A method and system for predicting the addition of lf refining slagging material

Abstract

This invention belongs to the field of iron and steel metallurgy technology, specifically a method and system for predicting the amount of slag-forming material added in LF refining. It calculates the total slag volume of LF refining based on historical furnace data; constructs a total slag volume dataset; and uses cluster analysis to cluster the data in the total slag volume dataset into N classes, building a total slag volume matrix. The cluster analysis is based on factors including refining path, slag turning amount, and initial sulfur content upon entry into the furnace. Based on the data of the furnace to be predicted, the corresponding total slag volume in the total slag volume matrix is ​​matched. Utilizing the principle of material conservation during smelting, the mass of CaO and Al2O3 required to be added during refining is calculated. Based on the required mass of CaO and Al2O3, the amount of slag-forming material added is calculated. This invention effectively improves the problems of unstable LF slag-forming effect, large fluctuations in auxiliary material consumption, and high labor intensity, and is conducive to standardized management, showing good application prospects in the iron and steel metallurgy field.

Description

Technical Field

[0001] This invention relates to the field of iron and steel metallurgy technology, specifically to a method and system for predicting the amount of LF refining slag-forming material to be added. Background Technology

[0002] LF refining is an important and commonly used steel refining method in steel production. In the steelmaking process, LF plays a crucial role, acting as a buffer between upstream and downstream processes. Currently, steel plants mainly rely on the operational experience of on-site operators to control the amount of slag-forming material added. This slag addition amount fluctuates significantly due to individual variations, leading to instability in slag basicity and fluidity. This negatively impacts the removal of inclusions in molten steel, composition fine-tuning, and ladle age, ultimately failing to meet continuous casting process requirements. Consequently, this can result in nozzle blockage and billet production that does not meet process requirements. Therefore, developing a method and system for predicting the amount of slag-forming material added in LF refining is of great significance for improving and stabilizing product quality and output in steel enterprises, and reducing production costs.

[0003] Currently, research on predicting slag feed addition in refining furnaces typically employs mechanistic models, statistical models, incremental models, and machine learning models. Mechanistic models calculate slag consumption based on the material balance during the smelting process; however, standalone mechanistic models struggle to reflect the actual conditions within the furnace, resulting in low prediction accuracy. Statistical models establish models by identifying relationships between large datasets using classification or multiple regression methods, achieving higher accuracy. However, considering only the connections between data points can lead to results contrary to the underlying mechanism, hindering practical production applications. Incremental models select historical furnaces with similar data conditions to the furnace to be predicted, mimicking and referencing the charging process of similar furnaces. However, the selection of reference furnaces and their adaptability to uncertainty are poor. Machine learning algorithms, such as those using neural networks, require a large, absolutely accurate, and complete dataset for model learning. For most small and medium-sized enterprises, due to data gaps between processes, data accuracy is questionable, and data collection is incomplete, necessitating significant investment in equipment. Summary of the Invention

[0004] To address the problems existing in the prior art, the main objective of this invention is to propose a method and system for predicting the amount of LF refining slag feed added.

[0005] To address the aforementioned technical problems, according to one aspect of the present invention, the present invention provides the following technical solution:

[0006] A method for predicting the amount of LF refining slag feed added includes the following steps:

[0007] S1. Calculate the total slag volume of LF refining based on historical furnace data;

[0008] S2. Construct a total refining slag dataset from the LF refining slag amount, and use cluster analysis to cluster the data in the total refining slag dataset into N classes to build a total refining slag matrix; the basis for cluster analysis includes: refining path, slag turning amount, and initial sulfur content upon entry into the station;

[0009] S3. Based on the data of the furnace to be predicted, match the corresponding total refining slag in the total refining slag matrix, and use the principle of material conservation in the smelting process to calculate the mass of CaO and Al2O3 that need to be added in the refining process.

[0010] S4. Calculate the amount of slag-forming material to be added based on the required mass of CaO and Al2O3.

