-blowing parameters, optimized stirring parameters and optimized bottom-blowing parameters.
The specific example of the special steel smelting process traceability optimization method based on product performance analysis in the first embodiment is also applicable to the special steel smelting process traceability optimization system based on product performance analysis in this embodiment, and by the foregoing detailed description of the special steel smelting process traceability optimization method based on product performance analysis, those skilled in the art can clearly know the special steel smelting process traceability optimization system based on product performance analysis in this embodiment, so that the description is omitted here for brevity.
It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps described in the present invention may be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution disclosed in the present invention can be achieved, and are not limited herein. The above embodiments do not limit the scope of the present invention. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of the present invention.
Claims (8)
1. The method for tracing and optimizing the special steel smelting process based on product performance analysis is characterized by comprising the following steps of:
Collecting metallographic images of special steel samples produced by smelting, performing convolution characteristic analysis, obtaining the inclusion content and coarse carbide content of a plurality of areas, and calculating to obtain the inclusion content, the coarse carbide content, the inclusion uniformity coefficient and the carbide uniformity coefficient, wherein the special steel is special steel for a shaver;
constructing a traceability analysis channel, wherein the traceability analysis channel comprises a top-blowing analysis branch, a stirring analysis branch and a bottom-blowing analysis branch;
When any one of the inclusion content, the coarse carbide content, the inclusion uniformity coefficient and the carbide uniformity coefficient is abnormal, respectively inputting the inclusion content, the coarse carbide content, the inclusion uniformity coefficient and the carbide uniformity coefficient into the top-blowing analysis branch, the stirring analysis branch and the bottom-blowing analysis branch to obtain a top-blowing tracing analysis result, a stirring tracing analysis result and a bottom-blowing tracing analysis result which comprise abnormal grades;
Performing optimization accuracy decision processing according to the top-blowing traceability analysis result, the stirring traceability analysis result and the abnormal grade in the bottom-blowing traceability analysis result to obtain a top-blowing optimization step length, a stirring optimization step length and a bottom-blowing optimization step length;
Constructing a gradient function according to the inclusion content, coarse carbide content, inclusion uniformity coefficient and carbide uniformity coefficient;
Sequentially adopting the top-blowing optimization step length, the stirring optimization step length and the bottom-blowing optimization step length, and sequentially carrying out tracing iteration on preset top-blowing parameters, preset stirring parameters and preset bottom-blowing parameters in a preset smelting process through gradient ascending iterative optimization according to the gradient function until a temporary gradient ascending optimization result is obtained and is used as a tracing result;
And carrying out optimization iteration for preset times on the corresponding process parameters in the tracing result, and continuing to carry out multi-round tracing iteration and optimization iteration until preset iteration conditions are reached, so as to obtain optimized top-blowing parameters, optimized stirring parameters and optimized bottom-blowing parameters, and generate an optimized smelting process.
2. The method of claim 1, wherein collecting metallographic images of the smelted specialty steel sample for convolution feature analysis comprises:
carrying out metallographic image acquisition on a plurality of areas of the special steel to obtain a plurality of metallographic images of the areas;
carrying out data mining and identification on smelting quality inspection data of the special steel to obtain a sample metallographic image set, a sample inclusion content set and a sample coarse carbide content set;
Based on a convolutional neural network, constructing a metallographic recognition path according to the sample metallographic image set, the sample inclusion content set and the sample coarse carbide content set;
inputting the metallographic images into the metallographic recognition path to obtain inclusion contents of a plurality of areas and coarse carbide contents of the plurality of areas;
And calculating the mean value and the mean square error according to the inclusion contents of the plurality of regions and the coarse carbide contents of the plurality of regions, and obtaining the inclusion contents, the coarse carbide contents, the inclusion uniformity coefficient and the carbide uniformity coefficient.
