Method, device, computer equipment and storage medium for predicting molten steel temperature in ladle

Abstract

The present invention provides a method, device, computer equipment and storage medium for predicting the temperature of molten steel in a ladle. The method comprises: collecting relevant information of the current heat and historical heats and establishing a historical database; classifying the relevant information; performing big data analysis on the relevant information to obtain a corresponding optimal temperature drop value; inputting the optimal temperature drop value into a symmetrically connected neural network for calculation to obtain a temperature drop prediction value T 网络 ; Perform feedforward neural network calculation to obtain the temperature drop curve under ladle conditions and the temperature drop prediction value T that changes with time 曲线 ; Measure the temperature value T of the end point TSO of this furnace 副枪 、T 网络 and T 曲线 Perform regression analysis to obtain the predicted value of the molten steel temperature in the ladle after tapping T 出钢后 According to T 出钢后 and T 曲线 Obtain the predicted temperature value T of the molten steel in the ladle at any moment after the steel is tapped t .

Description

Technical Field

[0001] The present invention relates to a method, a device, a computer device and a storage medium for predicting the temperature of molten steel in a ladle, and belongs to the technical field of converter steelmaking. Background Art

[0002] In metallurgical production, molten steel temperature directly impacts the smooth progress of continuous casting and other processes. Stable molten steel temperature is crucial for stable continuous casting, and appropriate molten steel superheat is crucial for obtaining high-quality ingots. Therefore, improving molten steel temperature control within the ladle throughout the entire process, from converter tapping to refining outside the furnace and onto the continuous casting turret, and stabilizing the molten steel temperature before continuous casting, is of great significance and research value. From converter tapping to the end of refining, molten steel undergoes lengthy transportation and handling processes, numerous steps, and significant temperature drops. This drop in molten steel temperature directly impacts refining, continuous casting, energy conservation, production process management, and ladle life. The steelmaking process is a typical high-temperature metallurgical process, and precise and effective molten steel temperature control is crucial for smooth production. The ladle, as a container for molten steel, serves as the primary connection between the steelmaking furnace and the continuous casting machine, fulfilling the dual tasks of transporting molten steel and performing refining outside the furnace. The effectiveness of molten steel temperature control within the ladle is a crucial economic and technical indicator in steelmaking operations.

[0003] The factors that influence molten steel temperature within the ladle are complex, including tapping temperature, converter bottom blowing, deoxidation and alloying, tapping time, ladle condition, ladle argon blowing, and ladle insulation. Only by fully understanding these various influencing factors and ladle conditions, and systematically analyzing the patterns that influence molten steel temperature, can we accurately predict and control molten steel temperature. Ladle condition is one of the primary factors affecting molten steel temperature stability. Different ladle conditions can affect the tapping temperature drop, the heating and cooling rates of the molten steel in the refining station, and other factors. These influences are primarily related to the heat absorption of the ladle lining and the heat dissipation of the slag layer above the ladle. Therefore, to effectively control the molten steel temperature, it is also necessary to understand the heat transfer behavior of the molten steel within the ladle and the factors that influence the temperature drop.

[0004] By accurately controlling the molten steel temperature during the steelmaking-continuous casting process, not only can a stable and suitable continuous casting temperature be achieved, but also costs can be reduced, energy can be saved, product quality can be improved, operations can be stabilized, and production efficiency can be increased. Since the molten steel remains in the ladle from the moment it is tapped until the continuous casting process, heat transfer within the ladle directly affects the temperature. Therefore, accurate prediction of the ladle temperature is essential. To this end, this application proposes a method for predicting the ladle temperature. Summary of the Invention

[0005] In order to solve the above problems, the present invention proposes a method, device, computer equipment and storage medium for predicting the temperature of molten steel in a ladle, which can accurately predict the temperature of molten steel in the ladle, realize effective control of the temperature of molten steel throughout the process, and ensure smooth and orderly production rhythm.

[0006] The technical solution adopted by the present invention to solve its technical problems is:

[0007] In a first aspect, an embodiment of the present invention provides a method for predicting the temperature of molten steel in a ladle, comprising the following steps:

[0008] Collect relevant information of the current heat and historical heats and establish a historical database; the relevant information includes tapping temperature, converter bottom blowing, deoxidation and alloying, ladle modifier, tapping conditions, ladle condition, ladle argon blowing and ladle insulation information;

[0009] Classify the relevant information according to the same or similar conditions;

[0010] Conduct big data analysis on relevant information under the same or similar conditions to obtain the corresponding optimal temperature drop value;

[0011] The optimal temperature drop value is used as the input value to perform symmetrical connection neural network calculation to obtain the temperature drop prediction value T 网络 ;

[0012] Perform feedforward neural network calculation to obtain the temperature drop curve under the same or similar ladle conditions as this heat, as well as the temperature drop prediction value T that changes with time under the ladle conditions of this heat 曲线 ;

[0013] The TSO temperature value T of the end point of this furnace is measured 副枪 , temperature drop prediction value T 网络 And the temperature drop prediction value T 曲线 Perform regression analysis to obtain the predicted value of the molten steel temperature in the ladle after tapping the heat, T 出钢后 ;

[0014] According to the predicted value of the molten steel temperature in the ladle after tapping the heat, T 出钢后 And the predicted temperature drop value T under the ladle conditions of this heat over time 曲线 Obtain the predicted temperature value T of the molten steel in the ladle at any moment after the steel is tapped t .