[0011] In a preferred embodiment of the method for predicting the amount of LF refining slag feed added according to the present invention, in step S1, the total amount of LF refining slag is calculated using the following formula:

[0012] M 总渣 =m1+m2+m3+m4+m5

[0013] In the formula, M 总渣 m1 is the total slag amount in LF refining; m2 is the amount of slag added during tapping; m3 is the amount of slag added to the converter; m4 is the amount of slag-forming materials added during refining (lime, high-Al slag, pre-melted slag); m5 is the amount of deoxidation products from the converter and refining (i.e., the amount of element oxidation during the smelting process).

[0014] As a preferred embodiment of the method for predicting the amount of LF refining slag feed added according to the present invention, wherein: in step S1,

[0015] m1 is obtained by weighing the residual slag after it has been distributed into different ladles by the overhead crane slag-turning worker.

[0016] m2 is obtained from converter report data;

[0017] m3 = a * G, where a is the converter slag discharge coefficient, kg / t, and G is the weight of molten steel, t;

[0018] m4 = m 4石灰 +m 4高Al渣 +m 4预熔渣 m 4石灰 m 4高Al渣 m 4预熔渣 Obtained from refining production reports;

[0019] m5 = m 5Al2O3 +m 5SiO2 m 5Al2O3 and m 5SiO2 The calculation method is shown in the following formula:

[0020] m 5Al2O3=(m BOF-Al-alloy +m BOF-Al-wire +m LF-Al-alloy +m LF-Al-wire -[%Al] 出站 *G*10)*102 / 54

[0021] In the formula, m BOF-Al-alloy The mass of Al element in the Al-containing alloy added after the converter furnace; m BOF-Al-wire The mass of Al element in the Al feed line after the converter furnace; m LF-Al-alloy The mass of Al element in the Al alloy and Al particles added for refining; m LF-Al-wire The mass of Al element fed into the Al wire for refining; [%Al] 出站 This refers to the mass percentage of Al in the molten steel leaving the station.

[0022] m 5SiO2 =(m BOF-Si-alloy +m LF-Si-alloy -[%Si] 出站 *G*10)*60 / 28

[0023] In the formula, m BOF-Si-alloy The mass of Si element in the Si-containing alloy added after the converter furnace; m LF-Si-alloy The mass of Si element added to the Si-containing alloy for refining; [%Si] 出站 This represents the mass percentage of Si element in the molten steel leaving the station.

[0024] In a preferred embodiment of the method for predicting the amount of LF refining slag feed added according to the present invention, in step S2, the total refining slag volume matrix obtained by clustering based on the refining path, slag turning amount, and initial sulfur content upon entry is as follows:

[0025]

[0026] In a preferred embodiment of the method for predicting the amount of LF refining slag-forming material added according to the present invention, wherein: in step S3, the method for calculating the mass of CaO and Al2O3 to be added during the refining process is as follows: m CaO =M 总渣 *(%CaO) 目标 -m1*1000*(%CaO) 目标 -m2*(%CaO) 石灰 -a*G*(%CaO) 下渣 m Al2O3 =M 总渣 *(%Al2O3) 目标 -m1*1000*(%Al2O3) 目标 -a*G*(%Al2O3) 下渣 -m 炉后脱氧Al2O3m 炉后脱氧Al2O3 =[m BOF-Al-alloy +m BOF-Al-wire -[%Al] 进站 [G*10]*102 / 54

[0027] In the formula, (%CaO) 目标 (%Al2O3) 目标 These represent the mass percentages of CaO and Al2O3 in the target slag system, respectively.

[0028] (%CaO) 下渣 (%SiO2) 下渣 These represent the mass percentages of CaO and Al2O3 in the converter slag.

[0029] (%CaO) 石灰 This represents the mass percentage of CaO in lime.

