Iron plant operation support method, operation support device, display device, and operation support program

By using machine learning models in an iron smelting plant, the comprehensive cost is inferred based on the actual data of blast furnace operating parameters, the problem of inaccurate cost management in the existing technology is solved, more accurate cost management and operation plan is achieved, and iron smelting is improved Factory operation support effect.

CN118901050APending Publication Date: 2024-11-05JFE STEEL CORP
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202380029111.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2022-03-28
Filing Date
2023-03-24
Publication Date
2024-11-05

AI Technical Summary

Technical Problem

The prior art is difficult to search for optimization conditions based on theoretical values ​​in iron smelting plants, resulting in the fact that the operation support technology of iron smelting plants cannot obtain the optimal solution suitable for reality, and the cost management is not accurate enough.

Method used

Using a machine learning-based learning model, the comprehensive costs of the blast furnace and energy sector are inferred and output by inputting actual data of blast furnace operating parameters, and the model is corrected using actual performance data to improve the accuracy of cost management.

Benefits of technology

It improves the accuracy of inference of comprehensive costs in iron smelting plants, provides more correct cost management and operational planning, and improves the operational support technology of iron smelting plants.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118901050A_ABST
    Figure CN118901050A_ABST
Patent Text Reader

Abstract

The present invention improves a technique for supporting operations in an iron making plant. This operation support method is used in an iron mill having a blast furnace department and an energy department, is executed by an information processing device, and inputs data of blast furnace operation parameters that affect the energy department to a learning model related to the iron mill. The integrated cost of the blast furnace department and the energy department is estimated by a learning model on the basis of the inputted data of the blast furnace operating parameters, and the estimated integrated cost is output, and the learning model is a model trained on the basis of actual performance data including the actual performance of the blast furnace operating parameters and the integrated cost.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present disclosure relates to an operation support method, an operation support device, a display device, and an operation support program for supporting operations in an ironworks. This application claims priority based on Japanese Patent Application No. 2022-052019 filed on March 28, 2022, the contents of which are hereby incorporated by reference. Background Art

[0002] In the past, there are known technologies for supporting operations in ironworks. For example, Patent Document 1 describes a method for determining conditions that minimize costs by taking the overall cost of an ironworks as an objective function and optimizing operating conditions such as crude steel production, molten iron blending ratio, and the type and amount of raw materials used as variables in a calculation model.

[0003] Patent Document 1: Japanese Patent Application Publication No. 2005-55997

[0004] In actual operation, the material heat balance of the blast furnace and the efficiency of the power plant often cause deviations from the theoretical values ​​derived from the calculation model. Therefore, the calculation model needs to be adjusted each time using the latest actual results. However, in the genetic algorithm processing used as the optimization processing method in Patent Document 1, this adjustment is difficult. Thus, in the prior art, it may not be possible to obtain the optimal solution suitable for actual operation by searching for optimization conditions based on theoretical values. Thus, there is room for improvement in the technology for supporting operations in iron and steel plants. Summary of the invention

[0005] The present disclosure has been made in view of the above circumstances and aims to improve the technology for supporting operations in an ironworks.

[0006] An operation support method according to one embodiment of the present disclosure is used in an ironworks having a blast furnace department and an energy department and is executed by an information processing device, wherein:

[0007] The operation support method inputs data of blast furnace operation parameters having an impact on the energy department into a learning model related to the ironworks,

[0008] Based on the input data of the blast furnace operating parameters, the comprehensive costs of the blast furnace department and the energy department are inferred by the learning model, and

[0009] Output the above inferred comprehensive cost,

[0010] The above learning model is a model trained based on actual performance data including actual performance of blast furnace operating parameters and comprehensive costs.

