A method, device and medium for optimizing linkage of LF refining quaternary final slag design and batching of aluminum killed steel

Abstract

The application relates to the technical field of steel metallurgical furnace outside refining, and discloses an aluminum killed steel LF refining four-element final slag design and charging linkage optimization method, equipment and medium. The method comprises the following steps: obtaining target molten steel components, a target refining temperature and CaO and MgO value ranges, generating four-element candidate final slag components based on steel slag balance relations; determining physical and chemical properties and steel slag balance evaluation parameters of the candidate final slag, obtaining a target final slag design result through process constraint screening and multi-target comprehensive scoring; based on the target final slag design result and charging material information, solving charging material addition amounts under material available amount constraints and slag component mass conservation constraints, and generating a charging scheme; obtaining actual executed charging material addition amounts and actual final slag components, updating charging calculation correction parameters according to the charging data, and using the charging calculation correction parameters for subsequent furnace charging scheme generation. The application is beneficial to improving the comprehensive matching degree of the target final slag design and the adaptability of the charging calculation to production condition changes.

Description

Technical Field

[0001] This application relates to the field of ladle refining technology in iron and steel metallurgy, and in particular to a method, equipment and medium for the design, batching linkage optimization of quaternary final slag in LF refining of aluminum killed steel. Background Technology

[0002] Ladle refining, also known as ladle refining, is a crucial step in the ladle refining process of molten steel. It is typically used to achieve purposes such as increasing steel temperature, fine-tuning composition, desulfurization, inclusion control, and improving steel cleanliness. During the LF refining of aluminum-killed steel, continuous deoxidation, desulfurization, and mass transfer and equilibrium reactions involving elements such as calcium and magnesium occur between the slag and molten steel. The slag composition not only affects the slag's viscosity, surface tension, and phase state but also the equilibrium state of oxygen, calcium, and magnesium in the molten steel. Therefore, rationally determining the final slag composition that matches the target molten steel composition is a critical step in controlling the LF refining process and the quality of the molten steel in aluminum-killed steel.

[0003] For aluminum-killed steel, the CaO-SiO2-Al2O3-MgO quaternary slag system is a commonly used LF refining slag system. In existing technologies, the slag composition in equilibrium with the molten steel, or parameters such as equilibrium oxygen, equilibrium calcium, and equilibrium magnesium in the molten steel, can be calculated based on the steel composition, refining temperature, and slag component content using the steel-slag equilibrium relationship. Furthermore, slag viscosity, slag surface tension, slag component activity, and slag phase state can be calculated to assist in slag composition design.

[0004] However, the design of final slag for aluminum-killed steel LF refining typically involves multiple interrelated evaluation indicators. For example, candidate final slags not only need to meet basic process requirements such as slag component content, viscosity, and phase region location, but may also need to consider slag surface tension and the equilibrium state of steel slag, including equilibrium oxygen, equilibrium calcium, and equilibrium magnesium in the molten steel. When multiple candidate final slags meet the basic process constraints, relying solely on a single evaluation parameter or setting the value range of each evaluation parameter separately is insufficient to further reflect the comprehensive matching degree of different candidate final slags with multiple process objectives. This makes it difficult to determine the target final slag design result that is more suitable for the current refining conditions from multiple feasible candidate final slags.

[0005] On the other hand, existing LF refining batching technology can perform batching calculations based on the target slag composition, batching material composition, material addition amount, and mass conservation relationship, and can determine the corresponding material addition amount through optimization calculations. However, the target final slag design and batching calculation are usually two relatively independent processing steps. When the type, composition, and availability of materials change on-site, how to directly use the target final slag design results determined by multiple metallurgical evaluations as constraints or optimization targets, and automatically determine the matching batching scheme based on the currently available materials, still needs further resolution.

[0006] Furthermore, when determining the amount of each ingredient added based on the target final slag design, it is necessary to establish a mass conservation relationship between the effective contribution of each ingredient to different slag components and the target final slag composition. Production factors such as actual material recovery and slag carryover during tapping can affect the correspondence between the amount of ingredients added and the final slag composition. When fixed batching calculation parameters are continuously used, the actual final slag composition obtained after LF refining according to the determined batching scheme may deviate from the target final slag design result. If the corresponding batching calculation parameters cannot be corrected based on the actual batching data and the actual final slag composition, the aforementioned changes in production conditions will be difficult to reflect in the batching calculation process of subsequent heats in a timely manner.

[0007] Therefore, how to comprehensively evaluate multiple candidate final slags for the CaO-SiO2-Al2O3-MgO quaternary slag system of aluminum-killed steel LF refining while meeting basic metallurgical process constraints, directly convert the determined target final slag design results into a batching scheme that adapts to on-site material conditions, and use actual production data to correct the calculation relationship between the target final slag design results and the amount of batching materials added, is a problem that still needs to be further solved by existing technologies. Summary of the Invention

[0008] This application provides a method, equipment, and medium for the design and batching optimization of quaternary final slag in the LF refining of aluminum-killed steel, in order to solve the problems in the existing quaternary final slag design process for LF refining of aluminum-killed steel, such as the difficulty in taking multiple metallurgical evaluation indicators into account in the determination of the target final slag, the difficulty in directly converting the target final slag design results into a batching scheme that adapts to the material conditions on site, and the difficulty in correcting the calculated relationship between the target final slag design results and the amount of batching materials added based on actual production data.

[0009] This application provides a method for the design and batching optimization of quaternary final slag in the LF refining of aluminum-killed steel, comprising: obtaining the target steel composition, target refining temperature, CaO value range, and MgO value range in the target final slag of the aluminum-killed steel to be refined, wherein the target steel composition includes at least Al content and Si content; determining multiple sets of candidate CaO content and candidate MgO content within the CaO value range and the MgO value range; and based on the target steel composition, the target refining temperature, and the relationship between the aluminum-killed steel and CaO-Si... The steel-slag equilibrium relationship between the O2-Al2O3-MgO quaternary slag is determined, and the Al2O3 and SiO2 candidate contents corresponding to the CaO and MgO candidate contents of each group are determined, forming multiple CaO-SiO2-Al2O3-MgO quaternary candidate final slag components. The slag viscosity, slag surface tension, and steel-slag equilibrium evaluation parameters corresponding to each candidate final slag component are determined. Based on preset process constraints, the multiple candidate final slag components are screened, considering the slag viscosity, slag surface tension, and steel-slag equilibrium evaluation parameters. At least two parameters are selected from the parameters included in the balanced evaluation parameters as scoring dimensions. Sub-scores are determined for each of the selected candidate final slag components under each scoring dimension. A comprehensive score is determined based on each sub-score and its corresponding weight. The target final slag design result is then determined based on the comprehensive score. The target final slag design result includes target final slag components or a target final slag component range. Batching material information is obtained, including at least the composition and available quantity of each batching material. Based on a batching calculation model with batching calculation correction parameters, target batching constraints corresponding to each slag component are determined according to the target final slag design result. Under the constraints of available material quantity and slag component mass conservation, the addition amount of each batching material satisfying the target batching constraints is calculated to obtain the batching scheme. The actual final slag components obtained after LF refining according to the batching scheme are obtained. The deviation between the actual final slag components and the target final slag design result is determined. At least one batching calculation correction parameter in the batching calculation model is updated, and the updated batching calculation correction parameter is used for generating batching schemes for subsequent furnace cycles.