[0015] As a possible implementation of this embodiment, performing big data analysis on relevant information under the same or similar conditions to obtain the optimal temperature drop value corresponding to any condition includes:

[0016] A big data analysis was conducted on the converter bottom blowing, deoxidation and alloying, ladle modifier, and tapping conditions under the same or similar conditions, and the linear regression equations of the converter bottom blowing flow rate, the weight of alloy used for deoxidation and alloying, the amount of ladle modifier added, the tapping condition information and the temperature drop under the same or similar conditions were obtained, thereby obtaining the corresponding optimal temperature drop values ​​T1, T2, T3, and T4 under any conditions.

[0017] As a possible implementation of this embodiment, the tapping condition information includes tapping time and tapping port life.

[0018] As a possible implementation of this embodiment, the optimal temperature drop value is used as an input value to perform symmetrical connection neural network calculation to obtain the temperature drop prediction value T 网络 ,include:

[0019] The optimal temperature drop values ​​T1, T2, T3, and T4 in the historical database with the same or similar conditions as this furnace are used as input values ​​for symmetrical connection neural network calculation, and the corresponding output values ​​are calculated by geometric mean to obtain the temperature drop prediction value T 网络 .

[0020] As a possible implementation of this embodiment, the feedforward neural network calculation is performed to obtain the temperature drop curve under the same or similar ladle conditions as the current heat, and the temperature drop prediction value that changes with time under the ladle conditions of the current heat, including:

[0021] The ladle condition, ladle argon blowing, ladle insulation and their corresponding temperature drop curves are analyzed and classified by big data, and the temperature drop curves corresponding to the historical database with the same or similar conditions as the ladle condition, ladle argon blowing and ladle insulation information of this heat are fitted to obtain their respective fitting curves.

[0022] The three fitting curves of ladle condition, ladle argon blowing and ladle insulation information are used as input layer for feedforward neural network calculation to obtain the temperature drop curve under the same or similar ladle conditions as the current heat;

[0023] Combined with the tapping time and temperature drop curve of this heat, the predicted temperature drop value changing with time under the ladle conditions of this heat is obtained.

[0024] As a possible implementation of this embodiment, the temperature value T of the TSO measurement of the end point of this heat is 副枪 , temperature drop prediction value T 网络 And the temperature drop prediction value T 曲线 Perform regression analysis to obtain the predicted value of the molten steel temperature in the ladle after tapping the heat, T 出钢后 ,include:

[0025] Measure the TSO temperature value T at the end of this furnace 副枪 and calculate the temperature drop prediction value T with the symmetrically connected neural network网络 And the predicted temperature drop value T under the ladle conditions of this heat over time 曲线 Perform regression analysis to obtain the predicted value of the molten steel temperature in the ladle after tapping the heat, T 出钢后 .

[0026] In a second aspect, an embodiment of the present invention provides a device for predicting the temperature of molten steel in a ladle, comprising:

[0027] The information collection module is used to collect relevant information of the current heat and historical heats and establish a historical database; the relevant information includes tapping temperature, converter bottom blowing, deoxidation and alloying, ladle modifier, tapping conditions, ladle status, ladle argon blowing and ladle insulation information;

[0028] An information classification module, used to classify the relevant information according to equal or similar conditions;

[0029] The optimal temperature drop value calculation module is used to perform big data analysis on relevant information under the same or similar conditions to obtain the corresponding optimal temperature drop value;

[0030] The connection neural network calculation module is used to perform symmetrical connection neural network calculation using the optimal temperature drop value as input value to obtain the temperature drop prediction value T 网络 ;

[0031] The feedforward neural network calculation module is used to perform feedforward neural network calculations to obtain the temperature drop curve under the same or similar ladle conditions as this heat, as well as the temperature drop prediction value T under the ladle conditions of this heat that changes with time. 曲线 ;

[0032] Regression analysis module is used to measure the TSO temperature value T at the end of this furnace. 副枪 , temperature drop prediction value T 网络 And the temperature drop prediction value T 曲线 Perform regression analysis to obtain the predicted value of the molten steel temperature in the ladle after tapping the heat, T 出钢后 ;

[0033] The molten steel temperature prediction module is used to predict the molten steel temperature in the ladle after the steel is tapped. 出钢后 And the predicted temperature drop value T under the ladle conditions of this heat over time 曲线 Obtain the predicted temperature value T of the molten steel in the ladle at any moment after the steel is tapped t .

[0034] As a possible implementation of this embodiment, the feedforward neural network calculation module includes:

[0035] The curve fitting module is used to perform big data analysis and classification on the ladle condition, ladle argon blowing, ladle insulation and their corresponding temperature drop curves, and to fit the temperature drop curves corresponding to the historical database with the same or similar conditions as the ladle condition, ladle argon blowing and ladle insulation information of this heat to obtain their own fitting curves.

[0036] The temperature drop curve module is used to use the three fitting curves of ladle condition, ladle argon blowing and ladle insulation information as the input layer to perform feedforward neural network calculation to obtain the temperature drop curve under the same or similar ladle conditions as the current heat;

[0037] The temperature drop prediction value calculation module is used to obtain the temperature drop prediction value that changes with time under the ladle conditions of this heat by combining the steel tapping time and the temperature drop curve of this heat.

