Converter smelting parameter optimization design method and system based on digital twinning

Abstract

The invention relates to the technical field of digital smelting, in particular to a converter smelting parameter optimization design method and system based on digital twinning. A converter smelting parameter optimization design system based on digital twinning comprises an information acquisition module, an image recognition module, a simulation correction module, a warning module and an output guidance module. According to the method, the stability and feasibility of the steelmaking process are analyzed in real time by analyzing the intelligent steelmaking control model of the steelmaking process control elements, so that the quality of molten steel is stable, the one-time drawing rate of tapping is improved, meanwhile, data simulation steelmaking can be carried out on new steel grade research and development, the cost is saved, and the research and development process is optimized.

Description

Technical Field

[0001] The present application relates to the field of digital smelting technology, and in particular to a converter smelting parameter optimization design method and system based on digital twin. Background Art

[0002] Currently, steel mills cannot judge the stability of molten steel and the end point of steelmaking in the converter process mainly by relying on the experience of workers, which requires long-term attention and lacks good control methods. Workers need to be in the production environment for a long time, and if they are not careful, it will affect the one-time drawing rate of steel. In addition, when developing new steel grades, they are often tested through small R&D laboratories or industrial trials. There is no good verification method and the cost is very high. Summary of the invention

[0003] In response to the above problems, the present application provides a converter smelting parameter optimization design method and system based on digital twin. By analyzing the intelligent steelmaking control model of the steelmaking process control elements, the stability and feasibility of the steelmaking process are analyzed in real time, thereby achieving stable quality of molten steel and improving the one-time drawing rate of steel. At the same time, data simulation steelmaking can be performed for the research and development of new steel grades, saving costs and optimizing the research and development process.

[0004] The technical solution of this application is: a converter smelting parameter optimization design method based on digital twin, comprising:

[0005] S1, interval time T, the IoT layer obtains the converter working data J recorded by the physical layer t And the photos of converter smelting taken t , t is the difference between the real time of recording converter working data and taking converter smelting photos and the initial time T0, that is, t is the working time, T0 is the initial time when the converter starts this smelting, T z The time threshold set for a single converter smelting is used to determine whether the working time t satisfies "t≥T z ”, if “t≥T z ", then the converter working data J t Photo of converter smelting t Stored in the smelting database K, if it does not satisfy "t≥T z ", a warning signal is issued and the smelting database K is cleared;

[0006] S2, twin layer traversal smelting database K converter smelting photos P t Corresponding working time t, extract the maximum working time t and record it as max, extract the smelting photo P corresponding to the working time max t , and sent to the empirical analysis model, the empirical analysis model output converter smelting photo P tAnalyze the empirical recognition time t0, and determine whether the empirical recognition time t0 satisfies "t0 = max ± γ", where γ is the set error time threshold. If it satisfies, enter S3, and if it does not, send a warning signal;

[0007] S3, the converter working data J in the smelting database K t The simulation correction model is sent to the simulation correction model, and the simulation correction model is based on the converter working data J t Simulate converter smelting, then output when T = T z Prediction of converter data L Tz , determine whether "|L Tz -X Tz |<β”, β is the set error data threshold, X Tz To set at T z The converter working data of successful smelting at this time, if it satisfies "|L Tz -X Tz |<β", no operation is performed. If "|L Tz -X Tz |<β”, then enter S4;

[0008] S4. All converter working data J in the smelting database K are t Enter the simulation correction model and input the virtual working data X Tz , for time max to T z Converter smelting simulation, output simulation virtual work data X n , n∈[t+T,N], N is the duration threshold T z The ratio of the interval time T, that is, N = T z / T;

[0009] S5. All converter working data J t and Virtual Work Data X n The twin layer simulation correction model is sent to complete the converter smelting simulation and output in T z Converter working data at J Tz , determine the converter working data J Tz Does it satisfy the "|J Tz -X Tz |<β”, if “|J Tz -X Tz |<β”, a warning signal is sent and the process goes to S6. If “|J Tz -X Tz |<β”, then the virtual working data X n Recorded as converter working data J t+T , stored in smelting database K, and enter S7;

[0010] S6, assign n+T to n, and go to S5;

[0011] S7. Traverse the smelting database K and extract all converter working data J that satisfy "t>max" t , the twin layer is based on the extracted converter working data J t Guide the IoT layer to control the converter equipment and complete the adjustment of smelting.

