A method of steelmaking in a converter based on limestone slagging

Abstract

The present application relates to the technical field of steelmaking, and provides a converter steelmaking method based on limestone slagging, which comprises the following steps: determining the feeding amount of each raw material corresponding to the first batch of slagging materials based on the mass percentage content of silicon element in molten iron and target slag basicity; predicting the smelting process of the next monitoring period in the middle of slagging, dynamically planning the second batch of slagging material adding strategy and the oxygen lance operation strategy coordinated therewith in the next monitoring period based on the smelting process of the next monitoring period; and calculating the total amount of required alkaline components and the total amount of slagging agent respectively based on the current basicity deviation and the current liquidity deviation of the molten pool at the beginning of the later slagging stage. The present application realizes scientific and quantitative control of the whole slagging process, avoids the blindness of traditional experience feeding, realizes predictive batch feeding of the second batch of slagging materials and oxygen lance operation coordination, solves the problem of traditional after-remedy feeding lag, and effectively avoids accidents such as molten pool temperature fluctuation, slag dry-back and spattering.

Description

Technical Field

[0001] This invention belongs to the field of steelmaking technology, specifically relating to a converter steelmaking method based on limestone slag formation. Background Technology

[0002] Converter steelmaking is the core process of modern steel production. Essentially, it involves the oxidation reaction of oxygen with elements such as carbon, silicon, manganese, and phosphorus in molten iron, and the formation of slag with specific physicochemical properties using slag-forming agents. This allows for the control of the steel's composition and temperature. Slag formation is a crucial step in converter steelmaking; limestone, as a traditional slag-forming material, decomposes into CaO and... CaO, as a basic oxide, can react with acidic oxides such as... The reaction forms a stable slag phase. Compared with the direct use of active lime, the use of limestone for slag making has potential advantages such as lower raw material costs and energy saving and consumption reduction in the calcination process.

[0003] Traditional converter steelmaking methods based on limestone slag formation generally rely on traditional experience-driven models, lacking scientific and quantitative basis for key decisions regarding slag-forming material input, and have many shortcomings, as follows:

[0004] (1) In the existing technology, the input of initial slag material in the early stage of slag making, the addition of slag material in the middle stage, and the adjustment of slag material in the later stage are all based on the experience of the operators, such as observing the flame, listening to the sound in the furnace, and relying on the memory of historical operations. A quantitative control system related to key smelting parameters, such as molten pool temperature, slag basicity, and decarburization process, has not been established. As a result, the slag making effect is greatly affected by the skills of the personnel and it is difficult to adapt to the needs of different molten iron composition and different target steel grades.

[0005] (2) In the existing technology, the post-remedial mode is adopted in the middle stage of slag making. Slag is added only after the parameter deviation is detected. It does not predict and control in advance based on the smelting process. Due to the fast reaction rate of the converter, the delayed feeding often misses the best time, resulting in aggravated temperature fluctuations, slag deterioration, and even splashing and drying accidents. At the same time, the amount and timing of each batch of feeding are still set based on experience, resulting in large fluctuations in the final product quality. Summary of the Invention

[0006] This invention provides a converter steelmaking method based on limestone slag formation to solve at least one of the aforementioned technical problems. This invention is achieved through the following technical solution:

[0007] A converter steelmaking method based on limestone slag formation includes the following steps:

[0008] S1. Pour all the molten iron from the ladle into the converter. Based on the mass percentage content of silicon in the molten iron and the target basicity of the slag, determine the amount of each raw material corresponding to the first batch of slag-forming material. Add the raw materials according to the amount of each raw material corresponding to the first batch of slag-forming material, and add all the raw materials corresponding to the first batch of slag-forming material into the converter at the beginning of the slag-forming stage.

[0009] S2. In the middle stage of slag formation, based on the multi-source information of the molten pool collected in each monitoring cycle, the smelting process of the next monitoring cycle is predicted by the converter dynamic thermochemical model. Based on the smelting process of the next monitoring cycle, the second batch of slag-forming material addition strategy and the oxygen lance operation strategy in coordination with it are dynamically planned. The feeding and oxygen lance adjustment are executed according to the planning strategy corresponding to each monitoring cycle.

[0010] S3. Based on the current basicity deviation and current fluidity deviation of the molten pool at the beginning of the later stage of slag forming, calculate the required total amount of basic components and the total amount of slag-forming agent, respectively. Based on the required total amount of basic components and the total amount of slag-forming agent, add the third batch of slag-forming material and add the third batch of slag-forming material to the converter.

[0011] Preferably, before step S1, a step S0 is included to pretreat the limestone, specifically including the following steps:

[0012] S01. Crushing and screening the limestone raw material entering the furnace to control its particle size within a preset range;

[0013] S02. Analyze the composition of the limestone after crushing and screening to determine its calcium carbonate mass percentage.

[0014] S03. Input the particle size distribution and composition analysis results of limestone, as well as the target steel grade information, into the converter dynamic thermochemical model as model initialization parameters.

[0015] Preferably, in step S1, the amount of each raw material to be fed into the first batch of slag-forming material is determined based on the mass percentage content of silicon in the molten iron and the target basicity of the slag. This specifically includes the following steps:

[0016] S11. Based on the mass percentage content of silicon in molten iron and the target basicity of slag, the total CaO requirement for steelmaking is obtained:

[0017] ;in, For the total CaO demand in steelmaking; The molecular weight ratio of silicon dioxide to silicon in molten iron is 2.14. The mass percentage content of silicon in molten iron; The target alkalinity of the slag is the target CaO to SiO2 mass ratio. This represents the amount of molten iron charged, in tons; 1000 is the conversion factor from tons to kilograms.

[0018] S12. Based on the particle size distribution and composition analysis results of limestone, calculate the effective calcium conversion rate of limestone:

[0019] ;in, The effective calcium conversion rate of limestone, The baseline calcium conversion rate for limestone. is the actual volume average particle size of limestone; e is a natural number with a value of 2.71; The reference volume average particle size of limestone is... This represents the mass percentage of calcium carbonate in limestone.

[0020] S13. Based on the molten iron temperature, total CaO demand in steelmaking, and effective calcium conversion rate of limestone, calculate the limestone feed amount corresponding to the first batch of slag-forming materials. ;

[0021] S14. Based on the total CaO demand in steelmaking, the effective calcium conversion rate of limestone, and the amount of limestone fed, calculate the amount of lime and lightly calcined dolomite fed for the first batch of slag-forming materials:

[0022] Lime feeding amount:

[0023] ;in, This refers to the amount of lime added for the first batch of slag-forming materials. The effective CaO content in lime is 85%-92%;

[0024] Lightly calcined dolomite feed amount:

[0025] ;

[0026] ;in, This refers to the amount of lightly calcined dolomite fed into the first batch of slag-forming materials. The coefficient for adding dolomite per ton of molten iron. Add coefficients to the base. This represents the magnesium-silicon correlation coefficient.

[0027] Preferably, in step S13, based on the molten iron temperature, the total CaO requirement for steelmaking, and the effective calcium conversion rate of limestone, a constrained optimization algorithm is used to calculate the limestone feed amount corresponding to the first batch of slag-forming materials:

[0028] Objective function: ;

[0029] Constraints: ;

[0030] in, Temperature is a factor that affects the environment. The temperature of the molten iron. The target temperature for molten iron, It means that one was found. Make the function The value is the smallest.

[0031] Preferably, step S2 includes:

[0032] S21. Collect the molten pool temperature, CO concentration in flue gas, and slag sonar characteristic values ​​for each monitoring cycle during the mid-slag-forming process;

[0033] S22. Input the molten pool temperature, CO concentration in flue gas and slag sonar characteristic values ​​into the trained converter dynamic thermochemical model, and predict the temperature change trajectory of the molten pool and the decarburization reaction process in the next monitoring cycle through the converter dynamic thermochemical model.