[0038] In a third aspect, an embodiment of the present invention provides a computer device comprising a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor. When the computer device is running, the processor communicates with the memory via the bus, and the processor executes the machine-readable instructions to perform the steps of any of the above-mentioned methods for predicting the temperature of molten steel in a ladle.

[0039] In a fourth aspect, an embodiment of the present invention provides a storage medium having a computer program stored thereon, and when the computer program is run by a processor, the steps of any of the above-mentioned methods for predicting the temperature of molten steel in a ladle are executed.

[0040] The beneficial effects of the technical solutions of the embodiments of the present invention are as follows:

[0041] The present invention collects various relevant information about the current and historical heats to establish a historical database. This process involves big data analysis, classification of equivalent or similar conditions, historical data optimization, and temperature drop curve fitting. This process combines the temperature drop values ​​calculated using a symmetrically connected neural network and a feedforward neural network, and finally, through regression analysis, predicts the temperature of the molten steel in the ladle after the current heat has been tapped. After tapping, the prediction model also predicts the temperature of the molten steel in the ladle at any given moment after the current heat has been tapped. This method enables accurate prediction and effective control of the molten steel temperature in the ladle, which is beneficial for reducing process costs, saving energy, improving product quality, stabilizing operations, and increasing production efficiency. It offers significant economic benefits and broad prospects for promotion.

[0042] The present invention adopts a combined prediction model based on the fusion of multiple models and multiple algorithms. It not only accurately predicts the temperature of the molten steel in the ladle after the tapping of the current heat, but also predicts the temperature of the molten steel in the ladle at any moment during the transfer, waiting or refining process after the tapping is completed. It realizes effective control of the molten steel temperature throughout the process and ensures a smooth and orderly production rhythm. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 This is a flow chart of a method for predicting the temperature of molten steel in a ladle according to an exemplary embodiment;

[0044] Figure 2 1 is a schematic structural diagram of a device for predicting the temperature of molten steel in a ladle according to an exemplary embodiment;

[0045] Figure 3 This is a schematic diagram of a specific implementation of predicting the temperature of molten steel in a ladle using the device for predicting the temperature of molten steel in a ladle according to the present invention;

[0046] Figure 4 A schematic diagram of a bottom blowing flow rate curve of a converter bottom blowing in a full argon blowing mode according to an exemplary embodiment;

[0047] Figure 5 A schematic diagram of a bottom blowing flow rate curve of a converter bottom blowing when a full nitrogen-argon switching mode is adopted is shown according to an exemplary embodiment. DETAILED DESCRIPTION

[0048] In order to more clearly illustrate the technical features of the present invention, the present invention is described in detail below through specific implementation methods and in conjunction with the accompanying drawings.

[0049] The thermal insulation performance of a ladle directly impacts the furnace's tapping temperature, ingot quality, ladle wall heat dissipation, ladle lining service life, and production costs. Therefore, studying the temperature drop of molten steel in the ladle during tapping, transportation, and pouring—in order to accurately predict tapping temperatures, maintain stable molten steel temperatures, and achieve low-temperature, constant-temperature pouring—is crucial for improving steel product quality and offers significant economic benefits.

[0050] like Figure 1 As shown, an embodiment of the present invention provides a method for predicting the temperature of molten steel in a ladle, comprising the following steps:

[0051] Collect relevant information of the current heat and historical heats and establish a historical database; the relevant information includes tapping temperature, converter bottom blowing, deoxidation and alloying, ladle modifier, tapping conditions, ladle condition, ladle argon blowing and ladle insulation information;

[0052] Classify the relevant information according to the same or similar conditions;

[0053] Conduct big data analysis on relevant information under the same or similar conditions to obtain the corresponding optimal temperature drop value;

[0054] The optimal temperature drop value is used as the input value to perform symmetrical connection neural network calculation to obtain the temperature drop prediction value T 网络 ;

[0055] Perform feedforward neural network calculation to obtain the temperature drop curve under the same or similar ladle conditions as this heat, as well as the temperature drop prediction value T that changes with time under the ladle conditions of this heat 曲线 ;

[0056] The TSO temperature value T of the end point of this furnace is measured 副枪 , temperature drop prediction value T 网络 And the temperature drop prediction value T 曲线 Perform regression analysis to obtain the predicted value of the molten steel temperature in the ladle after tapping the heat, T 出钢后 ;

[0057] According to the predicted value of the molten steel temperature in the ladle after tapping the heat, T 出钢后 And the predicted temperature drop value T under the ladle conditions of this heat over time 曲线 Obtain the predicted temperature value T of the molten steel in the ladle at any moment after the steel is tapped t .

[0058] As a possible implementation of this embodiment, performing big data analysis on relevant information under the same or similar conditions to obtain the optimal temperature drop value corresponding to any condition includes:

[0059] A big data analysis was conducted on the converter bottom blowing, deoxidation and alloying, ladle modifier, and tapping conditions under the same or similar conditions, and the linear regression equations of the converter bottom blowing flow rate, the weight of alloy used for deoxidation and alloying, the amount of ladle modifier added, the tapping condition information and the temperature drop under the same or similar conditions were obtained, thereby obtaining the corresponding optimal temperature drop values ​​T1, T2, T3, and T4 under any conditions.

[0060] As a possible implementation of this embodiment, the tapping condition information includes tapping time and tapping port life.

