Blast furnace airflow distribution prediction method and device, electronic equipment and storage medium

Abstract

The invention provides a blast furnace airflow distribution prediction method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining a crude fuel parameter historical data set and airflow parameter historical data at different moments, and associating the airflow parameter historical data and the crude fuel parameter historical data set at different moments according to reference time lag, the method comprises the following steps: acquiring a gas flow parameter and crude fuel parameter associated data set, establishing a fitting relationship between the gas flow parameter and the crude fuel parameter based on the gas flow parameter and crude fuel parameter associated data set, acquiring a current data set of the crude fuel parameter, and acquiring a current data set of the crude fuel parameter according to the current data set of the crude fuel parameter and the fitting relationship between the gas flow parameter and the crude fuel parameter; carrying out airflow distribution prediction; the air flow distribution in the blast furnace can be automatically predicted in advance, a scientific decision-making basis is provided for blast furnace operators, so that defense operation is carried out in time according to prediction information, the furnace condition is stabilized, the production efficiency and the production quality are improved, human intervention is reduced, and the manual labor load and human errors are reduced.

Description

Technical Field

[0001] The present application relates to the technical field of blast furnace smelting, and in particular to a method, device, electronic equipment and storage medium for predicting blast furnace gas flow distribution. Background Art

[0002] Blast furnace smelting is a very important link in steel production. It has an extremely complex industrial production process, involving physical and chemical reactions in multiple phases, and complex energy conversion and transfer. The quality of raw materials such as iron ore and coke will affect the distribution of blast furnace top gas. Usually, the uneven quality of raw materials will cause changes in the blast furnace gas flow distribution, which in turn affects the blast furnace permeability.

[0003] It is worth noting that there is a significant lag in the process from the raw fuel entering the blast furnace to the reaction that affects the blast furnace gas flow distribution. This means that when the furnace condition shows an unfavorable trend due to changes in the quality of the raw fuel, the best time for adjustment has often been missed. Therefore, how to predict the impact of changes in raw fuel on the blast furnace gas flow distribution in advance so that operators can perform defensive operations accordingly has become an important issue that needs to be urgently solved in the current blast furnace smelting technology. Summary of the invention

[0004] In view of the shortcomings of the prior art mentioned above, the present application provides a blast furnace airflow distribution prediction method, device, electronic equipment and storage medium to solve the above-mentioned technical problem of how to predict in advance the impact of changes in raw materials and fuels on the blast furnace airflow distribution.

[0005] The present application provides a method for predicting the airflow distribution of a blast furnace, the method comprising: acquiring historical data groups of raw fuel parameters and historical data of airflow parameters at different times; associating the historical data of airflow parameters and historical data groups of raw fuel parameters at different times according to a reference time lag to obtain a data set of airflow parameters and raw fuel parameters association, and establishing a fitting relationship between airflow parameters and raw fuel parameters based on the data set of airflow parameters and raw fuel parameters association; collecting a current data group of raw fuel parameters, and predicting the airflow distribution according to the current data group of raw fuel parameters and the fitting relationship between airflow parameters and raw fuel parameters.

[0006] In one embodiment of the present application, the historical data of airflow parameters at different times are associated with the historical data groups of raw fuel parameters according to a reference time lag, including: according to the reference time lag between various raw fuel parameters and gas utilization, the historical data of gas utilization at different times are associated with the historical data of raw fuel parameters of the same type, so as to obtain a data set of gas utilization and raw fuel parameters associated under each type of raw fuel parameters, and according to the reference time lag between various raw fuel parameters and total pressure difference, the historical data of total pressure difference at different times are associated with the historical data of the same type of raw fuel parameters, so as to obtain a data set of total pressure difference and raw fuel parameters associated under each type of raw fuel parameters, the raw fuel parameter historical data group includes different types of raw fuel parameter historical data, and the airflow parameter historical data includes the gas utilization historical data and the total pressure difference historical data; based on the data set of gas utilization and raw fuel parameters associated under all raw fuel parameters and the data set of total pressure difference and raw fuel parameters associated under all raw fuel parameters, the airflow parameter and raw fuel parameter associated data set is obtained.

[0007] In one embodiment of the present application, before associating the historical data of airflow parameters and the historical data groups of raw fuel parameters at different times according to the reference time lag, the method includes: calculating the correlation coefficient between the gas utilization rate and the same type of raw fuel parameters at different lag times according to the historical data of gas utilization rate and the historical data of the same type of raw fuel parameters at different times, and using the lag time corresponding to the highest correlation coefficient among the correlation coefficients between the gas utilization rate and the same type of raw fuel parameters at different lag times as the reference time lag between the same type of raw fuel parameters and the gas utilization rate, so as to obtain the reference time lag between various types of raw fuel parameters and the gas utilization rate; calculating the correlation coefficient between the total pressure difference and the same type of raw fuel parameters at different lag times according to the historical data of total pressure difference and the historical data of the same type of raw fuel parameters at different times, and using the lag time corresponding to the highest correlation coefficient among the correlation coefficients between the total pressure difference and the same type of raw fuel parameters at different lag times as the reference time lag between the same type of raw fuel parameters and the total pressure difference, so as to obtain the reference time lag between various types of raw fuel parameters and the total pressure difference.

[0008] In one embodiment of the present application, based on the data set associated with the airflow parameters and the raw fuel parameters, a fitting relationship between the airflow parameters and the raw fuel parameters is established, including: taking various raw fuel parameters as independent variables, and gas utilization rate and total pressure difference as dependent variables, and generating a training data set based on the data set associated with the airflow parameters and the raw fuel parameters; using the training data set to train a preset regression model to obtain the fitting relationship between the airflow parameters and the raw fuel parameters, and the fitting relationship between the airflow parameters and the raw fuel parameters includes the fitting relationship between the gas utilization rate and various raw fuel parameters, and the fitting relationship between the total pressure difference and various raw fuel parameters.

