Converter steelmaking molten pool carbon content detection method and system, medium and terminal

Abstract

The invention discloses a converter steelmaking molten pool carbon content detection method and system, a medium and a terminal, and the method comprises the steps: responding to a converter steelmaking signal, starting a three-section model, and obtaining the height of an oxygen lance and the oxygen flow of the oxygen lance in real time in the converter steelmaking process; a first carbon content detection value of a molten pool in the converter at the current moment is obtained through real-time analysis by utilizing a three-section type model and combining the height of an oxygen lance, the oxygen flow of the oxygen lance, the weight of waste steel added into the converter, the weight of molten iron and the carbon content; when the converter is in the last steelmaking stage, if a detection signal sent by the detection equipment is received, a second carbon content detection value of the molten pool in the converter at the current moment is obtained through real-time analysis based on the detection signal, and the first carbon content detection value and the second carbon content detection value are fused to obtain a carbon content detection result of the molten pool at the current moment. According to the method, the influence of the oxygen lance position, the oxygen flow and feeding on the decarburization rate is considered, the carbon content detection in the whole steelmaking process is realized, and the detection precision in the last steelmaking stage is optimized.

Description

Technical Field

[0001] The present invention relates to the field of converter steelmaking, and particularly to a method, system, medium and terminal for detecting the carbon content in the molten pool of converter steelmaking. Background Art

[0002] During the process of converter steelmaking, the control of the carbon content in the molten steel is an important content. At present, the methods for directly measuring or indirectly inferring the carbon content in the molten steel are: using the sublance measurement method, the furnace gas analysis inference method and the furnace mouth flame inference method. These three methods are only applicable to the later stage and the end point of converter steelmaking, and can measure or infer the carbon content at the end point of the converter, but cannot predict the carbon content in the molten steel during the steelmaking process. For the description of the decarburization rate of the molten steel during the steelmaking process, there is a classic three-stage decarburization rate equation. The classic three-stage decarburization rate equation describes the decarburization rate in the early, middle and later stages of converter smelting, which is convenient for engineers to understand the change law of the carbon content in the molten pool during the blowing process. However, it does not consider the influence of the lance position, oxygen flow rate and feeding on the decarburization rate, and cannot accurately describe the decarburization rate and the real-time carbon content, so it cannot be applied to the actual site. Summary of the Invention

[0003] Embodiments of the present invention provide a method, system, medium and terminal for detecting the carbon content in the molten pool of converter steelmaking. By using a three-stage model and considering the influence of the lance position, oxygen flow rate and feeding on the decarburization rate, the carbon content detection throughout the steelmaking process is realized, and further detection is performed using detection equipment at the end of steelmaking to improve the detection accuracy.

[0004] To solve the above technical problems, embodiments of the present invention provide a method for detecting the carbon content in the molten pool of converter steelmaking, including:

[0005] Responding to the converter steelmaking signal, starting the three-stage model, and during the converter steelmaking process, obtaining the lance height and the lance oxygen flow rate in real time;

[0006] Using the three-stage model, combining the lance height, the lance oxygen flow rate, and the weights of scrap steel, hot metal and carbon content added to the converter, and analyzing in real time to obtain a first carbon content detection value of the molten pool in the converter at the current moment;

[0007] When the converter is in the late stage of steelmaking, if a detection signal sent by the detection equipment is received, then based on the detection signal, analyzing in real time to obtain a second carbon content detection value of the molten pool in the converter at the current moment, and fusing the first carbon content detection value and the second carbon content detection value to obtain a third carbon content detection value of the molten pool in the converter at the current moment in the late stage of steelmaking, so as to complete the detection of the carbon content of the molten pool in the converter in the late stage of steelmaking.

[0008] Implementing the embodiments of the present invention, in response to the converter steelmaking signal, start the three-stage model, and during the converter steelmaking process, real-time obtain the lance height and lance oxygen flow rate. Then, use the three-stage model, combine the real-time obtained lance height and lance oxygen flow rate, as well as the scrap weight, hot metal weight and carbon content added to the converter, and real-time analyze to obtain the decarburization rate and the first carbon content detection value of the molten bath in the converter at the current moment. That is, when receiving the converter steelmaking signal, start the three-stage model, and considering the influence of the lance position, oxygen flow rate and feeding on the decarburization rate, use the three-stage model, combine the real-time obtained lance height and lance oxygen flow rate, as well as the scrap weight, hot metal weight and carbon content added to the converter, and real-time analyze to obtain the decarburization rate and the first carbon content detection value of the molten bath in the converter at each moment during the whole steelmaking process, so as to improve the detection accuracy of the molten bath decarburization rate and carbon content, and thus realize the precise control of the steelmaking process. In addition, when the converter is in the final stage of steelmaking, if receiving the detection signal sent by the detection device, then based on the detection signal, real-time analyze to obtain the second carbon content detection value of the molten bath in the converter at the current moment, and fuse the first carbon content detection value and the second carbon content detection value to obtain the third carbon content detection value of the molten bath in the converter at the current moment in the final stage of steelmaking, which can not only complete the carbon content detection of the molten bath in the converter in the final stage of steelmaking, but also further optimize the accuracy of the carbon content detection result of the molten bath in the converter through the detection result of the detection device.