[0061] As a possible implementation of this embodiment, the optimal temperature drop value is used as an input value to perform symmetrical connection neural network calculation to obtain the temperature drop prediction value T 网络 ,include:

[0062] The optimal temperature drop values ​​T1, T2, T3, and T4 in the historical database with the same or similar conditions as this furnace are used as input values ​​for symmetrical connection neural network calculation, and the corresponding output values ​​are calculated by geometric mean to obtain the temperature drop prediction value T 网络 .

[0063] As a possible implementation of this embodiment, the feedforward neural network calculation is performed to obtain the temperature drop curve under the same or similar ladle conditions as the current heat, and the temperature drop prediction value that changes with time under the ladle conditions of the current heat, including:

[0064] The ladle condition, ladle argon blowing, ladle insulation and their corresponding temperature drop curves are subjected to big data analysis and classification, and the temperature drop curves corresponding to the historical database with the same or similar conditions as the ladle condition, ladle argon blowing and ladle insulation information of this heat are fitted to obtain respective fitting curves: the temperature drop fitting curve corresponding to the historical database with the same or similar conditions as the ladle condition information of this heat, the temperature drop fitting curve corresponding to the historical database with the same or similar conditions as the ladle argon blowing information of this heat, and the temperature drop fitting curve corresponding to the historical database with the same or similar conditions as the ladle insulation information of this heat;

[0065] The three fitting curves of ladle condition, ladle argon blowing and ladle insulation information are used as input layer for feedforward neural network calculation to obtain the temperature drop curve under the same or similar ladle conditions as the current heat;

[0066] Combined with the tapping time and temperature drop curve of this heat, the predicted temperature drop value changing with time under the ladle conditions of this heat is obtained.

[0067] As a possible implementation of this embodiment, the temperature value T of the TSO measurement of the end point of this heat is 副枪 , temperature drop prediction value T 网络 And the temperature drop prediction value T 曲线 Perform regression analysis to obtain the predicted value of the molten steel temperature in the ladle after tapping the heat, T 出钢后 ,include:

[0068] Measure the TSO temperature value T at the end of this furnace 副枪 and calculate the temperature drop prediction value T with the symmetrically connected neural network 网络 And the predicted temperature drop value T under the ladle conditions of this heat over time 曲线 Perform regression analysis to obtain the predicted value of the molten steel temperature in the ladle after tapping the heat, T 出钢后 .

[0069] As a possible implementation of this embodiment, the steps of obtaining the optimal temperature drop values ​​T1, T2, T3, and T4 from a historical database with conditions identical or similar to those of the current batch are performed synchronously to improve work efficiency.

[0070] As a possible implementation method of this embodiment, the three execution steps of obtaining a temperature drop fitting curve corresponding to a historical database with conditions equal to or similar to the status information of the ladle of this heat, obtaining a temperature drop fitting curve corresponding to a historical database with conditions equal to or similar to the argon blowing information of the ladle of this heat, and obtaining a temperature drop fitting curve corresponding to a historical database with conditions equal to or similar to the insulation information of the ladle of this heat are performed simultaneously to improve work efficiency.

[0071] As a possible implementation method of this embodiment, the execution step of obtaining the optimal temperature drop values ​​T1, T2, T3, and T4 from the historical database with the same or similar conditions as the current heat and the execution step of fitting the temperature drop curves corresponding to the historical database with the same or similar conditions as the ladle condition, ladle argon blowing, and ladle insulation information of the current heat to obtain respective fitting curves are executed simultaneously to improve work efficiency.

[0072] As a possible implementation of this embodiment, the TSO temperature value T of the end point of this heat is measured. 副枪 The execution steps are used to calculate the temperature drop prediction value T with the symmetrically connected neural network. 网络 The execution steps and the predicted temperature drop value T under the ladle conditions of this heat that changes with time are obtained 曲线 The execution steps are executed synchronously to improve work efficiency.

[0073] As a possible implementation of this embodiment, the symmetrical connection neural network calculation step and the feedforward neural network calculation step are executed synchronously to improve work efficiency.

[0074] like Figure 2 As shown, an embodiment of the present invention provides a device for predicting the temperature of molten steel in a ladle, comprising:

[0075] The information collection module is used to collect relevant information of the current heat and historical heats and establish a historical database; the relevant information includes tapping temperature, converter bottom blowing, deoxidation and alloying, ladle modifier, tapping conditions, ladle status, ladle argon blowing and ladle insulation information;

[0076] An information classification module, used to classify the relevant information according to equal or similar conditions;

[0077] The optimal temperature drop value calculation module is used to perform big data analysis on relevant information under the same or similar conditions to obtain the corresponding optimal temperature drop value;

[0078] The connection neural network calculation module is used to perform symmetrical connection neural network calculation using the optimal temperature drop value as input value to obtain the temperature drop prediction value T 网络 ;

[0079] The feedforward neural network calculation module is used to perform feedforward neural network calculations to obtain the temperature drop curve under the same or similar ladle conditions as this heat, as well as the temperature drop prediction value T under the ladle conditions of this heat that changes with time. 曲线 ;

[0080] Regression analysis module is used to measure the TSO temperature value T at the end of this furnace. 副枪 , temperature drop prediction value T 网络 And the temperature drop prediction value T 曲线Perform regression analysis to obtain the predicted value of the molten steel temperature in the ladle after tapping the heat, T 出钢后 ;

[0081] The molten steel temperature prediction module is used to predict the molten steel temperature in the ladle after the steel is tapped. 出钢后 And the predicted temperature drop value T under the ladle conditions of this heat over time 曲线 Obtain the predicted temperature value T of the molten steel in the ladle at any moment after the steel is tapped t .