3. The method of claim 1, wherein constructing a trace-source analysis channel comprises:
Performing data mining on smelting quality inspection data of the special steel to obtain a sample inclusion content set, a coarse carbide content set, a sample inclusion uniformity coefficient set and a sample carbide uniformity coefficient set;
Respectively carrying out abnormal analysis of oxygen top blowing, molten pool stirring and bottom blowing according to the sample data in the sample inclusion content set, the coarse carbide content set, the sample inclusion uniformity coefficient set and the sample carbide uniformity coefficient set to obtain a sample top blowing traceable analysis result set, a sample stirring traceable analysis result set and a sample background traceable analysis result set;
The sample inclusion content set, the coarse carbide content set, the sample inclusion uniformity coefficient set and the sample carbide uniformity coefficient set are respectively combined with the sample top-blowing traceable analysis result set, the sample stirring traceable analysis result set and the sample background blowing traceable analysis result set to construct a top-blowing analysis branch, a stirring analysis branch and a bottom-blowing analysis branch;
and integrating the top-blowing analysis branch, the stirring analysis branch and the bottom-blowing analysis branch to obtain the traceability analysis channel.
4. The method of claim 3, wherein performing optimization accuracy decision processing according to anomaly levels in the top-blowing traceability analysis result, the stirring traceability analysis result and the bottom-blowing traceability analysis result comprises:
Performing optimization precision analysis according to the sample top-blowing traceability analysis result set, the sample stirring traceability analysis result set and the sample background traceability analysis result set to obtain a sample top-blowing optimization step length set, a sample stirring optimization step length set and a sample bottom-blowing optimization step length set, wherein the size of the optimization step length is positively correlated with the size of the abnormal grade;
Respectively adopting the sample top-blowing traceability analysis result set and the sample top-blowing optimization step length set, the sample stirring traceability analysis result set and the sample stirring optimization step length set, and the sample bottom-blowing traceability analysis result set and the sample bottom-blowing optimization step length set to construct an optimization precision decision channel based on a decision tree, wherein the optimization precision decision channel comprises a top-blowing decision path, a stirring decision path and a bottom-blowing decision path;
and inputting the abnormal grades in the top-blowing decision path, the stirring decision path and the bottom-blowing decision path into the top-blowing traceability analysis result, the stirring traceability analysis result and the bottom-blowing traceability analysis result to obtain the top-blowing optimization step length, the stirring optimization step length and the bottom-blowing optimization step length.
5. The method of claim 1, wherein a gradient function is constructed based on the inclusion content, coarse carbide content, inclusion uniformity coefficient, and carbide uniformity coefficient, according to the formula:
；
Wherein L is a gradient, 、/>、/>、/>Is weight,/>For the inclusion content,/>For the coarse carbide content,/>For the inclusion homogeneity coefficient,/>For the carbide homogeneity coefficient,/>For the number of regions for detecting the inclusion content and the coarse carbide content of the special steel sample produced after the iteration,/>For the detection of inclusion content in the ith region,/>For the detection of coarse carbide content in the ith zone,/>To detect the uniformity coefficient of inclusions in the iterated special steel sample,/>, the method comprises the following stepsThe carbide uniformity coefficient is used for detecting the iterated special steel sample.
6. The method according to claim 1, characterized in that the method comprises:
Sequentially adopting the top-blowing optimization step length, the stirring optimization step length and the bottom-blowing optimization step length, and adjusting and tracing iteration on the preset top-blowing parameter, the preset stirring parameter and the preset bottom-blowing parameter to obtain an iteration top-blowing parameter, an iteration stirring parameter and an iteration bottom-blowing parameter;
Sequentially adopting the iterative top-blowing parameter, the iterative stirring parameter and the iterative bottom-blowing parameter to perform trial smelting and sample detection on the special steel, and obtaining a first top-blowing gradient, a first stirring gradient and a first bottom-blowing gradient according to the gradient function detection calculation;
Judging whether the first top-blowing gradient, the first stirring gradient or the first bottom-blowing gradient is larger than 1, if not, continuing to adjust the tracing iteration, and if so, taking the corresponding iteration top-blowing parameter, iteration stirring parameter or iteration bottom-blowing parameter as the tracing result.