[0009] As a preferred solution, the detection device is a camera; when the converter is in the final stage of steelmaking, if receiving the detection signal sent by the detection device, then based on the detection signal, real-time analyze to obtain the second carbon content detection value of the molten bath in the converter at the current moment, specifically:

[0010] When the converter is in the final stage of steelmaking, if receiving the detection signal sent by the camera, then based on the detection signal, obtain the corresponding original image, and preprocess the original image to obtain the image to be analyzed; wherein, the original image is obtained by using the camera to real-time photograph the furnace mouth flame of the converter;

[0011] Perform feature description and extraction on the image to be analyzed to obtain the feature variables to be predicted;

[0012] Input the feature variables to be predicted into the carbon content detection model, so that the carbon content detection model outputs the second carbon content detection value of the molten bath in the converter at the current moment;

[0013] Among them, the carbon content detection model is obtained by training the pre-constructed GRNN model according to the sample data set.

[0014] Implementing the preferred solution of the embodiment of the present invention, using a camera to capture the flame at the furnace mouth of the converter in real time to obtain the original image, then preprocessing the original image, and performing feature description and extraction on the preprocessed image, can convert the key information in the image into feature variables that can be used for prediction, so that the model can better implement based on the feature variables, and input the feature variables into the trained GRNN model for carbon content detection, making the detection result more accurate and stable.

[0015] As a preferred solution, the feature description and extraction of the image to be analyzed to obtain the feature variables to be predicted are specifically as follows:

[0016] Extract and process the flame boundary of the image to be analyzed to separate the flame area and the background area in the image to be analyzed;

[0017] Solve the refined skeleton structure of the flame area and determine the center point of the flame area;

[0018] Obtain the polar coordinate description of the boundary of the flame area, and based on the polar coordinate description of the boundary of the flame area, solve the amplitude spectrum of the boundary of the flame area, and then calculate the average value, entropy and energy of the amplitude spectrum of the boundary of the flame area as the feature vector describing the boundary of the flame area;

[0019] Perform Gaussian normalization on the feature vector to obtain the feature variables to be predicted.

[0020] Implementing the preferred solution of the embodiment of the present invention, extracting and processing the flame boundary can effectively separate the flame area and the background area, reduce the influence of background noise on the extraction of flame features, so as to more concentratedly analyze the features of the flame area, improve the accuracy of subsequent analysis. Then, by solving the refined skeleton structure of the flame area, the shape and structural features of the flame can be more accurately described, and determining the center point of the flame area can provide an important spatial reference point for subsequent feature analysis, enhancing the spatial description ability of the features, thus facilitating subsequent feature analysis. In addition, by obtaining the polar coordinate description of the flame boundary and calculating the average value, entropy and energy of its amplitude spectrum, the average value, entropy and energy of the amplitude spectrum can reflect the shape and fluctuation characteristics of the flame boundary, so as to comprehensively describe the feature information of the flame boundary, provide rich input variables for subsequent carbon content prediction, enhance the diversity and robustness of the features, and make the prediction model more stable and accurate.

[0021] As a preferred solution, the detection device is a mass spectrometer; when the converter is in the late stage of steelmaking, if a detection signal sent by the detection device is received, then based on the detection signal, the second carbon content detection value of the molten pool in the converter at the current moment is analyzed in real time, specifically as follows:

[0022] When the converter is at the end of steelmaking, if a detection signal sent by the mass spectrometer is received, the CO and CO in the flue gas of the converter are obtained based on the detection signal. 2 The percentage content of CO and CO in the flue gas of the converter is 2 The percentage content of and the flue gas flow rate of the converter are obtained by detecting the flue gas of the converter using the mass spectrometer;

[0023] Based on the carbon balance relationship in the converter steelmaking process, and combined with the CO and CO in the converter flue gas 2 The percentage of and the flue gas flow rate of the converter are analyzed to obtain the decarburization rate of the molten pool in the converter at the current moment;

[0024] According to the preset carbon integral model, based on the decarburization rate of the molten pool in the converter at the current moment, the quality of the molten steel in the molten pool in the converter, and the minimum carbon content of the molten pool in the converter, the second carbon content detection value of the molten pool in the converter at the current moment is analyzed and obtained.

[0025] The preferred solution of the embodiment of the present invention is implemented by combining the CO and CO in the flue gas detected by the mass spectrometer. 2 The percentage of carbon, flue gas flow, and carbon balance in the steelmaking process can more comprehensively analyze the decarburization rate and carbon content of the molten pool, improve the accuracy and reliability of detection, and through the preset carbon integral model, combined with the decarburization rate of the molten pool, molten steel quality and minimum carbon content, the carbon content of the molten pool can be more accurately predicted.

[0026] As a preferred solution, the detection device is a secondary gun; when the converter is at the end of steelmaking, if a detection signal sent by the detection device is received, a second carbon content detection value of the molten pool in the converter at the current moment is obtained by real-time analysis based on the detection signal, specifically:

[0027] When the converter is at the end of steelmaking, if a detection signal sent by the auxiliary gun is received, a fourth carbon content detection value of the molten pool in the converter at the current moment is obtained based on the detection signal; wherein the fourth carbon content detection value of the molten pool in the converter at the current moment is obtained by detecting the molten pool in the converter using the auxiliary gun;

[0028] According to the preset dynamic stage carbon content algorithm, based on the real-time oxygen supply and oxygen supply measurement value of the oxygen lance, the quality of the molten steel in the converter, and the fourth carbon content detection value of the molten pool in the converter at the current moment, the second carbon content detection value of the molten pool in the converter at the current moment is analyzed; wherein, the oxygen supply measurement value is obtained by measuring the oxygen lance by the auxiliary gun.

[0029] Implementing the preferred solution of the embodiment of the present invention, the bath in the converter is detected in real time by means of a sublance to obtain the detected value of the fourth carbon content of the bath in the converter at the current moment. By combining the detected value of the fourth carbon content, as well as the real-time oxygen supply amount and the measured value of the oxygen supply amount of the oxygen lance, the oxygen supply amount of the oxygen lance can be optimized, and the carbon content of the bath can be analyzed more comprehensively, improving the accuracy and reliability of the detection.

[0030] As a preferred solution, the three-stage model is specifically:

[0031]

[0032] In the formula, represents the decarburization rate in the first stage of the converter steelmaking process; represents the decarburization rate in the second stage of the converter steelmaking process; represents the decarburization rate in the third stage of the converter steelmaking process; t represents time; W m represents the weight of the molten steel in the bath in the converter; k 1 、k 3 represent constant coefficients; α O represents the decarburization oxygen efficiency; V O represents the amount of oxygen entering the bath per unit time; C [C] represents the carbon content of the bath; C 0 represents the limiting carbon content; β O represents a constant coefficient; h represents the oxygen lance height; h min represents the lowest oxygen lance height; Q top 、Q topmax respectively represent the oxygen flow rate and the maximum oxygen flow rate of the oxygen lance.

[0033] Implementing the preferred solution of the embodiment of the present invention, the converter steelmaking process is divided into three stages, the decarburization rates are modeled separately, and time variables and the bath carbon content are introduced, which can dynamically capture the changes in the decarburization rate, reflect the actual decarburization behavior during the steelmaking process, and thus can more accurately describe the decarburization characteristics in different stages, improving the accuracy and applicability of the model.

[0034] As a preferred solution, the fusion of the first carbon content detection value and the second carbon content detection value to obtain the detected value of the third carbon content of the bath in the converter at the current moment during the late stage of steelmaking is specifically:

[0035] Assign a first weight coefficient to the first carbon content detection value to obtain the corresponding first weighted carbon content detection value, and assign a second weight coefficient to the second carbon content detection value to obtain the corresponding second weighted carbon content detection value;

[0036] Add the first weighted carbon content detection value and the second weighted carbon content detection value to obtain the third carbon content detection value of the molten bath in the converter at the end of steelmaking at the current moment.

[0037] Implementing the preferred solution of the embodiments of the present invention, respectively performing weighted processing on the first carbon content detection value and the second carbon content detection value, and adding the weighted results to obtain the final carbon content detection result of the molten bath in the converter at the end of steelmaking at the current moment, can effectively reduce the error that may be brought by a single detection method.

[0038] To solve the same technical problem, the embodiments of the present invention also provide a converter steelmaking molten bath carbon content detection system, including:

[0039] An acquisition module, configured to respond to the converter steelmaking signal, start the three-stage model, and in the process of converter steelmaking, real-time acquire the lance height and the lance oxygen flow rate;

[0040] A first detection module, configured to use the three-stage model, combine the lance height and the lance oxygen flow rate, as well as the weight of scrap steel, hot metal weight and carbon content added to the converter, and real-time analyze to obtain the first carbon content detection value of the molten bath in the converter at the current moment;

[0041] A second detection module, configured to when the converter is at the end of steelmaking, if receiving a detection signal sent by a detection device, then based on the detection signal, real-time analyze to obtain the second carbon content detection value of the molten bath in the converter at the current moment, and fuse the first carbon content detection value and the second carbon content detection value to obtain the third carbon content detection value of the molten bath in the converter at the end of steelmaking at the current moment, so as to complete the detection of the carbon content of the molten bath in the converter at the end of steelmaking.

[0042] To solve the same technical problem, the present invention also provides a computer-readable storage medium, the computer-readable storage medium includes a stored computer program; wherein, when the computer program runs, it controls the device where the computer-readable storage medium is located to execute the described method for detecting the carbon content of the molten bath in converter steelmaking.

[0043] To solve the same technical problem, the present invention also provides a terminal, including a processor, a memory and a computer program stored in the memory; wherein, the computer program can be executed by the processor to implement the described method for detecting the carbon content of the molten bath in converter steelmaking. Description of the Drawings

[0044] Figure 1 : A schematic flowchart of a method for detecting the carbon content of the molten bath in converter steelmaking provided in Embodiment 1 of the present invention;

[0045] Figure 2 : This is a schematic structural diagram of a converter steelmaking bath carbon content detection system provided in the first embodiment of the present invention. Specific embodiments

[0046] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0047] Embodiment 1:

[0048] Please refer to Figure 1 , which is a converter steelmaking bath carbon content detection method provided in the embodiment of the present invention. The method includes steps S1 to S3, and the specific steps are as follows:

[0049] Step S1, in response to the converter steelmaking signal, start the three-stage model, and during the converter steelmaking process, real-time obtain the oxygen lance height and the oxygen flow rate of the oxygen lance.

[0050] Step S2, use the three-stage model, combine the oxygen lance height and the oxygen flow rate of the oxygen lance, as well as the scrap weight, hot metal weight and carbon content added to the converter, and analyze in real time to obtain the decarburization rate of the bath in the converter at the current moment, and then analyze to obtain the first carbon content detection value of the bath at the current moment.

[0051] In this embodiment, during the converter steelmaking process, the secondary process control system (usually called the L2 system) is responsible for reading the weight data of various scraps of this heat from the PLC (Programmable Logic Controller). After tapping the hot metal, the L2 system reads the hot metal weight, hot metal temperature and hot metal composition (carbon C, silicon Si, manganese Mn, phosphorus P, sulfur S) of this heat from the PLC. After the converter charging is completed and the furnace is tilted upright, the oxygen lance descends to ignite and blow. The PLC will record the real-time oxygen flow rate, pressure and lance position / height of the oxygen lance. The cumulative value of the oxygen supply amount of the PLC will be available every 3 seconds. This system reads the real-time oxygen flow rate, lance position and oxygen supply amount of the oxygen lance from the PLC. During the smelting process, this system reads the addition time and addition weight of various fluxes (lime, light burned magnesia balls, fluorite, etc.) from the PLC. It is planned to perform program operations every 3s to calculate the decarburization rate and the first carbon content detection value of the bath in the converter at the current moment, and display them on the screen.

[0052] As an example, the PLC monitors the weights of various scrap steels added to the converter in real time through a weighing sensor. There are usually 3 - 5 types of scrap steels. Then, the collected scrap steel weight data is transmitted to the secondary process control system in real time through an industrial network (such as Profibus, Modbus, etc.).

[0053] As an optimal solution, for the "three - stage model" mentioned in step S2, please refer to equations (1), (2), and (3) specifically.

[0054]

[0055] In the formula, represents the decarburization rate in the first stage of the converter steelmaking process, with the unit of kg·s -1 ; represents the decarburization rate in the second stage of the converter steelmaking process, with the unit of kg·s -1 ; represents the decarburization rate in the third stage of the converter steelmaking process, with the unit of kg·s -1 ; t represents time, with the unit of s; W m represents the weight of the molten steel in the molten pool in the converter, with the unit of kg; k 1 、k 3 represent constant coefficients; α O represents the decarburization oxygen efficiency (the carbon content that can be removed by 1 kg of oxygen, with the unit of wt% / kg); V O represents the amount of oxygen entering the molten pool per unit time, with the unit of kg / s; C [C] represents the carbon content of the molten pool, with the unit of wt%; C 0 represents the limit carbon content, with the unit of wt%; β O represents a constant coefficient; h represents the height of the oxygen lance, with the unit of m; h min represents the lowest height of the oxygen lance, with the unit of m; Q top 、Q topmax respectively represent the oxygen flow rate and the maximum oxygen flow rate of the oxygen lance, with the unit of Nm 3 / h.

[0056] It should be noted that the early stage of converter steelmaking is the first stage in the converter steelmaking process, the middle stage of converter steelmaking is the second stage, and the late stage of converter steelmaking is the third stage. And in the third stage of the converter steelmaking process, when blowing is approaching the end, the converter is in the final stage of steelmaking. Among them, the judgment of the turning point between the first stage and the second stage: when the silicon content of the hot metal drops to 0.03%, the first stage ends. The judgment of the turning point between the second stage and the third stage: the carbon content is between 0.2 - 0.6%.

[0057] In this embodiment, when the converter is in the final stage of steelmaking, there are the following situations: no detection means, any one or more of the following detection and analysis methods: sublance measurement, flue gas analysis, and furnace mouth flame analysis. If there are no other detection means when the converter is in the final stage of steelmaking, then steps S1 to S2 are still carried out, and only the three-stage model is used, referring to Equation (3), to calculate in real time the decarburization rate and the first carbon content detection value of the molten bath in the converter at the current moment, and the first carbon content detection value is used as the final carbon content detection result.

[0058] Step S3, when the converter is in the final stage of steelmaking, if a detection signal sent by the detection device is received, then based on the detection signal, the second carbon content detection value of the molten bath in the converter at the current moment is analyzed in real time, and the first carbon content detection value and the second carbon content detection value are fused to obtain the third carbon content detection value of the molten bath in the converter in the final stage of steelmaking at the current moment, so as to complete the carbon content detection of the molten bath in the converter in the final stage of steelmaking.

[0059] As a preferred solution, if the detection device is a camera, then step S3 includes steps S301 to S303, and the specific steps are as follows:

[0060] Step S301, when the converter is in the final stage of steelmaking, if a detection signal sent by the camera is received, then based on the detection signal, the corresponding original image is obtained, and the original image is preprocessed to obtain an image to be analyzed.

[0061] Among them, the original image is obtained by using the camera to take a real-time picture of the furnace mouth flame of the converter.

[0062] In this embodiment, the preprocessing includes median filtering and color image segmentation. Among them, median filtering is a non-linear filtering method. While removing the noise in the image, it can better retain the edge information of the image and avoid image blurring; by segmenting the color image, the image can be divided into different regions or objects, and each region has similar color characteristics, which is crucial for applications such as target detection, object recognition, and scene understanding. And segmentation helps to highlight the important features in the image, enabling the algorithm to more accurately extract relevant information, thereby improving the overall analysis performance. The combination of median filtering and color image segmentation techniques enables the system to better adapt to complex image scenes and improve the stability and reliability in different environments.

[0063] Step S302, perform feature description and extraction on the image to be analyzed to obtain the feature variables to be predicted.

[0064] As a preferred solution, step S302 includes steps S3021 to S3024, and the specific steps are as follows:

[0065] Step S3021: Perform flame boundary extraction and processing on the image to be analyzed to separate the flame region and the background region within the image to be analyzed.

[0066] Step S3022: Solve the refined skeleton structure of the flame region and determine the center point of the flame region.

[0067] Step S3023: Obtain the polar coordinate description of the boundary of the flame region, and based on the polar coordinate description of the boundary of the flame region, solve the amplitude spectrum of the boundary of the flame region, and then calculate the average value, entropy, and energy of the amplitude spectrum of the boundary of the flame region as the feature vector describing the boundary of the flame region.

[0068] In this embodiment, with the flame center point as the center, each point on the boundary of the flame region is converted from Cartesian coordinates to polar coordinates as the polar coordinate description of the boundary of the flame region. Then, perform Fourier transform on the polar coordinate description of the boundary of the flame region to obtain the descriptors in the frequency domain, and these descriptors can represent the shape characteristics of the boundary. Then, extract the amplitude spectrum from the descriptors and calculate the average value, entropy, and energy of the amplitude spectrum as the feature vector describing the dynamic deformation of the flame boundary.

[0069] Step S3024: Perform Gaussian normalization on the feature vector to obtain the feature variable to be predicted.

[0070] Step S303: Input the feature variable to be predicted into the carbon content detection model so that the carbon content detection model outputs the second carbon content detection value C of the molten pool in the converter at the current moment f ;

[0071] Among them, the carbon content detection model is obtained by training a pre-constructed GRNN model according to the sample data set.

[0072] In this embodiment, the sample data set includes the feature variables corresponding to a number of furnace mouth flame images and the molten pool carbon content corresponding to each furnace mouth flame image. Specifically, the sample data set can be divided into a training set and a test set, and then use the training set to train the pre-constructed GRNN model, and use the test set to test and verify the trained model. If the test results meet the preset requirements, the current model is completed training.

[0073] Step S304: Fuse the first carbon content detection value and the second carbon content detection value to obtain the third carbon content detection value of the molten pool in the converter at the current moment in the late stage of steelmaking.

[0074] As a preferred solution, step S304 includes step S3041, and the specific steps are as follows:

[0075] Step S3041, referring to Equation (4), assign a first weight coefficient to the first carbon content detection value to obtain the corresponding first weighted carbon content detection value, and assign a second weight coefficient to the second carbon content detection value to obtain the corresponding second weighted carbon content detection value. Then add the first weighted carbon content detection value and the second weighted carbon content detection value to obtain the third carbon content detection value of the molten bath in the converter at the current moment in the late stage of steelmaking.

[0076] C 1 =W 6 C 6 +W f C f ;(4)

[0077] In the formula, C 1 represents the third carbon content detection value calculated when receiving the detection signal sent by the camera; C 6 represents the first carbon content detection value of the molten bath in the converter at the current moment calculated referring to Equation (6); W 6 represents the first weight coefficient; C f represents the second carbon content detection value calculated when receiving the detection signal sent by the camera; W f represents the second weight coefficient corresponding to the second carbon content detection value C f .

[0078] As a preferred solution, if the detection device is a mass spectrometer, Step S3 includes Step S305 to Step S308, and the specific steps are as follows:

[0079] Step S305, when the converter is in the late stage of steelmaking, if receiving the detection signal sent by the mass spectrometer, based on the detection signal, obtain the percentage contents of CO and CO 2 in the converter's flue gas, and the flue gas flow rate of the converter.

[0080] Among them, the percentage contents of CO and CO 2 in the converter's flue gas, and the flue gas flow rate of the converter are obtained by detecting the converter's flue gas using a mass spectrometer.

[0081] Step S306, referring to Equation (5), based on the carbon balance relationship in the converter steelmaking process, and combining the percentage contents of CO and CO 2 in the converter's flue gas, and the flue gas flow rate of the converter, analyze and obtain the decarburization rate of the molten bath in the converter at the current moment.

[0082]

[0083] In the formula, represents the decarburization rate of the molten bath in the converter at the current moment; Q gis the flue gas flow rate, with the unit of m 3 / s; and are the volume percentage contents of CO and CO 2 in the flue gas respectively.

[0084] For step S307, please refer to Equations (6), (7), and (8). According to the preset carbon integration model, based on the decarburization rate of the molten bath in the converter at the current moment, the mass of the molten steel in the molten bath in the converter, and the lowest carbon content of the molten bath in the converter, the second carbon content detection value of the molten bath in the converter at the current moment is analyzed.

[0085]

[0086] In the formula, ω″ C represents the cumulative decarburization amount of the molten bath; represents the decarburization rate of the molten bath at the current moment; ω C represents the instantaneous carbon content of the molten bath during converter steelmaking, that is, the second carbon content detection value C g calculated when receiving the detection signal sent by the mass spectrometer; ω′ C is the initial carbon content of the molten bath, with the unit of kg; W m is the mass of the molten steel in the molten bath, with the unit of t; k 1 , k 2 are the coefficients to be determined in the model; ω C (t) is the instantaneous carbon content of the molten bath at time t; is the lowest carbon content of the molten bath.

[0087] It should be noted that the carbon integration model is established on the basis of the mass balance principle: the total carbon amount of the steelmaking raw materials minus the carbon amount lost in the form of CO and CO 2 from the furnace gas is the carbon amount remaining in the steel.

[0088] For step S308, the first carbon content detection value and the second carbon content detection value are fused to obtain the third carbon content detection value of the molten bath in the converter at the current moment in the late stage of steelmaking.

[0089] As an optimal solution, step S308 includes step S3081, and the specific steps are as follows:

[0090] For step S3081, please refer to Equation (9). A first weight coefficient is assigned to the first carbon content detection value to obtain the corresponding first weighted carbon content detection value, and a second weight coefficient is assigned to the second carbon content detection value to obtain the corresponding second weighted carbon content detection value. Then, the first weighted carbon content detection value and the second weighted carbon content detection value are added together to obtain the third carbon content detection value of the molten bath in the converter at the current moment in the late stage of steelmaking.

[0091] C 2 = W 6 C 6 + W g C g ;(9)

[0092] In the formula, C 2 represents the detected value of the third carbon content calculated when receiving the detection signal sent by the mass spectrometer; C 6 represents the detected value of the first carbon content of the molten bath in the converter at the current moment calculated with reference to formula (6); W 6 represents the first weight coefficient; C g represents the detected value of the second carbon content calculated when receiving the detection signal sent by the mass spectrometer; W g represents the second weight coefficient corresponding to the detected value of the second carbon content C g .