[0030] m BOF-Al-alloy The mass of Al element in the Al-containing alloy added after the converter furnace;

[0031] m BOF-Al-wire The mass of Al element in the Al feed line after the converter furnace;

[0032] [%Al] 进站 The mass percentage of Al element in the molten steel entering the station;

[0033] As a preferred embodiment of the method for predicting the amount of LF refining slag-forming material added according to the present invention, in step S4, the addition of high-Al slag will lead to Ti reversion, and it is not suitable for steel grades with strict requirements on Ti content. Pre-melted slag will lead to B reversion, and it is not suitable for steel grades with strict requirements on B content. Based on historical feeding data and on-site constraints, the slag-forming material structure is divided into two modes: 1) lime and high-Al slag; 2) lime and pre-melted slag; the specific calculation method is as follows:

[0034] 1) Lime and high-Al slag

[0035] High Al slag addition amount m 高Al渣 =m Al2O3 / (%Al2O3) 高Al渣

[0036] Lime addition amount m 石灰 =(m CaO -(%CaO) 高Al渣 *m 高Al渣 ) / (%CaO) 石灰

[0037] 2) Lime and pre-melted slag

[0038] Pre-melted slag addition amount: m 预熔渣 =mAl2O3 / (%Al2O3) 预熔渣

[0039] Amount of lime added: m 石灰 =(m CaO -(%CaO) 预熔渣 *m 预熔渣 ) / (%CaO) 石灰

[0040] In the formula, (%Al2O3) 高Al渣 The mass percentage of Al2O3 in high-Al slag;

[0041] (%CaO) 高Al渣 The mass percentage of CaO in high-Al slag;

[0042] (%CaO) 石灰 This represents the mass percentage of CaO in lime.

[0043] (%Al2O3) 预熔渣 This refers to the mass percentage of Al2O3 in the pre-melted slag.

[0044] (%CaO) 预熔渣 This represents the mass percentage of CaO in the pre-melted slag.

[0045] To address the aforementioned technical problems, according to one aspect of the present invention, the present invention provides the following technical solution:

[0046] A prediction system for the amount of LF refining slag feed added includes:

[0047] The LF refining total slag calculation module calculates the LF refining total slag volume based on historical furnace data.

[0048] The clustering module constructs a total refining slag dataset from the LF refining slag volume. It then uses cluster analysis to cluster the data in the total refining slag dataset into N classes, building a total refining slag volume matrix. The basis for cluster analysis includes: refining path, slag turning amount, and initial sulfur content upon entry into the station.

[0049] The module for calculating the mass of CaO and Al2O3 that need to be added calculates the mass of CaO and Al2O3 that need to be added during the refining process by matching the total refining slag amount in the total refining slag amount matrix with the data of the furnace to be predicted, and using the principle of material conservation in the smelting process.

[0050] The slag-forming material addition calculation module calculates the amount of slag-forming material to be added based on the required mass of CaO and Al2O3.

[0051] The beneficial effects of this invention are as follows:

[0052] This invention proposes a method and system for predicting the amount of slag-forming material added in LF refining. The method calculates the total slag volume in LF refining based on historical furnace data. A dataset of total slag volume is constructed, and cluster analysis is used to group the data into N classes, creating a total slag volume matrix. The cluster analysis is based on factors including refining path, slag turning amount, and initial sulfur content upon entry into the furnace. Based on the data of the furnace to be predicted, the corresponding total slag volume in the total slag volume matrix is ​​matched. Using the principle of material conservation during smelting, the mass of CaO and Al2O3 required for refining is calculated. Based on the required mass of CaO and Al2O3, the amount of slag-forming material added is calculated. This invention effectively improves upon existing methods that address the problems of unstable LF slag-forming effects, large fluctuations in auxiliary material consumption, and high labor intensity. It facilitates standardized management and has promising application prospects in the iron and steel metallurgy field. Detailed Implementation

[0053] The technical solutions described below in conjunction with the embodiments will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0054] This invention proposes a method and system for predicting the amount of LF refining slag feed added. It uses a combination of mechanistic and statistical models to predict the amount of LF refining slag feed added. The total slag volume of historical furnaces is calculated relatively accurately. A clustering method is used to classify the slag volume into N categories according to refining path, slag turning amount, and initial sulfur content at the station to construct a total slag volume matrix. The corresponding total slag volume is matched according to the predicted furnace data. The amount of slag feed added is calculated based on a mechanistic model built according to material conservation.