7. The method of claim 6, wherein performing a preset number of optimization iterations on corresponding process parameters in the trace-back result comprises:
carrying out optimization iteration for preset times on iteration process parameters in the tracing result by adopting an optimization step length corresponding to the tracing result, and calculating to obtain a preset number of optimization gradients according to the gradient function, wherein in the optimization iteration, the size of the optimization step length is reduced along with the increase of the iteration times;
Selecting the optimized process parameters corresponding to the maximum gradient in the preset number of optimized gradients as an optimized result and as a basic process parameter of the next round of traceability iteration;
and continuing to perform multiple rounds of tracing iteration and optimization iteration until the global preset iteration times are reached, and outputting final top-blowing parameters, stirring parameters and bottom-blowing parameters to obtain the optimized top-blowing parameters, optimized stirring parameters and optimized bottom-blowing parameters.
8. A special steel smelting process traceability optimization system based on product performance analysis, characterized by comprising the steps of executing any one of the special steel smelting process traceability optimization methods based on product performance analysis as set forth in claims 1-7, wherein the system comprises:
the convolution characteristic analysis module is used for collecting metallographic images of special steel samples produced by smelting, carrying out convolution characteristic analysis, obtaining the inclusion content and coarse carbide content of a plurality of areas, and calculating to obtain the inclusion content, coarse carbide content, inclusion uniformity coefficient and carbide uniformity coefficient, wherein the special steel is a special steel of the shaver;
The traceability analysis channel construction module is used for constructing a traceability analysis channel, and the traceability analysis channel comprises a top-blowing analysis branch, a stirring analysis branch and a bottom-blowing analysis branch;
the traceability analysis module is used for respectively inputting the inclusion content, the coarse carbide content, the inclusion uniformity coefficient and the carbide uniformity coefficient into the top-blowing analysis branch, the stirring analysis branch and the bottom-blowing analysis branch when any one of the inclusion content, the coarse carbide content, the inclusion uniformity coefficient and the carbide uniformity coefficient is abnormal, so as to obtain a top-blowing traceability analysis result, a stirring traceability analysis result and a bottom-blowing traceability analysis result which comprise abnormal grades;
The optimization precision decision processing module is used for performing optimization precision decision processing according to the top-blowing traceability analysis result, the stirring traceability analysis result and the abnormal grade in the bottom-blowing traceability analysis result to obtain a top-blowing optimization step length, a stirring optimization step length and a bottom-blowing optimization step length;
The gradient function construction module is used for constructing a gradient function according to the inclusion content, the coarse carbide content, the inclusion uniformity coefficient and the carbide uniformity coefficient;
The tracing iteration module is used for sequentially adopting the top-blowing optimization step length, the stirring optimization step length and the bottom-blowing optimization step length, sequentially tracing iteration is carried out on preset top-blowing parameters, preset stirring parameters and preset bottom-blowing parameters in a preset smelting process through gradient ascending iteration optimization according to the gradient function until a temporary gradient ascending optimization result is obtained and is used as a tracing result;
and the optimized smelting process generation module is used for carrying out optimization iteration for preset times on the corresponding process parameters in the tracing result, and continuing to carry out multiple rounds of tracing iteration and optimization iteration until the preset iteration conditions are reached, so as to obtain optimized top-blowing parameters, optimized stirring parameters and optimized bottom-blowing parameters, and generate the optimized smelting process.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410169950.0A CN117745308B (en) | 2024-02-06 | 2024-02-06 | Special steel smelting process traceability optimization method based on product performance analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410169950.0A CN117745308B (en) | 2024-02-06 | 2024-02-06 | Special steel smelting process traceability optimization method based on product performance analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117745308A CN117745308A (en) | 2024-03-22 |