[0009] In one embodiment of the present application, the training data set is used to train the preset regression model, including: initializing the parameters of the preset regression model, taking the independent variables in the training data set as input and the dependent variables in the training data set as output, training the preset regression model to obtain the predicted value of the dependent variable; calculating the loss function based on the true value of the dependent variable and the predicted value of the dependent variable, and the true value of the dependent variable is obtained based on the dependent variable in the training data set; if the loss function meets the preset conditions, the training of the preset regression model is completed to obtain the fitting relationship between the airflow parameters and the raw fuel parameters; if the loss function does not meet the preset conditions, the parameters of the preset regression model are updated according to the loss function, so as to use the training data set to train the updated preset regression model until the loss function meets the preset conditions, or the number of parameter updates of the preset regression model reaches the preset update number.

[0010] In one embodiment of the present application, airflow distribution prediction is performed based on the current data group of raw fuel parameters and the fitting relationship between the airflow parameters and the raw fuel parameters, including: inputting the current data group of raw fuel parameters into a trained preset regression model, so that the trained preset regression model predicts gas utilization based on the current data group of raw fuel parameters and the fitting relationship between the gas utilization and various raw fuel parameters, and predicts total pressure difference based on the current data group of raw fuel parameters and the fitting relationship between the total pressure difference and various raw fuel parameters, so as to output the predicted value distribution of gas utilization and the predicted value distribution of total pressure difference.

[0011] In one embodiment of the present application, after performing airflow distribution prediction, the method includes: collecting current data of airflow parameters, the current data of airflow parameters including current data of gas utilization and current data of total pressure difference; calculating a confidence interval of gas utilization based on the predicted value distribution of gas utilization, and calculating a confidence interval of total pressure difference based on the predicted value distribution of total pressure difference; if the current data of gas utilization exceeds the confidence interval of gas utilization, and / or the current data of total pressure difference exceeds the confidence interval of total pressure difference, generating warning information, and issuing a warning based on the warning information.

[0012] In one embodiment of the present application, a blast furnace airflow distribution prediction device is also provided, and the device includes: a data acquisition module, which is used to obtain historical data groups of raw fuel parameters and historical data of airflow parameters at different times; an information processing module, which is used to associate the historical data of airflow parameters and historical data groups of raw fuel parameters at different times according to a reference time lag, to obtain a data set of airflow parameters and raw fuel parameters association, and based on the data set of airflow parameters and raw fuel parameters association, establish a fitting relationship between airflow parameters and raw fuel parameters; a data prediction module, which is used to predict airflow distribution according to a current data group of raw fuel parameters and the fitting relationship between airflow parameters and raw fuel parameters, and the current data group of raw fuel parameters is acquired by the data acquisition module.

[0013] In one embodiment of the present application, an electronic device is also provided, which includes: one or more processors; a storage device for storing one or more programs, and when the one or more programs are executed by the one or more processors, the electronic device implements the blast furnace airflow distribution prediction method as described above.

[0014] In one embodiment of the present application, a computer-readable storage medium is further provided, on which a computer program is stored. When the computer program is executed by a processor of a computer, the computer executes the blast furnace airflow distribution prediction method as described above.

[0015] Beneficial effects of the present invention: The present invention provides a method, device, electronic device and storage medium for predicting blast furnace airflow distribution. The method takes into account the time lag relationship between airflow parameters and raw fuel parameters, associates historical data of airflow parameters and historical data groups of raw fuel parameters at different times according to a reference time lag, and establishes a fitting relationship between airflow parameters and raw fuel parameters based on the associated data, so as to predict airflow distribution according to the fitting relationship and the current data group of raw fuel parameters collected, and can automatically predict the airflow distribution in the blast furnace in advance, providing a scientific decision-making basis for blast furnace operators, so that defensive operations can be performed in time according to the predicted information, thereby stabilizing the furnace condition, improving production efficiency and production quality, and reducing human intervention, thereby reducing manual labor load and human errors.

[0016] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a schematic diagram of an implementation environment of a blast furnace airflow distribution prediction method shown in an exemplary embodiment of the present application;

[0018] Figure 2 is a flow chart of a method for predicting blast furnace airflow distribution shown in an exemplary embodiment of the present application;

[0019] Figure 3 is a blast furnace gas flow distribution prediction flow chart shown in a specific embodiment of the present application;

[0020] Figure 4 is a block diagram of a blast furnace airflow distribution prediction device shown in an exemplary embodiment of the present application;

[0021] Figure 5 is a structural schematic diagram of a blast furnace airflow distribution prediction system shown in a specific embodiment of the present application;

[0022] Figure 6 It is a structural schematic diagram of an electronic device shown in an exemplary embodiment of the present application. DETAILED DESCRIPTION

[0023] The following describes the embodiments of the present application through specific examples, and those skilled in the art can easily understand other advantages and effects of the present application from the contents disclosed in this specification. The present application can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0024] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present application, and thus the drawings only show components related to the present application rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed at will, and the component layout may also be more complicated.

[0025] It should be noted that in this application, "first", "second", etc. are only used to distinguish similar objects, and are not used to limit the order or precedence of similar objects. The variations of "including", "having", etc. described above indicate that the scope covered by the subject of the word is not exclusive except for the examples shown by the word.

[0026] It is understood that the various numbers, step numbers, and other reference numerals recorded in this application are distinguished for the convenience of description and are not intended to limit the scope of this application. The size of the reference numerals in this application does not mean the order of execution. The execution order of each process should be determined by its function and internal logic.

[0027] In the following description, a large number of details are discussed to provide a more thorough explanation of the embodiments of the present application. However, it is obvious to those skilled in the art that the embodiments of the present application can be implemented without these specific details. In other embodiments, well-known structures and devices are shown in the form of block diagrams rather than in detail to avoid making the embodiments of the present application difficult to understand.

[0028] It should be noted that plates and strips usually refer to medium plates or strip steel, and rolled products are the raw materials of plates and strip steel. That is to say, after the rolled products are rolled through the rolling mill, the products obtained are medium plates or strip steel.

[0029] The embodiments of the present application respectively propose a blast furnace airflow distribution prediction method, a blast furnace airflow distribution prediction device, an electronic device, a computer-readable storage medium and a computer program product, which will be described in detail below.

[0030] See also Figure 1 , Figure 1 It is a schematic diagram of an implementation environment of a blast furnace airflow distribution prediction method shown as an exemplary embodiment of the present application.