[0010] Furthermore, based on the target steel composition, the target refining temperature, and the steel-slag equilibrium relationship between the aluminum-killed steel and the CaO-SiO2-Al2O3-MgO quaternary slag, the candidate Al2O3 content and candidate SiO2 content corresponding to each group of candidate CaO content and candidate MgO content are determined, including: determining the activity of at least one component in the steel according to the target steel composition; and determining the candidate Al2O3 content and candidate SiO2 content corresponding to each group of candidate CaO content and candidate MgO content based on the target refining temperature, the activity of at least one component in the steel, the activity of slag components, and the corresponding steel-slag equilibrium reaction relationship.

[0011] Further, the steel slag balance evaluation parameters include at least one of the balance oxygen content, balance calcium content, and balance magnesium content in molten steel; the preset process constraints include at least one of the following: slag component content constraints, slag viscosity constraints, steel slag balance evaluation parameter constraints, and phase diagram phase region constraints; the multiple candidate final slag components are screened according to the preset process constraints, including: mapping each candidate final slag component to the isothermal liquidus phase diagram of the CaO-SiO2-Al2O3-MgO quaternary slag system, and determining the phase region position corresponding to each candidate final slag component; candidate final slag components that do not meet the preset process constraints are excluded, and candidate final slag components that meet the preset process constraints are used as candidate final slag components for comprehensive scoring.

[0012] Furthermore, the scoring dimensions also include phase diagram evaluation parameters determined based on the phase region positions of the candidate final residue components in the isothermal liquidus phase diagram; determining sub-scores for each candidate final residue component participating in the comprehensive scoring under each scoring dimension, including: obtaining a corresponding target value or target range for each scoring dimension; for a scoring dimension with a target value, determining a corresponding sub-score based on the deviation between the evaluation parameters corresponding to the candidate final residue component and the target value; for a scoring dimension with a target range, determining a corresponding sub-score based on the degree of matching between the evaluation parameters corresponding to the candidate final residue component and the target range; normalizing each sub-score, and determining the comprehensive score based on the normalized sub-scores and their corresponding weights.

[0013] Furthermore, determining the target final slag design result based on the comprehensive score includes: sorting multiple candidate final slag components participating in the comprehensive score according to the comprehensive score; determining the candidate final slag component with the highest comprehensive score as the target final slag component; or, determining multiple candidate final slag components whose comprehensive scores meet preset scoring conditions as target final slag candidate components, and determining the target content range of the corresponding slag component based on the content distribution of each slag component in the multiple target final slag candidate components, so as to obtain the target final slag component range.

[0014] Further, based on the target final slag design results, the target batching constraints corresponding to each slag component are determined, and the addition amount of each batching material is solved, including: using the addition amount of each batching material as the variable to be solved; based on the contents of CaO, SiO2, Al2O3, and MgO in each batching material, the addition amount of each batching material, and the batching calculation correction parameters, the component contribution of each batching material to each slag component of CaO, SiO2, Al2O3, and MgO is determined according to the slag component mass conservation relationship; when the target final slag design results are obtained, the target batching constraints corresponding to each slag component are determined; and the target final slag design results are determined. When the slag design result includes the target final slag composition, the target batching constraint of the corresponding slag component is determined according to the target content of each slag component in the target final slag composition; when the target final slag design result includes the target final slag composition range, the target batching constraint of the corresponding slag component is determined according to the target content range of each slag component in the target final slag composition range; under the material availability constraint and the slag component mass conservation constraint, the amount of each batching material added that makes the component contribution of each slag component satisfy the corresponding target batching constraint is solved.

[0015] Furthermore, the ingredient information also includes the unit price of each ingredient; when solving for the amount of each ingredient to be added, at least one of the total slag amount constraint and the amount of a single ingredient added is set; when there are multiple feasible solutions for the amount of each ingredient to be added that satisfies the target ingredient constraint, the corresponding total material cost is determined according to the amount of each ingredient added and the corresponding unit price in each feasible solution, and at least one recommended ingredient scheme is determined from the multiple feasible solutions with the goal of reducing the total material cost.

[0016] Further, after obtaining the actual final slag composition, the method further includes: determining the final slag composition deviation between the actual final slag composition and the target final slag design result, and generating the target final slag achievement evaluation result for the current heat based on the final slag composition deviation; obtaining the actual addition amount of each of the batching materials in the current heat or multiple historical heats, and determining the parameter estimate value of the batching calculation correction parameter according to the slag component mass conservation relationship based on the composition of each batching material, the actual addition amount of each batching material, and the corresponding actual final slag composition; updating the batching material yield coefficient and / or tapping slag correction parameter in the batching calculation model according to the parameter estimate value; and using the updated batching material yield coefficient and / or tapping slag correction parameter in the batching calculation model for subsequent heats to re-establish the correspondence between the target final slag design result and the addition amount of each of the batching materials, and generating the batching scheme for subsequent heats.

[0017] A second aspect of this application provides an electronic device, including a processor and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, it implements any of the above-described methods for the design and batching linkage optimization of quaternary final slag in the LF refining of aluminum-killed steel.

[0018] A third aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements any of the above-described methods for the design and batching linkage optimization of quaternary final slag in the LF refining of aluminum-killed steel.

[0019] Compared with the prior art, the above technical solution has at least the following beneficial effects: This application targets the CaO-SiO2-Al2O3-MgO quaternary final slag of aluminum-killed steel LF refining. Based on the target molten steel state and the steel-slag balance relationship, multiple candidate final slags are formed. Among the candidate final slags that meet the preset process constraints, multiple metallurgical evaluation parameters are combined for comprehensive evaluation to determine the target final slag design result that matches the current refining conditions. Furthermore, the target final slag design result is directly converted into the batching constraints of each slag component. Based on the composition, available quantity, and batching calculation correction parameters of the batching materials on site, the addition amount of each batching material is solved under the mass conservation relationship of the slag components. The batching calculation correction parameters are updated using the actual production results. By linking the comprehensive evaluation of candidate final slag, the determination of target final slag, the calculation of the amount of feed materials added, and the feedback of actual production results into the same technical process, we can reduce the one-sidedness of the evaluation caused by determining the final slag based on a single slag parameter. On the other hand, we can make the target final slag determined by metallurgical evaluation directly correspond to the batching scheme under the current on-site material conditions, and make the batching calculation relationship of subsequent furnaces corrected according to the actual production conditions, thereby improving the fit between the final slag design results and the actual batching process.