[0055] This invention provides a method for predicting the amount of LF refining slag feed added, comprising the following steps:

[0056] S1. Calculate the total slag volume of LF refining based on historical furnace data;

[0057] S2. Construct a total refining slag dataset from the LF refining slag amount, and use cluster analysis to cluster the data in the total refining slag dataset into N classes to build a total refining slag matrix; the basis for cluster analysis includes: refining path, slag turning amount, and initial sulfur content upon entry into the station;

[0058] S3. Based on the data of the furnace to be predicted, match the corresponding total refining slag in the total refining slag matrix, and use the principle of material conservation in the smelting process to calculate the mass of CaO and Al2O3 that need to be added in the refining process.

[0059] S4. Calculate the amount of slag-forming material to be added based on the required mass of CaO and Al2O3.

[0060] In one embodiment of the present invention, in step S1, the total amount of LF refining slag is calculated using the following formula:

[0061] M 总渣 =m1+m2+m3+m4+m5

[0062] In the formula, M 总渣 m1 is the total slag amount in LF refining; m2 is the amount of slag added during tapping; m3 is the amount of slag added to the converter; m4 is the amount of slag-forming materials added during refining (lime, high-Al slag, pre-melted slag); m5 is the amount of deoxidation products from the converter and refining (i.e., the amount of element oxidation during the smelting process).

[0063] In one embodiment of the present invention, in step S1,

[0064] m1 is obtained by weighing the residual slag after it has been distributed into different ladles by the overhead crane slag-turning worker.

[0065] m2 is obtained from converter report data;

[0066] m3 = a * G, where a is the converter slag charge coefficient, kg / t, and G is the weight of molten steel, t;

[0067] m4 = m 4石灰 +m 4高Al渣 +m 4预熔渣 m 4石灰 m 4高Al渣 m 4预熔渣 Obtained from refining production reports;

[0068] m5 = m 5Al2O3 +m 5SiO2 m 5Al2O3 and m 5SiO2 The calculation method is shown in the following formula:

[0069] m 5Al2O3 =(m BOF-Al-alloy +m BOF-Al-wire +m LF-Al-alloy +m LF-Al-wire -[%Al] 出站 *G*10)*102 / 54

[0070] In the formula, m BOF-Al-alloy The mass of Al element in the Al-containing alloy added after the converter furnace; m BOF-Al-wire The mass of Al element in the Al feed line after the converter furnace; m LF-Al-alloy The mass of Al element in the Al alloy and Al particles added for refining; m LF-Al-wireThe mass of Al element fed into the Al wire for refining; [%Al] 出站 This refers to the mass percentage of Al in the molten steel leaving the station.

[0071] m 5SiO2 =(m BOF-Si-alloy +m LF-Si-alloy -[%Si] 出站 *G*10)*60 / 28

[0072] In the formula, m BOF-Si-alloy The mass of Si element in the Si-containing alloy added after the converter furnace; m LF-Si-alloy The mass of Si element added to the Si-containing alloy for refining; [%Si] 出站 This represents the mass percentage of Si element in the molten steel leaving the station.

[0073] In one embodiment of the present invention, in step S2, the total refining slag volume matrix obtained by clustering based on refining path, slag turning amount, and initial sulfur content upon entry is as follows:

[0074]

[0075] In one embodiment of the present invention, the method for calculating the mass of CaO and Al2O3 to be added during the refining process in step S3 is as follows:

[0076] m CaO =M 总渣 *(%CaO) 目标 -m1*1000*(%CaO) 目标 -m2*(%CaO) 石灰 -a*G*(%CaO) 下渣

[0077] m Al2O3 =M 总渣 *(%Al2O3) 目标 -m1*1000*(%Al2O3) 目标 -a*G*(%Al2O3) 下渣 -m 炉后脱氧Al2O3

[0078] m 炉后脱氧Al2O3 =[m BOF-Al-alloy +m BOF-Al-wire -[%Al] 进站 [G*10]*102 / 54

[0079] In the formula, (%CaO) 目标 (%Al2O3) 目标 These represent the mass percentages of CaO and Al2O3 in the target slag system, respectively.