[0034] S23. Based on the predicted temperature change trajectory of the molten pool and the decarburization reaction process in the next monitoring cycle, dynamic planning of feeding is carried out for the next monitoring cycle to obtain the optimal total amount of the second batch of slag-forming material, the feeding sequence, and the matching oxygen lance position adjustment strategy for the next monitoring cycle.

[0035] S24. Based on the results of dynamic feeding planning, the second batch of slag-forming material is added to the converter in batches according to the addition sequence, and the oxygen lance position adjustment strategy is executed simultaneously.

[0036] Preferably, the dynamic planning of feeding for the next monitoring cycle in step S23, based on the predicted temperature change trajectory of the molten pool and the decarburization reaction process within the next monitoring cycle, is achieved through the optimization decision module built into the converter dynamic thermochemical model, specifically including the following steps:

[0037] S231. Based on the predicted temperature change trajectory of the molten pool in the next monitoring cycle, the total excess heat load of the next monitoring cycle is calculated by integration, and the amount of limestone added for the second batch of slag-forming material is calculated by combining the endothermic decomposition characteristics of limestone.

[0038] Meanwhile, based on the predicted decarbonization reaction process of the next monitoring cycle, the pre-stored decarbonization rate-slag viscosity relationship model is queried to determine whether sinter needs to be added. If the determination result is that it is necessary, the amount of sinter to be added in the second batch of slag-forming material is calculated; otherwise, no sinter is added.

[0039] S232. Based on the predicted temperature change trajectory of the molten pool and the absolute value of the maximum instantaneous slope of the decarburization reaction process in the next monitoring cycle, calculate the reaction intensity index for the next monitoring cycle. Based on the value of the reaction intensity index and the decision lookup table, dynamically determine the number of batches to be added, the timing of each batch, and the amount of each batch.

[0040] S233. For each batch of feeding action planned in step S232, an oxygen lance position adjustment command is automatically matched. The oxygen lance position adjustment command stipulates that the oxygen lance position must be lowered to the preset enhanced stirring position when feeding begins, and the oxygen lance position must be raised to the preset optimized reaction position after a set delay after feeding ends.

[0041] Preferably, the formula for calculating the amount of limestone added in the second batch of slag-forming material in step S231 is as follows:

[0042] ;in, The amount of limestone added in the second batch of slag-forming material during the i-th monitoring period. Let be the total amount of excess heat that needs to be removed in the i-th monitoring period. The effective calcium conversion rate of limestone, The enthalpy of heat absorption per unit mass during the decomposition of limestone;

[0043] The formula for calculating the amount of sinter added is:

[0044] ;in, The amount of sinter added in the second batch of slag-forming material during the i-th monitoring period. Let be the average rate of the decarbonization reaction process during the i-th monitoring period. The quantitative relationship coefficient for the amount of sinter added is given.

[0045] The formula for calculating the reaction intensity index is:

[0046] ;in, Let i be the index of the severity of the reaction in the i-th monitoring period. The absolute value of the maximum instantaneous slope of the predicted temperature change trajectory over the entire next monitoring period. The absolute value of the maximum instantaneous slope of the predicted decarbonization reaction process over the entire next monitoring period. and These are the weighting coefficients corresponding to the preset temperature change rate and decarburization rate based on the furnace volume and steel grade.

[0047] Preferably, step S3 includes:

[0048] S31. Calculate the total amount of alkaline components required based on the current basicity deviation of the molten pool at the beginning of the later stage of slag formation;

[0049] S32. Calculate the total amount of slagging agent required based on the current fluidity deviation of the molten pool at the beginning of the later stage of slagging;

[0050] S33. Based on the preset distribution ratio of the required total amount of alkaline components, total amount of slag-forming agent, and amount of feed, calculate the amount of limestone, fluorite, and iron oxide scale feed corresponding to the third batch of slag-forming material.

[0051] Preferably, the process also includes step S4: after adding the third batch of slag-forming material, the slag state is finally optimized, and based on the judgment result of whether the physicochemical properties of the final slag meet the target requirements, a decision is made to tap the steel or perform a compensation operation. After performing the steel tapping or compensation operation based on the decision result, the process is verified again until the decision result is to tap the steel, at which point the steel tapping operation is performed.

[0052] Compared with the prior art, the present invention has the following beneficial effects:

[0053] (1) This invention achieves scientific and quantitative control of the entire slag-making process through the complete limestone slag-making steelmaking steps, avoids the blindness of traditional experience-based feeding, ensures the supply and demand balance of CaO in the initial slag-making stage, and lays a good slag phase foundation for subsequent smelting. In the middle stage of slag-making, by collecting multi-source information of the molten pool such as molten pool temperature, CO concentration in flue gas, and slag sonar characteristic value, and combining the converter dynamic thermochemical model to predict the smelting process of the next cycle, the predictive batch feeding of the second batch of slag-making material and the coordinated operation of oxygen lance are realized, which solves the problem of delayed feeding in the traditional "post-event remediation" method and effectively avoids accidents such as molten pool temperature fluctuation, slag drying and splashing. In the later stage of slag-making, the total amount of alkaline components and the total amount of slag-forming agent are calculated based on the molten pool basicity deviation and fluidity deviation, so as to realize the precise feeding of the third batch of slag-making material and ensure that the physicochemical properties of the slag meet the standards. Compared with the traditional experience-driven mode, it significantly improves the stability of steel quality and smelting efficiency, reduces raw material consumption and energy consumption, reduces production accidents caused by operational errors, and enhances the controllability and adaptability of the smelting process.

[0054] (2) The present invention adds a limestone pretreatment step, which not only ensures that the limestone has a sufficient specific surface area in the converter to accelerate the decomposition reaction, but also effectively avoids the problem of the particles being carried away by the airflow if they are too small or the particles being too large and the decomposition is incomplete, thus improving the utilization rate of raw materials. The composition analysis of each batch of limestone is performed by X-ray fluorescence analyzer to accurately obtain the mass percentage of calcium carbonate, which provides accurate data for the calculation of the limestone substitution ratio for lime and the total basicity contribution in the first batch of slag-making materials. This avoids the feeding deviation caused by the different quality and unknown composition of raw materials. The limestone particle size distribution, composition analysis results and target steel grade information are input into the converter dynamic thermochemical model as initialization parameters, so that the prediction and dynamic planning calculation in the middle stage of slag making are more in line with the characteristics of the current batch of raw materials, significantly improving the model prediction accuracy and control reliability, thereby making the feeding decision of the entire slag-making process more accurate, reducing raw material waste, reducing slag-making costs, and improving the stability of the smelting process and the consistency of molten steel quality. Attached Figure Description

[0055] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0056] Figure 1 A flowchart of a converter steelmaking method based on limestone slag formation is provided for an embodiment of the present invention;

[0057] Figure 2 This is a schematic diagram of the temperature change trajectory in an embodiment of the present invention;

[0058] Figure 3 This is a schematic diagram of the decarbonization reaction process in an embodiment of the present invention. Detailed Implementation

[0059] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0060] Example 1: This embodiment of the invention provides a converter steelmaking method based on limestone slag formation, such as... Figure 1-3 As shown, it includes the following steps:

[0061] S1. Pour all the molten iron from the ladle into the converter. Based on the mass percentage content of silicon in the molten iron and the target basicity of the slag, determine the amount of each raw material corresponding to the first batch of slag-forming material. Add the raw materials according to the amount of each raw material corresponding to the first batch of slag-forming material, and add all the raw materials corresponding to the first batch of slag-forming material into the converter at the beginning of the slag-forming stage.

[0062] S2. In the middle stage of slag formation, based on the multi-source information of the molten pool collected in each monitoring cycle, the smelting process of the next monitoring cycle is predicted by the converter dynamic thermochemical model. Based on the smelting process of the next monitoring cycle, the second batch of slag-forming material addition strategy and the oxygen lance operation strategy in coordination with it are dynamically planned. The feeding and oxygen lance adjustment are executed according to the planning strategy corresponding to each monitoring cycle.