[0012] As a further preferred solution, the specific steps of recording converter working data and taking converter smelting photos at the physical layer are as follows:

[0013] The IoT layer connects production equipment, collects various data, and calculates them through the weight formula: t =μ1J 1t +μ2J 2t +μ3J 3t +······+μ m J mt , m=1, 2, 3······M, M is the total number of all collected data, μ m The weight threshold set for the mth item of collected data is used to obtain the overall data, which is recorded as converter working data J t ;

[0014] There is a camera in the converter, which takes photos of the smelting conditions in the converter at intervals T, recorded as converter smelting photos P t The converter smelting photo is a photo of the state information of the metal solution surface when the metal is smelted in the converter.

[0015] As a further preferred solution, the method for constructing the empirical analysis model in step S2 is:

[0016] The empirical analysis model is established based on a convolutional neural network and is equipped with a basic empirical database. The basic empirical database stores a large number of converter smelting photos and the workers' conclusions on each converter smelting photo. The empirical analysis model completes training and learning based on the basic empirical database, and inputs converter smelting photos that are not stored in the basic empirical database for verification. The workers judge whether the empirical analysis model is fully trained based on the verification results and adjust the empirical analysis model to ensure the good performance of the empirical analysis model.

[0017] The analysis process is mainly feature analysis. After the converter smelting photos are sent to the empirical analysis model, the empirical analysis model extracts the features that appear in the smelting photos, and outputs the time corresponding to the extracted features according to the feature-time correspondence table. The mean of all feature times is calculated to obtain the working time of the smelting photos. The feature-time correspondence table is established by the workers, which is a list of the relationship between the different features that appear on the surface of the metal solution during the converter smelting process and the corresponding time.

[0018] As a further preferred solution, the method for constructing the simulation correction model in step S3 is:

[0019] First, three-dimensional modeling is performed according to the geometric structure of the physical entity to obtain a solid model. Then, simulation tools are used to simulate the entity behavior of the solid model. Combined with the reaction mechanism during the working process of the converter, a simulation correction model is constructed.

[0020] The simulation correction model is a data model based on a neural network. It can simulate the working process of the converter smelting according to the input data, so that it can respond quickly after adjusting the input parameters, and simulate reactions close to the actual situation to complete the simulation work of the converter, replicate the same working process as the real physical equipment, and complete the simulated metal smelting according to the data; it can also complete the missing converter smelting process in the middle according to the simulated smelting results, extract the corresponding data of the simulated smelting, and calculate the converter working data J t .

[0021] As a further preferred solution, the specific steps of the twin layer guiding the IoT layer to control the converter equipment in S7 are:

[0022] The simulation correction model is used to extract the converter working data J t Mapping is performed to obtain the corresponding data of the simulated smelting, and the IoT layer obtains the converter working data extracted by the twin layer. t , sorted from small to large according to the working time t, the reverse controller obtains the various data corresponding to the simulated smelting mapped by the simulation correction model according to the sorting, and sends the various data corresponding to the simulated smelting to the control units of various entity devices in the physical layer through the connection between the Internet of Things layer and the physical layer, thereby completing the guidance and control of the converter equipment.

[0023] As a further preferred solution, an application layer is also included, which can receive warning signals sent by the twin layer, and can visualize the simulation correction model simulating the converter smelting process through a display device, and can adjust the parameters of each link for manual control.

[0024] As a further preferred solution, it also includes judging the function of the converter equipment.

[0025] Extract the converter working data J corresponding to max in the smelting database K max , and the remaining converter work data J of the smelting database K t The simulation correction model is sent to the simulation correction model, and the simulation correction model is based on the converter working data J t Simulate converter smelting and obtain the predicted converter data L when t = max max , determine the converter working data Jmax Does it satisfy the "|J max -L max |<β”, if “|J max -L max |<β", no operation is performed. If the "|J max -L max |<β”, an alarm is issued, indicating that the converter equipment is malfunctioning.