[0011] An operation support device according to one embodiment of the present disclosure includes a control unit and is an operation support device in an ironworks having a blast furnace department and an energy department, wherein:

[0012] The control unit inputs data of blast furnace operating parameters having an impact on the energy department into the learning model related to the ironworks,

[0013] Based on the input data of the blast furnace operating parameters, the comprehensive costs of the blast furnace department and the energy department are inferred by the learning model, and

[0014] Output the above inferred comprehensive cost,

[0015] The learning model is a model trained using actual performance data including actual performance of blast furnace operating parameters and overall costs as teaching data.

[0016] A display device according to one embodiment of the present disclosure is a display device for displaying the output of an operation support device having a control unit in an ironworks having a blast furnace department and an energy department, wherein:

[0017] The control unit of the operation support device displays the comprehensive costs of the blast furnace department and the energy department, which are inferred by the learning model of the ironworks based on the data of the blast furnace operation parameters that have an impact on the energy department.

[0018] The learning model is a model trained using actual performance data including actual performance of blast furnace operating parameters and overall costs as teaching data.

[0019] An operation support program according to one embodiment of the present disclosure is an operation support program executed by an information processing device in an ironworks having a blast furnace department and an energy department, wherein:

[0020] Make the computer execute the following processing, namely:

[0021] inputting data of blast furnace operating parameters having an impact on the energy sector into the learning model related to the ironworks,

[0022] Based on the input data of the blast furnace operating parameters, the comprehensive costs of the blast furnace department and the energy department are inferred by the learning model, and

[0023] Output the above inferred comprehensive cost,

[0024] The learning model is a model trained using actual performance data including actual performance of blast furnace operating parameters and overall costs as teaching data.

[0025] According to one embodiment of the present disclosure, it is possible to improve the technology for supporting operations in an ironworks. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 This is a block diagram showing a schematic configuration of an operation support device according to one embodiment of the present disclosure.

[0027] Figure 2 This is a flowchart showing the operation of the operation support device according to one embodiment of the present disclosure.

[0028] Figure 3 This is a diagram showing the structure of a learning model according to one embodiment of the present disclosure.

[0029] Figure 4 This is a flowchart showing the operation of the operation support device according to one embodiment of the present disclosure.

[0030] Figure 5 The results of a case study were performed under 6 different operating conditions.

[0031] Figure 6 These are the case study results when the antecedent conditions are different. DETAILED DESCRIPTION

[0032] Hereinafter, an operation support technology in an ironworks according to an embodiment of the present disclosure will be described with reference to the drawings.

[0033] In each figure, the same reference numerals are attached to the same or corresponding parts. In the description of this embodiment, the description of the same or corresponding parts will be appropriately omitted or simplified.

[0034] Reference Figure 1 The outline and structure of the operation support technology in the ironworks involved in this embodiment are described. The operation support technology in the ironworks involved in this embodiment is executed by an information processing device 10 (hereinafter also referred to as the operation support device 10). This support method is used in an ironworks having a blast furnace department and an energy department.

[0035] First, the outline of the present embodiment is explained, and then described in detail. Usually, in an iron and steel plant, a production plan including the amount of steel produced, the amount of iron produced, the amount of coke produced, etc. is prepared based on the needs of steel products. The production plan is prepared for each factory for the purpose of continuing production activities without excess or deficiency. In addition, in the energy department that manages all energy, a supply and demand plan is prepared for supplying the utilities required in each factory without causing them to be insufficient. Here, utilities include gas, electricity, steam, water, compressed air, etc. When making an operation plan, in each department including the energy department, not only the production plan is implemented, but also cost management is performed from the perspective of minimizing manufacturing costs. Specifically, under the condition of satisfying the production plan, the operation plan is prepared in a manner that minimizes manufacturing costs by optimizing various operating conditions such as the types and amounts of raw materials and utilities used, and equipment operating conditions. In an iron and steel plant, the combustible gases (B gas, C gas, LD gas) produced as by-products in blast furnaces, coke ovens, and converters are used directly or as mixed gases (M gas) in the plant, and the remaining part is used in the self-use power generation equipment managed by the energy department. In the self-use power generation equipment, the combustion exhaust gas produced by burning the above-mentioned by-product gases and a part of the purchased fuel is subjected to heat exchange with pure water, etc., thereby generating steam, which is used for power generation in the steam turbine. In addition, in this process, the required amount of steam for the iron and steel plant is also taken out (exhaust). Based on the above, the supply and demand management of various utilities managed by the energy department, especially gas, electricity, and steam, is closely related not only to the energy equipment conditions such as turbines, but also to the operating conditions managed in each factory, especially the conditions for the generation of by-product gases. However, the supply and demand plans for gas, electricity, and steam are usually made by receiving detailed operating plans for each factory, and it is difficult to say that the cost situation of the energy department is fully considered in the operating plan. Therefore, the result obtained by summarizing the optimal costs of each department is not necessarily the optimal cost of the entire iron and steel plant. Therefore, the operation support technology in the ironworks according to the present embodiment aims to provide a blast furnace and energy comprehensive simulation technology that enables more accurate cost management.