[0063] S3. Based on the current basicity deviation and current fluidity deviation of the molten pool at the beginning of the later stage of slag forming, calculate the required total amount of basic components and the total amount of slag-forming agent, respectively. Based on the required total amount of basic components and the total amount of slag-forming agent, add the third batch of slag-forming material and add the third batch of slag-forming material to the converter.

[0064] In this embodiment, the raw materials corresponding to the first batch of slag-forming materials include lime, lightly calcined dolomite, and limestone.

[0065] In this embodiment, the second batch of slag-forming material is limestone, or limestone and sintered ore.

[0066] In this embodiment, the multi-source information of the molten pool includes the molten pool temperature, CO concentration in the flue gas, and slag sonar characteristic value. The molten pool temperature is collected by directly inserting a thermocouple at the end of the auxiliary lance below the surface of the molten iron. The CO concentration in the flue gas is continuously collected online by a laser gas analyzer or mass spectrometer located on the converter flue. The slag sonar characteristic value is collected by a sonar slagging monitoring system installed at the converter mouth. The sonar slagging monitoring system emits sound waves to the surface of the molten pool and receives the echoes. By analyzing the echo intensity, frequency attenuation, and other characteristics, a dimensionless fluidity index is calculated in real time, which is the slag sonar characteristic value. When the slag viscosity is high and the fluidity is poor, the sound waves attenuate significantly when propagating in the viscous slag, resulting in a weak signal, and the sonar slagging monitoring system will output a higher slag sonar characteristic value. When the slag viscosity is low and the fluidity is good, the sound waves can penetrate the fluid slag better, resulting in a strong signal, and the sonar slagging monitoring system will output a lower slag sonar characteristic value.

[0067] In this embodiment, the converter dynamic thermochemical model is a software module embedded in the converter process control computer. It integrates the mass conservation equation, energy conservation equation, and reaction kinetic equation for converter steelmaking, and its key thermodynamic and kinetic parameters have been trained and corrected based on a large amount of historical smelting data. The model takes the multi-source information and operating parameters of the molten pool collected in the current monitoring cycle as the initial input, and takes the next monitoring cycle as the time domain to predict and output the smelting process of the next monitoring cycle. The smelting process includes the temperature change trajectory of the molten pool and the decarburization reaction process.

[0068] In this embodiment, the second batch of slag-forming material addition strategy includes the total amount and timing of the second batch of slag-forming material addition. The oxygen lance operation strategy mainly refers to the oxygen lance position adjustment strategy. Coordination means adjusting the oxygen lance position accordingly and in real time according to the batch and quantity of slag-forming material addition, so as to optimize the stirring intensity of the molten pool, promote the rapid melting and assimilation of the added cold material, and ensure that the decarburization reaction proceeds efficiently and stably.

[0069] In this embodiment, the specific execution process of step S2 includes: based on the temperature change trajectory and decarburization reaction process predicted by the converter dynamic thermochemical model, calculating the total amount of coolant and slag required to control the temperature rise and promote slag formation, and then determining the total amount of limestone and sinter added in the next monitoring cycle. According to the predicted reaction intensity, the total amount added is divided into several batches and the timing and quantity of each batch are determined. At the same time, the optimal oxygen lance position and holding time are matched for each batch of addition action to form a complete dynamic planning strategy. Finally, according to this planning strategy, the batch feeding and oxygen lance linkage operation are automatically executed.

[0070] In this embodiment, the current basicity deviation of the molten pool is the difference between the real-time slag basicity value obtained by rapid analysis of the slag sample extracted by the secondary gun at the current moment and the target slag basicity.

[0071] In this embodiment, the current fluidity deviation of the molten pool is the difference between the slag sonar characteristic value collected in real time by the sonar slag monitoring system and the target fluidity sonar characteristic value set based on the steel grade and smelting stage.

[0072] In this embodiment, the third batch of slag-forming materials includes limestone, fluorite, and iron oxide scale.

[0073] In this embodiment, the early stage of slag formation refers to the period from the start of oxygen blowing in the converter to the point where the oxidation of silicon and manganese elements in the furnace is basically completed and the intense carbon-oxygen reaction has not yet begun. The beginning of this stage is marked by "start of oxygen blowing". The criterion for its end and entry into the middle stage of slag formation is: the concentration of carbon monoxide in the flue gas monitored by the flue gas analysis system ends its initial stable low state and shows a first significant and continuous upward trend. For example, the CO volume concentration rises from less than 5% to more than 10% within 30 seconds. This signifies that easily oxidizable elements such as silicon and manganese in the molten iron have been oxidized in large quantities, the decarburization reaction has begun to dominate, and the furnace heat load is about to reach its peak.

[0074] The mid-stage of slag formation refers to the period from the onset of intense carbon-oxygen reaction to its significant weakening and the molten pool temperature approaching the target endpoint. The beginning of this stage marks the end of the aforementioned early-stage slag formation. The criterion for its end and entry into the late-stage slag formation is: the carbon monoxide concentration monitored by the flue gas analysis system shifts from the peak plateau period to a continuous and rapid decline. For example, the CO volume concentration drops from a peak of over 20% to below 15% within 1-2 minutes. This signifies that the carbon content in the molten steel has been significantly reduced, the decarburization reaction rate has slowed down, the molten pool heating momentum has weakened, and the smelting process has entered the final adjustment stage.

[0075] The late stage of slag formation refers to the period from when the intensity of the carbon-oxygen reaction significantly weakens to when oxygen blowing ends and steel is ready to be tapped. The beginning of this stage marks the end of the aforementioned mid-stage of slag formation. During this stage, fine-tuning the physicochemical properties of the slag, such as alkalinity and fluidity, becomes the core task to ensure the final quality of the molten steel and successful tapping.

[0076] The beneficial effects of the above technical solution are as follows: Through a complete limestone slag-making steelmaking process, scientific and quantitative control of the entire slag-making process is achieved. Specifically, the initial batch of slag-making material is fed based on the mass percentage of silicon in the molten iron and the target basicity of the slag, avoiding the blindness of traditional experience-based feeding and ensuring a balance between CaO supply and demand in the initial slag-making stage, laying a good slag phase foundation for subsequent smelting. In the middle stage of slag-making, multi-source information from the molten pool, such as molten pool temperature, CO concentration in flue gas, and slag sonar characteristics, is collected and combined with a converter dynamic thermochemical model to predict the smelting process of the next cycle, enabling the second batch of slag-making material to be produced. Predictive batch feeding of slag and coordinated oxygen lance operation solve the problem of delayed feeding in traditional "post-event remedial" methods, effectively avoiding accidents such as molten pool temperature fluctuations, slag re-drying, and splashing. In the later stage of slag formation, the total amount of alkaline components and slag-forming agent is calculated based on the deviation of molten pool basicity and fluidity, enabling precise feeding of the third batch of slag-forming material. This ensures that the physicochemical properties of the slag meet the standards. Compared with the traditional experience-driven model, this significantly improves the stability of molten steel quality and smelting efficiency, reduces raw material consumption and energy consumption, reduces production accidents caused by operational errors, and enhances the controllability and adaptability of the smelting process.

[0077] Example 2: Based on Example 1, before step S1, a step S0 is added to perform limestone pretreatment, specifically including the following steps:

[0078] S01. Crushing and screening the limestone raw material entering the furnace to control its particle size within a preset range;

[0079] S02. Analyze the composition of the limestone after crushing and screening to determine its calcium carbonate mass percentage.

[0080] S03. Input the particle size distribution and composition analysis results of limestone, as well as the target steel grade information, into the converter dynamic thermochemical model as model initialization parameters.