[0026] A converter smelting parameter optimization design system based on digital twin, characterized by comprising:

[0027] The information collection module is used to collect converter operation data, take converter smelting photos, and determine the converter operation time;

[0028] Image recognition module, used to analyze converter smelting photos and determine the analysis results;

[0029] The simulation correction module is used to simulate and analyze the operating data of the collected mechanical equipment, and to judge the smelting results and simulate the operating data of the mechanical equipment;

[0030] The warning module issues corresponding warnings according to the running time judgment result of the information collection module, the analysis result of the image recognition module and the smelting result of the simulation correction module;

[0031] The output guidance module outputs the operation data of the mechanical equipment simulated by the simulation correction module according to the smelting results.

[0032] As a further preferred solution, it also includes:

[0033] The visual control module is used to display the simulation process of the simulation correction module and control the simulation process.

[0034] This application has the following advantages:

[0035] 1. The present invention analyzes the stability and feasibility of the steelmaking process in real time by analyzing the intelligent steelmaking control model of the steelmaking process control elements, thereby achieving stable quality of molten steel and improving the one-time drawing rate of steel. At the same time, data simulation steelmaking can be performed for the research and development of new steel grades, saving costs and optimizing the research and development process.

[0036] 2. The present invention visualizes the process of simulated converter smelting by providing an application layer, and controls the equipment of the physical layer through the application layer. At the same time, it can receive warning signals at any time to ensure timely processing. It also detects the converter equipment through the simulation correction model to discover the impact of the converter smelting equipment on the steelmaking process, reduce the occurrence of accidents, and ensure the success rate of steelmaking. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying any creative work.

[0038] Figure 1 This is a structural schematic diagram of a converter smelting parameter optimization design system based on digital twin provided in an embodiment of the present application. DETAILED DESCRIPTION

[0039] In order to make the purpose, technical scheme and advantages of the present application clearer, some embodiments of the present application are further described in detail below in conjunction with the accompanying drawings and Examples. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. However, it will be appreciated by those skilled in the art that in the various embodiments of the present application, many technical details are proposed in order to enable the reader to better understand the present application. However, even without these technical details and various changes and modifications based on the following embodiments, the technical scheme claimed in the present application can also be implemented.

[0040] Example 1

[0041] A converter smelting parameter optimization design method based on digital twin, comprising:

[0042] S1, interval time T, the IoT layer obtains the converter working data J recorded by the physical layer t And the photos of converter smelting taken t ,1),Wherein, the physical layer refers to the physical entity objects and their associated objects that need to be studied. The physical layer generates various types of data and information, including basic equipment data, operation data, process data, etc.,2) The IoT layer provides unified data and model resource management services, which can obtain the data of physical entity objects in the physical layer, and store, analyze and process these data. The IoT layer also provides a model resource library built for various characteristics of physical entities. t is the difference between the real-time time of recording the working data of the converter and taking the photos of the converter smelting and the initial time T0, that is, t is the working time, T0 is the initial time when the converter starts this smelting, and T z The time threshold set for a single converter smelting is used to determine whether the working time t satisfies "t≥T z ”, if “t≥T z ", then the converter working data J t Photo of converter smelting t Stored in the smelting database K, if it does not satisfy "t≥T z", a warning signal is issued, the smelting database K is cleared, the physical layer records the converter working data and takes converter smelting photos. The specific steps are: the IoT layer connects the production equipment, collects various data, and calculates them through the weight formula: J t =μ1J 1t +μ2J 2t +μ3J 3t +······+μ m J mt , m=1, 2, 3······M, M is the total number of all collected data, μ m The weight threshold set for the mth item of collected data is used to obtain the overall data, which is recorded as converter working data J t The data include the output of various sensors, changes in the machine's operating status, environmental parameters, etc. There are also calculated parameters for the ratios of various auxiliary materials, slag-forming agents and other materials. The ratios affect the removal of elements such as P and S in the molten steel and the target molten steel quality.

[0043] There is a camera in the converter, which takes photos of the smelting conditions in the converter at intervals T, recorded as converter smelting photos P t The converter smelting photo is a photo of the state information of the metal solution surface when the metal is smelted in the converter.

[0044] For example, in the blowing stage, the duration is generally half an hour to two hours. According to the situation of the steel plant, it usually takes one hour to add. For the data collection interval of 3 minutes, then T z =1 hour, T=3 minutes, then the corresponding data collection time t should be between 0 and 1 hour. When the recorded time t is greater than 1 hour and the blowing has not been completed, the worker should check and intervene.

[0045] S2, twin layer traversal smelting database K converter smelting photos P t Corresponding working time t, extract the maximum working time t and record it as max, extract the smelting photo P corresponding to the working time max t , and sent to the empirical analysis model, the empirical analysis model output converter smelting photo P t Analyze the empirical recognition time t0, and determine whether the empirical recognition time t0 satisfies "t0 = max ± γ", where γ is the set error time threshold. If it satisfies, enter S3, and if it does not, send a warning signal;

[0046] It should be added that the construction method of the empirical analysis model is as follows: the empirical analysis model is established based on a convolutional neural network and is provided with a basic empirical database. The basic empirical database stores a large number of converter smelting photos and the conclusions of the workers on each converter smelting photo. The empirical analysis model completes training and learning based on the basic empirical database, and inputs converter smelting photos that are not stored in the basic empirical database for verification. The workers judge whether the empirical analysis model is trained perfectly based on the verification results, and adjust the empirical analysis model to ensure the good performance of the empirical analysis model.

[0047] The analysis process is mainly feature analysis. After the converter smelting photos are sent to the empirical analysis model, the empirical analysis model extracts the features that appear in the smelting photos, and outputs the time corresponding to the extracted features according to the feature-time correspondence table. The mean of all feature times is calculated to obtain the working time of the smelting photos. The feature-time correspondence table is established by the workers, which is a list of the relationship between the different features that appear on the surface of the metal solution during the converter smelting process and the corresponding time.

[0048] The twin layer is used to build digital content that maps physical entity objects, access various data and models of physical entity objects from the IoT layer, and also provide a model building environment based on business needs. The twin simulates the operation of the physical entity by integrating data and models, and drives the operation of the twin with the data of the physical entity to maintain consistency between the virtual and the real.

[0049] S3, the converter working data J in the smelting database K t The simulation correction model is sent to the simulation correction model, and the simulation correction model is based on the converter working data J t Simulate converter smelting, then output when T = T z Prediction of converter data L Tz , determine whether "|L Tz -X Tz |<β”, β is the set error data threshold, X Tz To set at T z The converter working data of successful smelting at this time, if it satisfies "|L Tz -X Tz |<β", no operation is performed. If "|L Tz -X Tz |<β”, then enter S4;

[0050] The construction method of the simulation correction model is as follows: firstly, three-dimensional modeling is performed according to the geometric structure of the physical entity to obtain a solid model, and then the simulation tool is used to simulate the entity behavior of the solid model, and the simulation correction model is constructed in combination with the reaction mechanism during the working process of the converter;

[0051] The simulation correction model is a data model based on a neural network. It can simulate the working process of the converter smelting according to the input data, so that it can respond quickly after adjusting the input parameters, and simulate reactions close to the actual situation to complete the simulation work of the converter, replicate the same working process as the real physical equipment, and complete the simulated metal smelting according to the data; it can also complete the missing converter smelting process in the middle according to the simulated smelting results, extract the corresponding data of the simulated smelting, and calculate the converter working data J t .