[0036] In the operation of a blast furnace, first, the operating conditions are determined using a model constructed from material and heat balance data. However, in actual operation, due to the ever-changing state of the blast furnace or the influence of changes in the surrounding environment, the theoretical value calculated by the model and the actual data obtained by actual operation diverge. In this embodiment, in order to suppress this divergence, machine learning is performed based on the data of actual operation to correct the model.

[0037] Specifically, in the operation support technology in the ironworks involved in the present embodiment, the operation support device 10 uses a physical model (hereinafter also referred to as a learning model) involved in the ironworks to perform the inference processing required for the operation support in the ironworks. The learning model is a learning model that makes the input the data of the blast furnace operation parameters and the output the cost of the blast furnace department and the energy department (hereinafter also referred to as the comprehensive cost). The learning model trains the actual performance data including the data of the blast furnace operation parameters and the actual performance of the comprehensive cost as teaching data. Here, in the present embodiment, the operation support device 10 inputs the data of the blast furnace operation parameters that have an impact on the energy department into the learning model. In addition, the operation support device 10 infers the comprehensive cost through the learning model based on the input data of the blast furnace operation parameters. In addition, the operation support device 10 outputs the inferred comprehensive cost.

[0038] Thus, according to the present embodiment, the operation support device 10 inputs the data of the blast furnace operation parameters that have an impact on the energy department into the learning model and outputs the comprehensive cost. In addition, the learning model is trained with the actual performance data as teaching data. That is, in the present embodiment, in order to correct the deviation between the actual operation and the theoretical value derived from the model, there is a feedback function based on the actual performance data. Thus, in the present embodiment, the accuracy of the inference of the comprehensive cost in the iron and steel plant is improved. In other words, according to the present embodiment, a blast furnace and energy comprehensive simulation technology that can perform more accurate cost management can be provided, which can improve the technology for supporting operations in the iron and steel plant.

[0039] Here, the data of blast furnace operating parameters that have an impact on the energy department include, for example, reduction material conditions and air supply conditions. The operating conditions that have an impact on the comprehensive cost of the blast furnace department and the energy department are various. For example, in the blast furnace, reduction material conditions such as coke ratio and fine powder carbon ratio, which are closely related to B gas generation conditions, air supply conditions such as air humidity and air supply temperature, and air supply oxygen concentration also have a great impact on the comprehensive cost. In this embodiment, by considering the reduction material conditions and air supply conditions as variables, more accurate cost management and highly executable operation plan preparation are achieved.

[0040] Next, each structure of the operation support device 10 will be described in detail.

[0041] The operation support device 10 is any device used by a user (for example, an administrator or a person in charge of a steel plant). For example, a general-purpose electronic device or a dedicated electronic device can be adopted as the operation support device 10. Figure 1 As shown, the operation support device 10 includes a control unit 11 , a storage unit 12 , a communication unit 13 , an input unit 14 , and an output unit 15 .