[0020] The above-mentioned technical solutions also have the following advantages: By further determining the composition of candidate final slag based on the activity of steel components, the activity of slag components, and the equilibrium reaction relationship between steel and slag, and by combining the slag component content, viscosity, steel-slag equilibrium evaluation parameters, and phase regions of the phase diagram for pre-screening, candidate final slags that do not meet the basic metallurgical process requirements can be eliminated before comprehensive scoring; by setting target values ​​or target ranges for different evaluation parameters and normalizing and weighting the corresponding sub-scores, the impact of differences in the dimensions and numerical ranges of different evaluation parameters on the comprehensive evaluation results can be reduced, and it supports the determination of a single target final slag composition from multiple feasible candidate final slags. Or the target final slag composition range; by establishing target batching constraints based on the contribution of each batching material to the target slag component, and determining the recommended batching scheme in combination with material costs when multiple feasible batching solutions exist, the on-site material constraints can be further considered while meeting the target final slag composition requirements; by estimating and updating the batching material recovery rate coefficient and / or tapping slag correction parameters by combining the actual material addition amount, material composition and actual final slag composition, the adjustment of batching calculation parameters can be based on corresponding actual production data, improving the adaptability of subsequent furnace batching calculations to changes in material recovery and tapping slag conditions. Attached Figure Description

[0021] Figure 1 A flowchart illustrating the design and batching linkage optimization method for the quaternary final slag of aluminum-killed steel LF refining provided in this application; Figure 2 A schematic diagram illustrating the candidate final residue process screening and multi-objective comprehensive scoring provided in this application; Figure 3 This is a schematic diagram illustrating the actual final slag feedback and the correction of the batching calculation parameters provided in this application. Detailed Implementation

[0022] The technical solution of this application will be further described below with reference to the accompanying drawings. It should be understood that the following embodiments are used to illustrate the technical solution of this application and are not intended to limit the scope of protection of this application. Unless otherwise stated, the content of each component in this application may be expressed as a mass percentage.

[0023] In this application, LF refining can refer to ladle furnace refining. Aluminum-killed molten steel can refer to molten steel that has been deoxidized with aluminum and whose composition, temperature, and slag are controlled during the LF refining process. The target molten steel composition can refer to the molten steel composition requirements that need to be achieved after LF refining, including at least Al content and Si content, and may further include the target content or target content range of one or more elements selected from C, Mn, P, S, Cr, O, Ca, and Mg.

[0024] The quaternary final slag in this application refers to final slag with CaO-SiO2-Al2O3-MgO as the main design and evaluation components, and does not mean that the actual slag contains only the above four components. Actual slag may also contain FeO, MnO, CaF2, or other components introduced from raw materials, tapping slag, furnace lining reactions, etc. For quaternary final slag design, the contents of the four main design components—CaO, SiO2, Al2O3, and MgO—can be normalized and used for candidate final slag generation, target final slag design, and batching calculations.

[0025] Candidate final slag composition refers to the optional final slag composition obtained through steel slag balance calculations based on the target molten steel composition, refining temperature, and candidate contents of CaO and MgO. The target final slag design result can include a single target final slag composition or a range of target final slag compositions determined by multiple candidate final slag compositions.

[0026] Target batching constraints refer to constraints formed based on target final slag composition or target final slag composition range, used to limit the target content or target content range that each slag component, CaO, SiO2, Al2O3, and MgO, should meet after batching. Component contribution refers to the effective contribution of the batching materials to the formation of the corresponding slag components under the action of batching calculation correction parameters after being added in the corresponding amounts.

[0027] The actual amount of raw materials added refers to the amount of each raw material actually added during the current furnace production process. The actual final slag composition refers to the final slag composition obtained through sampling, online testing, or other slag composition detection methods after LF refining according to the corresponding batching scheme. The final slag composition deviation refers to the deviation between the actual final slag composition and the target final slag design result, used to evaluate the degree to which the current furnace meets the target final slag design requirements.

[0028] Batching calculation correction parameters can refer to parameters used in the calculation of the amount of batching materials added, which are incorporated into the target final slag design results, to correct the difference between the theoretical material contribution and the actual production results. Examples include the batching material yield coefficient, the effective contribution correction coefficient, and / or the slag correction parameter for tapping. Parameter estimates can refer to the batching calculation correction parameter values ​​obtained by back-calculating or estimating based on the batching material composition of one or more historical heats, the actual amount of batching materials added, and the actual final slag composition, according to the mass conservation relationship of slag components.

[0029] Figure 1 This is a flowchart illustrating the design and batching optimization method for the quaternary final slag in the LF refining of aluminum-killed steel provided in this application. (See also...) Figure 1The method provided in this application embodiment may include steps S100 to S800. Steps S100 to S500 are used to form the target final slag design result; step S600 uses the target final slag design result as the batching target, and directly calculates the amount of batching materials added based on the batching material information and batching calculation correction parameters; step S700 obtains the actual amount of batching materials added and the actual final slag composition and determines the final slag composition deviation; step S800 updates the batching calculation correction parameters based on the actual batching data and the actual final slag composition. The updated batching calculation correction parameters are returned to step S600 for subsequent furnace cycles.

[0030] Step S100: Obtain the target steel composition, target refining temperature, CaO value range, and MgO value range.

[0031] In step S100, the target steel composition can be obtained from the steel grade process specifications, production plan, internal control composition standards for molten steel, or the process settings for the current heat. The refining temperature can be the target refining temperature, the design temperature, or the temperature used for slag balance calculations. The CaO and MgO value ranges can be determined based on one or more of the following: steel grade process requirements, historical qualified heats, slag fluidity requirements, inclusion control requirements, and furnace lining protection requirements.

[0032] In one specific embodiment, taking GCr15 aluminum-killed bearing steel as an example, the refining temperature can be 1873 K, the target Al content in the molten steel can be 0.03 wt%, the target Si content in the molten steel can be 0.25 wt%, the MgO content can be 4 wt%, and the candidate CaO content can vary within the range of 50 wt% to 61 wt%. The above values ​​are used to illustrate a specific calculation process; under other aluminum-killed steel grades or other production conditions, they can be re-determined according to the corresponding process requirements.

[0033] Step S200: Based on the steel slag balance relationship, multiple quaternary candidate final slag components are formed.

[0034] In step S200, multiple candidate CaO contents can be selected within the CaO value range, and multiple candidate MgO contents can be selected within the MgO value range. Then, multiple combinations of candidate CaO and candidate MgO contents are formed according to preset combination rules. For each candidate content combination, based on the target steel composition, target refining temperature, and the steel-slag balance relationship between aluminum-killed steel and the CaO-SiO2-Al2O3-MgO quaternary slag, the corresponding Al2O3 and SiO2 candidate contents are determined, thereby forming multiple quaternary candidate final slag compositions.