[0081] In this embodiment, controlling the particle size within a preset range specifically involves: using a double-roll crusher to perform primary crushing of the blocky limestone, followed by grading via a vibrating screen, to ensure that the limestone particle size entering the furnace is concentrated within the preset range, such as the middle range of 15-60mm. The preset range ensures that the limestone has sufficient specific surface area in the converter to accelerate the decomposition reaction, while effectively preventing the particles from being carried away by the airflow due to being too small or the particles from being too large, resulting in incomplete decomposition.

[0082] In this embodiment, the composition analysis of the limestone after crushing and screening is specifically carried out by: rapidly detecting each batch of limestone raw materials using an X-ray fluorescence analyzer, and using the calcium carbonate mass percentage data directly in step S1 to calculate the limestone substitution ratio for lime, as well as to calculate the total alkalinity contribution of the first batch of slag-forming material.

[0083] In this embodiment, the particle size distribution and composition analysis results of limestone are used as the initialization parameters of the converter dynamic thermochemical model. Specifically, the actual calcium carbonate content, average particle size and other parameters of limestone are entered into the converter dynamic thermochemical model to replace the original default parameters of general limestone in the converter dynamic thermochemical model. This makes the prediction and dynamic programming calculation in step S2 based on a high degree of matching with the characteristics of the current batch of raw materials, which significantly improves the prediction accuracy and control reliability of the model.

[0084] The beneficial effects of the above technical solution are as follows: Based on Example 1, an additional limestone pretreatment step is added. By using a double-roll crusher and vibrating screen, the limestone particle size is controlled within a preset range. This ensures that the limestone has sufficient specific surface area in the converter to accelerate the decomposition reaction, while effectively avoiding the problems of particles that are too small being carried away by the airflow or particles that are too large resulting in incomplete decomposition. This improves the raw material utilization rate. Furthermore, X-ray fluorescence analysis is used to analyze the composition of each batch of limestone, accurately obtaining the percentage of calcium carbonate by mass. This provides accurate data for calculating the limestone substitution ratio and total alkalinity contribution in the first batch of slag-forming material, avoiding issues caused by raw material quality. To address the feeding deviations caused by inconsistent quality and unknown composition, limestone particle size distribution, composition analysis results, and target steel grade information are input into the converter dynamic thermochemical model as initialization parameters, replacing the original general default parameters in the model. This makes the mid-stage prediction and dynamic programming calculations of slag formation more closely aligned with the characteristics of the current batch of raw materials, significantly improving the model's prediction accuracy and control reliability. Consequently, the feeding decisions throughout the slag formation process become more precise, reducing raw material waste and lowering slag formation costs. At the same time, it improves the stability of the smelting process and the consistency of molten steel quality, solving problems such as poor raw material compatibility and inaccurate model predictions caused by the lack of traditional pretreatment.

[0085] Example 3: Based on Example 2, in step S1, the amount of each raw material to be fed into the first batch of slag-forming material is determined based on the mass percentage content of silicon in the molten iron and the target basicity of the slag. This specifically includes the following steps:

[0086] S11. Based on the mass percentage content of silicon in molten iron and the target basicity of slag, the total CaO requirement for steelmaking is obtained:

[0087] ;in, For the total CaO demand in steelmaking; The molecular weight ratio of silicon dioxide to silicon in molten iron is 2.14. The mass percentage content of silicon in molten iron; The target alkalinity of the slag is the target CaO to SiO2 mass ratio. This represents the amount of molten iron charged, in tons; 1000 is the conversion factor from tons to kilograms.

[0088] S12. Based on the particle size distribution and composition analysis results of limestone, calculate the effective calcium conversion rate of limestone:

[0089] ;in, The effective calcium conversion rate of limestone, The baseline calcium conversion rate for limestone. is the actual volume average particle size of limestone; e is a natural number with a value of 2.71; The reference volume average particle size of limestone is... This represents the mass percentage of calcium carbonate in limestone.

[0090] S13. Based on the molten iron temperature, total CaO demand in steelmaking, and effective calcium conversion rate of limestone, calculate the limestone feed amount corresponding to the first batch of slag-forming materials. ;

[0091] S14. Based on the total CaO demand in steelmaking, the effective calcium conversion rate of limestone, and the amount of limestone fed, calculate the amount of lime and lightly calcined dolomite fed for the first batch of slag-forming materials:

[0092] Lime feeding amount:

[0093] ;in, This refers to the amount of lime added for the first batch of slag-forming materials. The effective CaO content in lime is 85%-92%;

[0094] Lightly calcined dolomite feed amount:

[0095] ;

[0096] ;in, This refers to the amount of lightly calcined dolomite fed into the first batch of slag-forming materials. The coefficient for adding dolomite per ton of molten iron. Add coefficients to the base. This represents the magnesium-silicon correlation coefficient.

[0097] In this embodiment, the total CaO requirement for steelmaking is: the mass of silicon dioxide generated from the oxidation of silicon in molten iron, and the total mass of calcium oxide required to neutralize to the target alkalinity.

[0098] In this embodiment, the effective calcium conversion rate of limestone represents the proportion of calcium carbonate in limestone that can ultimately be converted into effective CaO.

[0099] In this embodiment, the baseline calcium conversion rate of limestone is the theoretical mass ratio of pure calcium carbonate decomposition, assuming... The value is 0.56, which means that 100kg of pure calcium carbonate can decompose into a maximum of 56kg of CaO.

[0100] In this embodiment, the reference volume average particle size of limestone represents the optimal limestone particle size, at which the CaO conversion rate is highest.

[0101] In this embodiment, The particle size influence factor is represented by |d-30|. When the actual volume average particle size d of limestone deviates from the reference volume average particle size of limestone by 30 mm, this value decreases, thereby reducing the total conversion rate μ. |d-30| represents the absolute deviation. 0.02 is the attenuation coefficient, which is used to control the attenuation rate.

[0102] In this embodiment, The purity factor is 1.0 when the limestone is 100% pure (α=100). As the purity decreases, the factor decreases linearly.

[0103] In this embodiment, the formula for calculating the amount of lime fed is as follows: the total CaO demand for steelmaking minus the effective CaO contributed by limestone gives the amount of CaO that needs to be provided by lime. Then, divide by the effective CaO content of lime to obtain the actual amount of lime fed.

[0104] In this embodiment, lightly calcined dolomite is mainly used to provide MgO to protect the furnace lining. Its addition amount is proportional to the amount of molten iron charged and the silicon content, thereby... The amount of lightly calcined dolomite to be fed can be calculated.

[0105] In this embodiment, the dolomite addition coefficient per ton of molten iron represents the calculated mass of lightly calcined dolomite required per ton of molten iron, expressed in kg / t. The basic addition coefficient represents the basic requirement of lightly calcined dolomite per ton of molten iron, regardless of the silicon content, expressed in kg / t. The magnesium-silicon correlation coefficient quantifies the influence of the silicon content of molten iron on the demand for lightly calcined dolomite. Its physical meaning is: for every 0.1% increase in the mass percentage of silicon in molten iron, an additional mass of lightly calcined dolomite is required per ton of molten iron. The introduction of the magnesium-silicon correlation coefficient is based on the following metallurgical principle: after silicon in molten iron oxidizes, it generates acidic oxide SiO2, which chemically corrodes the MgO converter lining. The higher the silicon content, the more SiO2 is generated, and the greater the risk of corrosion to the lining. Therefore, it is necessary to add more MgO-containing dolomite to bring the MgO content in the slag to or near saturation, thereby inhibiting corrosion of the lining and even forming a solid protective layer on the lining surface.