[0031] like Figure 1 As shown, the implementation environment may include a data acquisition device 110 and a computer device 120, wherein the data acquisition device 110 may be a terminal for collecting raw fuel parameters and airflow parameters, such as at least one of a computer, a mobile phone, a tablet, etc., and the computer device 120 may be at least one of a microcomputer, an embedded computer, a neural network computer, etc., which are not limited here. The computer device 120 may be used to process the raw fuel parameter historical data groups and the airflow parameter historical data at different times to realize the prediction of blast furnace airflow distribution. The user may input the raw fuel parameter historical data groups and the airflow parameter historical data at different times into the data acquisition device 110, and send them to the computer device 120 through the data acquisition device 110. The user may also directly input the raw fuel parameter historical data groups and the airflow parameter historical data at different times into the computer device 120, which are not limited here.

[0032] Schematically, the computer device 120 obtains the raw fuel parameter historical data group and the airflow parameter historical data at different times through the data acquisition device 110; the airflow parameter historical data and the raw fuel parameter historical data group at different times are associated according to the reference time lag to obtain the airflow parameter and the raw fuel parameter associated data set, and based on the airflow parameter and the raw fuel parameter associated data set, the fitting relationship between the airflow parameter and the raw fuel parameter is established; the raw fuel parameter current data group is collected through the data acquisition device 110, and the airflow distribution is predicted according to the current data group of the raw fuel parameter and the fitting relationship between the airflow parameter and the raw fuel parameter. It can be seen that the technical solution of the embodiment of the present application can automatically predict the airflow distribution in the blast furnace in advance, and provide a scientific decision-making basis for the blast furnace operators, so as to carry out defensive operations in time according to the predicted information, thereby stabilizing the furnace condition, improving production efficiency and production quality, and reducing human intervention, thereby reducing manual labor load and human errors.

[0033] It should be noted that the blast furnace airflow distribution prediction method provided in the embodiment of the present application can be specifically executed by the computer device 120, and accordingly, the blast furnace airflow distribution prediction device can be set in the computer device 120.

[0034] See also Figure 2 , Figure 2 is a flowchart of a method for predicting blast furnace airflow distribution shown in an exemplary embodiment of the present application. The method for predicting blast furnace airflow distribution can be applied to Figure 1 The implementation environment shown in FIG. 1 is specifically implemented by the computer device 120 in the implementation environment. It should be understood that the blast furnace airflow distribution prediction method can also be applied to other exemplary implementation environments and be specifically implemented by devices in other implementation environments. This embodiment does not limit the implementation environment to which the blast furnace airflow distribution prediction method is applicable. Figure 2 As shown, in an exemplary embodiment, the blast furnace gas flow distribution prediction method includes at least steps S210 to S230, which are described in detail as follows:

[0035] Step S210, obtaining the raw fuel parameter historical data set and the airflow parameter historical data at different times.

[0036] In one embodiment of the present application, raw fuel parameters refer to the key components and performance parameters of sintered ore and coke, and the raw fuel parameter data group includes TFe (total iron), MgO (magnesium oxide), Al2O3 (aluminum oxide), SiO2 (silicon dioxide), TiO2 (titanium dioxide), FeO (ferrous oxide), sintered ore drum index, particle size, ash content, sulfur content, moisture, M10 (coke wear resistance index), M40 (coke crushing strength index), CSR (post-reaction strength), CRI (reactivity) and other different categories of raw fuel parameters. Correspondingly, the raw fuel parameter historical data group includes historical data of raw fuel parameters of different categories. Users can record and store the raw fuel parameter historical data groups at different times in advance, wherein the data of various raw fuel parameters can be obtained in time through the inspection and testing system after the inspection and testing is completed. The airflow parameters include at least one of the gas utilization rate and the total pressure difference, etc. Correspondingly, the historical data of the airflow parameters include the historical data of the gas utilization rate and / or the historical data of the total pressure difference. The user can record and store the historical data of the airflow parameters at different times in advance, wherein the gas utilization rate data can be obtained through the gas analyzer, for example: collect the CO (carbon monoxide) and CO2 (carbon dioxide) data of the gas analyzer, and obtain the gas utilization rate data through CO2 / (CO+CO2) calculation. The wind pressure and top pressure data can also be collected through the pressure sensor, and the difference between the wind pressure and the top pressure is calculated to obtain the total pressure difference data. Then the raw fuel parameter historical data group and the airflow parameter historical data at different times are stored in the database.

[0037] It should be understood that the historical data sets of raw material and fuel parameters at different times and the historical data sets of airflow parameters at different times should be in the same production cycle of blast furnace smelting.

[0038] Step S220, associating the historical data of airflow parameters and the historical data groups of raw fuel parameters at different times according to the reference time lag to obtain a data set of airflow parameters and raw fuel parameters association, and establishing a fitting relationship between the airflow parameters and the raw fuel parameters based on the data set of airflow parameters and raw fuel parameters association.

[0039] In one embodiment of the present application, a reference time lag may be preset, for example, the time when the data of the raw fuel parameter changes and the time when the data of the airflow parameter changes significantly may be recorded in advance, and the difference between the time when the data of the raw fuel parameter changes and the time when the data of the airflow parameter changes significantly may be calculated as the reference time lag. Of course, the reference time lag may also be obtained by other means, which are not limited here.

[0040] Before associating the historical data of airflow parameters and the historical data group of raw fuel parameters, the data granularity of the historical data of airflow parameters and the historical data group of raw fuel parameters can also be unified, that is, these data are processed into data of the same time level, which can be hourly, minutely or secondly, and there is no restriction here. Schematically, taking the processing of the historical data of airflow parameters and the historical data group of raw fuel parameters as hourly data as an example, for data with a collection frequency higher than the hourly level, the hourly average can be taken, and for data with a collection frequency lower than the hourly level, linear interpolation can be calculated, and finally the data granularity of the historical data of airflow parameters and the historical data group of raw fuel parameters is unified to the hourly level.

[0041] Then, according to the reference time lag, the historical data of the airflow parameters and the historical data group of the raw fuel parameters at each moment after the data granularity is unified are associated to obtain a data set of airflow parameters and raw fuel parameters association, and data fitting is performed on the data set of airflow parameters and raw fuel parameters association to obtain a fitting relationship between the airflow parameters and the raw fuel parameters.