[0042] The control unit 11 includes at least one processor, at least one dedicated circuit, or a combination thereof. The processor is a general-purpose processor such as a CPU (central processing unit) or a GPU (graphics processing unit) or a dedicated processor dedicated to specific processing. The dedicated circuit is, for example, an FPGA (field-programmable gate array) or an ASIC (application specific integrated circuit). The control unit 11 controls each part of the operation support device 10 and executes processing related to the operation of the operation support device 10.

[0043] The storage unit 12 includes at least one semiconductor memory, at least one magnetic memory, at least one optical memory, or a combination of at least two of them. The semiconductor memory is, for example, a RAM (random access memory) or a ROM (read only memory). The RAM is, for example, an SRAM (static random access memory) or a DRAM (dynamic random access memory). The ROM is, for example, an EEPROM (electrically erasable programmable read only memory). The storage unit 12 functions as, for example, a main storage device, an auxiliary storage device, or a cache memory. The storage unit 12 stores data used for the operation of the operation support device 10 and data obtained by the operation of the operation support device 10. For example, the storage unit 12 stores a learning model. In addition, the learning model may also be stored in an external device different from the operation support device 10. In this case, the operation support device 10 uses the learning model by accessing the external device storing the learning model via the communication unit 13.

[0044] The communication unit 13 includes at least one external communication interface. The communication interface can be any interface of wired communication or wireless communication. In the case of wired communication, the communication interface is, for example, a LAN (Local Area Network) interface or a USB (Universal Serial Bus). In the case of wireless communication, the communication interface is, for example, an interface corresponding to mobile communication standards such as LTE (Long Term Evolution), 4G (4th generation) or 5G (5th generation), or an interface corresponding to short-range wireless communication such as Bluetooth (registered trademark). The communication unit 13 receives data used for the operation of the operation support device 10, or sends data obtained through the operation of the operation support device 10.

[0045] The input unit 14 includes at least one input interface. The input interface is, for example, a physical key, an electrostatic capacitance key, a pointing device, or a touch screen integrated with the display. In addition, the input interface may be, for example, a microphone for accepting voice input or a camera for accepting gesture input. The input unit 14 accepts an operation to input data for the action of the operation support device 10. The input unit 14 may be replaced by the operation support device 10 and connected to the operation support device 10 as an external input device. As a connection method, for example, any method such as USB (Universal Serial Bus), HDMI (registered trademark) (High-Definition Multimedia Interface) or Bluetooth (registered trademark) can be used.

[0046] The output unit 15 includes at least one output interface. The output interface is, for example, a display that outputs information through images or a speaker that outputs information through sounds. The display is, for example, an LCD (liquid crystal display) or an organic EL (electro luminescence) display. The output unit 15 displays and outputs data obtained through the operation of the operation support device 10. The output unit 15 may be replaced by the operation support device 10 and connected to the operation support device 10 as an external output device. As a connection method, any method such as USB, HDMI (registered trademark) or Bluetooth (registered trademark) can be used.

[0047] The functions of the operation support device 10 are realized by executing the program involved in this embodiment by a processor corresponding to the operation support device 10. That is, the functions of the operation support device 10 are realized by software. The program causes the computer to function as the operation support device 10 by causing the computer to execute the operation of the operation support device 10. That is, the computer functions as the operation support device 10 by executing the operation of the operation support device 10 according to the program.

[0048] In the present embodiment, the program can be recorded on a recording medium that can be read by a computer. The recording medium that can be read by a computer includes a non-temporary computer-readable medium, such as a magnetic recording device, an optical disc, an optical magneto-recording medium, or a semiconductor memory. The circulation of the program is carried out, for example, by selling, transferring, or renting a removable recording medium such as a DVD (digital versatile disc) or CD-ROM (compact disc read only memory) on which the program is recorded. In addition, the circulation of the program can also be carried out by storing the program in a storage device of an external server and sending the program from the external server to other computers. In addition, the program can also be provided as a program product.