[0082] As a possible implementation of this embodiment, the feedforward neural network calculation module includes:

[0083] The curve fitting module is used to perform big data analysis and classification on the ladle condition, ladle argon blowing, ladle insulation and their corresponding temperature drop curves, and to fit the temperature drop curves corresponding to the historical database with the same or similar conditions as the ladle condition, ladle argon blowing and ladle insulation information of this heat to obtain their own fitting curves.

[0084] The temperature drop curve module is used to use the three fitting curves of ladle condition, ladle argon blowing and ladle insulation information as the input layer to perform feedforward neural network calculation to obtain the temperature drop curve under the same or similar ladle conditions as the current heat;

[0085] The temperature drop prediction value calculation module is used to obtain the temperature drop prediction value that changes with time under the ladle conditions of this heat by combining the steel tapping time and the temperature drop curve of this heat.

[0086] like Figure 3 As shown, the converter bottom blowing information includes the converter bottom blowing mode, bottom blowing gas type, bottom blowing flow and pressure, bottom blowing time and other related information; the deoxidation alloying information includes the type, addition amount, yield and other related information of the converter tapping deoxidizer and alloy; the ladle modifier information includes the lime, fluorite, other modifiers added to the ladle during tapping and other related information; the tapping condition information includes the tapping port life, tapping port status, tapping slag blocking effect and tapping time and other related information; the ladle status information includes the ladle age, turnover time, residue in the ladle and other related information; the ladle argon blowing information includes the ventilation effect of the ladle air brick, gas flow and pressure, argon blowing time and other related information; the ladle insulation information includes the type of insulation and the addition amount; T 副枪 The temperature value measured by the auxiliary gun TSO at the end of blowing in this furnace; T 网络 The temperature drop historical database corresponding to the same or similar conditions as the relevant information of this furnace is used as the input layer, and the temperature drop prediction value is calculated through the symmetrical connection neural network; T 曲线The temperature drop fitting curve corresponding to the historical database of the same or similar conditions as the ladle condition, ladle argon blowing, and ladle insulation information of this heat is used as the input layer, and the "temperature drop prediction value of the ladle condition of this heat that changes with time" corresponding to the temperature drop curve under the ladle condition is obtained through feedforward neural network calculation; T 出钢后 is the TSO measured temperature value at the end of this heat, the temperature drop prediction value calculated by the symmetrically connected neural network, the temperature drop prediction value that changes with time under the ladle conditions of this heat, and the predicted value of the molten steel temperature in the ladle after tapping obtained by regression analysis; T t After tapping is completed, during the ladle transfer, waiting or refining process, the temperature forecast value of the molten steel in the ladle at any moment after tapping of this heat is derived from the "predicted value of the molten steel temperature in the ladle after tapping of this heat" and the "predicted value of the temperature drop over time under the conditions of the ladle of this heat".

[0087] The temperature change of molten steel in the ladle involves many complex factors such as tapping temperature, converter bottom blowing, deoxidation and alloying, ladle modifier, tapping conditions, ladle condition, ladle argon blowing, ladle insulation, etc. The present invention first collects various relevant information of the current heat and historical heats to establish a historical database. Various relevant information of this heat and historical heats include various factors such as converter bottom blowing, deoxidation and alloying, ladle modifier, tapping conditions, ladle condition, ladle argon blowing, ladle insulation, etc. For specific content, please refer to the relevant information in the accompanying drawings. At the same time, they are classified according to the same or similar conditions, which helps to compare and correspond with the conditions of this heat; then, big data analysis is performed on the converter bottom blowing, deoxidation and alloying, ladle modifier, and tapping conditions under the same or similar conditions, and linear regression equations are obtained for the converter bottom blowing flow rate, the weight of alloy used for deoxidation and alloying, the amount of ladle modifier added, tapping condition information (tapping time, tapping port life, etc.) and temperature drop under the same or similar conditions, thereby obtaining the optimal temperature drop values ​​T1, T2, T3, and T4 corresponding to any conditions, where:

[0088] T1=f(F,P,t...), F is the bottom blowing gas flow rate; P is the bottom blowing gas pressure; t is the bottom blowing time...

[0089] T2=f(W i ,ψ i ...),W i -i type of steel deoxidizer or alloy addition amount; ψ i -I type steel deoxidizer or alloy yield...

[0090] T3=f(G i ,Φ i ...),G i -Amount of type i ladle modifier added; Φ i -I type ladle modifier yield...

[0091] T4 = f(a, b, m...), where a is the life of the tapping hole; b is the level of slag blocking effect during tapping; and m is the tapping time.

[0092] Then, the optimal temperature drop values ​​T1, T2, T3, and T4 in the historical database with the same or similar conditions as this furnace are used as input values ​​for symmetrical connection neural network calculation, and the corresponding output values ​​are geometrically averaged to obtain the temperature drop prediction value T 网络 , further improving the prediction accuracy.

[0093] The ladle condition, ladle argon blowing, ladle insulation and their corresponding temperature drop curves are subjected to big data analysis and classification, and the temperature drop curves corresponding to the historical database with the same or similar conditions as the ladle condition, ladle argon blowing and ladle insulation information of this heat are fitted. These three fitting curves are then used as the input layer for feedforward neural network calculation to obtain the temperature drop curves under the same or similar ladle conditions as this heat. Combined with the specific information of this heat (steel tapping time, etc.), the temperature drop prediction value that changes with time under the ladle conditions of this heat is obtained.