[0035] In one implementation, the activity of at least one component in the molten steel can be determined first based on the target molten steel composition. Then, based on the target refining temperature, the activity of the molten steel component, the activity of the slag component, and the corresponding steel-slag equilibrium reaction relationship, the candidate contents of Al2O3 and SiO2 corresponding to the current candidate contents of CaO and MgO can be obtained. The activity of the molten steel component can be determined using the Wagner interaction coefficient model; the activity of the slag component can be determined using a slag activity model based on the molecular-ion coexistence theory.

[0036] Specifically, corresponding steel-slag equilibrium reaction relationships can be established for Al and Si in molten steel. Based on the activities of Al and Si in the target molten steel, the equilibrium relationship corresponding to the refining temperature, and the activities of the corresponding slag components in the candidate final slag, the Al₂O₃ and SiO₂ contents that satisfy the aforementioned steel-slag equilibrium reaction relationships can be calculated. The above calculations are performed for each group of CaO and MgO candidate contents, thereby obtaining the Al₂O₃ and SiO₂ candidate contents corresponding to different combinations of CaO and MgO candidate contents.

[0037] The purpose of the above steel slag balance calculation is to determine the corresponding Al2O3 and SiO2 contents under the condition of varying CaO and MgO contents, given a target molten steel state and refining temperature, so that each group of candidate final slag components has a steel slag balance basis corresponding to the current aluminum killed steel refining conditions.

[0038] Step S300: Determine the slag viscosity, slag surface tension, and steel slag balance evaluation parameters.

[0039] In step S300, for each candidate final slag composition obtained in step S200, the slag viscosity, slag surface tension, and steel-slag balance evaluation parameters can be calculated separately. The steel-slag balance evaluation parameters may include at least one of the following: equilibrium oxygen content in molten steel, equilibrium calcium content in molten steel, and equilibrium magnesium content in molten steel.

[0040] In one embodiment, slag viscosity can be obtained using a slag viscosity calculation model. This model can be established by fitting thermodynamic calculation results, or it can employ other models capable of determining slag viscosity based on slag component content and temperature. Slag surface tension can be obtained using a slag surface tension calculation model; in one example, this model can be established by combining slag coexistence theory and the Butler equation. The equilibrium oxygen, equilibrium calcium, and equilibrium magnesium in the molten steel can be obtained using the aforementioned steel-slag equilibrium relationship calculation model.

[0041] Under the GCr15 example conditions described above, some candidate final slag compositions and their physicochemical properties are shown in Table 1 below. The data in Table 1 below are used to show the corresponding relationships between Al2O3, SiO2, slag surface tension, and slag viscosity as the CaO candidate content changes with the candidate final slag composition.

[0042] Table 1 Examples of candidate final slag compositions and physicochemical properties for GCr15 aluminum-killed steel

[0043] Note: The MgO content in the candidate final residues shown in Table 1 is fixed at 4 wt%.

[0044] As shown in Table 1, while keeping the target refining temperature, target Al content, target Si content, and MgO content constant, changing the candidate CaO content yields different combinations of Al2O3 and SiO2, corresponding to different slag viscosities and surface tensions. Therefore, steps S200 to S300 can form a set of candidate final slags with different physicochemical characteristics, providing evaluation objects for subsequent process constraint screening and multi-objective comprehensive evaluation.

[0045] Step S400: Screen candidate final residue components according to preset process constraints.

[0046] Figure 2 A schematic diagram illustrating the screening of candidate final residue processes and the multi-objective comprehensive scoring method provided in this application. See also... Figure 2 Before conducting multi-objective comprehensive evaluation, multiple quaternary candidate final slag components are screened based on preset process constraints. These preset process constraints determine whether the candidate final slags possess the basic process feasibility for subsequent comprehensive evaluation and may include at least one of the following: slag component content constraints, slag viscosity constraints, steel slag equilibrium evaluation parameter constraints, and phase diagram phase region constraints. Specifically, slag component content constraints can be used to limit the allowable content range of at least one component among CaO, SiO2, Al2O3, and MgO; slag viscosity constraints can be used to exclude candidate final slags with excessively high or low viscosity that are difficult to meet the current refining process requirements; and steel slag equilibrium evaluation parameter constraints can be used to determine whether the equilibrium oxygen, equilibrium calcium, or equilibrium magnesium in the molten steel is within the corresponding process allowable range. For phase diagram phase region constraints, each candidate final slag component can be mapped to an isothermal liquidus phase diagram of the CaO-SiO2-Al2O3-MgO quaternary slag system corresponding to the current refining temperature, determining the corresponding phase region position, and judging whether the phase region position meets the preset target phase region requirements. Phase diagrams can be obtained from a pre-established phase diagram database or through thermodynamic calculations.

[0047] It should be noted that the preset process constraints and the subsequent comprehensive scoring serve different purposes. The preset process constraints are used to eliminate candidate final residues that do not possess basic process feasibility; the subsequent comprehensive scoring is used to compare the overall matching degree of candidate final residues that meet the basic process constraints with multiple process objectives. Therefore, higher scores in each scoring dimension are not used to compensate for candidate final residues that do not meet the necessary process constraints.

[0048] Step S500: Perform multi-objective comprehensive scoring and determine the final target design result.

[0049] See also Figure 2 In step S500, candidate final slag components that meet preset process constraints are selected as candidate final slag components for comprehensive scoring. At least two evaluation parameters can be selected as scoring dimensions from the parameters included in the slag viscosity, slag surface tension, and steel slag equilibrium evaluation parameters; in some embodiments, phase diagram evaluation parameters determined based on the phase region position of the candidate final slag in the isothermal liquidus phase diagram can also be used as scoring dimensions.

[0050] For each scoring dimension, a corresponding target value or target range can be preset. For scoring dimensions with target values, sub-scores can be determined based on the deviation between the evaluation parameters corresponding to the candidate final slag and the target value; for scoring dimensions with target ranges, sub-scores can be determined based on the degree of matching between the evaluation parameters corresponding to the candidate final slag and the target range.

[0051] Since slag viscosity, surface tension, equilibrium element content in molten steel, and phase diagram evaluation parameters may have different dimensions and numerical ranges, the sub-scores of each scoring dimension can be normalized to ensure they are on a uniform scoring scale. The following provides an optional implementation method to illustrate the scoring process. For the first sub-score with a target value... Each scoring dimension can determine sub-scores based on the relative deviation between the candidate final residue evaluation parameters and the target value.

[0052]

[0053] in, Indicates the first The final candidate was in Normalized sub-scores under each scoring dimension; Indicates the first The final candidate was in Evaluation parameters under each scoring dimension; Indicates the first The target values ​​corresponding to each scoring dimension; This represents the normalized scaling parameter used to unify the bias scale, and >0; Represents the sensitivity coefficient and >0; In subsequent embodiments, the normalized sub-scores at a unified scoring scale are directly obtained through the above scoring mapping. The above exponential mapping is only one implementation method; linear mapping, piecewise mapping, or other mapping methods that can convert different evaluation parameters into a unified scoring scale can also be used.