| CN117745308B true CN117745308B (en) | 2024-06-14 |
Family
ID=90253035
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410169950.0A Active CN117745308B (en) | 2024-02-06 | 2024-02-06 | Special steel smelting process traceability optimization method based on product performance analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117745308B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118691165A (en) * | 2024-08-26 | 2024-09-24 | 南京先维信息技术有限公司 | Aluminum smelting product quality monitoring and tracing method and system based on big data modeling |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110109811A (en) * | 2019-05-10 | 2019-08-09 | 北京物资学院 | A kind of source tracing method towards GPU calculated performance problem |
| CN111157620A (en) * | 2020-01-03 | 2020-05-15 | 广东韶钢松山股份有限公司 | Traceability analysis method for large-size inclusions in steel |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113139814B (en) * | 2021-04-13 | 2023-03-14 | 广东工业大学 | Ceramic production transaction traceability system |
| CN115576285A (en) * | 2022-10-23 | 2023-01-06 | 唐山惠唐物联科技有限公司 | Full-automatic production method and system for LF refining |
-
2024
- 2024-02-06 CN CN202410169950.0A patent/CN117745308B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110109811A (en) * | 2019-05-10 | 2019-08-09 | 北京物资学院 | A kind of source tracing method towards GPU calculated performance problem |
| CN111157620A (en) * | 2020-01-03 | 2020-05-15 | 广东韶钢松山股份有限公司 | Traceability analysis method for large-size inclusions in steel |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117745308A (en) | 2024-03-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117745308B (en) | Special steel smelting process traceability optimization method based on product performance analysis | |
| CN108676955B (en) | A kind of BOF Steelmaking Endpoint carbon content and temprature control method | |
| US8002871B2 (en) | Methods and apparatus for an oxygen furnace quality control system | |
| WO2018012257A1 (en) | Method for estimating phosphorus concentration in molten steel and converter blowing control device | |
| CN106096788B (en) | Converter steelmaking process cost control method and system based on PSO _ ELM neural network | |
| CN104630410A (en) | Real-time dynamic converter steelmaking quality prediction method based on data analysis | |
| CN110781944B (en) | Automatic molten iron slag-off control method based on deep learning | |
| US11545239B1 (en) | Method for predicting yield of calcium in a calcium treatment process based on deep neural network | |
| CN110309608B (en) | Blast furnace molten iron silicon content forecasting method aiming at time lag uncertain information | |
| JP4093934B2 (en) | Model parameter determination method and program thereof, model prediction method and program thereof | |
| CN112465223A (en) | Blast furnace temperature state prediction method | |
| KR20200118200A (en) | Molten metal component estimation device, molten metal component estimation method, and molten metal manufacturing method | |
| CN117707096B (en) | Dynamic optimization method and system for special steel processing technology based on performance requirements | |
| CN116976148B (en) | Method and system for monitoring ion content change in copper electrolysis process | |
| CN116822342A (en) | Process identification and performance prediction method for zirconium alloy laser cutting | |
| CN118016203B (en) | LF refining temperature forecasting method based on mechanism model and XGBoost algorithm | |
| Tang et al. | An estimation of distribution algorithm with resampling and local improvement for an operation optimization problem in steelmaking process | |
| JP2010007150A (en) | Method for estimating carbon concentration in molten steel | |
| KR102437794B1 (en) | Molten metal component estimation device, method of estimating molten metal component, and method of manufacturing molten metal | |
| CN116911057A (en) | Real-time prediction method for temperature of molten pool in converter steelmaking converting process | |
| Murua et al. | A slag prediction model in an electric arc furnace process for special steel production | |
| CN114841277A (en) | Method and system for predicting yield of converter tapping alloy | |
| CN114822723B (en) | Method and device for predicting end point of refining anode copper reaction | |
| JP2012149341A (en) | Estimation method of molten metal component and estimation apparatus of molten metal component | |
| CN118553361A (en) | Prediction method of nickel-boron-aluminum intermediate alloy based on machine learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |