An oxygen lance control method, system, electronic device and storage medium

Abstract

The present application provides a lance control method, system, electronic device and storage medium. The lance control method includes obtaining historical smelting images of the converter mouth to construct a training set and a test set, obtaining historical steelmaking data of converter steelmaking, training a state prediction model with the training set, and testing the state prediction model with the test set to obtain a trained state prediction model. Determine the initial control parameters of the lance according to the historical steelmaking data of converter steelmaking, control the lance based on the initial control parameters, and obtain the mouth smelting image of the converter during the smelting process in real time. According to the mouth smelting image, predict the smelting situation of the converter through the state prediction model, correct the initial control parameters based on the smelting situation of the converter, and adjust the lance according to the corrected control parameters. The present application improves the intelligent level.

Description

Technical Field

[0001] This application relates to the technical field of steelmaking, and particularly to a method, a system, an electronic device and a storage medium for controlling an oxygen lance. Background Art

[0002] Converter steelmaking uses molten iron, scrap steel, and ferroalloys as the main raw materials. The steelmaking process is completed in the converter using the physical heat of the molten iron itself and the heat generated by chemical reactions between the components of the molten iron. During this process, harmful elements such as P and S in the molten iron are removed, and beneficial elements such as Mn and Cr are added to the molten iron to improve the properties of the produced steel. In the converter steelmaking process, the oxygen lance is one of the essential main process equipment. The control quality of the oxygen lance directly affects the smelting effect and blowing time, and thus affects the output and quality of the steel.

[0003] Currently, in the existing converter steelmaking process, the operation of the oxygen lance is mainly completed manually. The operator raises or lowers the oxygen lance based on experience. However, when producing in this way, it is difficult for the insertion position and depth of the oxygen lance to meet the specified requirements each time, which affects the accuracy and stability of production. At the same time, the on-site operation is complex, the labor intensity is high, and there are many situations to handle.

[0004] In summary, there is an urgent need for an intelligent control method for the oxygen lance to improve accuracy and stability and reduce labor intensity. Summary of the Invention

[0005] In view of the above-mentioned disadvantages of the prior art, the present invention provides a method, a system, an electronic device and a storage medium for controlling an oxygen lance to improve accuracy and stability and reduce labor intensity.

[0006] This application provides an oxygen lance control method, including:

[0007] Obtain historical smelting images of the converter furnace mouth to construct a training set and a test set, and obtain historical steelmaking data of converter steelmaking. The historical steelmaking data of converter steelmaking includes historical initial smelting parameters, historical oxygen lance height, historical time points corresponding to the historical oxygen lance height, and historical steelmaking duration;

[0008] Train a state prediction model using the training set, and test the state prediction model using the test set to obtain a trained state prediction model. The state prediction model takes the historical furnace mouth image of converter steelmaking as the input and the converter steelmaking state as the output;

[0009] Determine the initial control parameters of the oxygen lance according to the historical steelmaking data of converter steelmaking;

[0010] Control the oxygen lance based on the initial control parameters, and obtain the furnace mouth smelting image of the converter in real time during the smelting process;

[0011] Based on the smelting image at the furnace mouth, predict the smelting condition of the converter through a state prediction model, where the smelting condition includes abnormal states;

[0012] Modify the initial control parameters based on the smelting condition of the converter, and adjust the oxygen lance according to the modified control parameters.

[0013] In an exemplary embodiment of the present application, determining the initial control parameters of the oxygen lance according to the historical steelmaking data of the converter includes:

[0014] Obtain the initial parameters of the current converter steelmaking;

[0015] According to the current initial converter steelmaking parameters and the historical steelmaking data, obtain the steelmaking furnace numbers corresponding to the historical steelmaking data similar to the current initial converter steelmaking parameters;

[0016] Determine the oxygen lance height corresponding to the steelmaking furnace number and the time point corresponding to the oxygen lance height as the initial control parameters.

[0017] In an exemplary embodiment of the present application, obtaining the steelmaking furnace numbers corresponding to the historical steelmaking data similar to the current initial converter steelmaking conditions according to the current initial converter steelmaking conditions and the historical steelmaking data includes:

[0018] Perform similarity analysis on the historical initial smelting parameters to obtain several steelmaking furnace numbers corresponding to the historical initial smelting parameters similar to the current initial converter steelmaking initial smelting parameters;

[0019] Cluster the several steelmaking furnace numbers according to the oxygen lance height, obtain the evaluation index of the historical finished steel quality index corresponding to each category of steelmaking furnace numbers, and determine the steelmaking furnace numbers with the evaluation index as the steelmaking furnace numbers corresponding to the historical initial smelting parameters similar to the current initial converter steelmaking conditions.

[0020] In an exemplary embodiment of the present application, before obtaining the evaluation index of the historical finished steel quality index corresponding to each category of steelmaking furnace numbers and determining the steelmaking furnace number with the highest evaluation index as the steelmaking furnace number corresponding to the historical initial smelting parameters similar to the current initial converter steelmaking conditions, it further includes:

[0021] Remove the steelmaking furnace numbers whose historical finished steel quality indicators are not within the preset quality index range.

[0022] In an exemplary embodiment of the present application, performing similarity analysis on the historical initial smelting parameters includes:

[0023] Perform normalization processing on each index data of the historical initial smelting parameters to obtain the historical initial smelting parameters after normalization processing;

[0024] Calculate the difference between the index data in the current initial parameters of converter steelmaking and the index data corresponding to the normalized historical initial smelting parameters.

[0025] Determine the similarity according to the difference.

[0026] Determine several steelmaking heats corresponding to the historical initial smelting parameters with similarity ranking in the front as several steelmaking heats corresponding to the historical initial smelting parameters similar to the current initial parameters of converter steelmaking.

[0027] In an exemplary embodiment of the present application, clustering the several steelmaking heats according to the lance height includes:

[0028] Perform normalization processing on the steelmaking duration and lance height of the several steelmaking heats.

[0029] Calculate the lance height difference corresponding to the same time point of two steelmaking heats after normalization processing.

[0030] Cluster according to the maximum distance level according to the lance height difference and a preset clustering radius.

[0031] In an exemplary embodiment of the present application, correct the initial control parameters based on the smelting condition of the converter, and adjust the lance according to the corrected control parameters, including:

[0032] If the predicted smelting condition of the converter is an abnormal state, determine the lance control parameters according to the corresponding relationship between the preset abnormal state - lance height adjustment method.

[0033] Control the lance to adjust according to the lance control parameters.

[0034] In a second aspect, the present application provides a lance control system, including:

[0035] An acquisition module, configured to obtain historical smelting images of the converter mouth to construct a training set and a test set, and obtain historical steelmaking data of converter steelmaking, where the historical steelmaking data of converter steelmaking includes historical initial smelting parameters, historical lance height, historical time points corresponding to the historical lance height, and historical steelmaking duration.

[0036] A state prediction model construction module, configured to train a state prediction model through the training set, and test the state prediction model with the test set to obtain a trained state prediction model, where the state prediction model takes the historical converter mouth image of converter steelmaking as an input and the converter steelmaking state as an output.

[0037] An initial control parameter determination module, configured to determine the initial control parameters of the lance according to the historical steelmaking data of converter steelmaking.

[0038] A processing module, configured to control the oxygen lance based on initial control parameters, and configured to obtain a smelting image of the converter mouth during the smelting process in real time;

[0039] A smelting condition prediction module, configured to predict the smelting condition of the converter according to the smelting image of the converter mouth through a state prediction model, where the smelting condition includes an abnormal state;

[0040] A control module, configured to correct the initial control parameters based on the smelting condition of the converter, and adjust the oxygen lance according to the corrected control parameters.

[0041] In another aspect, the present application further provides an electronic device, including: one or more processors; a storage device, configured to store one or more programs, when the one or more programs are executed by the one or more processors, enabling the electronic device to implement the oxygen lance control method provided above.

[0042] In still another aspect, the present application 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, enabling the computer to implement the oxygen lance control method described above.

[0043] Advantageous effects of the present invention:

[0044] In the present application, the state prediction model associates the converter mouth image in converter steelmaking with the converter steelmaking state, and predicts the smelting condition based on the state prediction model. If the smelting condition is an abnormal state, the oxygen lance height is adjusted according to the preset correspondence between the abnormal state and the oxygen lance height adjustment method, thereby realizing the intelligent control of the oxygen lance. On the premise of ensuring reliable smelting, the smelting process is made more accurate and stable. At the same time, the labor intensity of operators is reduced, and the intelligent production level is improved.

[0045] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present application. Description of the drawings

[0046] The drawings here are incorporated into the specification and constitute a part of this specification, showing embodiments consistent with the present application, and are used together with the specification to explain the principles of the present application. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts. In the drawings:

[0047] Figure 1 Is a flowchart of the oxygen lance control method shown in an exemplary embodiment of the present application;

[0048] Figure 2 Is Figure 1Flowchart of step S130 in the illustrated embodiment in an exemplary embodiment;

[0049] Figure 3 is Figure 2 Flowchart of step S220 in the illustrated embodiment in an exemplary embodiment;

[0050] Figure 4 is Figure 3 Flowchart of step S310 in the illustrated embodiment in an exemplary embodiment;

[0051] Figure 5 is Figure 3 Flowchart of clustering several steelmaking furnace batches according to the lance height in step S320 in the illustrated embodiment in an exemplary embodiment;

[0052] Figure 6 is Figure 1 Flowchart of step S160 in the illustrated embodiment in an exemplary embodiment;

[0053] Figure 7 Flowchart of the lance control method shown in a specific embodiment;

[0054] Figure 8 is Figure 7 Clustering result diagram of clustering according to the lance height in the illustrated embodiment;

[0055] Figure 9 Block diagram of the lance control system shown in an exemplary embodiment of the present application;

[0056] Figure 10 Shows a schematic structural diagram of a computer system of an electronic device suitable for implementing the embodiments of the present application. Detailed implementation manners

[0057] The following will describe the embodiments of the present invention with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are only for explaining the present invention, rather than for limiting the protection scope of the present invention.

[0058] It should be noted that the drawings provided in the following embodiments only illustrate the basic concept of the present invention schematically. Therefore, only the components related to the present invention are shown in the drawings, rather than being drawn according to the number, shape, and size of the components in actual implementation. The type, quantity, and ratio of each component in actual implementation can be arbitrarily changed, and the component layout type may also be more complex.

[0059] In the following description, a large number of details are explored to provide a more thorough explanation of the embodiments of the present invention. However, it is obvious to those skilled in the art that the embodiments of the present invention 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 invention difficult to understand.

[0060] Please refer to Figure 1 , Figure 1 which is a flowchart of an oxygen lance control method shown for an exemplary embodiment of the present application.

[0061] As Figure 1 shown, in an exemplary embodiment of the present application, the oxygen lance control method at least includes steps S110, S120, S130, S140, S150, and S160, which are introduced in detail as follows:

[0062] Step S110. Obtain historical smelting images of the converter mouth to construct a training set and a test set, and obtain historical converter steelmaking data;

[0063] It should be noted that the historical converter steelmaking data includes historical initial smelting parameters, historical oxygen lance height, historical time points corresponding to the historical oxygen lance height, and historical steelmaking duration.

[0064] Exemplarily, randomly divide 80% of the historical smelting images and the converter steelmaking states identified manually or by equipment according to the changes in the historical smelting images into the training set, and the remaining 20% of the historical smelting images and the converter steelmaking states identified manually or by equipment according to the changes in the historical smelting images into the test set.

[0065] The division method of the training set and the test set is not set and will not be elaborated here.

[0066] Step S120. Train the state prediction model with the training set and test the state prediction model with the test set to obtain a trained state prediction model;

[0067] The prediction model takes the historical converter mouth images of converter steelmaking as the input and the converter steelmaking state as the output.

[0068] Specifically, input the historical smelting images in the test set into the state prediction model, output the converter steelmaking state (i.e., the prediction result), and compare the converter steelmaking state predicted by the prediction model with the converter steelmaking state identified manually (i.e., the actual result) corresponding to the historical smelting images in the test set.

[0069] If the accuracy is not less than a preset accuracy threshold, the state prediction model is determined as a trained state prediction model. Exemplarily, if the proportion of the prediction results in the test set that match the actual results is not less than the preset accuracy threshold (such as 95%), the accuracy of the state prediction model meets the requirements, and the state prediction model is determined as a trained state prediction model.

[0070] Step S130. Determine the initial control parameters of the oxygen lance according to the historical steelmaking data of converter steelmaking.

[0071] It should be noted that the historical steelmaking data includes historical initial smelting parameters, historical oxygen lance height, historical time points corresponding to the historical oxygen lance height, and historical steelmaking duration.

[0072] Step S140. Control the oxygen lance based on the initial control parameters, and obtain the furnace mouth smelting image of the converter in real time during the smelting process.

[0073] Step S150. Predict the smelting condition of the converter through the state prediction model according to the furnace mouth smelting image.

[0074] It should be noted that the smelting condition includes abnormal states, and the abnormal states include splashing, back drying, excessive high temperature (exceeding the preset temperature threshold), etc.

[0075] Step S160. Correct the initial control parameters based on the smelting condition of the converter, and adjust the oxygen lance according to the corrected control parameters.

[0076] Exemplarily, if the smelting condition of the converter predicted by the state prediction model is an abnormal state, correct the initial control parameters according to the preset abnormal state - oxygen lance height adjustment method, and adjust the oxygen lance according to the corrected control parameters.

[0077] The preset abnormal state - oxygen lance height adjustment method associates different abnormal situations with corresponding oxygen lance height adjustment methods. Therefore, when the steelmaking state is an abnormal state, the corresponding oxygen lance height adjustment method can be determined according to the preset correspondence between the abnormal state and the oxygen lance height adjustment method.

[0078] In the related art, the operation process of the oxygen lance is mainly completed manually. The operator raises or lowers the oxygen lance according to experience. After analyzing the above solution, the inventor found that when producing according to the related art, it is difficult to make the insertion position and depth of the oxygen lance meet the specified requirements every time, which affects the accuracy and stability of temperature measurement and sampling. At the same time, the on-site operation is complex, the labor intensity is high, and many situations need to be dealt with. Therefore, it is considered to associate the converter steelmaking furnace mouth image with the smelting state through a state prediction model, and based on the state prediction model, the smelting situation of the converter is predicted in real time. Furthermore, the initial control parameters are corrected based on the smelting situation of the converter, and the oxygen lance is adjusted according to the corrected control parameters to achieve the intelligent control of the oxygen lance. On the premise of ensuring reliable smelting, the smelting process is made more accurate and stable. At the same time, the labor intensity of the operator is reduced, and the intelligent production level is improved.

[0079] Please refer to Figure 2 , Figure 2 For Figure 1 the flowchart of step S130 in the exemplary embodiment shown.

[0080] As Figure 2 shown, in an exemplary embodiment of the present application, Figure 1 the process of determining the initial control parameters of the oxygen lance according to the historical steelmaking data of the converter in the shown embodiment includes step S210, step S220, and step S230, which are introduced in detail as follows:

[0081] Step S210. Obtain the initial parameters of the current converter steelmaking;

[0082] Step S220. According to the initial parameters of the current converter steelmaking and the historical steelmaking data, obtain the steelmaking furnace numbers corresponding to the historical steelmaking data similar to the initial parameters of the current converter steelmaking;

[0083] Step S230. Determine the oxygen lance height corresponding to the steelmaking furnace number and the time point corresponding to the oxygen lance height as the initial control parameters.

[0084] Please refer to Figure 3 , Figure 3 For Figure 2 the flowchart of step S220 in the exemplary embodiment shown.

[0085] As Figure 3 shown, in an exemplary embodiment of the present application, Figure 2 the process of obtaining the steelmaking furnace numbers corresponding to the historical steelmaking data similar to the initial conditions of the current converter steelmaking according to the initial conditions of the current converter steelmaking and the historical steelmaking data in the shown embodiment includes step S310 and step S320, which are introduced in detail as follows:

[0086] Step S310. Perform a similarity analysis on the historical initial smelting parameters to obtain several steelmaking heats corresponding to the historical initial smelting parameters that are similar to the current initial smelting parameters of converter steelmaking.

[0087] Step S320. Cluster several steelmaking heats according to the lance height, obtain the evaluation index of the historical finished steel quality index corresponding to each category of steelmaking heats, and determine the steelmaking heats corresponding to the historical initial smelting parameters that are similar to the current initial conditions of converter steelmaking based on the evaluation index.

[0088] Exemplarily, obtaining the evaluation index of the historical finished steel quality index corresponding to each category of steelmaking heats can be performed in the following manner.

[0089] Determine the quality index corresponding to each finished steel quality index according to the preset finished steel quality index - quality index correspondence relationship, then multiply the quality index corresponding to each finished steel quality index by its corresponding preset weight, and then sum them to obtain the evaluation index.

[0090] Please refer to Figure 4 , Figure 4 For Figure 3 the flowchart of step S310 in the exemplary embodiment shown.

[0091] As Figure 4 shown, in an exemplary embodiment of the present application, Figure 3 the process of performing a similarity analysis on the historical initial smelting parameters in the shown embodiment includes step S410, step S420, step 430, and step S440, which are introduced in detail as follows:

[0092] Step S410. Perform normalization processing on each index data of the historical initial smelting parameters to obtain the historical initial smelting parameters after normalization processing.

[0093] Specifically, perform normalization processing on each index data of the historical initial smelting parameters (such as scrap weight, initial carbon element content, etc.), that is, convert each index data into a specific value between 0 and 1, specifically: (data - Min) / (Max - Min);

[0094] where data is the index data in the initial conditions, Min is the minimum value of the corresponding index data in the initial conditions, and Max is the maximum value of the corresponding index data in the initial conditions.

[0095] Step S420. Calculate the difference between each index data in the current initial parameters of converter steelmaking and the index data corresponding to the historical initial smelting parameters after normalization processing.

[0096] Step S430. Determine the similarity according to the difference.

[0097] Exemplarily, according to the difference, the way to determine the similarity can be to take the square root of the weighted sum of the differences of each index data, and the obtained value can be determined as the similarity.

[0098] Step S440. Determine several steelmaking furnace charges corresponding to several historical initial smelting parameters with the top similarity rankings as several steelmaking furnace charges corresponding to historical initial smelting parameters similar to the current converter steelmaking initial parameters.

[0099] Please refer to Figure 5 , Figure 5 as Figure 3 the flowchart of step S320 in the exemplary embodiment shown.

[0100] As Figure 5 shown, in an exemplary embodiment of the present application, Figure 3 the process of clustering the several steelmaking furnace charges according to the lance height in the shown embodiment includes step S510, step S520, and step 530, which are introduced in detail as follows:

[0101] Step S510. Perform normalization processing on the steelmaking duration and lance height of several steelmaking furnace charges;

[0102] Specifically, the normalization processing can be performed according to experience or preset conditions. For example, take the median value of the steelmaking duration of several steelmaking furnace charges as the steelmaking duration after normalization processing, and take the median value of the lance height of several steelmaking furnace charges as the lance height after normalization processing.

[0103] Step S520. Calculate the lance height difference corresponding to the same time point of each pair of steelmaking furnace charges after normalization processing;

[0104] It should be noted that in the present application, the term "time point" takes the starting time of converter steelmaking as the 0 moment, and the calculated time point (that is, the difference between each moment in the steelmaking process and the starting time of converter steelmaking). For example, the 600s refers to the time point in the steelmaking process that is 600s away from the starting time point of converter steelmaking.

[0105] Step 530. Cluster according to the lance height difference and the preset clustering radius in the maximum distance hierarchy.

[0106] In another exemplary embodiment of the present application, before obtaining the evaluation index of the historical finished steel quality index corresponding to the steelmaking furnace charges of each category and determining the steelmaking furnace charge with the highest evaluation index as the steelmaking furnace charge corresponding to the historical initial smelting parameters similar to the current converter steelmaking initial conditions, the following steps are further included:

[0107] Remove the steelmaking furnace charges whose historical finished steel quality indexes are not within the preset quality index range.

[0108] The preset quality index range can be set by oneself and will not be elaborated here.

[0109] By removing the steelmaking heats whose historical finished steel quality indexes are not within the preset quality index range, the accuracy of oxygen lance control can be improved, and the quality of finished steel can be guaranteed.

[0110] Please refer to Figure 6 , Figure 6 for Figure 1 the flowchart of step S160 in the exemplary embodiment shown.

[0111] As Figure 6 shown, in an exemplary embodiment of the present application, Figure 1 in the shown embodiment, the process of correcting the initial control parameters based on the smelting condition of the converter and adjusting the oxygen lance according to the corrected control parameters includes step S610 and step S620, which are introduced in detail as follows:

[0112] Step S610. If the predicted smelting condition of the converter is an abnormal state, determine the oxygen lance control parameters according to the preset correspondence between abnormal state - oxygen lance height adjustment method;

[0113] The preset correspondence between abnormal state - oxygen lance height adjustment method can be set by oneself and will not be elaborated here.

[0114] Step S620. Control the oxygen lance to be adjusted according to the oxygen lance control parameters.

[0115] Please refer to Figure 7 , Figure 7 for the flowchart of the oxygen lance control method shown in a specific embodiment.

[0116] As Figure 7 shown, in a specific embodiment, the specific steps of the oxygen lance control method are as follows:

[0117] S1. Obtain the historical data of converter steelmaking. The historical data includes the heat number of each heat in the converter smelting in the steel plant in the recent three months, the initial smelting parameters (such as scrap weight, initial P content, initial C content, target temperature, target C content, etc.), the height data of the oxygen lance during smelting collected every second, the state data during smelting (abnormal conditions such as splashing information, dry return, etc.), and the actual composition, temperature, performance, etc. of the molten steel after smelting; the historical data is arranged, stored, and indexed in the order of heat smelting, and the heat number is the primary key corresponding to all data, and the corresponding data can be found by retrieving the heat number;

[0118] S2. Calculate the similarity, then cluster the lance operating height, calculate the score of the smelting result, and obtain the historical heats with initial conditions very similar to those of the current heat and good smelting effects in the historical heat data.

[0119] Specifically, perform normalization operations on the data of each index (such as scrap weight, initial C content, etc.) in the initial conditions, that is, uniformly scale the range to between 0 and 1, specifically: (data - Min) / (Max - Min);

[0120] Among them, data is the index data in the initial conditions, Min is the minimum value of the corresponding index data in the initial conditions, and Max is the maximum value of the corresponding index data in the initial conditions.

[0121] Then calculate the difference from the data of each index of the initial parameters of converter steelmaking in the current heat (normalized first according to the above calculation formula), and take the square root of the weighted sum of the squares of the differences of each index data as the similarity. The weighting coefficient is preset according to the importance of the elements;

[0122] Select the top 50 heats with the highest similarity rankings for further analysis;

[0123] Cluster the lance heights of the top 50 heats with the highest similarity rankings selected. Since the steelmaking time of each heat is inconsistent, make up for the deficiencies in the steelmaking time and lance height, so that the lance height data of each heat is consistent in time series. Specifically, take the data of the first 600 seconds, and then reduce the dimension to the average value of every 6 - second data as the new data, reducing the dimension to 100 data;

[0124] Then calculate the average value of the distances of the 100 - point lance height time - series data between every two heats, and classify them into 5 categories according to the maximum - distance hierarchical clustering with a given preset clustering radius as the clustering radius.

[0125] Evaluate the smelting results of each heat in each of the 5 categories respectively; calculate the evaluation value according to the sum of the weights (preset) of each finished - steel quality index. The weights of each finished - steel quality index are different. When the data of a very important finished - steel quality index does not meet the standard, a one - vote veto is implemented and the score is 0; among the 5 - category data, if a certain category of data contains very few heats, such as less than or equal to 3 heats, then this category of data is excluded to improve the representativeness of the clustering - recommended data. Among them, the clustering results are as Figure 8 shown, where the abscissa time is time, with the unit of s; the ordinate high is the lance height, with the unit of m.

[0126] Determine the steel - making heat with the highest evaluation value as the steel - making heat corresponding to the historical initial smelting parameters similar to the initial parameters of the current converter steelmaking;

[0127] S3. Assign the timing data of the oxygen lance height operation (including the steelmaking time points and the corresponding oxygen lance height data) of the selected steelmaking furnace to the operation data of the oxygen lance automation of the current furnace as the initial control parameters of the oxygen lance, and control the oxygen lance according to the initial control parameters to carry out production;

[0128] S4. Divide the historical converter steelmaking hearth images (image information of 50 frames in the first 10 seconds) and the historical steelmaking states of the converter steelmaking (classification data with labels such as splashing and drying back) into a training set and a test set, where the training set accounts for 80%; define the number of layers of the LSTM network model as 3 layers, train the data, and obtain a state prediction model;

[0129] Specifically, use the historical smelting images in the training set as the input, and use the converter steelmaking states identified manually or by equipment according to the changes in the historical smelting images as the output to train the state prediction model.

[0130] Use the data in the test set to test the accuracy of the state prediction model. Specifically, input the historical hearth images in the test set into the state prediction model, and output the historical steelmaking states of the converter steelmaking corresponding to the historical hearth images (i.e., the prediction results). Compare the prediction results with the corresponding converter steelmaking states (i.e., the actual results) identified manually or by equipment according to the changes in the historical smelting images in the test set. If the proportion of the prediction results in the test set that match the actual results is not less than the preset accuracy threshold (such as 95%), the accuracy of the state prediction model meets the requirements, and the state prediction model is determined as the trained state prediction model.

[0131] During the smelting process, obtain the converter steelmaking hearth images in real time, input the converter steelmaking hearth images into the state prediction model, and predict the smelting conditions that will occur in the next short period of time;

[0132] S5. According to the predicted smelting conditions, if the predicted smelting conditions are abnormal states, determine the oxygen lance control parameters according to the corresponding relationship between the preset abnormal states and the oxygen lance height adjustment methods, and control the oxygen lance to adjust according to the oxygen lance control parameters to automatically correct the oxygen lance height and prevent adverse phenomena such as splashing and drying back;

[0133] S6. After the smelting is completed, collect the smelting data of this furnace and store it in the historical database, waiting for the next smelting.

[0134] Please refer to Figure 9 , this application embodiment also provides an oxygen lance control system 900.

[0135] As Figure 9 shown, the oxygen lance control system 900 of this application embodiment includes:

[0136] The acquisition module 910 is used to obtain the historical smelting images of the converter furnace mouth to construct a training set and a test set, and obtain the historical steelmaking data of converter steelmaking. The historical steelmaking data of converter steelmaking includes historical initial smelting parameters, historical oxygen lance height, the historical time point corresponding to the historical oxygen lance height, and historical steelmaking duration;

[0137] The state prediction model construction module 920 is used to train the state prediction model through the training set and test the state prediction model with the test set to obtain a trained state prediction model. The prediction model takes the historical furnace mouth image of converter steelmaking as the input and the converter steelmaking state as the output;

[0138] The initial control parameter determination module 930 is used to determine the initial control parameters of the oxygen lance according to the historical steelmaking data of converter steelmaking;

[0139] The processing module 940 is used to control the oxygen lance based on the initial control parameters and is used to obtain the furnace mouth smelting image of the converter in real time during the smelting process;

[0140] The smelting situation prediction module 950 is used to predict the smelting situation of the converter through the state prediction model according to the furnace mouth smelting image;

[0141] The control module 960 corrects the initial control parameters based on the smelting situation of the converter and adjusts the oxygen lance according to the corrected control parameters.

[0142] It should be noted that the oxygen lance control system provided in the above embodiment and the oxygen lance control method provided in the above embodiment belong to the same concept. The specific ways in which each module and unit perform operations have been described in detail in the method embodiment and will not be repeated here. In practical applications, the oxygen lance control system provided in the above embodiment can, according to needs, allocate the above functions to different functional modules, that is, divide the internal structure of the device into different functional modules to complete all or part of the functions described above. This is not limited here.

[0143] An embodiment of the present application further 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 realizes the oxygen lance control method provided in each of the above embodiments.

[0144] Figure 10 The structural schematic diagram of the computer system of the electronic device suitable for implementing the embodiment of the present application is shown. It should be noted that Figure 10 The computer system 1000 of the electronic device shown is only an example and should not bring any limitations to the functions and usage scope of the embodiment of the present application.

[0145] Such asFigure 10 As shown in Figure 10 , the computer system 1000 includes a Central Processing Unit (CPU) 1001, which can perform various appropriate actions and processes according to the program stored in the Read-Only Memory (ROM) 1002 or the program loaded from the storage section into the Random Access Memory (RAM) 1003, such as executing the methods described in the above embodiments. In the RAM 1003, various programs and data required for system operation are also stored. The CPU 1001, ROM 1002, and RAM 1003 are connected to each other via a bus 1004. An Input / Output (I / O) interface 1005 is also connected to the bus 1004.

[0146] The following components are connected to the I / O interface 1005: an input section 1006 including a keyboard, a mouse, etc.; an output section 1007 including, for example, a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), etc., and a speaker, etc.; a storage section including a hard disk, etc.; and a communication section 1009 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 1009 performs communication processing via a network such as the Internet. A drive 1010 is also connected to the I / O interface 1005 as needed. A removable medium 1011, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 1010 as needed, so that the computer program read from it can be installed into the storage section as needed.

[0147] Specifically, according to the embodiments of the present application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, the embodiments of the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a computer program for executing the methods shown in the flowcharts. In such an embodiment, the computer program can be downloaded and installed from the network through the communication section 1009, and / or installed from the removable medium 1011. When the computer program is executed by the Central Processing Unit (CPU) 1001, various functions defined in the system of the present application are executed.

[0148] It should be noted that the computer-readable medium shown in the embodiments of the present application may be a computer-readable signal medium, a computer-readable storage medium, or any combination of the two. A computer-readable storage medium may be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of the computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable computer program. Such a propagated data signal may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device. The computer program contained on the computer-readable medium may be transmitted by any appropriate medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0149] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present application. Among them, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code, and the above module, program segment, or part of code contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order from that marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram or flowchart, as well as the combination of blocks in the block diagram or flowchart, may be implemented by a dedicated hardware-based system for performing the specified functions or operations, or may be implemented by a combination of dedicated hardware and computer instructions.

[0150] The units involved in the embodiments described in this application can be implemented in software or in hardware, and the described units can also be provided in a processor. Among them, the names of these units do not, in some cases, constitute a limitation on the units themselves.

[0151] Another aspect of this application 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 is caused to execute the oxygen lance control method as described above. The computer-readable storage medium may be included in the electronic device described in the above embodiments, or may exist separately without being assembled into the electronic device.

[0152] Another aspect of this application also provides a computer program product or a computer program. The computer program product or the computer program includes computer instructions, and the computer instructions are 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 oxygen lance control method provided in the above various embodiments.

[0153] The above embodiments only exemplarily illustrate the principles and effects of the present invention, rather than limiting the present invention. Any person familiar with this technology can modify or change the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or changes completed by those with ordinary knowledge in the technical field without departing from the spirit and technical ideas disclosed by the present invention should still be covered by the claims of the present invention.