[0094] The predicted value of the molten steel temperature in the ladle after tapping is obtained through regression analysis based on the TSO measured temperature value at the end of this heat, the temperature drop prediction value calculated by the symmetrically connected neural network, and the temperature drop prediction value that changes with time under the ladle conditions of this heat.

[0095] After tapping is completed, during the ladle transfer, waiting or refining process, the "predicted value of the molten steel temperature in the ladle after tapping this heat" and the "predicted value of the temperature drop over time under the conditions of the ladle of this heat" can also be used to obtain the predicted value of the temperature of the molten steel in the ladle at any moment after tapping this heat.

[0096] The present invention adopts a combined prediction model based on the fusion of multiple models and multiple algorithms. It not only accurately predicts the temperature of the molten steel in the ladle after the tapping of the current heat, but also predicts the temperature of the molten steel in the ladle at any moment during the transfer, waiting or refining process after the tapping is completed. It realizes effective control of the molten steel temperature throughout the process and ensures a smooth and orderly production rhythm.

[0097] Specific calculation example 1:

[0098] Heat 1: The converter bottom blowing adopts the full argon blowing mode, and the bottom blowing flow curve is as follows Figure 4 As shown, the pressure is 1.2GPa; 502kg of aluminum manganese iron, 2382kg of silicon manganese iron, and 1425kg of high carbon manganese iron; the small-grained lime and fluorite in the ladle are 360kg and 210kg respectively; the tapping port is in normal condition, with a service life of 82 furnaces, the slide plate has a good slag blocking effect, and the tapping time is 4.5min; the ladle is 15 furnaces old, the turnover ladle is normal, and there is no residue in the ladle; the ladle air brick has a good ventilation effect, and the gas flow rate is 70Nm 3 / h, pressure 0.5GPa, argon blowing time 4.5min; ladle insulation 50kg; blowing end auxiliary gun TSO measurement temperature T 副枪 The temperature drop historical database recommended values ​​T1, T2, T3, and T4 corresponding to the same or similar conditions of the relevant information of this furnace are 1632.8℃, 1633.5℃, 1636.7℃, and 1638.1℃ respectively. The temperature drop prediction value T calculated by the symmetrically connected neural network is 网络 The temperature drop prediction value T is 1635.3℃, and the feedforward neural network is calculated according to the steel tapping time of 4.5min. 曲线 The predicted temperature value T obtained by regression analysis is 36.1℃. 出钢后 : 1600.3℃, the temperature measured manually after tapping is 1600℃, which shows that the temperature predicted by the model is the same as the actual measured temperature; after tapping, the ladle is hoisted to the LF refining process 11 minutes later, according to the "predicted value of the molten steel temperature in the ladle after tapping this heat T 出钢后 ” and “Temperature drop prediction value T under the ladle conditions of this heat over time” 曲线 The model calculates the predicted value of the molten steel temperature in the ladle 11 minutes after the steel is tapped. t It is 1582.5℃. The manual temperature measurement of LF refining process is 1582℃ after arriving at the station. It can be seen that the prediction accuracy of the model is relatively accurate.

[0099] Specific calculation example 2:

[0100] Heat 2: The converter bottom blowing adopts the full nitrogen and argon switching mode, and the bottom blowing flow curve is as follows Figure 5 As shown, the pressure is 1.2GPa; 481kg of aluminum manganese iron, 2477kg of silicon manganese iron, and 1462kg of high-carbon manganese iron; the small-grained lime and fluorite in the ladle are 420kg and 230kg respectively; the tapping port is in normal condition, with a service life of 38 furnaces, the slide plate has a good slag blocking effect, and the tapping time is 3.8min; the ladle has a 23-furnace age, a normal turnover ladle, and no residue in the ladle; the ladle air brick has a good ventilation effect, and the gas flow rate is 63Nm 3 / h, pressure 0.5GPa, argon blowing time 3.6min; ladle insulation 60kg; blowing end auxiliary gun TSO measurement temperature T 副枪 The temperature drop historical database recommended values ​​T1, T2, T3, and T4 corresponding to the same or similar conditions as the relevant information of this furnace are 1652.1℃, 1650.8℃, 1653.5℃, and 1651.7℃ respectively. The temperature drop prediction value T calculated by the symmetrically connected neural network is 网络 The temperature drop prediction value T is 1652.3℃, and the feedforward neural network is calculated according to the steel tapping time of 3.8min. 曲线 The predicted temperature value T obtained by regression analysis is 40.3℃. 出钢后: 1612.2℃, the temperature measured manually after tapping is 1612℃, which shows that the temperature predicted by the model differs from the actual measured temperature by 0.2℃; after tapping, the ladle is hoisted to the LF refining process 18 minutes later, according to the "predicted value of the molten steel temperature in the ladle after tapping this heat, T 出钢后 ” and “Temperature drop prediction value T under the ladle conditions of this heat over time” 曲线 The model calculates the predicted value of the molten steel temperature in the ladle 18 minutes after the steel is tapped. t It is 1580.4℃. The manual temperature measurement of LF refining process is 1580℃ after arriving at the station. It can be seen that the prediction accuracy of the model is relatively accurate, with a difference of 0.4℃.