[0106] The beneficial effects of the above technical solutions are as follows: The formula for the total CaO requirement in steelmaking is based on parameters such as the silicon content of molten iron, the target basicity of slag, and the amount of molten iron charged, scientifically quantifying the total amount of CaO required for neutralizing SiO2 generated by silicon oxidation, thus avoiding the problem of insufficient or excessive CaO caused by traditional empirical estimation; the formula for the effective calcium conversion rate of limestone comprehensively considers the deviation between the actual particle size and the reference particle size of limestone and the influence of calcium carbonate purity on calcium conversion efficiency, making the conversion rate calculation more consistent with the actual raw material characteristics and improving the accuracy of subsequent feed amount calculations; based on molten iron temperature, total CaO requirement in steelmaking, and effective calcium content of limestone... The conversion rate determines the amount of limestone to be fed. By combining the lime feed formula and the light-burned dolomite feed formula, the total CaO requirement for steelmaking is met, and the slag-making cost is reduced by replacing lime with lime. At the same time, the amount of light-burned dolomite to be fed is linked to the amount of molten iron and the silicon content, which can effectively provide MgO to protect the furnace lining, reduce furnace lining erosion, and extend the service life of the furnace. The entire calculation process forms a complete quantitative logic, which solves the problems of relying on experience for the first batch of slag-making material, inaccurate composition control, and insufficient furnace lining protection in the traditional method. This improves slag-making efficiency and molten steel quality, and reduces production costs and equipment wear.

[0107] Example 4: Based on Example 3, step S13 uses a constrained optimization algorithm to calculate the limestone feed amount corresponding to the first batch of slag-forming materials, based on the molten iron temperature, total CaO demand for steelmaking, and effective calcium conversion rate of limestone.

[0108] Objective function: ;

[0109] Constraints: ;

[0110] in, Temperature is a factor that affects the environment. The temperature of the molten iron. The target temperature for molten iron, It means that one was found. Make the function The value is the smallest.

[0111] In this embodiment, the constrained optimization algorithm is a multi-objective optimization function used to find an optimal limestone feed rate. This aims to minimize the overall deviation between the cost target and the heat balance target; the cost target is to maximize the proportion of limestone replacing lime; the heat balance target is to ensure that the molten iron temperature is as close as possible to the target temperature. As small as possible.

[0112] In this embodiment, This indicates the proportion of CaO that needs to be supplemented by lime. The smaller the value, the better the limestone substitution effect and the lower the cost.

[0113] In this embodiment, the temperature influence factor is an empirical coefficient used to adjust the weight of the two objectives, cost and temperature, in the optimization process. When the temperature deviation is large, a larger value of the temperature influence factor tends to prioritize stabilizing the temperature. When the temperature is close to the target, a smaller value of λ can be used, which focuses more on reducing costs.

[0114] The beneficial effects of the above technical solution are as follows: Objective function: It balances two core objectives: slag formation cost and molten pool heat balance. The temperature influence factor can be adjusted according to actual operating conditions to ensure a flexible balance between prioritizing temperature stabilization and cost reduction. Constraints: This algorithm limits the reasonable range of CaO contribution from limestone, ensuring effective substitution of limestone for lime while avoiding difficulties in slag formation or insufficient slag production due to excessive substitution. The algorithm can find the optimal amount of limestone to feed, enabling the slag-making process to reduce costs while maintaining stable molten pool temperature. It solves the problem that traditional limestone feeding methods cannot simultaneously balance cost and heat balance, improves the economics of slag-making and the stability of the smelting process, further enhances the controllability of molten steel quality, and lays a better initial foundation for subsequent slag-making stages.

[0115] Example 5: Based on Example 1, step S2 includes:

[0116] S21. Collect the molten pool temperature, CO concentration in flue gas, and slag sonar characteristic values ​​for each monitoring cycle during the mid-slag-forming process;

[0117] S22. Input the molten pool temperature, CO concentration in flue gas and slag sonar characteristic values ​​into the trained converter dynamic thermochemical model, and predict the temperature change trajectory of the molten pool and the decarburization reaction process in the next monitoring cycle through the converter dynamic thermochemical model.

[0118] S23. Based on the predicted temperature change trajectory of the molten pool and the decarburization reaction process in the next monitoring cycle, dynamic planning of feeding is carried out for the next monitoring cycle to obtain the optimal total amount of the second batch of slag-forming material, the feeding sequence, and the matching oxygen lance position adjustment strategy for the next monitoring cycle.

[0119] S24. Based on the results of dynamic feeding planning, the second batch of slag-forming material is added to the converter in batches according to the addition sequence, and the oxygen lance position adjustment strategy is executed simultaneously.

[0120] The beneficial effects of the above technical solution are as follows: By refining the intermediate slag-forming steps, precise dynamic control of intermediate slag-forming is achieved. Step S21 collects multi-source information of the molten pool, such as molten pool temperature, CO concentration in flue gas, and slag sonar characteristic values, providing comprehensive and reliable input data for subsequent model prediction and avoiding the limitations of single-parameter monitoring. Step S22 inputs the multi-source information of the molten pool into the trained converter dynamic thermochemical model, which can accurately predict the trajectory of molten pool temperature change and decarburization reaction process in the next cycle, providing a scientific basis for feeding planning and solving the problem of lack of predictability in traditional intermediate feeding. Step S23 performs dynamic feeding planning based on the prediction results, ensuring... The optimal total amount, timing, and oxygen lance position adjustment strategy for the second batch of slag-forming material are determined to ensure a high degree of compatibility between the feeding and smelting process. Step S24 is executed according to the plan, with batch feeding and oxygen lance adjustment, so that the newly added cold material can be quickly melted and assimilated, avoiding accumulation and re-drying. At the same time, it ensures that the decarburization reaction proceeds efficiently and stably. The entire process forms a closed loop of "monitoring-prediction-planning-execution", realizing the intelligent and collaborative mid-term slag-forming. Compared with the traditional experience-based mid-term feeding, it significantly improves the accuracy of molten pool temperature control, slag fluidity stability, and decarburization reaction efficiency, reduces raw material waste and production accidents, and provides a key guarantee for improving the quality of molten steel.

[0121] Example 6: Based on Example 5, step S23, which involves dynamic planning of the feed for the next monitoring cycle based on the predicted temperature change trajectory of the molten pool and the decarburization reaction process within the next monitoring cycle, is implemented through the optimization decision-making module built into the converter dynamic thermochemical model. Specifically, it includes the following steps:

[0122] S231. Based on the predicted temperature change trajectory of the molten pool in the next monitoring cycle, the total excess heat load of the next monitoring cycle is calculated by integration, and the amount of limestone added for the second batch of slag-forming material is calculated by combining the endothermic decomposition characteristics of limestone.

[0123] Meanwhile, based on the predicted decarbonization reaction process of the next monitoring cycle, the pre-stored decarbonization rate-slag viscosity relationship model is queried to determine whether sinter needs to be added. If the determination result is that it is necessary, the amount of sinter to be added in the second batch of slag-forming material is calculated; otherwise, no sinter is added.

[0124] S232. Based on the predicted temperature change trajectory of the molten pool and the absolute value of the maximum instantaneous slope of the decarburization reaction process in the next monitoring cycle, calculate the reaction intensity index for the next monitoring cycle. Based on the value of the reaction intensity index and the decision lookup table, dynamically determine the number of batches to be added, the timing of each batch, and the amount of each batch.

[0125] S233. For each batch of feeding action planned in step S232, an oxygen lance position adjustment command is automatically matched. The oxygen lance position adjustment command stipulates that the oxygen lance position must be lowered to the preset enhanced stirring position when feeding begins, and the oxygen lance position must be raised to the preset optimized reaction position after a set delay after feeding ends.

[0126] In this embodiment, the optimization decision module is a software algorithm unit embedded in the converter dynamic thermochemical model, which is used to receive the prediction data of the converter dynamic thermochemical model and output dynamic programming strategy.