[0054] For those with target range The The rating dimension, when the first Evaluation parameters corresponding to each candidate final residue When the target range is within, the corresponding sub-score can be... Set to 1; when the evaluation parameter When the target range is outside the specified range, the corresponding sub-score can be determined based on the deviation between the evaluation parameter and the nearest range boundary. For example, it can be determined according to the following formula:

[0055] in, and They represent the first The lower and upper limits of the target range corresponding to each scoring dimension. This represents the normalized scaling parameter used to unify deviations from the scale, and >0, Represents the sensitivity coefficient and >0. Therefore, a higher sub-score is obtained when the evaluation parameter is within the target range, and the sub-score decreases as the deviation from the nearest range boundary increases when the evaluation parameter deviates from the target range. The above exponential mapping is only one possible implementation; in other embodiments, linear mapping, piecewise linear mapping, or other mapping methods that decrease the sub-score as the deviation increases can also be used.

[0056] After normalizing the sub-scores, the comprehensive score of the candidate final slag can be obtained based on the weight combination corresponding to each scoring dimension. In one implementation, the first... The overall score of each candidate final scorer can be expressed as: ,and ,in, Indicates the first The overall score of each candidate for final scum This represents the corresponding normalized sub-score. Indicates the first The weights corresponding to each rating dimension This represents the total number of scoring dimensions involved in the overall score, with each weight satisfying 0 ≤ 0. ≤1. The weight of each scoring dimension can be preset according to the current refining process target of aluminum-killed steel, or it can be calibrated according to the statistical results of historical qualified heats.

[0057] The target final slag design result can be determined based on the comprehensive score. When a single target final slag component needs to be output, the candidate final slag components participating in the comprehensive score can be sorted according to the comprehensive score, and the candidate final slag component with the highest comprehensive score can be determined as the target final slag component. When a range of target final slag components needs to be output, multiple candidate final slag components whose comprehensive scores meet the preset scoring conditions can be determined as target final slag candidate components, and then the target content range of the corresponding component can be determined according to the content distribution of each slag component in the multiple target final slag candidate components.

[0058] For example, the minimum and maximum contents of CaO, SiO2, Al2O3, and MgO in multiple target final residue candidate components can be obtained separately, and the minimum to maximum contents corresponding to each component can be determined as the target content range for that component. The above comprehensive score reflects the comprehensive matching degree between the candidate final residue and the preset process target under the current candidate set, scoring dimension, and weight configuration, and is used to compare each candidate final residue under the candidate set and preset evaluation rules.

[0059] Step S600: Based on the target final slag design results and batching material information, calculate the amount of batching materials to be added and generate a batching scheme.

[0060] Figure 3 This is a schematic diagram illustrating the actual final slag feedback and the correction of the batching calculation parameters provided in this application. See also... Figure 1 and Figure 3 In step S600, the target final slag design result is input into the batching calculation model. The batching calculation model also acquires the batching material information and the currently used batching calculation correction parameters, and forms target batching constraints corresponding to each slag component (CaO, SiO2, Al2O3, and MgO) based on the target final slag design result. Under the constraints of available material quantity and slag component mass conservation, the addition amount of each batching material is directly solved. Therefore, the target final slag design result is directly used as the target condition for batching solution, eliminating the need for separate steps to predict the generation of final slag components and compare them with the target final slag design result.

[0061] The information on the ingredients should at least include the type, composition, and available quantity of each ingredient, and may also include unit price, allowable addition range, etc. Ingredients may include lime, refining slag, pre-melted slag, fluorite, dolomite, bauxite, or other slag-forming materials that can provide the target slag components. For the first... For this type of ingredient, the mass fraction of each component, CaO, SiO2, Al2O3, and MgO, can be recorded.

[0062] To illustrate how the target final slag design result establishes a direct calculation relationship with the amount of each ingredient added, we can... The amount of each ingredient added is denoted as ,in, , This indicates the number of material types involved in the current batching calculation, and will be the number of batching materials in the current batch. The first ingredient in the batch The mass fraction of the slag components is denoted as , Indicates the number of the target slag component, and will the first The first ingredient material for the first The effective recovery rate or effective contribution correction factor of the slag component is denoted as: The slag from the tapping belt or other existing slag will be used to treat the first... The modified contribution of each slag component is denoted as In one implementation, the first... Contribution of slag components It can be represented as:

[0063] in, This is a dimensionless correction factor. and Adopted and The corresponding mass units. The above relationship indicates that, given the composition of the ingredients and the correction parameters for ingredient calculation, the amount of each ingredient added collectively determines the effective contribution of each slag component: CaO, SiO2, Al2O3, and MgO. For the final slag design expressed in quaternary normalized form, the sum of the contributions of the components corresponding to CaO, SiO2, Al2O3, and MgO can be taken as the total contribution of the quaternary component, and the number of components can be determined. Contribution ratio of various slag components :

[0064] in, This represents the set of components composed of four target slag elements: CaO, SiO2, Al2O3, and MgO. This indicates the first element in the set of elements. Slag components, For the first The contribution of each component in the slag. Component contribution ratio. It is used to establish target batching constraints in the batching solution process and is not used as an independent prediction of the final slag output.

[0065] When the target final slag design result is a single target final slag component, it can be based on the first component in the target final slag. Target content of slag components Establish target ingredient constraints. For example, it can require... equal Or, if process tolerances are allowed, require Located in Within the allowable range centered on the process. When the process simultaneously provides a target total slag quantity, the target content can also be converted into the target component mass, and the target batching constraint is used when the contribution of each component reaches the corresponding target component mass.

[0066] When the target final slag design result is within the target final slag composition range, if the first The target content range of the slag components is as follows: Then it can be established The target ingredient constraints. (The above) , , and All measurements use the same mass fraction as the metric. When the target final slag design result is expressed as a mass percentage, it can be converted to the corresponding mass fraction before establishing the target batching constraints. Therefore, both a single target final slag component and a target final slag component range can be directly converted into constraints on the amount of each batching material added.

[0067] When solving the batching problem, at least one of the following constraints can be set: available material quantity constraint, total slag quantity constraint, and single ingredient addition quantity constraint. For example, the first... Amount of each ingredient added It can satisfy ,in Indicates the number of furnaces in the current batch. The available quantity or upper limit of allowable addition of each ingredient. Under the conditions of satisfying the above constraints and the mass conservation relationship of slag components, solve for the addition quantity of one or more ingredients that make CaO, SiO2, Al2O3 and MgO meet the corresponding target ingredient constraints.