[0101] A computer device provided by an embodiment of the present invention includes a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor. When the device is running, the processor and the memory communicate through the bus, and the processor executes the machine-readable instructions to perform the steps of any of the above-mentioned methods for predicting the temperature of molten steel in a ladle.

[0102] Specifically, the above-mentioned memory and processor can be general-purpose memory and processor, which are not specifically limited here. When the processor runs the computer program stored in the memory, it can execute the above-mentioned method for predicting the temperature of molten steel in the ladle.

[0103] Those skilled in the art will understand that the structure of the computer device does not constitute a limitation of the computer device, and may include more or fewer components than shown in the figure, or combine certain components, or split certain components, or arrange the components differently.

[0104] In some embodiments, the computer device may further include a touch screen that can be used to display a graphical user interface (e.g., a startup interface for an application) and receive user operations on the graphical user interface (e.g., startup operations for an application). The specific touch screen may include a display panel and a touch panel. The display panel may be configured in the form of an LCD (Liquid Crystal Display), an OLED (Organic Light-Emitting Diode), or the like. The touch panel may collect contact or contactless operations performed by the user on or near it, and generate pre-set operation instructions, for example, the user uses any suitable object or accessory such as a finger or stylus to operate on or near the touch panel. In addition, the touch panel may include two parts: a touch detection device and a touch controller. The touch detection device detects the user's touch position and posture, detects the signal caused by the touch operation, and transmits the signal to the touch controller; the touch controller receives the touch information from the touch detection device, converts it into information that the processor can process, and then sends it to the processor, and can receive commands sent by the processor and execute them. In addition, the touch panel can be implemented using various types such as resistive, capacitive, infrared, and surface acoustic wave, and any technology developed in the future can also be used to implement the touch panel. Furthermore, the touch panel can cover the display panel, and the user can operate on or near the touch panel covered on the display panel according to the graphical user interface displayed on the display panel. After the touch panel detects the operation on or near it, it transmits it to the processor to determine the user input, and then the processor provides a corresponding visual output on the display panel in response to the user input. In addition, the touch panel and the display panel can be implemented as two independent components or integrated.

[0105] Corresponding to the method for starting the above-mentioned application, an embodiment of the present invention further provides a storage medium on which a computer program is stored. When the computer program is run by a processor, the steps of any of the above-mentioned methods for predicting the temperature of molten steel in a ladle are executed.

[0106] The startup device of the application provided in the embodiment of the present application can be specific hardware on the device or software or firmware installed on the device. The implementation principle and technical effects of the device provided in the embodiment of the present application are the same as those of the aforementioned method embodiment. For the sake of brief description, for any part not mentioned in the device embodiment, reference can be made to the corresponding content in the aforementioned method embodiment. Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can all refer to the corresponding processes in the aforementioned method embodiment, and will not be repeated here.

[0107] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0108] In the embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of modules is only a logical function division. There may be other division methods in actual implementation. For example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interface, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms.

[0109] Modules described as separate components may or may not be physically separate, and components shown as modules may or may not be physical modules, that is, they may be located in one place or distributed across multiple network modules. Some or all of these modules may be selected to achieve the purpose of this embodiment based on actual needs.

[0110] In addition, each functional module in the embodiments provided in the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0111] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0112] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0113] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0114] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A method for predicting the temperature of molten steel in a ladle, characterized in that: The steps include: Collect relevant information of the current heat and historical heats and establish a historical database; the relevant information includes tapping temperature, converter bottom blowing, deoxidation and alloying, ladle modifier, tapping conditions, ladle condition, ladle argon blowing and ladle insulation information; Classify the relevant information according to the same or similar conditions; Conduct big data analysis on relevant information under the same or similar conditions to obtain the corresponding optimal temperature drop value; The optimal temperature drop value is used as the input value to perform symmetrical connection neural network calculation to obtain the temperature drop prediction value T 网络 ; Perform feedforward neural network calculation to obtain the temperature drop curve under the same or similar ladle conditions as this heat, as well as the temperature drop prediction value T that changes with time under the ladle conditions of this heat 曲线 ; The TSO temperature value T of the end point of this furnace is measured 副枪 , temperature drop prediction value T 网络 And the temperature drop prediction value T 曲线 Perform regression analysis to obtain the predicted value of the molten steel temperature in the ladle after tapping the heat, T 出钢后 ; According to the predicted value of the molten steel temperature in the ladle after tapping the heat, T 出钢后 And the predicted temperature drop value T under the ladle conditions of this heat over time 曲线 Obtain the predicted temperature value T of the molten steel in the ladle at any moment after the steel is tapped t ; The aforementioned large data analysis of relevant information under the same or similar conditions to obtain the optimal temperature drop value corresponding to any condition includes: A big data analysis was conducted on converter bottom blowing, deoxidation and alloying, ladle modifiers, and tapping conditions under the same or similar conditions. The linear regression equations for converter bottom blowing flow, alloy weight used for deoxidation and alloying, ladle modifier addition, tapping condition information, and temperature drop were obtained under the same or similar conditions, thereby obtaining the optimal temperature drop values ​​T1, T2, T3, and T4 under any conditions. The tapping condition information includes tapping time and tapping port life; The optimal temperature drop value is used as the input value to perform symmetrical connection neural network calculation to obtain the temperature drop prediction value T 网络 ,include: The optimal temperature drop values ​​T1, T2, T3, and T4 in the historical database with the same or similar conditions as this furnace are used as input values ​​for symmetrical connection neural network calculation, and the corresponding output values ​​are calculated by geometric mean to obtain the temperature drop prediction value T 网络 ; The feedforward neural network calculation is performed to obtain the temperature drop curve under the same or similar ladle conditions as the current heat, and the temperature drop prediction value that changes with time under the ladle conditions of the current heat, including: The ladle condition, ladle argon blowing, ladle insulation and their corresponding temperature drop curves are analyzed and classified by big data, and the temperature drop curves corresponding to the historical database with the same or similar conditions as the ladle condition, ladle argon blowing and ladle insulation information of this heat are fitted to obtain their respective fitting curves. The three fitting curves of ladle condition, ladle argon blowing and ladle insulation information are used as input layer for feedforward neural network calculation to obtain the temperature drop curve under the same or similar ladle conditions as the current heat; Combined with the tapping time and temperature drop curve of this heat, the predicted temperature drop value changing with time under the ladle conditions of this heat is obtained.