[0042] In one embodiment of the present application, the historical data of airflow parameters at different times are associated with the historical data groups of raw fuel parameters according to a reference time lag, including: according to the reference time lag between various raw fuel parameters and gas utilization, the historical data of gas utilization at different times are associated with the historical data of the same type of raw fuel parameters to obtain a data set of gas utilization and raw fuel parameters association under each type of raw fuel parameters, and according to the reference time lag between various raw fuel parameters and total pressure difference, the historical data of total pressure difference at different times are associated with the historical data of the same type of raw fuel parameters to obtain a data set of total pressure difference and raw fuel parameters association under each type of raw fuel parameters, the raw fuel parameter historical data group includes different types of raw fuel parameter historical data, and the airflow parameter historical data includes gas utilization historical data and total pressure difference historical data; based on the data set of gas utilization and raw fuel parameter association under all raw fuel parameters and the data set of total pressure difference and raw fuel parameter association under all raw fuel parameters, an airflow parameter and raw fuel parameter association data set is obtained.

[0043] In this embodiment, taking the raw fuel parameter FeO of different categories as an example, according to the reference time lag between FeO and gas utilization rate, the historical data of gas utilization rate at different times and the historical data of FeO at different times are associated to obtain the data set associated with gas utilization rate and FeO, and so on, the data set associated with gas utilization rate and raw fuel parameter under each category of raw fuel parameter is obtained; according to the reference time lag between FeO and total pressure difference, the historical data of total pressure difference at different times and the historical data of FeO at different times are associated to obtain the data set associated with total pressure difference and FeO, and so on, the data set associated with total pressure difference and raw fuel parameter under each category of raw fuel parameter is obtained. Then, the data set associated with gas utilization rate and raw fuel parameter under all raw fuel parameters and the data set associated with total pressure difference and raw fuel parameter under all raw fuel parameters are used as the data set associated with airflow parameter and raw fuel parameter.

[0044] Indicatively, the reference time lag between various raw fuel parameters and gas utilization, and the reference time lag between various raw fuel parameters and the total pressure difference can be preset. For example, experiments can be conducted in advance to adjust the data change of only one raw fuel parameter each time, record the data change moment of the raw fuel parameter, the moment when the gas utilization data changes significantly, and the moment when the total pressure difference data changes significantly, calculate the difference between the data change moment of the raw fuel parameter and the moment when the gas utilization data changes significantly, and use it as the reference time lag between this type of raw fuel parameter and the gas utilization, and calculate the difference between the data change moment of the raw fuel parameter and the moment when the total pressure difference data changes significantly, and use it as the reference time lag between this type of raw fuel parameter and the total pressure difference. Of course, the reference time lag between various raw fuel parameters and the gas utilization, and the reference time lag between various raw fuel parameters and the total pressure difference can also be obtained by other means, which are not limited here.

[0045] In one embodiment of the present application, before associating the historical data of airflow parameters and the historical data groups of raw fuel parameters at different times according to the reference time lag, the method includes: calculating the correlation coefficient between the gas utilization rate and the same type of raw fuel parameters at different lag times according to the historical data of gas utilization rate and the same type of raw fuel parameters at different times, and using the lag time corresponding to the highest correlation coefficient among the correlation coefficients between the gas utilization rate and the same type of raw fuel parameters at different lag times as the reference time lag between the same type of raw fuel parameters and the gas utilization rate, so as to obtain the reference time lag between various types of raw fuel parameters and the gas utilization rate; calculating the correlation coefficient between the total pressure difference and the same type of raw fuel parameters at different lag times according to the historical data of total pressure difference and the same type of raw fuel parameters at different times, and using the lag time corresponding to the highest correlation coefficient among the correlation coefficients between the total pressure difference and the same type of raw fuel parameters at different lag times as the reference time lag between the same type of raw fuel parameters and the total pressure difference, so as to obtain the reference time lag between various types of raw fuel parameters and the total pressure difference.

[0046] In this embodiment, taking different categories of raw fuel parameters FeO as an example, according to the historical data of gas utilization rate at different times and the historical data of FeO at different times, the correlation coefficient between gas utilization rate and FeO at different lag times such as 1 hour, 2 hours, 3 hours, etc. is calculated, the highest correlation coefficient is taken, and the lag time of the highest correlation coefficient is used as the reference lag between FeO and gas utilization rate, and so on, to obtain the reference lag between different categories of raw fuel parameters and gas utilization rate; according to the historical data of total pressure difference at different times and the historical data of FeO at different times, the correlation coefficient between total pressure difference and FeO at different lag times such as 1 hour, 2 hours, 3 hours, etc. is calculated, the highest correlation coefficient is taken, and the lag time of the highest correlation coefficient is used as the reference lag between FeO and total pressure difference, and so on, to obtain the reference lag between different categories of raw fuel parameters and total pressure difference.

[0047] Illustratively, there are many specific ways to calculate the correlation coefficient, for example, the Pearson correlation coefficient or the Spearman rank correlation coefficient can be used to calculate according to the correlation coefficient matrix, which is not limited here.

[0048] In one embodiment of the present application, based on a data set associated with airflow parameters and raw fuel parameters, a fitting relationship between airflow parameters and raw fuel parameters is established, including: taking various raw fuel parameters as independent variables, gas utilization rate and total pressure difference as dependent variables, and generating a training data set based on the data set associated with airflow parameters and raw fuel parameters; using the training data set to train a preset regression model to obtain a fitting relationship between airflow parameters and raw fuel parameters, the fitting relationship between airflow parameters and raw fuel parameters including a fitting relationship between gas utilization rate and various raw fuel parameters, and a fitting relationship between total pressure difference and various raw fuel parameters.

[0049] In this embodiment, a training data set is generated based on a data set associated with airflow parameters and raw fuel parameters, including independent variables (various raw fuel parameters) and dependent variables (gas utilization rate and total pressure difference). Considering that there is a certain degree of multicollinearity in the raw fuel parameters, a ridge regression model, a lasso regression model, an elastic network regression model, a partial least squares regression model or other multivariate regression models that can handle multicollinearity can be selected as a preset regression model to find the relationship between the independent variable and the dependent variable. Based on the selected multiple regression model, i.e., the preset regression model and the training data set, the relationship between various raw fuel parameters and gas utilization rate, and the relationship between various raw fuel parameters and total pressure difference are fitted respectively, including: the relationship between different types of raw fuel parameters such as TFe, MgO, Al2O3, SiO2, TiO2, FeO, sinter drum index, particle size, ash content, sulfur content, moisture, M10, M40, CSR, CRI and gas utilization rate, and the relationship between different types of raw fuel parameters such as TFe, MgO, Al2O3, SiO2, TiO2, FeO, sinter drum index, particle size, ash content, sulfur content, moisture, M10, M40, CSR, CRI and total pressure difference. All these relationships are used as the fitting relationship between airflow parameters and raw fuel parameters.