[0093] As a preferred solution, if the detection device is a sublance, step S3 includes steps S309 to S311, and the specific steps are as follows:

[0094] Step S309, when the converter is in the final stage of steelmaking, if a detection signal sent by the sublance is received, based on the detection signal, obtain the detected value of the fourth carbon content of the molten bath in the converter at the current moment.

[0095] Among them, the detected value of the fourth carbon content of the molten bath in the converter at the current moment is obtained by using the sublance to detect the molten bath in the converter.

[0096] Step S310, please refer to formula (10), and according to the preset dynamic stage carbon content algorithm, based on the real-time oxygen supply amount and the measured value of the oxygen supply amount of the oxygen lance, the molten steel quality of the molten bath in the converter, and the detected value of the fourth carbon content of the molten bath in the converter at the current moment, analyze and obtain the detected value C SL of the second carbon content of the molten bath in the converter at the current moment.

[0097] Among them, the measured value of the oxygen supply amount is obtained by using the sublance to measure the oxygen lance.

[0098]

[0099] In the formula, C SL represents the real-time carbon content in the dynamic stage, that is, the detected value of the second carbon content, with the unit of 10 -2 %; C 0 is the theoretical limit carbon content, with the unit of 10 -2 %; C m is the carbon content measured by the sublance, with the unit of 10 -2 %; α DThe decarburization coefficient 1, which represents how many kilograms of carbon can react with each cubic meter of pure oxygen, with the unit of kg / m 3 ; β D is the decarburization coefficient 2, which represents the speed of the decarburization reaction, with the unit of %; VO m is the oxygen supply amount during the sublance measurement, with the unit of Nm 3 ; VO is the real-time oxygen supply amount, with the unit of Nm 3 ; W m is the mass of molten steel, with the unit of kg; B i is the oxygen-carrying coefficient of flux i, with the unit of m 3 / kg; W i is the addition amount of flux i after the sublance measurement, with the unit of kg.

[0100] Step S311: Integrate the first carbon content detection value and the second carbon content detection value to obtain the third carbon content detection value of the molten bath in the converter at the end of steelmaking at the current moment.

[0101] As an optimal solution, step S311 includes step S3111, which is specifically as follows:

[0102] Step S3111: Refer to formula (11). Assign the first weight coefficient to the first carbon content detection value to obtain the corresponding first weighted carbon content detection value, and assign the second weight coefficient to the second carbon content detection value to obtain the corresponding second weighted carbon content detection value. Then add the first weighted carbon content detection value and the second weighted carbon content detection value to obtain the third carbon content detection value of the molten bath in the converter at the end of steelmaking at the current moment.

[0103] C 3 =W 6 C 6 +W SL C SL ; (11)

[0104] In the formula, C 3 represents the third carbon content detection value calculated when receiving the detection signal sent by the sublance; C 6 represents the first carbon content detection value of the molten bath in the converter at the current moment calculated by referring to formula (6); W 6 represents the first weight coefficient; C SL represents the second carbon content detection value calculated when receiving the detection signal sent by the sublance; W SL represents the second weight coefficient corresponding to the second carbon content detection value C SL .

[0105] In this embodiment, when the converter is in the final stage of steelmaking, if detection signals sent by two or more detection devices are received, based on the detection signals corresponding to each detection device, the second carbon content detection value corresponding to each detection device is obtained by real-time analysis. Then, by fusing the first carbon content detection value and the second carbon content detection values corresponding to all detection devices, the third carbon content detection value of the molten bath in the converter at the current moment in the final stage of steelmaking can be obtained. As an example, when the converter is in the final stage of steelmaking, if detection signals sent by a camera, a mass spectrometer, and a sublance are received, the foregoing method can be used to respectively obtain the first carbon content detection value C of the molten bath in the converter at the current moment calculated according to Equation (6) 6 and the second carbon content detection value C f calculated when the detection signal sent by the camera is received, g the second carbon content detection value C SL calculated when the detection signal sent by the mass spectrometer is received, 6 and the second carbon content detection value C f calculated when the detection signal sent by the sublance is received. Then, referring to Equation (12), by fusing C g , C SL , the third carbon content detection value C of the molten bath in the converter at the current moment in the final stage of steelmaking can be obtained 4 .

[0106] C 4 =W 6 C 6 +W SL C SL +W g C g +W f C f ;(12)

[0107] Please refer to Figure 2 , which is a schematic structural diagram of a molten bath carbon content detection system for converter steelmaking provided by an embodiment of the present invention. The system includes an acquisition module M1, a first detection module M2, and a second detection module M3. The specific functions of each module are as follows:

[0108] The acquisition module M1 is configured to respond to the converter steelmaking signal, start the three-stage model, and in the process of converter steelmaking, acquire the oxygen lance height and the oxygen flow rate of the oxygen lance in real time;

[0109] The first detection module M2 is configured to use the three-stage model, combine the oxygen lance height and the oxygen flow rate of the oxygen lance, as well as the weight of scrap steel, hot metal weight, and carbon content added to the converter, and analyze in real time to obtain the first carbon content detection value of the molten bath in the converter at the current moment;

[0110] The second detection module M3 is configured to, when the converter is in the late stage of steelmaking, if a detection signal sent by a detection device is received, analyze in real time, based on the detection signal, a second carbon content detection value of the molten bath in the converter at the current moment, and fuse the first carbon content detection value and the second carbon content detection value to obtain a third carbon content detection value of the molten bath in the converter in the late stage of steelmaking at the current moment, so as to complete the detection of the carbon content of the molten bath in the converter in the late stage of steelmaking.

[0111] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described system can refer to the corresponding process in the foregoing method embodiment, and will not be elaborated herein.