[0127] In this embodiment, in step S231, based on the predicted temperature change trajectory of the molten pool in the next monitoring cycle, the total excess heat load for the next monitoring cycle is calculated by integration. Specifically, the temperature change trajectory of the molten pool in the next monitoring cycle is used as input, and the integral area of ​​the portion of the temperature change trajectory that exceeds the preset target temperature control upper limit in the time domain of the next monitoring cycle is calculated. The physical meaning of this integral area is the total excess heat that needs to be removed in the next monitoring cycle. Let the total excess heat that needs to be removed in the i-th monitoring cycle be... ,like Figure 2 As shown, Figure 2 The curve in the figure represents the trajectory of temperature change. Figure 2 The straight line in the figure represents the preset target temperature. Figure 2 The shaded area represents the integral area between 100s and 150s, which is the total amount of excess heat that needs to be removed in the third monitoring cycle. Figure 2 The monitoring cycle is set to 50 seconds.

[0128] In this embodiment, the pre-stored decarburization rate-slag viscosity relationship model is an empirical model trained based on a large amount of historical smelting data from the converter. It establishes a mapping relationship between the decarburization rate and the slag fluidity index η, where the decarburization rate is the slope of the curve corresponding to the decarburization reaction process. Figure 3The specific process for calculating the amount of sinter to be added is as follows: The average rate of the decarbonization reaction process in the next monitoring cycle, i.e., the slope of the line connecting the points on the curve corresponding to the start and end times of the next monitoring cycle, is input into the decarbonization rate-slag viscosity relationship model. If the output slag viscosity value indicates that the predicted slag viscosity in the next monitoring cycle will be higher than the target threshold for ensuring good dephosphorization and desulfurization efficiency, then the amount of sinter to be added is calculated according to the preset sinter addition amount quantification relationship coefficient, where the sinter addition amount quantification relationship coefficient is the recommended amount of sinter to be added per unit decarbonization rate. If the output slag viscosity value indicates that the predicted slag viscosity in the next monitoring cycle is less than or equal to the target threshold for ensuring good dephosphorization and desulfurization efficiency, then no sinter is added.

[0129] In this embodiment, the decision lookup table defines the feeding patterns corresponding to different numerical ranges of the reaction intensity index, including the number of batches of feeding, the time interval between each batch, and the feeding amount allocation ratio of each batch.

[0130] In this embodiment, the enhanced stirring position is a relatively low oxygen lance position, typically between 1.2 meters and 1.4 meters. The purpose of lowering the lance position is to enhance the impact depth and stirring intensity of the oxygen stream on the molten pool, thereby accelerating the melting and dissolution of newly added cold materials, i.e., speeding up the melting and dissolution of limestone and sinter, and preventing them from accumulating on the slag surface and causing "re-drying".

[0131] In this embodiment, the optimized reaction position is a relatively high oxygen lance position, typically ranging from 1.5 meters to 1.7 meters. The purpose of raising the lance position is to optimize the depth of the impact pit, reduce splashing and iron loss, and ensure that the decarburization reaction can proceed smoothly and efficiently in the later stages of blowing.

[0132] In this embodiment, the delay time refers to the time interval between the completion of a single batch feeding action and the execution of the gun lifting operation, which is usually between 50 seconds and 120 seconds. The purpose of this delay is to provide sufficient time for the newly added cold material to be absorbed and assimilated by the molten pool, and to avoid premature gun lifting which would weaken the stirring intensity and result in insufficient slag formation.

[0133] The beneficial effects of the above technical solution are as follows: This embodiment further improves the scientificity and accuracy of the dynamic planning of slag feeding in the middle stage by using the optimization decision module built into the converter dynamic thermochemical model; Step S231 calculates the total excess heat load based on the integral of the predicted molten pool temperature change trajectory, and determines its addition amount by combining the endothermic characteristics of limestone decomposition, ensuring that excess heat can be accurately removed and avoiding excessively high molten pool temperature. At the same time, it judges whether to add sintered ore by using the decarburization rate-slag viscosity relationship model, and adds slag as needed to improve slag viscosity and improve dephosphorization and desulfurization efficiency; Step S232 is based on the maximum temperature change trajectory and the decarburization reaction process. The instantaneous slope absolute value is used to calculate the reaction intensity index, and combined with the decision lookup table, the batch, timing, and amount of material added are dynamically determined to adapt the material addition to different reaction intensities. Step S233 matches the oxygen lance position adjustment command for each batch of material addition. When adding material, the lance position is lowered to enhance stirring and promote the melting of cold material. After adding material, the lance is raised later to ensure a stable reaction, which solves the problem of the disconnect between the traditional mid-term material addition and oxygen lance operation. This optimized decision module makes the mid-term material addition planning more in line with the dynamic process of smelting, improves the flexibility and adaptability of slag making, further reduces energy consumption and raw material consumption, reduces slag defects, and ensures the stability and consistency of molten steel quality.

[0134] Example 7: Based on Example 6, the total excess heat load of the next monitoring cycle in step S231, combined with the endothermic decomposition characteristics of limestone, is used to calculate the amount of limestone added in the second batch of slag-forming material. The formula is as follows:

[0135] ;in, The amount of limestone added in the second batch of slag-forming material during the i-th monitoring period. Let be the total amount of excess heat that needs to be removed in the i-th monitoring period. The effective calcium conversion rate of limestone, The enthalpy of heat absorption per unit mass during the decomposition of limestone;

[0136] The formula for calculating the amount of sinter added is:

[0137] ;in, The amount of sinter added in the second batch of slag-forming material during the i-th monitoring period. Let be the average rate of the decarbonization reaction process during the i-th monitoring period. The quantitative relationship coefficient for the amount of sinter added is given.

[0138] Based on the predicted temperature change trajectory of the molten pool and the absolute value of the maximum instantaneous slope during the next monitoring cycle, the reaction intensity index for the next monitoring cycle is calculated:

[0139] ;in, Let i be the index of the severity of the reaction in the i-th monitoring period. The absolute value of the maximum instantaneous slope of the predicted temperature change trajectory over the entire next monitoring period. The absolute value of the maximum instantaneous slope of the predicted decarbonization reaction process over the entire next monitoring period. and These are the weighting coefficients corresponding to the preset temperature change rate and decarburization rate based on the furnace volume and steel grade.

[0140] The beneficial effects of the above technical solution are as follows: This embodiment improves the accuracy and operability of feeding decisions by quantifying the calculation of key parameters in the mid-term of slag making through specific formulas; the calculation formula for the amount of limestone added in the second batch of slag making material is based on the total amount of excess heat to be removed in the i-th monitoring cycle, the effective calcium conversion rate of limestone, and the enthalpy of endothermic heat per unit mass of limestone decomposition, which can accurately calculate the required amount of limestone, ensuring that the molten pool temperature is controlled within the target range and avoiding excessive temperature fluctuations; the calculation formula for the amount of sinter added is based on the average rate of the decarburization reaction process in the monitoring cycle and the preset quantitative relationship coefficient of the amount of sinter added, which determines the amount of sinter added as needed, so that the slag viscosity always matches the decarburization reaction requirements and improves the slag making efficiency. The reaction intensity index calculation formula accurately quantifies the reaction intensity by combining the maximum absolute values ​​of the temperature change rate and decarburization rate (i.e., the maximum instantaneous slope absolute value of the temperature change trajectory and the maximum instantaneous slope absolute value of the decarburization reaction process in the entire next monitoring cycle) with weighting coefficients preset according to furnace capacity and steel grade. This provides a reliable basis for planning the batch, timing, and quantity of feed. These calculation formulas transform abstract smelting parameters into directly calculable quantitative indicators, avoiding the subjectivity and errors of traditional experience-based judgments. This makes the mid-term feed calculation more scientific and accurate, further improving the precision of slag condition control, reducing production fluctuations, and ensuring the quality of molten steel.