[0080] (%CaO) 下渣 (%Al2O3) 下渣 These represent the mass percentages of CaO and Al2O3 in the converter slag.

[0081] (%CaO) 石灰 This represents the mass percentage of CaO in lime.

[0082] m BOF-Al-alloy The mass of Al element in the Al-containing alloy added after the converter furnace;

[0083] m BOF-Al-wire The mass of Al element in the Al feed line after the converter furnace;

[0084] [%Al] 进站 This represents the percentage by mass of Al in the molten steel entering the station.

[0085] In one embodiment of the present invention, in step S4, the addition of high-Al slag leads to Ti reversion, and it is not suitable for steel grades with strict requirements on Ti content. Pre-melted slag reverts to B, and it is not suitable for steel grades with strict requirements on B content. Based on historical feeding data and on-site constraints, the slag-forming material structure is divided into two modes: 1) lime and high-Al slag; 2) lime and pre-melted slag; the specific calculation method is as follows:

[0086] 1) Lime and high-Al slag

[0087] High Al slag addition amount m 高Al渣 =m Al2O3 / (%Al2O3) 高Al渣

[0088] Lime addition amount m 石灰 =(m CaO -(%CaO) 高Al渣 *m 高Al渣 ) / (%CaO) 石灰

[0089] 2) Lime and pre-melted slag

[0090] Pre-melted slag addition amount: m 预熔渣 =m Al2O3 / (%Al2O3) 预熔渣

[0091] Amount of lime added: m 石灰 =(m CaO -(%CaO) 预熔渣 *m 预熔渣 ) / (%CaO) 石灰

[0092] In the formula, (%Al2O3) 高Al渣The mass percentage of Al2O3 in high-Al slag;

[0093] (%CaO) 高Al渣 The mass percentage of CaO in high-Al slag;

[0094] (%CaO) 石灰 This represents the mass percentage of CaO in lime.

[0095] (%Al2O3) 预熔渣 This refers to the mass percentage of Al2O3 in the pre-melted slag.

[0096] (%CaO) 预熔渣 This represents the mass percentage of CaO in the pre-melted slag.

[0097] According to one aspect of the present invention, the present invention provides the following technical solution:

[0098] A prediction system for the amount of LF refining slag feed added includes:

[0099] The LF refining total slag calculation module calculates the LF refining total slag volume based on historical furnace data.

[0100] The clustering module constructs a total refining slag dataset from the LF refining slag volume. It then uses cluster analysis to cluster the data in the total refining slag dataset into N classes, building a total refining slag volume matrix. The basis for cluster analysis includes: refining path, slag turning amount, and initial sulfur content upon entry into the station.

[0101] The module for calculating the mass of CaO and Al2O3 that need to be added calculates the mass of CaO and Al2O3 that need to be added during the refining process by matching the total refining slag amount in the total refining slag amount matrix with the data of the furnace to be predicted, and using the principle of material conservation in the smelting process.

[0102] The slag-forming material addition calculation module calculates the amount of slag-forming material to be added based on the required mass of CaO and Al2O3.

[0103] Using the above embodiments of the present invention, the amount of slag-forming material added to the LF refining process of a certain plant is predicted. Taking a 135t refining furnace as an example, 1000 complete historical data are taken. The mass percentage of each component in the target slag system, converter slag and slag-forming material is shown in Table 1. Table 2 is the clustered matrix of total refining slag volume. The experimental furnace data used for prediction are listed in Table 3. Table 4 shows the implementation effect of each furnace.

[0104] Table 1. Mass percentage (wt%) of each component in the target slag system, converter slag, and slag-forming material.

[0105] Element CaO <![CDATA[Al2O3]]> <![CDATA[SiO2]]> MgO FeO+MnO Target Scum 54-63 25-33 6-11 5-8 ＜1 Converter slag feeding 48 2.5 16 6 18 lime 85 0 2.5 0 0 High Al slag 9.2 68.5 1.4 1.2 0 pre-melted slag 46 40.5 4.3 3.4 0

[0106] Table 2. Clustered Total Refined Slag Volume Matrix

[0107]

[0108] Table 3. Data on the experimental furnaces to be predicted

[0109]

[0110] Table 4. Implementation Results for Each Furnace

[0111]

[0112]

[0113] As can be seen from Table 4, the prediction method for the amount of LF refining slag feed added in this invention can effectively calculate the amount of refining slag feed added. After on-site implementation, it was found that the slag composition after adding the feed according to the amount calculated by this invention was within the target range, and the S content at the outlet met the standard (the measured S content at the outlet was less than the target S content). This is conducive to standardized operation, stable control of slag forming effect, and avoidance of unnecessary cost expenditures.

[0114] The above description is merely a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural transformations made using the contents of the present invention under the inventive concept of the present invention, or direct / indirect applications in other related technical fields, are included within the patent protection scope of the present invention.

Claims

1. A method for predicting the amount of LF refining slag feed added, characterized in that, Includes the following steps: S1. Calculate the total slag volume of LF refining based on historical furnace data; S2. Construct a total refining slag dataset from the total LF refining slag amount, and use cluster analysis to cluster the data in the total refining slag dataset into N classes to build a total refining slag amount matrix. The basis for cluster analysis includes: refining path, slag turning amount, and initial sulfur content upon entry into the station; S3. Based on the data of the furnace to be predicted, match the corresponding total refining slag in the total refining slag matrix, and use the principle of material conservation in the smelting process to calculate the mass of CaO and Al2O3 that need to be added in the refining process. S4. Calculate the amount of slag-forming material to be added based on the required mass of CaO and Al2O3. In step S1, the total amount of LF refining slag is calculated using the following formula: M 总渣 =m1+m2+m3+m4+m5 In the formula, M 总渣 m1 is the total slag amount in LF refining; m2 is the slag turning amount; m3 is the slag added during tapping; m4 is the slag added to the converter; m5 is the amount of slag added during refining; m5 is the amount of deoxidation products from the converter and refining. In step S1 m1 is obtained by weighing the residual slag after it has been distributed into different ladles by the overhead crane slag-turning worker. m2 is obtained from converter report data; m3=a*G, where a is the converter slag amount coefficient, kg / t, and G is the weight of molten steel, t; m4=m 4石灰 + m 4高Al渣 + m 4预熔渣 m 4石灰 m 4高Al渣 m 4预熔渣 Obtained from refining production reports; m5=m 5Al2O3 + m 5SiO2 m 5Al2O3 and m 5SiO2 The calculation method is shown in the following formula: m 5Al2O3 =(m BOF-Al-alloy +m BOF-Al-wire +m LF-Al-alloy +m LF-Al-wire -[%Al] 出站 *G*10)*102 / 54 In the formula, m BOF-Al-alloy The mass of Al element in the Al-containing alloy added after the converter furnace; m BOF-Al-wire The mass of Al element in the Al feed line after the converter furnace; m LF-Al-alloy The mass of Al element in the Al alloy and Al particles added for refining; m LF-Al-wire The mass of Al element fed into the Al feedstock for refining; [%Al] 出站 This refers to the mass percentage of Al in the molten steel leaving the station. m 5SiO2 =(m) BOF-Si-alloy +m LF-Si-alloy -[%Si] 出站 *G*10)*60 / 28 In the formula, m BOF-Si-alloy The mass of Si element in the Si-containing alloy added after the converter furnace; m LF-Si-alloy The mass of Si element added to the Si-containing alloy for refining; [%Si] 出站 This represents the mass percentage of Si element in the molten steel leaving the station.