[0052] S4. All converter working data J in the smelting database K are t Enter the simulation correction model and input the virtual working data X Tz , for time max to T z Converter smelting simulation, output simulation virtual work data X n , n∈[t+T,N], N is the duration threshold T z The ratio of the interval time T, that is, N = T z / T;

[0053] S5. All converter working data J t and Virtual Work Data X n The twin layer simulation correction model is sent to complete the converter smelting simulation and output in T z Converter working data at J Tz , determine the converter working data J Tz Does it satisfy the "|J Tz -X Tz |<β”, if “|J Tz -X Tz |<β”, a warning signal is sent and the process goes to S6. If “|J Tz -X Tz |<β”, then the virtual working data X n Recorded as converter working data J t+T , stored in smelting database K, and enter S7;

[0054] S6, assign n+T to n, and go to S5;

[0055] S7. Traverse the smelting database K and extract all converter working data J that satisfy "t>max" t , the twin layer is based on the extracted converter working data J t Guide the IoT layer to control the converter equipment and complete the adjustment of smelting.

[0056] It should be noted that the specific steps of the twin layer guiding the IoT layer to control the converter equipment are:

[0057] The simulation correction model is used to extract the converter working data J t Mapping is performed to obtain the corresponding data of the simulated smelting, and the IoT layer obtains the converter working data extracted by the twin layer. t , sorted from small to large according to the working time t, the reverse controller obtains the various data corresponding to the simulated smelting mapped by the simulation correction model according to the sorting, and sends the various data corresponding to the simulated smelting to the control units of various entity devices in the physical layer through the connection between the Internet of Things layer and the physical layer, thereby completing the guidance and control of the converter equipment.

[0058] The present invention analyzes the stability and feasibility of the steelmaking process in real time by analyzing the intelligent steelmaking control model of the steelmaking process control elements, thereby achieving stable quality of molten steel and improving the one-time drawing rate of steel. At the same time, data simulation steelmaking can be performed for the research and development of new steel grades, saving costs and optimizing the research and development process.

[0059] In order to enable workers and managers to better observe and use it, an application layer is also set up. The application layer can receive warning signals sent by the twin layer, and can visualize the simulation correction model to simulate the converter smelting process through the display device, and can adjust the parameters of each link for manual control.

[0060] In order to simultaneously detect the impact of converter equipment on smelting, it also includes functional judgment of converter equipment.

[0061] Extract the converter working data J corresponding to max in the smelting database K max , and the remaining converter work data J of the smelting database K t The simulation correction model is sent to the simulation correction model, and the simulation correction model is based on the converter working data J t Simulate converter smelting and obtain the predicted converter data L when t = max max , determine the converter working data J max Does it satisfy the "|J max -L max |<β”, if “|J max -L max |<β", no operation is performed. If the "|J max -L max |<β”, a warning is issued, indicating that the device is malfunctioning.

[0062] When the blowing process has lasted for half an hour, the maximum is 30 minutes, the corresponding data is extracted, and the model analysis and prediction are performed on the data recorded 9 times before. The data differences are compared to determine whether the equipment is normal, ensuring that the smelting process can be completed according to the accurate data relationship.

[0063] The present invention visualizes the process of simulating converter smelting by providing an application layer, controls the equipment of the physical layer through the application layer, and can receive warning signals at any time to ensure timely processing. It also detects the converter equipment through the simulation correction model to discover the influence of the converter smelting equipment on the steelmaking process in time, reduce the occurrence of accidents, and ensure the success rate of steelmaking.

[0064] The present application performs data delay detection on the operation data before sending the operation data into the standard fault clustering network. When some operation data have a large delay, it is eliminated; then it is predicted and supplemented, which can eliminate the influence of data delay on the detection results to a certain extent and improve the accuracy of the detection results.

[0065] Example 2

[0066] Figure 1 For a converter smelting parameter optimization design system based on digital twin provided in the embodiment of this application, please refer to Figure 1 The system can be specifically applied to relevant clustering devices or equipment, and the clustering devices or equipment store corresponding execution programs. It is conceivable that the execution subject may also have other options, which is not limited here.

[0067] A converter smelting parameter optimization design system based on digital twin, characterized by comprising:

[0068] The information collection module is used to collect converter operation data, take converter smelting photos, and determine the converter operation time;

[0069] Image recognition module, used to analyze converter smelting photos and determine the analysis results;

[0070] The simulation correction module is used to simulate and analyze the operating data of the collected mechanical equipment, and to judge the smelting results and simulate the operating data of the mechanical equipment;

[0071] The warning module issues corresponding warnings according to the running time judgment result of the information collection module, the analysis result of the image recognition module and the smelting result of the simulation correction module;

[0072] The output guidance module outputs the operation data of the mechanical equipment simulated by the simulation correction module according to the smelting results.

[0073] As a further preferred solution, it also includes:

[0074] The visual control module is used to display the simulation process of the simulation correction module and control the simulation process.

[0075] It should be understood that those skilled in the art can make improvements or changes based on the above description, and all such improvements and changes should fall within the scope of protection of the claims attached to this application. Parts not described in detail in this specification belong to the prior art known to professional and technical personnel in this field.

Claims

1. A converter smelting parameter optimization design method based on digital twin, characterized in that: include: S1, interval time T, the IoT layer obtains the converter working data J recorded by the physical layer t And the photos of converter smelting taken t , t is the difference between the real time of recording converter working data and taking converter smelting photos and the initial time T0, that is, t is the working time, T0 is the initial time when the converter starts this smelting, T z The time threshold set for a single converter smelting is used to determine whether the working time t satisfies "t≥T z ", if "t≥T z ", then the converter working data J t Photo of converter smelting t Stored in the smelting database K, if it does not satisfy "t≥T z ", a warning signal is issued and the smelting database K is cleared; S2, twin layer traversal smelting database K converter smelting photos P t Corresponding working time t, extract the maximum working time t and record it as max, extract the smelting photo P corresponding to the working time max t , and sent to the empirical analysis model, the empirical analysis model output converter smelting photo P t Analyze the empirical recognition time t0, and determine whether the empirical recognition time t0 satisfies "t0 = max ± γ", γ is the set error time threshold, if it satisfies, enter S3, if not, send a warning signal; S3, the converter working data J in the smelting database K t The simulation correction model is sent to the simulation correction model, and the simulation correction model is based on the converter working data J t Simulate converter smelting, then output when T = T z Prediction of converter data L Tz , determine whether "|L Tz -X Tz |<β”, β is the set error data threshold, X Tz To set at T z The converter working data of successful smelting is Tz -X Tz |<β", no operation is performed. If "|L Tz -X Tz |<β”, then enter S4; S4. All converter working data J in the smelting database K are t Enter the simulation correction model and input the virtual working data X Tz , for time max to T z Converter smelting simulation, output simulation virtual work data X n , n∈[t+T,N], N is the duration threshold T z The ratio of the interval time T, that is, N = T z / T; S5. All converter working data J t and Virtual Work Data X n The twin layer simulation correction model is sent to complete the converter smelting simulation and output in T z Converter working data at J Tz , determine the converter working data J Tz Is it satisfied? Tz -X Tz |<β", if it does not satisfy "|J Tz -X Tz |<β", a warning signal is sent and S6 is entered. If "|J Tz -X Tz |<β”, then the virtual working data X n Recorded as converter working data J t+T , stored in smelting database K, and enter S7; S6, assign n+T to n, and go to S5; S7. Traverse the smelting database K and extract all converter working data J that satisfy "t>max" t , the twin layer is based on the extracted converter working data J t Guide the IoT layer to control the converter equipment and complete the adjustment of smelting.

2. A converter smelting parameter optimization design method based on digital twinning according to claim 1, characterized in that: The specific steps for recording converter working data and taking converter smelting photos at the physical layer are as follows: The IoT layer connects production equipment, collects various data, and calculates them through the weight formula: t =μ1J 1t +μ2J 2t +μ3J 3t +······+μ m J mt , m=1, 2, 3······M, M is the total number of all collected data, μ m The weight threshold set for the mth item of collected data is used to obtain the overall data, which is recorded as converter working data J t ; There is a camera in the converter, which takes photos of the smelting conditions in the converter at intervals T, recorded as converter smelting photos P t The converter smelting photo is a photo of the state information of the metal solution surface when the metal is smelted in the converter.