[0141] Example 8: Based on Example 1, step S3 includes:

[0142] S31. Calculate the total amount of alkaline components required based on the current basicity deviation of the molten pool at the beginning of the later stage of slag formation;

[0143] S32. Calculate the total amount of slagging agent required based on the current fluidity deviation of the molten pool at the beginning of the later stage of slagging;

[0144] S33. Based on the required total amount of alkaline components, the required total amount of slag-forming agent, and the preset distribution ratio of the feed amount, calculate the limestone feed amount, fluorite feed amount, and iron oxide scale feed amount corresponding to the third batch of slag-forming material.

[0145] In this embodiment, the formula for calculating the total amount of alkaline components required is as follows:

[0146] ;in, The total amount of alkaline components required, 1000 is the conversion factor from tons to kilograms; The alkalinity adjustment coefficient is an empirical coefficient preset based on the target steel grade and smelting stage. This represents the current basicity deviation of the molten pool at the beginning of the later stages of slag formation. This refers to the amount of molten iron added.

[0147] In this embodiment, the formula for calculating the total amount of slag-reducing agent required is as follows:

[0148] ;in, The total amount of slag-reducing agent required, where 1000 is the conversion factor from tons to kilograms; The fluidity adjustment coefficient is an empirical coefficient preset based on the target steel grade and smelting stage. This represents the current fluidity deviation of the molten pool at the beginning of the later stages of slag formation.

[0149] In this embodiment, the preset allocation ratio is assumed to be 6:3:1; therefore, the limestone feed amount corresponding to the third batch of slag-forming material is: The amount of fluorite fed in the third batch of slag-forming material: ;

[0150] The amount of iron oxide scale to be fed in the third batch of slag-forming material: .

[0151] The beneficial effects of the above technical solution are as follows: by replacing traditional experience-based adjustments with quantitative calculations, slag defects caused by inaccurate adjustments in the later stages are avoided, ensuring that the physicochemical properties of the slag meet the requirements for steel tapping, guaranteeing the smoothness of the steel tapping process, reducing problems such as slag blocking failure, and avoiding increased costs and steel pollution caused by excessive use of slag-reducing agents, thereby improving the quality of molten steel and reducing production losses.

[0152] Example 9: Based on Example 1, it also includes step S4: After adding the third batch of slag-forming material, the slag state is finally optimized, and based on the judgment result of whether the physicochemical properties of the final slag meet the target requirements, a decision is made to tap the steel or perform a compensation operation. After performing the steel tapping or compensation operation based on the decision result, the process is verified again until the decision result is to tap the steel, at which point the steel tapping operation is performed.

[0153] In this embodiment, step S4 can be implemented in the following way:

[0154] S41. After the third batch of slag-forming material is added to the converter, maintain the current oxygen lance position and continue oxygen blowing. Simultaneously, activate the sonar feedback-oxygen lance linkage control program. This program takes the real-time slag sonar characteristic value collected by the sonar slag-forming monitoring system as input and dynamically fine-tunes the oxygen lance position in a 10-15 second control cycle. The control rules are as follows: when the real-time slag sonar characteristic value is consistently higher than the upper limit of the target value (indicating that the slag is too sticky), lower the oxygen lance position by 0.1-0.2 meters in the next control cycle to enhance the stirring intensity at the bottom of the molten pool and promote the homogenization of slag composition; when the real-time slag sonar characteristic value is consistently lower than the lower limit of the target value (indicating that the slag may be too thin), raise the oxygen lance position by 0.1-0.2 meters to slow down stirring and prevent splashing. Continue running for 2-3 minutes until the slag sonar characteristic value stabilizes within the target range.

[0155] After steps S42 and S41 are completed, the current slag sample is extracted using a secondary gun. The key chemical components of the slag, mainly the slag alkalinity and ferrous oxide content, are rapidly analyzed within 60 seconds using a rapid X-ray fluorescence spectrometer installed near the furnace. At the same time, the sonar slag monitoring system records the current stable sonar characteristic values ​​of the slag.

[0156] S43. The process control computer receives the slag basicity, ferrous oxide content, and slag sonar characteristic values ​​measured in step S42, automatically compares them with the optimal slag target range preset in the database for the current target steel grade, and makes a decision based on the following rules:

[0157] Decision A (Direct Steel Tap): A "Direct Steel Tap" command is generated if and only if the three indicators of slag basicity, ferrous oxide content and slag sonar characteristic value are simultaneously within their respective target ranges.

[0158] Decision B (Making Optimization): If any indicator (such as slag basicity) deviates slightly from the target range, and the other indicators are qualified, a "Making Optimization Required" instruction is generated. Making aims to fine-tune the slag state by utilizing the residual reaction in the molten pool through short-term continuous blowing.

[0159] Decision C (Add slag conditioner): If any indicator (such as ferrous oxide content) deviates significantly from the target range, a "Add slag conditioner" instruction is generated. The system automatically determines the type of slag conditioner to be added and calculates the amount to be added based on the deviated indicator and the amount of deviation. Fluorite is used to adjust alkalinity, and iron oxide scale is used to adjust ferrous oxide content.

[0160] S44. If decision A is made, the oxygen lance should be raised immediately and the blowing should be stopped. Then the converter should be tilted to tap the steel.

[0161] If decision B is made, the oxygen lance is kept in the current optimized position and oxygen is blown for another 40 seconds. After the supplementary blowing is completed, the system automatically returns to step S42, extracts the slag sample again for analysis, and enters the next decision cycle to form a closed-loop control.

[0162] If the decision is C, then based on the calculated amount added, control the feeding system to add the corresponding fluorite or iron oxide scale. After adding, control the oxygen lance position to drop to the enhanced stirring position and blow oxygen to stir for 50 seconds to ensure that the additive melts and reacts quickly. After stirring, the system automatically returns to step S42, re-extracts the slag sample for analysis, and enters the next round of decision cycle.

[0163] The beneficial effects of the above technical solution are as follows: This embodiment achieves final optimization and closed-loop control of the slag state by adding step S4. The sonar feedback-oxygen lance linkage control program takes the real-time slag sonar characteristic value as input and dynamically fine-tunes the oxygen lance position to ensure that the slag fluidity is stable within the target range, avoiding the problem of slag being too sticky or too thin. After optimization, slag samples are extracted by the secondary lance and the composition analysis is completed within 60 seconds using a rapid X-ray fluorescence spectrometer, ensuring that key slag parameters can be obtained in a timely manner, providing a basis for decision-making based on slag alkalinity and ferrous oxide content. The comparison results between the slag sonar characteristic values ​​and the target range are used to make decisions based on three scenarios: direct tapping, optimized supplementary blowing, and addition of slag conditioning agents. Closed-loop control ensures that the slag ultimately meets the standards. This step solves the problem of lack of systematic optimization and precise decision-making in the later stages of traditional slag making, enabling the slag state to reach its optimal state before tapping, further improving the stability of molten steel quality, reducing the risk of steel downgrading due to slag problems, reducing unnecessary supplementary blowing and the use of slag conditioning agents, lowering energy consumption and production costs, and enhancing the reliability and controllability of the smelting process.