2. The method for predicting the amount of LF refining slag feed added according to claim 1, characterized in that, In step S2, the total refining slag volume matrix obtained by clustering based on refining path, slag turning amount, and initial sulfur content upon entry is as follows: 。 3. The method for predicting the amount of LF refining slag feed added according to claim 1, characterized in that, In step S3, the method for calculating the mass of CaO and Al2O3 to be added during the refining process is as follows: m CaO =M 总渣 *(%High) 目标 -m1*1000*(%CaO) 目标 -m2*(%CaO) 石灰 -a*G*(%CaO) 下渣 m Al2O3 =M 总渣 *(%Al2O3) 目标 -m1*1000*(%Al2O3) 目标 -a*G*(%Al2O3) 下渣 -m 炉后脱氧Al2O3 m 炉后脱氧Al2O3 =[m BOF-Al-alloy +m BOF-Al-wire -[%Al] 进站 *G*10]*102 / 54 In the formula, (%CaO) 目标 (%Al2O3) 目标 These represent the mass percentages of CaO and Al2O3 in the target slag system, respectively. (%CaO) 下渣 (%Al2O3) 下渣 These represent the mass percentages of CaO and Al2O3 in the converter slag. (%CaO) 石灰 This represents the mass percentage of CaO in lime. m BOF-Al-alloy The mass of Al element in the Al-containing alloy added after the converter furnace; m BOF-Al-wire The mass of Al element in the Al feed line after the converter furnace; [%Al] 进站 This represents the percentage by mass of Al in the molten steel entering the station.

4. The method for predicting the amount of LF refining slag feed added according to claim 1, characterized in that, In step S4, the addition of high-Al slag will cause Ti to return, so it is not suitable for steel grades with strict requirements on Ti content. Pre-melted slag will cause B to return, so it is not suitable for steel grades with strict requirements on B content. Based on historical feeding data and on-site constraints, the slag-forming material structure is divided into two modes: 1) lime and high-Al slag; 2) lime and pre-melted slag. The specific calculation method is as follows: 1) Lime and high-Al slag High Al slag addition amount: m 高Al渣 =m Al2O3 / (%Al2O3) 高Al渣 Amount of lime added: m 石灰 =(m CaO -(%CaO) 高Al渣 *m 高Al渣 ) / (%CaO) 石灰 2) Lime and pre-melted slag Pre-melted slag addition amount: m 预熔渣 =m Al2O3 / (%Al2O3) 预熔渣 Amount of lime added: m 石灰 =(m CaO -(%CaO) 预熔渣 *m 预熔渣 ) / (%CaO) 石灰 In the formula, (%Al2O3) 高Al渣 The mass percentage of Al2O3 in high-Al slag; (%CaO) 高Al渣 The mass percentage of CaO in high-Al slag; (%CaO) 石灰 This represents the mass percentage of CaO in lime. (%Al2O3) 预熔渣 This refers to the mass percentage of Al2O3 in the pre-melted slag. (%CaO) 预熔渣 This represents the mass percentage of CaO in the pre-melted slag.

5. A prediction system for the amount of LF refining slag feed added, used to implement the prediction method for the amount of LF refining slag feed added according to any one of claims 1-4, characterized in that, include: The LF refining total slag calculation module calculates the LF refining total slag volume based on historical furnace data. The clustering module constructs a total refined slag dataset from the LF total refined slag amount, and uses cluster analysis to cluster the data in the total refined slag dataset into N classes, thus constructing a total refined slag amount matrix. The basis for cluster analysis includes: refining path, slag turning amount, and initial sulfur content upon entry into the station; The module for calculating the mass of CaO and Al2O3 that need to be added calculates the mass of CaO and Al2O3 that need to be added during the refining process by matching the total refining slag amount in the total refining slag amount matrix with the data of the furnace to be predicted, and using the principle of material conservation in the smelting process. The slag-forming material addition calculation module calculates the amount of slag-forming material to be added based on the required mass of CaO and Al2O3.

Images