[0050] In one embodiment of the present application, a training data set is used to train a preset regression model, including: initializing parameters of the preset regression model, taking the independent variables in the training data set as input and the dependent variables in the training data set as output, training the preset regression model to obtain a predicted value of the dependent variable; calculating a loss function based on the true value of the dependent variable and the predicted value of the dependent variable, the true value of the dependent variable being obtained based on the dependent variable in the training data set; if the loss function meets the preset conditions, the training of the preset regression model is completed to obtain a fitting relationship between the airflow parameters and the raw fuel parameters; if the loss function does not meet the preset conditions, the parameters of the preset regression model are updated according to the loss function, so as to train the updated preset regression model using the training data set, until the loss function meets the preset conditions, or the number of parameter updates for the preset regression model reaches a preset number of updates.

[0051] In this embodiment, the training preset regression model is taken as an example of a ridge regression model, and the ridge regression steps are as follows:

[0052] 1. Initialization parameters: Initialize the regression coefficient and the weight of the regularization term. The regression coefficient can be initialized to 0, and the weight of the regularization term can be initialized to a smaller value, such as 0.01.

[0053] 2. Calculate the loss function: Use the current regression coefficient and the weight of the regularization term to calculate the loss function, namely SSE (Sum of Squared Errors).

[0054] 3. Update regression coefficients: Use methods such as gradient descent to update the regression coefficients to minimize the loss function.

[0055] 4. Update the weight of the regularization term: Update the weight of the regularization term based on the updated regression coefficient. The weight of the regularization term can be increased by a very small value, such as 0.001, to avoid overfitting.

[0056] 5. Repeat steps 2-4 until the loss function converges, that is, the loss function meets the preset conditions (that is, the loss function reaches the preset loss threshold), or, the predetermined number of iterations is reached, that is, the number of parameter updates of the ridge regression model reaches the preset number of updates.

[0057] In addition, after the model training is completed, the predicted value of the dependent variable can be compared with the actual value of the dependent variable, and the MSE (Mean Squared Error) and R can be calculated. 2 (R-squared, determination coefficient), evaluates the model effect of the preset regression model after training, where the predicted value of the dependent variable is the predicted value of gas utilization rate and the predicted value of total pressure difference output by the preset regression model after training, and the true value of the dependent variable is the data corresponding to the gas utilization rate and the total pressure difference in the training data set.

[0058] Step S230, collecting the current data set of raw fuel parameters, and performing airflow distribution prediction based on the current data set of raw fuel parameters and the fitting relationship between the airflow parameters and the raw fuel parameters.

[0059] In one embodiment of the present application, after obtaining the fitting relationship between the airflow parameters and the raw fuel parameters, the current raw fuel parameter data group (i.e., the current raw fuel parameter data group) can be collected in real time or periodically through the inspection and testing system, and the future airflow parameter data distribution at a certain time interval can be predicted based on the current raw fuel parameter data group and the obtained fitting relationship, so as to provide a scientific decision-making basis for blast furnace operators, so that future furnace conditions can be determined according to the predicted information, and defensive operations can be performed in time or in advance.

[0060] In one embodiment of the present application, airflow distribution prediction is performed based on the current data group of raw fuel parameters and the fitting relationship between airflow parameters and raw fuel parameters, including: inputting the current data group of raw fuel parameters into a trained preset regression model, so that the trained preset regression model predicts gas utilization based on the current data group of raw fuel parameters and the fitting relationship between gas utilization and various raw fuel parameters, and predicts total pressure difference based on the current data group of raw fuel parameters and the fitting relationship between total pressure difference and various raw fuel parameters, so as to output the predicted value distribution of gas utilization and the predicted value distribution of total pressure difference.

[0061] In this embodiment, the current data group of raw fuel parameters includes current data of various raw fuel parameters. The trained preset regression model predicts the gas utilization corresponding to the same type of raw fuel parameters based on the current data of various raw fuel parameters and the fitting relationship between the gas utilization and the same type of raw fuel parameters to obtain the predicted values ​​of the gas utilization corresponding to various raw fuel parameters. The total pressure difference corresponding to the same type of raw fuel parameters is predicted based on the current data of various raw fuel parameters and the fitting relationship between the total pressure difference and the same type of raw fuel parameters to obtain the predicted values ​​of the total pressure difference corresponding to various raw fuel parameters. Finally, the predicted value distribution of gas utilization and the predicted value distribution of total pressure difference are output, wherein the predicted value distribution of gas utilization includes the predicted values ​​of gas utilization corresponding to various raw fuel parameters, and the predicted value distribution of total pressure difference includes the predicted values ​​of total pressure difference corresponding to various raw fuel parameters.

[0062] Taking different categories of raw fuel parameters FeO as an example, the trained preset regression model predicts the gas utilization rate corresponding to FeO according to the current data of FeO and the fitting relationship between the gas utilization rate and FeO, and obtains the predicted value of the gas utilization rate corresponding to FeO. Similarly, the predicted values ​​of the gas utilization rate corresponding to various raw fuel parameters are obtained. The total pressure difference corresponding to FeO is predicted according to the current data of FeO, the fitting relationship between the total pressure difference and FeO, and the predicted value of the total pressure difference corresponding to FeO is obtained. Similarly, the predicted values ​​of the total pressure difference corresponding to various raw fuel parameters are obtained.

[0063] Before inputting the current data of various raw fuel parameters into the trained preset regression model, the current data of various raw fuel parameters can also be uniformly processed into data of the same time level, such as hourly data.