[0164] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A converter steelmaking method based on limestone slag formation, characterized in that, Includes the following steps: S1. Pour all the molten iron from the ladle into the converter. Based on the mass percentage content of silicon in the molten iron and the target basicity of the slag, determine the amount of each raw material corresponding to the first batch of slag-forming material. Add the raw materials according to the amount of each raw material corresponding to the first batch of slag-forming material, and add all the raw materials corresponding to the first batch of slag-forming material into the converter at the beginning of the slag-forming stage. S2. In the middle stage of slag formation, based on the multi-source information of the molten pool collected in each monitoring cycle, the smelting process of the next monitoring cycle is predicted by the converter dynamic thermochemical model. Based on the smelting process of the next monitoring cycle, the second batch of slag-forming material addition strategy and the oxygen lance operation strategy in coordination with it are dynamically planned. The feeding and oxygen lance adjustment are executed according to the planning strategy corresponding to each monitoring cycle. S3. Based on the current basicity deviation and current fluidity deviation of the molten pool at the beginning of the later stage of slag forming, calculate the required total amount of basic components and the total amount of slag-forming agent, respectively. Based on the required total amount of basic components and the total amount of slag-forming agent, add the third batch of slag-forming material and add the third batch of slag-forming material to the converter. Step S2 includes: S21. Collect the molten pool temperature, CO concentration in flue gas, and slag sonar characteristic values ​​for each monitoring cycle during the mid-slag-forming process; S22. Input the molten pool temperature, CO concentration in flue gas and slag sonar characteristic values ​​into the trained converter dynamic thermochemical model, and predict the temperature change trajectory of the molten pool and the decarburization reaction process in the next monitoring cycle through the converter dynamic thermochemical model. S23. Based on the predicted temperature change trajectory of the molten pool and the decarburization reaction process in the next monitoring cycle, dynamic planning of feeding is carried out for the next monitoring cycle to obtain the optimal total amount of the second batch of slag-forming material, the feeding sequence, and the matching oxygen lance position adjustment strategy for the next monitoring cycle. S24. Based on the results of dynamic feeding planning, the second batch of slag-forming material is added to the converter in batches according to the addition sequence, and the oxygen lance position adjustment strategy is executed simultaneously. In step S23, based on the predicted temperature change trajectory of the molten pool and the decarburization reaction process in the next monitoring cycle, dynamic planning of the feed for the next monitoring cycle is performed. This is achieved through the optimization decision-making module built into the converter dynamic thermochemical model, and specifically includes the following steps: S231. Based on the predicted temperature change trajectory of the molten pool in the next monitoring cycle, the total excess heat load of the next monitoring cycle is calculated by integration, and the amount of limestone added for the second batch of slag-forming material is calculated by combining the endothermic decomposition characteristics of limestone. Meanwhile, based on the predicted decarbonization reaction process of the next monitoring cycle, the pre-stored decarbonization rate-slag viscosity relationship model is queried to determine whether sinter needs to be added. If the determination result is that it is necessary, the amount of sinter to be added in the second batch of slag-forming material is calculated; otherwise, no sinter is added. S232. Based on the predicted temperature change trajectory of the molten pool and the absolute value of the maximum instantaneous slope of the decarburization reaction process in the next monitoring cycle, calculate the reaction intensity index for the next monitoring cycle. Based on the value of the reaction intensity index and the decision lookup table, dynamically determine the number of batches to be added, the timing of each batch, and the amount of each batch. S233. For each batch of feeding action planned in step S232, an oxygen lance position adjustment command is automatically matched. The oxygen lance position adjustment command stipulates that the oxygen lance position must be lowered to the preset enhanced stirring position when feeding begins, and the oxygen lance position must be raised to the preset optimized reaction position after a set delay after feeding ends.

2. The converter steelmaking method based on limestone slag formation according to claim 1, characterized in that, Before step S1, step S0 is included to pretreat the limestone, specifically including the following steps: S01. Crushing and screening the limestone raw material entering the furnace to control its particle size within a preset range; S02. Analyze the composition of the limestone after crushing and screening to determine its calcium carbonate mass percentage. S03. Input the particle size distribution and composition analysis results of limestone, as well as the target steel grade information, into the converter dynamic thermochemical model as model initialization parameters.

3. The converter steelmaking method based on limestone slag formation according to claim 2, characterized in that, Step S1 determines the amount of each raw material to be fed into the first batch of slag-forming material based on the mass percentage content of silicon in the molten iron and the target basicity of the slag. This includes the following steps: S11. Based on the mass percentage content of silicon in molten iron and the target basicity of slag, the total CaO requirement for steelmaking is obtained: ;in, For the total CaO demand in steelmaking; The molecular weight ratio of silicon dioxide to silicon in molten iron is 2.

14. The mass percentage content of silicon in molten iron; The target alkalinity of the slag is the target CaO to SiO2 mass ratio. This represents the amount of molten iron charged, in tons; 1000 is the conversion factor from tons to kilograms. S12. Based on the particle size distribution and composition analysis results of limestone, calculate the effective calcium conversion rate of limestone: ;in, The effective calcium conversion rate of limestone, The baseline calcium conversion rate for limestone. is the actual volume average particle size of limestone; e is a natural number with a value of 2.71; The reference volume average particle size of limestone is... This represents the mass percentage of calcium carbonate in limestone. S13. Based on the molten iron temperature, total CaO demand in steelmaking, and effective calcium conversion rate of limestone, calculate the limestone feed amount corresponding to the first batch of slag-forming materials. ; S14. Based on the total CaO demand in steelmaking, the effective calcium conversion rate of limestone, and the amount of limestone fed, calculate the amount of lime and lightly calcined dolomite fed for the first batch of slag-forming materials: Lime feeding amount: ;in, This refers to the amount of lime added for the first batch of slag-forming materials. The effective CaO content in lime is 85%-92%; Lightly calcined dolomite feed amount: ； ;in, This refers to the amount of lightly calcined dolomite fed into the first batch of slag-forming materials. The coefficient for adding dolomite per ton of molten iron. Add coefficients to the base. This represents the magnesium-silicon correlation coefficient.

4. The converter steelmaking method based on limestone slag formation according to claim 3, characterized in that, Step S13: Based on the molten iron temperature, the total CaO requirement for steelmaking, and the effective calcium conversion rate of limestone, a constrained optimization algorithm is used to calculate the limestone feed amount corresponding to the first batch of slag-forming materials. Objective function: ; Constraints: ; in, Temperature is a factor that affects the environment. The temperature of the molten iron. The target temperature for molten iron, It means that one was found. Make the function The value is the smallest.

5. A converter steelmaking method based on limestone slag formation according to claim 1, characterized in that, The formula for calculating the amount of limestone added in the second batch of slag-forming material in step S231 is as follows: ;in, The amount of limestone added in the second batch of slag-forming material during the i-th monitoring period. Let be the total amount of excess heat that needs to be removed in the i-th monitoring period. The effective calcium conversion rate of limestone, The enthalpy of heat absorption per unit mass during the decomposition of limestone; The formula for calculating the amount of sinter added is: ;in, The amount of sinter added in the second batch of slag-forming material during the i-th monitoring period. Let be the average rate of the decarbonization reaction process during the i-th monitoring period. The quantitative relationship coefficient for the amount of sinter added is given. The formula for calculating the reaction intensity index is: ;in, Let i be the index of the severity of the reaction in the i-th monitoring period. The absolute value of the maximum instantaneous slope of the predicted temperature change trajectory over the entire next monitoring period. The absolute value of the maximum instantaneous slope of the predicted decarbonization reaction process over the entire next monitoring period. and These are the weighting coefficients corresponding to the preset temperature change rate and decarburization rate based on the furnace volume and steel grade.

6. The converter steelmaking method based on limestone slag formation according to claim 1, characterized in that, Step S3 includes: S31. Calculate the total amount of alkaline components required based on the current basicity deviation of the molten pool at the beginning of the later stage of slag formation; S32. Calculate the total amount of slagging agent required based on the current fluidity deviation of the molten pool at the beginning of the later stage of slagging; S33. Based on the preset distribution ratio of the required total amount of alkaline components, total amount of slag-forming agent, and amount of feed, calculate the amount of limestone, fluorite, and iron oxide scale feed corresponding to the third batch of slag-forming material.

7. A converter steelmaking method based on limestone slag formation according to claim 1, characterized in that, It also includes step S4: after adding the third batch of slag-forming material, the slag state is finally optimized, and based on the judgment result of whether the physicochemical properties of the final slag meet the target requirements, a decision is made to tap the steel or perform a compensation operation. After performing the steel tapping or compensation operation based on the decision result, the process is verified again until the decision result is to tap the steel, at which point the steel tapping operation is performed.

Images