[0068] When there are multiple feasible solutions for the amount of ingredients to be added that satisfy the target batching constraint, a recommended batching scheme can be further determined based on the cost of the ingredients. Let the first... The unit price per unit mass of each ingredient is: Then the total material cost corresponding to a feasible solution It can be represented as:

[0069] The optimization objective can be to reduce the total material cost. At least one recommended batching scheme can be determined from multiple feasible solutions that satisfy the target batching constraints. Under different mathematical forms, the specific solution can be obtained using linear programming, mixed integer programming, nonlinear programming, or other constrained optimization methods. This application is not limited to any specific solution algorithm.

[0070] It should be noted that when the target final slag composition range is wide or multiple alternative ingredients exist, the number of feasible solutions satisfying the target batching constraints may increase; under the combined effect of the target batching constraints, available material quantities, and restrictions on the addition of individual materials, only one feasible solution may exist. The batching calculation model determines the corresponding feasible solutions based on the actual constraints and outputs the batching scheme.

[0071] Step S700: Obtain the actual amount of ingredients added and the actual composition of the final residue, and determine the deviation of the final residue composition.

[0072] When performing LF refining according to the batching plan generated in step S600, the actual amount of each batching material added in the current furnace can be recorded. The actual amount added can be obtained from the weighing system, feeding equipment, production execution system, or manually confirmed data. In cases where materials are added, reduced, or replaced during on-site execution, it is preferable to use the actual amount added after execution as the data input for subsequent parameter estimation.

[0073] After LF refining is completed, the actual composition of the final residue can be obtained through sampling and testing, online testing, or other component analysis methods. See also... Figure 3 The target final slag design result and the actual final slag composition are used together to determine the final slag composition deviation, and the target final slag achievement evaluation result for the current furnace is generated based on the final slag composition deviation.

[0074] When the target final slag design result is a single target final slag component, for the first... The composition of a slag component can be determined by the actual content of that component in the final slag. With target content The difference between them is taken as the final residue composition deviation:

[0075] When the target final slag design result is within the target final slag composition range When the actual content is within the target content range, the final residue composition deviation of the component can be set to 0; when the actual content is below the lower limit... At that time, As a deviation in the final residue composition; when the actual content exceeds the upper limit. At that time, This serves as a deviation in the final residue composition. Therefore, the degree of achievement of both single-value and range objectives can be evaluated uniformly.

[0076] In one embodiment, evaluation results can be generated based on the final slag composition deviations of each component (CaO, SiO2, Al2O3, and MgO), or the target final slag achievement evaluation result for the current furnace can be generated based on the absolute value, weighted deviation, or whether the deviation falls within the target range. The evaluation results are used to characterize the achievement of the final slag target for the current furnace.

[0077] Table 2 below shows a recording format for the target final slag design result, actual final slag composition, and final slag composition deviation. This table is used to illustrate the data correspondence and does not require that each furnace must use the same data storage format.

[0078] Table 2 Recording format of target final slag design results, actual final slag composition, and deviation of final slag composition

[0079] The data shown in Table 2 can be compiled according to the historical records of each furnace. The final slag composition deviation is used to evaluate the achievement of the target final slag for the current furnace; when the final slag composition deviation reaches the preset update conditions, the batching calculation correction parameter update in step S800 can be triggered.

[0080] Step S800: Update the batching calculation correction parameters based on the actual batching data and the actual final residue composition.

[0081] See also Figure 3 The batching calculation correction parameter update branch uses the composition of the batching materials, the actual amount of batching materials added, and the actual final slag composition as parameter estimation inputs. The final slag composition deviation is mainly used for evaluating the achievement of the target final slag in the current furnace batch and can also be used to determine whether a parameter update is needed. When performing a parameter update, the parameter estimation values ​​of the batching calculation correction parameters are determined according to the slag component mass conservation relationship based on the actual batching data and the actual final slag composition.

[0082] In one implementation, the ingredient yield coefficient or effective contribution correction coefficient can be used. Parameters for correcting slag stripping during steel tapping At least one of them is used as the parameter to be estimated. For the first Each historical furnace, according to the furnace number Actual amount of each ingredient added The first ingredient in this batch Mass fraction of slag components And the parameters to be estimated, can be used to establish the first Mass conservation relationships of slag components:

[0083] in, This represents a set of parameters consisting of correction parameters calculated from one or more ingredients to be estimated. Indicates in the parameter set According to the following The actual batching data of each historical furnace corresponds to the first Effective contribution of various slag components Indicates the first The first of the historical furnaces The actual amount of each ingredient added; Indicates the first The first of the historical furnaces The first batch of ingredients The mass fraction of each component in the slag. For the implementation of quaternary component normalization, the proportion of each component can be further determined based on its effective contribution.

[0084] in, Indicates in the parameter set Next, according to the first The first batch determined by the actual batching data of the historical furnace. The contribution ratio of the first slag component to the total effective contribution of the quaternary component is given; the meanings of the other symbols are the same as described above. The first slag component is obtained based on actual final slag testing. Each furnace in turn Actual content of slag components After that, it can be used as and The difference between them is used as a parameter to estimate the residuals. (The above...) This is merely an intermediate calculation in the parameter estimation process, used to identify ingredients and calculate correction parameters based on actual production data, and does not constitute an independent prediction step for final slag generation in step S600.

[0085] When estimating parameters based on multiple historical furnace cycles, an objective function for parameter estimation can be constructed. For example, the residuals of multiple historical furnace cycles and multiple slag components can be combined according to the following formula:

[0086] in, Represents the parameter set The corresponding objective function for parameter estimation, This represents the set of historical furnaces that participated in parameter estimation. Indicates the first The weights of various slag components in parameter estimation and ≥0. To reduce To estimate the parameters, we solve for the estimated values ​​of the corrected parameters for the current ingredient calculation. When the actual total slag volume or actual slag mass is available, the actual final slag composition can be directly converted into the actual mass of each slag component, and parameter estimation can be performed based on the difference between the component mass and the component contribution calculated according to the mass conservation relationship.

[0087] When there are many parameters to be estimated, multiple historical heats can be used to jointly determine the parameter estimates. Alternatively, some batching calculation correction parameters can be pre-fixed, and only the yield coefficients of the batching materials that need correction and / or the slag correction parameters of the tapping strip need to be estimated. Parameter estimation can be performed using least squares, recursive least squares, parameter optimization, or other methods that can determine the parameter estimates based on actual batching data and actual final slag composition.

[0088] To avoid excessive changes in the batching calculation correction parameters due to fluctuations in a single furnace batch, one implementation method involves smoothly updating the current parameters and their estimated values. For example:

[0089] in, Indicates the first Correction parameters for batching calculation used in each furnace. Indicates according to the first Furnace number or including the first The parameters estimated based on actual feed data and actual final slag composition from multiple historical furnace batches, including the furnace batches themselves, were determined. This indicates that the updated version is available for the first time. Correction parameters for batching calculation used in each furnace. A smoothing coefficient between 0 and 1. Adjusted... It allows adjustment of the influence of historical parameters and current parameter estimates on the updated parameters.