[0112] The embodiment of the present invention further provides a computer-readable storage medium, where the computer-readable storage medium includes a stored computer program; wherein, when the computer program runs, it controls the device where the computer-readable storage medium is located to execute the method for detecting the carbon content of the molten bath in converter steelmaking described in Embodiment 1.

[0113] The embodiment of the present invention further provides a terminal, including a processor, a memory, and a computer program stored in the memory; wherein, the computer program can be executed by the processor to implement the method for detecting the carbon content of the molten bath in converter steelmaking described in Embodiment 1.

[0114] Preferably, the computer program can be divided into one or more modules / units (such as computer programs, computer programs), and one or more modules / units are stored in the memory and executed by the processor to complete the present invention. One or more modules / units can be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program in the terminal.

[0115] The processor may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor, or the processor may also be any conventional processor. The processor is the control center of the terminal and connects various parts of the terminal through various interfaces and lines.

[0116] The memory mainly includes a program storage area and a data storage area. Among them, the program storage area can store an operating system, application programs required for at least one function, etc., and the data storage area can store relevant data, etc. In addition, the memory can be a high-speed random access memory, or a non-volatile memory, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, etc., or the memory can also be other volatile solid-state storage devices.

[0117] It should be noted that the above terminal may include, but is not limited to, a processor and a memory. Those skilled in the art can understand that the above terminal is only an example and does not constitute a limitation on the terminal. It may include more or fewer components, or combine some components, or different components.

[0118] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0119] The present invention provides a method, a system, a medium and a terminal for detecting the carbon content in the molten pool of converter steelmaking. In response to the converter steelmaking signal, a three-stage model is started, and during the converter steelmaking process, the height of the oxygen lance and the oxygen flow rate of the oxygen lance are obtained in real time. Then, using the three-stage model, in combination with the height of the oxygen lance and the oxygen flow rate of the oxygen lance obtained in real time, as well as the weight of scrap steel, the weight of hot metal and the carbon content added to the converter, the decarburization rate and the first carbon content detection value of the molten pool in the converter at the current moment are analyzed in real time. That is, when the converter steelmaking signal is received, the three-stage model is started, and considering the influence of the oxygen lance position, oxygen flow rate and feeding on the decarburization rate, using the three-stage model, in combination with the height of the oxygen lance and the oxygen flow rate of the oxygen lance obtained in real time, as well as the weight of scrap steel, the weight of hot metal and the carbon content added to the converter, the decarburization rate and the first carbon content detection value of the molten pool in the converter at each moment during the whole steelmaking process are analyzed in real time to improve the detection accuracy of the decarburization rate and carbon content of the molten pool, so as to achieve precise control of the steelmaking process. In addition, when the converter is in the late stage of steelmaking, if a detection signal sent by the detection device is received, then based on the detection signal, the second carbon content detection value of the molten pool in the converter at the current moment is analyzed in real time, and the first carbon content detection value and the second carbon content detection value are fused to obtain the third carbon content detection value of the molten pool in the converter at the current moment in the late stage of steelmaking, which can not only complete the detection of the carbon content of the molten pool in the converter in the late stage of steelmaking, but also further optimize the accuracy of the carbon content detection result of the molten pool in the converter through the detection result of the detection device.

[0120] The specific embodiments described above further elaborate on the objectives, technical solutions, and beneficial effects of the present invention. It should be understood that the above are only specific embodiments of the present invention and are not used to limit the protection scope of the present invention. In particular, for those skilled in the art, any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for detecting carbon content in a converter steelmaking molten pool, characterized in that: include: In response to the converter steelmaking signal, the three-stage model is started, and the oxygen lance height and oxygen flow rate of the oxygen lance are obtained in real time during the converter steelmaking process; By using the three-stage model, in combination with the oxygen lance height and the oxygen flow rate of the oxygen lance, as well as the weight of scrap steel, the weight of molten iron and the carbon content added to the converter, the decarburization rate and the first carbon content detection value of the molten pool in the converter at the current moment are obtained through real-time analysis; When the converter is at the final stage of steelmaking, if a detection signal sent by the detection equipment is received, a second carbon content detection value of the molten pool in the converter at the current moment is obtained through real-time analysis based on the detection signal, and the first carbon content detection value and the second carbon content detection value are fused to obtain a third carbon content detection value of the molten pool in the converter at the final stage of steelmaking at the current moment, so as to complete the carbon content detection of the molten pool in the converter at the final stage of steelmaking.

2. A method for detecting carbon content in a converter steelmaking molten pool as claimed in claim 1, characterized in that: The detection device is a camera; when the converter is at the end of steelmaking, if a detection signal sent by the detection device is received, a second carbon content detection value of the molten pool in the converter at the current moment is obtained by real-time analysis based on the detection signal, specifically: When the converter is at the end of steelmaking, if a detection signal sent by the camera is received, a corresponding original image is acquired based on the detection signal, and the original image is preprocessed to obtain an image to be analyzed; wherein the original image is obtained by using the camera to take real-time photos of the flame at the mouth of the converter; Describing and extracting features of the image to be analyzed to obtain feature variables to be predicted; Inputting the characteristic variable to be predicted into a carbon content detection model so that the carbon content detection model outputs a second carbon content detection value of the molten pool in the converter at the current moment; The carbon content detection model is obtained by training a pre-built GRNN model based on a sample data set.