Claims

1. An oxygen lance control method, characterized in that, Including: Obtain historical smelting images of the converter mouth to construct a training set and a test set, and obtain historical steelmaking data of converter steelmaking. The historical steelmaking data of converter steelmaking includes historical initial smelting parameters, historical lance heights, historical time points corresponding to the historical lance heights, and historical steelmaking durations; Train a state prediction model with the training set and test the state prediction model with the test set to obtain a trained state prediction model. The state prediction model takes the historical converter mouth images of converter steelmaking as input and the converter steelmaking state as output; Obtain the current initial smelting parameters of converter steelmaking; Conduct a similarity analysis on the historical initial smelting parameters to obtain several steelmaking heats corresponding to the historical initial smelting parameters similar to the current initial smelting parameters of converter steelmaking; Cluster the several steelmaking heats according to the lance height, obtain the evaluation index of the historical finished steel quality index corresponding to each class of steelmaking heats, and determine the steelmaking heats with the evaluation index as the steelmaking heats corresponding to the historical initial smelting parameters similar to the current initial conditions of converter steelmaking; Determine the lance height corresponding to the steelmaking heat and the time point corresponding to the lance height as the initial control parameters; Control the lance based on the initial control parameters and obtain the mouth smelting image of the converter during smelting in real time; According to the mouth smelting image, predict the smelting situation of the converter through the state prediction model. The smelting situation includes abnormal states; Correct the initial control parameters based on the smelting situation of the converter and adjust the lance according to the corrected control parameters.

2. The oxygen lance control method according to claim 1, wherein Before obtaining the evaluation index of the historical finished steel quality index corresponding to each class of steelmaking heats and determining the steelmaking heat with the highest evaluation index as the steelmaking heat corresponding to the historical initial smelting parameters similar to the current initial conditions of converter steelmaking, it further includes: Remove the steelmaking heats whose historical finished steel quality indexes are not within the preset quality index range.

3. The oxygen lance control method according to claim 1, characterized in that, The similarity analysis of the historical initial smelting parameters includes: Normalize the index data of the historical initial smelting parameters to obtain the normalized historical initial smelting parameters; Calculate the difference between the index data in the current initial smelting parameters of converter steelmaking and the index data corresponding to the normalized historical initial smelting parameters; Determine the similarity according to the difference; Determine several steelmaking heats corresponding to several historical initial smelting parameters with the top similarity rankings as several steelmaking heats corresponding to the historical initial smelting parameters similar to the current initial smelting parameters of converter steelmaking.

4. The oxygen lance control method according to claim 1, characterized in that, Clustering the several steelmaking heats according to the lance height includes: Normalize the steelmaking durations and lance heights of the several steelmaking heats; Calculate the lance height difference corresponding to the same time point of two normalized steelmaking heats; Cluster according to the maximum distance level according to the lance height difference and the preset clustering radius.

5. The oxygen lance control method according to claim 1, characterized in that Correct the initial control parameters based on the smelting situation of the converter and adjust the lance according to the corrected control parameters, including: If the predicted smelting condition of the converter is an abnormal state, determine the oxygen lance control parameters according to the preset correspondence between abnormal states and oxygen lance height adjustment methods; Control the oxygen lance to adjust according to the oxygen lance control parameters.

6. An oxygen lance control system, characterized in that, It includes: A collection module for obtaining historical smelting images of the converter mouth to construct a training set and a test set, and obtaining historical converter steelmaking data, where the historical converter steelmaking data includes historical initial smelting parameters, historical oxygen lance height, historical time points corresponding to the historical oxygen lance height, and historical steelmaking duration; A state prediction model construction module for training the state prediction model through the training set and testing the state prediction model with the test set to obtain a trained state prediction model, where the state prediction model takes the historical converter mouth image as the input and the converter steelmaking state as the output; An initial control parameter determination module for obtaining the current initial smelting parameters of the converter steelmaking; performing a similarity analysis on the historical initial smelting parameters to obtain several steelmaking heats corresponding to the historical initial smelting parameters similar to the current initial smelting parameters of the converter steelmaking; clustering the several steelmaking heats according to the oxygen lance height, obtaining the evaluation index of the historical finished steel quality index corresponding to each category of steelmaking heats, and determining the steelmaking heat with the evaluation index as the steelmaking heat corresponding to the historical initial smelting parameters similar to the current initial conditions of the converter steelmaking; determining the oxygen lance height corresponding to the steelmaking heat and the time point corresponding to the oxygen lance height as the initial control parameter; A processing module for controlling the oxygen lance based on the initial control parameter and for obtaining the smelting image of the converter mouth during the smelting process in real time; A smelting condition prediction module for predicting the smelting condition of the converter through the state prediction model according to the smelting image of the converter mouth; A control module for correcting the initial control parameter based on the smelting condition of the converter and adjusting the oxygen lance according to the corrected control parameter.

7. An electronic device, characterized in that, It includes: One or more processors; A storage device for storing one or more programs, which when executed by the one or more processors, cause the electronic device to implement the oxygen lance control method according to any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, A computer program is stored thereon, which when executed by the processor of the computer, causes the computer to execute the oxygen lance control method according to any one of claims 1-5.

Images