[0090] The updated yield coefficients of the batching materials and / or the correction parameters for slag removal during tapping are returned to the batching calculation model. For subsequent heats, the batching calculation model uses the updated batching calculation correction parameters to re-establish the mass conservation relationship and target batching constraints between the target final slag design result and the amount of each batching material added, and generates the batching scheme for subsequent heats, thus forming... Figure 1 and Figure 3 The parameter feedback relationship across furnaces is shown.

[0091] Figure 3This further illustrates the relationship between the target final slag achievement evaluation and the updating of batching calculation parameters. The target final slag design result is input into the batching calculation model. Under the action of the current batching calculation correction parameters, the batching calculation model generates a batching scheme. After LF refining, the actual final slag composition is obtained. The target final slag design result and the actual final slag composition are used together to determine the final slag composition deviation and generate the target final slag achievement evaluation result for the current heat. On the other hand, the batching material composition, the actual batching material addition amount, and the actual final slag composition are input into the parameter estimation process. Based on the mass conservation relationship, the parameter estimate value is determined and the batching calculation correction parameters are updated. The two branches are used for the target achievement evaluation of the current heat and the batching calculation parameter correction for subsequent heats, respectively. The updated batching calculation correction parameters are then returned to the batching calculation model.

[0092] The following uses GCr15 aluminum-killed bearing steel as an example to illustrate the data connection between the above steps. First, at a refining temperature of 1873K, the target Al content of 0.03 wt%, the target Si content of 0.25 wt%, the MgO content of 4 wt%, and the CaO candidate range of 50 wt% to 61 wt% are used as inputs. Subsequently, multiple sets of CaO candidate contents are determined within the CaO candidate range, and the corresponding Al2O3 and SiO2 candidate contents are obtained through the steel-slag balance relationship, forming multiple quaternary candidate final slag compositions.

[0093] For each candidate final slag composition, slag viscosity, slag surface tension, and evaluation parameters such as equilibrium oxygen, equilibrium calcium, and / or equilibrium magnesium in the molten steel are calculated. Pre-screening is then performed by combining component content, viscosity, steel-slag equilibrium evaluation parameters, and at least one necessary process constraint from the phase diagram phase regions. For candidate final slags that meet the necessary process constraints, sub-scores for multiple scoring dimensions are calculated and normalized. A comprehensive score is then determined based on the corresponding weights, thereby determining a single target final slag composition or a target final slag composition range. The data shown in Table 1 can be used as input for the composition and physicochemical properties of some of the candidate final slags.

[0094] The batching calculation model obtains the determined target final slag design results, the composition and availability of on-site batching materials, and the current batching calculation correction parameters. For each slag component—CaO, SiO2, Al2O3, and MgO—target batching constraints are established based on the target final slag design results, and mass conservation relationships are constructed according to the content of each material component, the amount of material added, the yield coefficient of batching materials, and the correction parameters for slag removal from the tapping strip. Under the condition of satisfying the available material quantity and other process constraints, the amount of each batching material added is directly solved. When multiple feasible batching schemes exist, the total material cost can be calculated based on the unit price and amount added of each material, and the recommended batching scheme that conforms to the preset selection rules can be selected.

[0095] When completing LF refining according to the recommended batching scheme, record the actual amount of each batching material added on site, and obtain the actual final slag composition after refining. Compare the actual final slag composition with the target final slag design result to obtain the final slag composition deviation corresponding to each component of CaO, SiO2, Al2O3, and MgO, which is used to form the evaluation result of the target final slag achievement for the current furnace.

[0096] For parameter updates, the actual amount of raw materials added, the composition of raw materials, and the corresponding actual final slag composition in the current heat or multiple historical heats can be used as parameter estimation data. The estimated values ​​of the raw material yield coefficient and / or the slag correction parameters for tapping are determined according to the mass conservation relationship described in step S800, and then smoothly updated as needed. The next heat's batching plan is generated using the updated batching calculation correction parameters, thereby allowing the calculated relationship between the target final slag design result and the on-site raw material addition amount to be corrected based on actual production data.

[0097] It should be noted that the target steel composition, target refining temperature, and candidate final slag data in the above GCr15 implementation example are used to illustrate the processing flow of this application; the types of raw materials, inventory, unit price, actual addition amount, and actual final slag test results can be determined according to the specific production heat. This application does not use a fixed combination of raw materials or a fixed addition amount as an implementation condition.

[0098] This application also provides an electronic device. In some embodiments, the electronic device may include a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it can implement the method for designing and optimizing the quaternary final slag and batching of aluminum-killed steel LF refining as described in any of the foregoing embodiments.

[0099] The processor may be one or more of a central processing unit, digital signal processor, application-specific integrated circuit, field-programmable gate array, or other processing unit capable of executing computer programs; the memory may include volatile memory and / or non-volatile memory. The electronic device may communicate with a steel composition data source, slag composition detection equipment, material database, or production management system to obtain the target steel composition, batching material information, and actual final slag composition.

[0100] In some embodiments, this application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the methods in any of the foregoing embodiments. The computer-readable storage medium may be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, solid-state drive, removable memory card, or other tangible medium capable of storing program code.

[0101] The above embodiments are merely illustrative of the technical concept and features of this application, intended to enable those skilled in the art to understand the content of this application and implement it accordingly, and should not be construed as limiting the scope of protection of this application. For those skilled in the art, corresponding modifications, equivalent substitutions, or improvements can be made to the above embodiments without departing from the spirit and principles of this application. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection defined by the claims of this application.

Claims

1. A method for designing and optimizing the batching of quaternary final slag in the LF refining of aluminum-killed steel, characterized in that, include: The target steel composition, target refining temperature, and range of CaO and MgO values ​​in the target final slag are obtained. The target steel composition includes at least Al and Si content. Multiple candidate contents of CaO and MgO are determined within the range of CaO and MgO values. Based on the target steel composition, the target refining temperature, and the steel-slag balance relationship between the aluminum-killed steel and the CaO-SiO2-Al2O3-MgO quaternary slag, the candidate contents of Al2O3 and SiO2 corresponding to each group of candidate contents of CaO and MgO are determined, thus forming multiple candidate final slag compositions of CaO-SiO2-Al2O3-MgO quaternary slag. The slag viscosity, slag surface tension, and steel-slag balance evaluation parameters corresponding to each candidate final slag component are determined respectively. Multiple candidate final slag components are screened according to preset process constraints. At least two parameters are selected as scoring dimensions from the parameters included in the slag viscosity, slag surface tension, and steel-slag balance evaluation parameters. Sub-scores of each candidate final slag component under each scoring dimension are determined respectively. A comprehensive score is determined according to each sub-score and its corresponding weight. The target final slag design result is determined according to the comprehensive score. The target final slag design result includes the target final slag component or the target final slag component range. Obtain the ingredient information, which includes at least the composition and available quantity of each ingredient. Based on the ingredient calculation model with ingredient calculation correction parameters, determine the target ingredient constraints corresponding to each slag component according to the target final slag design results. Under the constraints of available material quantity and slag component mass conservation, solve for the addition amount of each ingredient that satisfies the target ingredient constraints to obtain the ingredient scheme. Obtain the actual final slag composition after LF refining according to the batching scheme, determine the final slag composition deviation between the actual final slag composition and the target final slag design result, update at least one batching calculation correction parameter in the batching calculation model, and use the updated batching calculation correction parameter for the generation of batching schemes for subsequent furnaces.