3. A converter smelting parameter optimization design method based on digital twinning as claimed in claim 2, characterized in that: The method for constructing the empirical analysis model in step S2 is: The empirical analysis model is established based on a convolutional neural network and is equipped with a basic empirical database. The basic empirical database stores a large number of converter smelting photos and the workers' conclusions on each converter smelting photo. The empirical analysis model completes training and learning based on the basic empirical database, and inputs converter smelting photos that are not stored in the basic empirical database for verification. The workers judge whether the empirical analysis model is fully trained based on the verification results and adjust the empirical analysis model to ensure the good performance of the empirical analysis model. The analysis process is mainly feature analysis. After the converter smelting photos are sent to the empirical analysis model, the empirical analysis model extracts the features that appear in the smelting photos, and outputs the time corresponding to the extracted features according to the feature-time correspondence table, and calculates the mean of all feature times to obtain the working time of the smelting photos. The feature-time correspondence table is established by the workers and is a list of the relationships between the different features that appear on the surface of the molten metal during the converter smelting process and the corresponding times.

4. A converter smelting parameter optimization design method based on digital twinning as claimed in claim 3, characterized in that: The construction method of the simulation correction model in step S3 is: First, three-dimensional modeling is performed according to the geometric structure of the physical entity to obtain a solid model. Then, simulation tools are used to simulate the entity behavior of the solid model. Combined with the reaction mechanism during the converter working process, a simulation correction model is constructed. The simulation correction model is a data model based on a neural network. It can simulate the working process of the converter smelting according to the input data, so that it can respond quickly after adjusting the input parameters, and simulate reactions close to the actual situation to complete the simulation work of the converter, replicate the same working process as the real physical equipment, and complete the simulated metal smelting according to the data; it can also complete the missing converter smelting process in the middle according to the simulated smelting results, extract the corresponding data of the simulated smelting, and calculate the converter working data J t .

5. The converter smelting parameter optimization design method based on digital twinning according to claim 4, characterized in that: The specific steps of the twin layer guiding the IoT layer to control the converter equipment in S7 are: The simulation correction model is used to extract the converter working data J t Mapping is performed to obtain the corresponding data of the simulated smelting, and the IoT layer obtains the converter working data extracted by the twin layer. t , sorted from small to large according to the working time t, the reverse controller obtains the various data corresponding to the simulated smelting mapped by the simulation correction model according to the sorting, and sends the various data corresponding to the simulated smelting to the control units of various entity devices in the physical layer through the connection between the Internet of Things layer and the physical layer, thereby completing the guidance and control of the converter equipment.

6. A converter smelting parameter optimization design method based on digital twinning as claimed in claim 5, characterized in that: It also includes an application layer, which can receive warning signals sent by the twin layer, and can visualize the simulation correction model simulating the converter smelting process through a display device, and can adjust the parameters of each link for manual control.

7. The converter smelting parameter optimization design method based on digital twinning according to claim 6, characterized in that: It also includes functional judgment of converter equipment, the specific steps are: Extract the converter working data J corresponding to max in the smelting database K max , and the remaining converter work data J of the smelting database K t The simulation correction model is sent to the simulation correction model, and the simulation correction model is based on the converter working data J t Simulate converter smelting and obtain the predicted converter data L when t = max max , determine the converter working data J max Is it satisfied? max -L max |<β", if "|J max -L max |<β", no operation is performed. If "|J max -L max |<β”, an alarm is issued, indicating that the converter equipment is malfunctioning.

8. A converter smelting parameter optimization design system based on digital twin, characterized in that: Included are: The information collection module is used to collect converter operation data, take converter smelting photos, and determine the converter operation time; Image recognition module, used to analyze converter smelting photos and determine the analysis results; The simulation correction module is used to simulate and analyze the operating data of the collected mechanical equipment, and to judge the smelting results and simulate the operating data of the mechanical equipment; The warning module issues corresponding warnings according to the running time judgment result of the information collection module, the analysis result of the image recognition module and the smelting result of the simulation correction module; The output guidance module outputs the operation data of the mechanical equipment simulated by the simulation correction module according to the smelting results.

9. The converter smelting parameter optimization design system based on digital twin according to claim 7, characterized in that: Also included are: The visual control module is used to display the simulation process of the simulation correction module and control the simulation process.

Images