3. A method for detecting carbon content in a converter steelmaking molten pool as claimed in claim 2, characterized in that: The feature description and extraction of the image to be analyzed to obtain the feature variables to be predicted is specifically: Extracting and processing the flame boundary of the image to be analyzed to separate the flame area and the background area in the image to be analyzed; Solving the refined skeleton structure of the flame region and determining the center point of the flame region; Obtaining a polar coordinate description of the boundary of the flame region, and solving the amplitude spectrum of the boundary of the flame region based on the polar coordinate description of the boundary of the flame region, and then calculating the average value, entropy and energy of the amplitude spectrum of the boundary of the flame region as a feature vector describing the boundary of the flame region; Gaussian normalization is performed on the feature vector to obtain the feature variable to be predicted.

4. A method for detecting carbon content in a converter steelmaking molten pool as claimed in claim 1, characterized in that: The detection device is a mass spectrometer; when the converter is at the end of steelmaking, if a detection signal sent by the detection device is received, a second carbon content detection value of the molten pool in the converter at the current moment is obtained by real-time analysis based on the detection signal, specifically: When the converter is at the end of steelmaking, if a detection signal sent by the mass spectrometer is received, the percentage of CO and CO2 in the flue gas of the converter and the flue gas flow rate of the converter are obtained based on the detection signal; wherein the percentage of CO and CO2 in the flue gas of the converter and the flue gas flow rate of the converter are obtained by detecting the flue gas of the converter using the mass spectrometer; Based on the carbon balance relationship in the converter steelmaking process, combined with the percentage of CO and CO2 in the converter flue gas and the converter flue gas flow rate, the decarburization rate of the molten pool in the converter at the current moment is analyzed; According to the preset carbon integral model, based on the decarburization rate of the molten pool in the converter at the current moment, the quality of the molten steel in the molten pool in the converter, and the minimum carbon content of the molten pool in the converter, the second carbon content detection value of the molten pool in the converter at the current moment is analyzed and obtained.

5. A method for detecting carbon content in a converter steelmaking molten pool as claimed in claim 1, characterized in that: The detection device is a secondary gun; when the converter is in the final stage of steelmaking, if a detection signal sent by the detection device is received, a second carbon content detection value of the molten pool in the converter at the current moment is obtained by real-time analysis based on the detection signal, specifically: When the converter is at the end of steelmaking, if a detection signal sent by the auxiliary gun is received, a fourth carbon content detection value of the molten pool in the converter at the current moment is obtained based on the detection signal; wherein the fourth carbon content detection value of the molten pool in the converter at the current moment is obtained by detecting the molten pool in the converter using the auxiliary gun; According to the preset dynamic stage carbon content algorithm, based on the real-time oxygen supply and oxygen supply measurement value of the oxygen lance, the quality of the molten steel in the converter, and the fourth carbon content detection value of the molten pool in the converter at the current moment, the second carbon content detection value of the molten pool in the converter at the current moment is analyzed; wherein, the oxygen supply measurement value is obtained by measuring the oxygen lance by the auxiliary gun.

6. A method for detecting carbon content in a converter steelmaking molten pool as claimed in claim 1, characterized in that: The three-stage model is specifically: In the formula, Indicates the decarburization rate in the first stage of the converter steelmaking process; Indicates the decarburization rate in the second stage of the converter steelmaking process; represents the decarburization rate in the third stage of converter steelmaking; t represents time; W m Indicates the weight of molten steel in the molten pool in the converter; k1 and k3 represent constant coefficients; α O Indicates the decarbonization oxygen efficiency; V O Indicates the amount of oxygen entering the molten pool per unit time; C [C] Indicates the carbon content of the molten pool; C0 indicates the limit carbon content; β O represents the constant coefficient; h represents the oxygen lance height; h min Indicates the minimum oxygen lance height; Q top , Q topmax They represent the oxygen flow rate and maximum oxygen flow rate of the oxygen gun respectively.

7. A method for detecting carbon content in a converter steelmaking molten pool according to any one of claim 1, characterized in that: The third carbon content detection value of the molten pool in the converter at the end of steelmaking is obtained by fusing the first carbon content detection value and the second carbon content detection value, which is specifically: Assigning a first weight coefficient to the first carbon content detection value to obtain a corresponding first weighted carbon content detection value, and assigning a second weight coefficient to the second carbon content detection value to obtain a corresponding second weighted carbon content detection value; The first weighted carbon content detection value and the second weighted carbon content detection value are added to obtain a third carbon content detection value of the molten pool in the converter at the end of steelmaking at the current moment.

8. A converter steelmaking molten pool carbon content detection system, characterized in that: include: The acquisition module is used to respond to the converter steelmaking signal, start the three-stage model, and obtain the oxygen lance height and oxygen flow rate of the oxygen lance in real time during the converter steelmaking process; A first detection module is used to use the three-stage model, in combination with the oxygen lance height and the oxygen flow rate of the oxygen lance, as well as the weight of scrap steel added to the converter, the weight of molten iron and the carbon content, to analyze in real time to obtain a first carbon content detection value of the molten pool in the converter at the current moment; The second detection module is used for, when the converter is at the final stage of steelmaking, if a detection signal sent by the detection equipment is received, then, based on the detection signal, real-time analysis is performed to obtain a second carbon content detection value of the molten pool in the converter at the current moment, and the first carbon content detection value and the second carbon content detection value are integrated to obtain a third carbon content detection value of the molten pool in the converter at the final stage of steelmaking at the current moment, so as to complete the carbon content detection of the molten pool in the converter at the final stage of steelmaking.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program; wherein, when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute a method for detecting carbon content in a converter steelmaking molten pool as described in any one of claims 1 to 7.

10. A terminal, characterized in that: It comprises a processor, a memory and a computer program stored in the memory; wherein the computer program can be executed by the processor to implement a method for detecting the carbon content of a converter steelmaking molten pool as described in any one of claims 1 to 7.

Images