2. The method for designing and optimizing the quaternary final slag in the LF refining of aluminum-killed steel according to claim 1, characterized in that, Based on the target steel composition, the target refining temperature, and the steel-slag equilibrium relationship between the aluminum-killed steel and the CaO-SiO2-Al2O3-MgO quaternary slag, the candidate Al2O3 content and candidate SiO2 content corresponding to each group of candidate CaO content and candidate MgO content are determined, including: The activity of at least one component in the molten steel is determined based on the target molten steel composition. Based on the target refining temperature, the activity of at least one component in the molten steel, the activity of the slag components, and the corresponding steel-slag equilibrium reaction relationship, the candidate contents of Al2O3 and SiO2 corresponding to the candidate contents of CaO and MgO in each group are determined.

3. The method for designing and optimizing the quaternary final slag in the LF refining of aluminum-killed steel according to claim 1, characterized in that, The steel slag balance evaluation parameters include at least one of the following: balance oxygen content in molten steel, balance calcium content in molten steel, and balance magnesium content in molten steel. The preset process constraints include at least one of the following: slag component content constraints, slag viscosity constraints, steel slag balance evaluation parameter constraints, and phase diagram phase region constraints. The candidate final residue components are screened according to the preset process constraints, including: Map each of the candidate final slag components to the isothermal liquidus phase diagram of the CaO-SiO2-Al2O3-MgO quaternary slag system to determine the phase region position corresponding to each of the candidate final slag components. Candidate final slag components that do not meet the preset process constraints are excluded, while candidate final slag components that meet the preset process constraints are selected as candidate final slag components for comprehensive evaluation.

4. The method for designing and optimizing the quaternary final slag in the LF refining of aluminum-killed steel according to claim 3, characterized in that, The scoring dimensions also include phase diagram evaluation parameters determined based on the phase region position of the candidate final residue component in the isothermal liquidus phase diagram; Determine the sub-scores of each candidate final residue component participating in the comprehensive scoring under each scoring dimension, including: For each of the aforementioned scoring dimensions, obtain the corresponding target value or target range; For a scoring dimension with a target value, the corresponding sub-score is determined based on the deviation between the evaluation parameters corresponding to the candidate final residue components and the target value. For a scoring dimension with a target range, the corresponding sub-score is determined based on the degree of matching between the evaluation parameters corresponding to the candidate final residue components and the target range. The sub-scores are normalized, and the comprehensive score is determined based on the normalized sub-scores and their corresponding weights.

5. The method for designing and optimizing the quaternary final slag in the LF refining of aluminum-killed steel according to claim 4, characterized in that, The target final slag design result is determined based on the comprehensive score, including: The multiple candidate final residue components participating in the comprehensive score are ranked according to the comprehensive score. The candidate final slag component with the highest comprehensive score is determined as the target final slag component; or, multiple candidate final slag components whose comprehensive scores meet the preset scoring conditions are determined as target final slag candidate components, and the target content range of the corresponding slag component is determined according to the content distribution of each slag component in the multiple target final slag candidate components, so as to obtain the target final slag component range.

6. The method for designing and optimizing the quaternary final slag in the LF refining of aluminum-killed steel according to claim 1, characterized in that, Based on the target final slag design results, determine the target batching constraints corresponding to each slag component, and solve for the addition amount of each batching material, including: The amount of each ingredient added is taken as the variable to be solved; Based on the contents of CaO, SiO2, Al2O3 and MgO in each of the ingredients, the amount of each ingredient added and the ingredient calculation correction parameters, the contribution of each ingredient to the slag components CaO, SiO2, Al2O3 and MgO is determined according to the mass conservation relationship of slag components. When the target final slag design result includes the target final slag composition, the target feed constraint of the corresponding slag component is determined according to the target content of each slag component in the target final slag composition. When the target final slag design result includes the target final slag composition range, the target feed constraint of the corresponding slag component is determined according to the target content range of each slag component in the target final slag composition range. Under the constraints of available material quantity and mass conservation of slag components, the amount of each batching material added is determined so that the contribution of each slag component satisfies the corresponding target batching constraint.

7. The method for designing and optimizing the quaternary final slag in the LF refining of aluminum-killed steel according to claim 6, characterized in that, The ingredient information also includes the unit price of each ingredient; When determining the amount of each of the ingredients to be added, at least one of the following constraints is also set: total slag amount constraint and single ingredient addition amount constraint. When there are multiple feasible solutions for the addition amount of each ingredient that satisfies the target ingredient constraint, the corresponding total material cost is determined according to the addition amount and corresponding unit price of each ingredient in each feasible solution. With the goal of reducing the total material cost, at least one recommended ingredient scheme is determined from the multiple feasible solutions.

8. The method for designing and optimizing the quaternary final slag in the LF refining of aluminum-killed steel according to claim 1, characterized in that, After obtaining the actual final residue composition, the process further includes: Determine the deviation of the final slag composition between the actual final slag composition and the target final slag design result, and generate the target final slag achievement evaluation result for the current furnace based on the final slag composition deviation; Obtain the actual addition amount of each of the above-mentioned ingredients in the current furnace or multiple historical furnaces, and based on the composition of each of the above-mentioned ingredients, the actual addition amount of each of the above-mentioned ingredients, and the corresponding actual final slag composition, determine the parameter estimate value of the ingredient calculation correction parameter according to the slag component mass conservation relationship. Update the batching material yield coefficient and / or the steel tapping slag correction parameter in the batching calculation model according to the parameter estimate; The updated yield coefficients of the batching materials and / or the correction parameters for the slag in the tapping strip are used in the batching calculation model for subsequent heats to re-establish the correspondence between the target final slag design result and the amount of each batching material added, and to generate the batching scheme for subsequent heats.

9. An electronic device, characterized in that, It includes a processor and a memory, wherein the memory stores a computer program, and when the computer program is executed by the processor, it implements the method for designing and optimizing the quaternary final slag of aluminum-killed steel LF refining as described in any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the method for designing and optimizing the quaternary final slag of aluminum-killed steel LF refining as described in any one of claims 1 to 8.

Images