[0064] In one embodiment of the present application, after performing airflow distribution prediction, the method includes: collecting current data of airflow parameters, the current data of airflow parameters including current data of gas utilization and current data of total pressure difference; calculating a confidence interval of gas utilization based on the predicted value distribution of gas utilization, and calculating a confidence interval of total pressure difference based on the predicted value distribution of total pressure difference; if the current data of gas utilization exceeds the confidence interval of gas utilization, and / or the current data of total pressure difference exceeds the confidence interval of total pressure difference, generating warning information, and issuing a warning based on the warning information.

[0065] In this embodiment, the current data of gas utilization rate is obtained by a gas analyzer, the current wind pressure and top pressure are detected by a pressure sensor, and the current data of total pressure difference is obtained according to the current wind pressure and top pressure. The 95% confidence interval or 99% confidence interval or other confidence interval of gas utilization rate can be calculated based on the predicted value of gas utilization rate corresponding to various raw fuel parameters, and the 95% confidence interval or 99% confidence interval or other confidence interval of total pressure difference can be calculated based on the predicted value of total pressure difference corresponding to various raw fuel parameters, so as to obtain the upper and lower limits of the predicted dependent variable. Then, the current data of gas utilization rate is compared with the confidence interval of gas utilization rate, and the current data of total pressure difference is compared with the confidence interval of total pressure difference, to determine whether the current data of gas utilization rate and total pressure difference fall within the corresponding confidence intervals. If any of the current data of gas utilization rate and total pressure difference exceeds the corresponding confidence intervals, an early warning message is generated to alarm, so that experienced operators can select appropriate strategies for defensive adjustments according to the difference between the early warning information and the confidence interval and the current data.

[0066] Before comparing the current data of the gas utilization rate and the total pressure difference with their respective confidence intervals, the current data of the gas utilization rate and the total pressure difference may be uniformly processed into data of the same time level, such as hourly data.

[0067] See also Figure 3 , Figure 3 FIG. 1 is a flow chart showing the prediction of blast furnace gas flow distribution according to a specific embodiment of the present application. Figure 3 As shown in the figure, the blast furnace airflow distribution prediction process is as follows:

[0068] 1. Collect and store historical data of key parameters: including historical data of raw fuel parameters and historical data of air flow parameters. Raw fuel parameters include TFe, MgO, Al2O3, SiO2, TiO2, FeO, sinter drum index, particle size, ash content, sulfur content, moisture, M10, M40, CSR, CRI, etc. Air flow parameters include gas utilization rate and total pressure difference;

[0069] 2. Correlate the historical data of airflow parameters and raw fuel parameters: process the historical data of airflow parameters and the historical data of raw fuel parameters into hourly data, calculate the time lag according to the correlation coefficient matrix, and complete the correlation;

[0070] 3. Fitting the relationship between airflow parameters and raw fuel parameters: Based on the multivariate regression model, fit the relationship between each raw fuel parameter and gas utilization rate, and the relationship between each raw fuel parameter and total pressure difference;

[0071] 4. Monitor the changes in raw fuel and predict the changes in airflow: Monitor the changes in raw fuel in real time. When the real-time data of raw fuel parameters changes, call the multivariate regression model to predict the data distribution of airflow parameters, compare the data of current airflow parameters, and alarm for any excess.

[0072] For the detailed process of the specific embodiments of the present application, please refer to the records in the aforementioned embodiments, which will not be repeated here.

[0073] See also Figure 4 , Figure 4 is a block diagram of a blast furnace gas flow distribution prediction device shown in an exemplary embodiment of the present application. The device can be applied to Figure 1 The system can also be applied to other exemplary implementation environments, and this embodiment does not limit the implementation environment to which the system is applied. Figure 4 As shown, the exemplary blast furnace airflow distribution prediction device includes: a data acquisition module 410, which is used to obtain historical data groups of raw fuel parameters and historical data of airflow parameters at different times; an information processing module 420, which is used to associate the historical data of airflow parameters and the historical data groups of raw fuel parameters at different times according to a reference time lag, to obtain a data set of airflow parameters and raw fuel parameters association, and to establish a fitting relationship between airflow parameters and raw fuel parameters based on the data set of airflow parameters and raw fuel parameters association; a data prediction module 430, which is used to predict airflow distribution according to the current data group of raw fuel parameters and the fitting relationship between airflow parameters and raw fuel parameters, and the current data group of raw fuel parameters is acquired by the data acquisition module 410.

[0074] In this embodiment, the data acquisition module 410 may include an inspection and testing system, a gas analyzer, a pressure sensor, and a preprocessing unit, wherein the preprocessing unit may be a computer device such as a microcomputer, an embedded computer, a neural network computer, etc., and the raw fuel parameter historical data group at different times is obtained through the inspection and testing system, the CO and CO2 historical data at different times are obtained through the gas analyzer, and the wind pressure and top pressure historical data at different times are collected through the pressure sensor. The preprocessing unit processes the CO and CO2 historical data at different times through CO2 / (CO+CO2) to obtain the gas utilization rate historical data at different times, and processes the wind pressure and top pressure historical data at different times by calculating the difference between the wind pressure and the top pressure to obtain the total pressure difference historical data at different times, and then uses the gas utilization rate historical data at different times and / or the total pressure difference historical data at different times as the airflow parameter historical data at different times. At present, the data acquisition module 410 may also be a hardware device that directly collects or stores the raw fuel parameter historical data group and the airflow parameter historical data at different times. The information processing module 420 and the data prediction module 430 can both be computer devices such as microcomputers, embedded computers, neural network computers, or servers.

[0075] See also Figure 5 , Figure 5 FIG. 1 is a schematic diagram of a blast furnace gas flow distribution prediction system according to a specific embodiment of the present application. Figure 5 As shown, the blast furnace airflow distribution prediction system includes a data acquisition module and a server, wherein the data acquisition module includes a plurality of data sensor devices for collecting data on raw fuel parameters and airflow parameters; the server includes a data association module, a data storage module, a data analysis module, a real-time monitoring module and a data display module, wherein the data association module is used to process data on raw fuel parameters and airflow parameters of different acquisition frequencies, process the data into hourly data, and complete the time-lag association between the historical data of the raw fuel parameters and the historical data of the airflow parameters; the data storage module stores the historical data of the raw fuel parameters and the historical data of the airflow parameters after association; the analysis module is used to fit the relationship between the historical data of the raw fuel parameters and the historical data of the airflow parameters after association based on a preset regression model, and store the trained preset regression model; the real-time monitoring module is used to predict the airflow distribution when the data of the raw fuel parameters changes, and calculate the confidence interval of the airflow parameters, and compare it with the real-time data of the airflow parameters; the data display module is used to display the real-time data and prediction results of the airflow parameters, and to alarm for situations exceeding the upper and lower limits of the confidence interval.