2. The method for predicting the temperature of molten steel in a ladle according to claim 1, wherein: The TSO measurement temperature value T of the end point of this furnace 副枪 , temperature drop prediction value T 网络 And the temperature drop prediction value T 曲线 Perform regression analysis to obtain the predicted value of the molten steel temperature in the ladle after tapping the heat, T 出钢后 ,include: Measure the TSO temperature value T at the end of this furnace 副枪 and calculate the temperature drop prediction value T with the symmetrically connected neural network 网络 And the predicted temperature drop value T under the ladle conditions of this heat over time 曲线 Perform regression analysis to obtain the predicted value of the molten steel temperature in the ladle after tapping the heat, T 出钢后 .

3. A device for predicting the temperature of molten steel in a ladle, characterized in that: include: The information collection module is used to collect relevant information of the current heat and historical heats and establish a historical database; the relevant information includes tapping temperature, converter bottom blowing, deoxidation and alloying, ladle modifier, tapping conditions, ladle status, ladle argon blowing and ladle insulation information; An information classification module, used to classify the relevant information according to equal or similar conditions; The optimal temperature drop value calculation module is used to perform big data analysis on relevant information under the same or similar conditions to obtain the corresponding optimal temperature drop value; The connection neural network calculation module is used to perform symmetrical connection neural network calculation using the optimal temperature drop value as input value to obtain the temperature drop prediction value T 网络 ; The feedforward neural network calculation module is used to perform feedforward neural network calculations to obtain the temperature drop curve under the same or similar ladle conditions as this heat, as well as the temperature drop prediction value T under the ladle conditions of this heat that changes with time. 曲线 ; Regression analysis module is used to measure the TSO temperature value T at the end of this furnace. 副枪 , temperature drop prediction value T 网络 And the temperature drop prediction value T 曲线 Perform regression analysis to obtain the predicted value of the molten steel temperature in the ladle after tapping the heat, T 出钢后 ; The molten steel temperature prediction module is used to predict the molten steel temperature in the ladle after the steel is tapped. 出钢后 And the predicted temperature drop value T under the ladle conditions of this heat over time 曲线 Obtain the predicted temperature value T of the molten steel in the ladle at any moment after the steel is tapped t ; The aforementioned large data analysis of relevant information under the same or similar conditions to obtain the optimal temperature drop value corresponding to any condition includes: A big data analysis was conducted on converter bottom blowing, deoxidation and alloying, ladle modifiers, and tapping conditions under the same or similar conditions. The linear regression equations for converter bottom blowing flow, alloy weight used for deoxidation and alloying, ladle modifier addition, tapping condition information, and temperature drop were obtained under the same or similar conditions, thereby obtaining the optimal temperature drop values ​​T1, T2, T3, and T4 under any conditions. The tapping condition information includes tapping time and tapping port life; The optimal temperature drop value is used as the input value to perform symmetrical connection neural network calculation to obtain the temperature drop prediction value T 网络 ,include: The optimal temperature drop values ​​T1, T2, T3, and T4 in the historical database with the same or similar conditions as this furnace are used as input values ​​for symmetrical connection neural network calculation, and the corresponding output values ​​are calculated by geometric mean to obtain the temperature drop prediction value T 网络 ; The feedforward neural network calculation module includes: The curve fitting module is used to perform big data analysis and classification on the ladle condition, ladle argon blowing, ladle insulation and their corresponding temperature drop curves, and to fit the temperature drop curves corresponding to the historical database with the same or similar conditions as the ladle condition, ladle argon blowing and ladle insulation information of this heat to obtain their own fitting curves. The temperature drop curve module is used to use the three fitting curves of ladle condition, ladle argon blowing and ladle insulation information as the input layer to perform feedforward neural network calculation to obtain the temperature drop curve under the same or similar ladle conditions as the current heat; The temperature drop prediction value calculation module is used to obtain the temperature drop prediction value that changes with time under the ladle conditions of this heat by combining the steel tapping time and the temperature drop curve of this heat.

4. A computer device, characterized in that: It includes a processor, a memory and a bus, the memory stores machine-readable instructions executable by the processor, and when the computer device is running, the processor and the memory communicate through the bus, and the processor executes the machine-readable instructions to perform the steps of the method for predicting the temperature of molten steel in the ladle as described in claim 1 or 2.

5. A storage medium, characterized in that: The storage medium stores a computer program, which, when executed by a processor, executes the steps of the method for predicting the temperature of molten steel in a ladle as claimed in claim 1 or 2.

Images