[0076] Schematically, the workflow of the blast furnace gas flow distribution prediction system is as follows:

[0077] 1. The data acquisition module collects the historical data groups of raw fuel parameters and historical data of airflow parameters at different times;

[0078] 2. Since the collection frequencies of different data may be different, the data association module uniformly processes the raw fuel parameter historical data groups and airflow parameter historical data at different times into hourly data, and then performs time-lag association on the raw fuel parameter historical data groups and airflow parameter historical data at different times processed into hourly data, and stores the associated data through the data storage module;

[0079] 3. The data analysis module obtains the associated data through the data storage module, and performs relationship fitting on the associated data based on the preset regression model to obtain the relationship between each raw fuel parameter and the airflow parameter;

[0080] 4. The data acquisition module collects the real-time data of the raw fuel parameters and the airflow parameters, and the data association module processes the real-time data of the raw fuel parameters and the airflow parameters into hourly data, sends the real-time data of the raw fuel parameters and the airflow parameters processed into hourly data to the real-time monitoring module, and sends the real-time data of the airflow parameters processed into hourly data to the data display module;

[0081] 5. The real-time monitoring module uses the preset regression model and the real-time data of the raw fuel parameters processed at the hourly level to predict the airflow distribution, calculates the confidence interval based on the prediction results, compares the real-time data of the airflow parameters processed at the hourly level with the confidence interval, and generates warning information when it exceeds the confidence interval;

[0082] 6. The data display module displays the real-time data and prediction results of the airflow parameters processed at the hourly level, and displays the warning information when it is received.

[0083] The blast furnace airflow distribution prediction system in the specific embodiment of the present application obtains the data of raw fuel parameters and the data of blast furnace gas utilization and total pressure difference respectively through the data acquisition module, realizes the association of various raw fuel parameter data with blast furnace gas utilization and total pressure difference data through the data association module, fits the relationship between each raw fuel parameter and the gas utilization and total pressure difference respectively through the data analysis module, monitors the changes of raw fuel parameter data in real time through the real-time monitoring module, and predicts the changes of gas utilization and total pressure difference while maintaining the current operation when the raw fuel situation changes, and alarms for possible gas utilization or pressure difference exceeding the limit, so that experienced operators can perform defensive operations in advance in combination with process logic after obtaining the alarm information, avoid the deterioration of blast furnace condition, implement preventive operations against raw fuel changes, scientifically guide the direction of blast furnace operation adjustment, and achieve an ideal production state that takes into account the balance between smooth furnace condition and gas utilization.

[0084] It should be noted that the blast furnace airflow distribution prediction device and system provided in the above embodiment and the blast furnace airflow distribution prediction method provided in the above embodiment belong to the same concept, wherein the specific manner in which each module and unit performs the operation has been described in detail in the method embodiment and will not be repeated here. In practical applications, the blast furnace airflow distribution prediction device and system provided in the above embodiment can distribute the above functions to different functional modules as needed, that is, divide the internal structure of the device and system into different functional modules to complete all or part of the functions described above, and this is not limited here.

[0085] This embodiment also provides an electronic device, including: one or more processors; a storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the electronic device implements the blast furnace airflow distribution prediction method provided in the above-mentioned embodiments.

[0086] See also Figure 6 , Figure 6 is a structural diagram of an electronic device shown in an exemplary embodiment of the present application. It should be noted that: Figure 6 The electronic device 600 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0087] like Figure 6 As shown, the electronic device 600 includes a processor 601, a memory 602 and a communication bus 603; the communication bus 603 is used to connect the processor 601 and the memory 602; the processor 601 is used to execute the computer program stored in the memory 602 to implement one or more methods in the above embodiments.

[0088] This embodiment also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor of a computer, the computer executes the blast furnace gas flow distribution prediction method as described above. The computer-readable storage medium may be included in the electronic device described in the above embodiment, or may exist independently without being assembled into the electronic device.

[0089] This embodiment also provides a computer program product or computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the blast furnace gas flow distribution prediction method provided in each of the above embodiments.

[0090] The electronic device provided in this embodiment includes a processor, a memory, a transceiver and a communication interface. The memory and the communication interface are connected to the processor and the transceiver and complete communication with each other. The memory is used to store computer programs, the communication interface is used to communicate, and the processor and the transceiver are used to run the computer program so that the electronic device executes each step of the above method.

[0091] In this embodiment, the memory may include a random access memory (RAM), and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0092] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0093] The computer-readable storage medium in this embodiment can be understood by ordinary technicians in this field: all or part of the steps of implementing the above-mentioned method embodiments can be completed by hardware related to the computer program. The aforementioned computer program can be stored in a computer-readable storage medium. When the program is executed, the execution includes the steps of the above-mentioned method embodiments; and the aforementioned storage medium includes: ROM (read-only memory), RAM (random access memory), magnetic disk or optical disk and other media that can store program codes.

[0094] The above embodiments are merely illustrative of the principles and effects of the present application, and are not intended to limit the present application. Anyone familiar with the technology may modify or change the above embodiments without violating the spirit and scope of the present application. Therefore, all equivalent modifications or changes made by a person of ordinary skill in the art without departing from the spirit and technical ideas disclosed in the present application shall still be covered by the claims of the present application.

Claims

1. A method for predicting blast furnace airflow distribution, characterized in that: The method comprises: Obtaining historical data sets of raw fuel parameters and historical data of airflow parameters at different times; Correlating the historical data of airflow parameters and the historical data groups of raw fuel parameters at different times according to the reference time lag to obtain an airflow parameter and raw fuel parameter correlation data set, and establishing a fitting relationship between the airflow parameter and the raw fuel parameter based on the airflow parameter and raw fuel parameter correlation data set; The current data group of raw fuel parameters is collected, and the airflow distribution is predicted according to the current data group of raw fuel parameters and the fitting relationship between the airflow parameters and the raw fuel parameters.

2. The method for predicting blast furnace airflow distribution according to claim 1, characterized in that: The historical data of airflow parameters and the historical data of raw fuel parameters at different times are associated according to the reference time lag, including: According to the reference time lag between various raw fuel parameters and gas utilization, historical data of gas utilization at different times are associated with historical data of raw fuel parameters of the same type, so as to obtain a data set of gas utilization and raw fuel parameters associated with each type of raw fuel parameters; and according to the reference time lag between various raw fuel parameters and total pressure difference, historical data of total pressure difference at different times are associated with historical data of raw fuel parameters of the same type, so as to obtain a data set of total pressure difference and raw fuel parameters associated with each type of raw fuel parameters, wherein the raw fuel parameter historical data group includes different types of raw fuel parameter historical data, and the gas flow parameter historical data includes the gas utilization historical data and the total pressure difference historical data; The data set associated with the gas utilization rate and the raw fuel parameters under all raw fuel parameters and the data set associated with the total pressure difference and the raw fuel parameters under all raw fuel parameters are obtained.

3. The method for predicting blast furnace airflow distribution according to claim 2, characterized in that: Before associating the historical data of airflow parameters and the historical data groups of raw fuel parameters at different times according to the reference time lag, the method includes: According to the historical data of gas utilization rate at different times and the historical data of the parameters of the same type of raw fuel, the correlation coefficients between the gas utilization rate and the parameters of the same type of raw fuel at different lag times are calculated, and the lag time corresponding to the highest correlation coefficient among the correlation coefficients between the gas utilization rate and the parameters of the same type of raw fuel at different lag times is used as the reference time lag between the parameters of the same type of raw fuel and the gas utilization rate, so as to obtain the reference time lag between the parameters of various types of raw fuel and the gas utilization rate; According to the historical data of total pressure difference at different times and the historical data of parameters of the same type of raw fuel, the correlation coefficient between the total pressure difference and the parameters of the same type of raw fuel at different lag times is calculated, and the lag time corresponding to the highest correlation coefficient among the correlation coefficients between the total pressure difference and the parameters of the same type of raw fuel at different lag times is used as the reference time lag between the parameters of the same type of raw fuel and the total pressure difference, so as to obtain the reference time lag between various types of raw fuel parameters and the total pressure difference.

4. The method for predicting blast furnace airflow distribution according to claim 2, characterized in that: Based on the data set associated with the airflow parameters and the raw fuel parameters, a fitting relationship between the airflow parameters and the raw fuel parameters is established, including: Taking various raw fuel parameters as independent variables, gas utilization rate and total pressure difference as dependent variables, and generating a training data set based on the data set associated with the airflow parameters and the raw fuel parameters; The training data set is used to train the preset regression model to obtain the fitting relationship between the airflow parameters and the raw fuel parameters, and the fitting relationship between the airflow parameters and the raw fuel parameters includes the fitting relationship between the gas utilization rate and various raw fuel parameters, and the fitting relationship between the total pressure difference and various raw fuel parameters.

5. The method for predicting blast furnace airflow distribution according to claim 4, characterized in that: The training data set is used to train the preset regression model, including: Initializing parameters of the preset regression model, taking the independent variables in the training data set as input and the dependent variables in the training data set as output, training the preset regression model to obtain a predicted value of the dependent variable; Calculating a loss function based on a true value of a dependent variable and a predicted value of the dependent variable, wherein the true value of the dependent variable is obtained based on the dependent variable in the training data set; If the loss function meets the preset conditions, the training of the preset regression model is completed to obtain the fitting relationship between the airflow parameters and the raw fuel parameters; If the loss function does not meet the preset conditions, the parameters of the preset regression model are updated according to the loss function, so as to train the updated preset regression model using the training data set until the loss function meets the preset conditions, or the number of parameter updates for the preset regression model reaches a preset number of updates.

6. The method for predicting blast furnace airflow distribution according to any one of claims 4 or 5, characterized in that: According to the current data group of the raw fuel parameters and the fitting relationship between the airflow parameters and the raw fuel parameters, airflow distribution prediction is performed, including: The current data group of raw fuel parameters is input into the trained preset regression model, so that the trained preset regression model predicts the gas utilization rate according to the current data group of raw fuel parameters and the fitting relationship between the gas utilization rate and various raw fuel parameters, and predicts the total pressure difference according to the current data group of raw fuel parameters and the fitting relationship between the total pressure difference and various raw fuel parameters, so as to output the predicted value distribution of the gas utilization rate and the predicted value distribution of the total pressure difference.

7. The method for predicting blast furnace airflow distribution according to claim 6, characterized in that: After the airflow distribution prediction is performed, the method includes: Collecting current data of airflow parameters, wherein the current data of airflow parameters include current data of gas utilization rate and current data of total pressure difference; Calculating a confidence interval of the gas utilization rate based on the predicted value distribution of the gas utilization rate, and calculating a confidence interval of the total pressure difference based on the predicted value distribution of the total pressure difference; If the current data of the gas utilization rate exceeds the confidence interval of the gas utilization rate, and / or the current data of the total pressure difference exceeds the confidence interval of the total pressure difference, an early warning message is generated, and an early warning is issued according to the early warning message.

8. A blast furnace airflow distribution prediction device, characterized in that: The device comprises: A data acquisition module is used to obtain historical data sets of raw fuel parameters and historical data of airflow parameters at different times; An information processing module, used for associating the historical data of airflow parameters and the historical data groups of raw fuel parameters at different times according to the reference time lag to obtain an airflow parameter and raw fuel parameter associated data set, and establishing a fitting relationship between the airflow parameter and the raw fuel parameter based on the airflow parameter and raw fuel parameter associated data set; The data prediction module is used to predict the airflow distribution according to the current data group of raw fuel parameters and the fitting relationship between the airflow parameters and the raw fuel parameters. The current data group of raw fuel parameters is collected by the data collection module.

9. An electronic device, characterized in that: The electronic device comprises: one or more processors; A storage device for storing one or more programs, which, when executed by the one or more processors, enables the electronic device to implement the blast furnace airflow distribution prediction method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed by a processor of a computer, the computer is enabled to execute the blast furnace airflow distribution prediction method as described in any one of claims 1-7.

Images