[0049] A part or all of the functions of the operation support device 10 may be realized by a dedicated circuit corresponding to the control unit 11. In other words, a part or all of the functions of the operation support device 10 may be realized by hardware.

[0050] (Operation Example 1 of Operation Support Device 10)

[0051] Next, operation example 1 of the operation support device 10 according to the present embodiment will be described. Figure 2 This is a flowchart showing an example of an operation support method executed by the operation support device 10 according to the present embodiment.

[0052] Step S100: The control unit 11 of the operation support device 10 trains the learning model based on the actual performance data. Specifically, the control unit 11 uses the actual performance data including the actual performance of the blast furnace operation parameters and the comprehensive cost as teaching data to train the learning model. As described above, there may be a deviation between the theoretical value calculated by the physical model and the actual data obtained from the actual operation. In this embodiment, in order to suppress the deviation, the learning model is trained based on the data of the actual operation.

[0053] In the present embodiment, the blast furnace operating parameters include reducing material parameters and air supply parameters. The reducing material parameters include the coke ratio and the fine carbon ratio. In addition, the reducing material parameters may also include the composition and carbon content of the coke. The air supply parameters include air supply humidity, air supply temperature, and air supply oxygen concentration. These blast furnace operating parameters all have a great impact on the energy department. In other words, these blast furnace operating parameters have a great impact on the cost management in the energy department, that is, the cost in the energy department. Therefore, in the present embodiment, a learning model that takes the blast furnace operating parameters that have an impact on the energy department as input is adopted. In the present embodiment, by adopting the learning model, the accuracy of inference of the comprehensive cost in the iron and steel plant is improved. Below, in the present embodiment, the case where the blast furnace operating parameters include the coke ratio, the fine carbon ratio, the air supply humidity, the air supply temperature, and the air supply oxygen concentration is explained.

[0054] The comprehensive cost includes the energy cost of the blast furnace department and the main cost of the energy department. The energy cost of the blast furnace department includes the cost related to the reduced materials and the utility cost. The utility cost refers to the cost of the utilities required in each factory, etc. The main cost of the energy department includes the cost of purchased fuel and the cost of buying and selling electricity. In this embodiment, the comprehensive cost is determined by the following mathematical formula (1).

[0055] [Mathematical formula 1]

[0056] Total cost = Energy cost of blast furnace department + Main cost of energy department = (reduced material related cost + utility cost) + (purchased fuel cost + electricity purchase and sale cost)…(1)

[0057] The learning model is, for example, a machine learning model generated based on a multilayer perceptron consisting of an input layer, a hidden layer, and an output layer, but is not limited thereto. The learning model may also be any model generated based on a machine learning algorithm such as a convolutional neural network (CNN), a recurrent neural network (RNN), or other deep learning. In addition, the learning model is not limited to a model generated by machine learning. For example, the learning model may also be a multiple regression model, a model based on a Kalman filter, or the like. In addition, for example, the learning model may also be a theoretical model constructed based on data on material balance and heat balance. In addition, the "training" in this embodiment includes training in a machine learning algorithm, but is not limited thereto. The "training" in this embodiment includes, for example, optimization or correction processing of various parameters in a multiple regression model, a theoretical model, or the like.

[0058] Step S110: The control unit 11 determines the blast furnace operating parameters based on the conditions (hereinafter also referred to as preconditions) determined according to the production plan of the ironmaking plant. The preconditions are set, for example, by user input. The preconditions include the iron output (or iron output ratio), the unit price of coke, the rolling operation plan, the coal market conditions, and the crude oil market conditions. Specifically, the control unit 11 determines the blast furnace operating parameters, for example, using a lookup table corresponding to the iron output and the blast furnace operating parameters. In this case, the storage unit 12 stores the lookup table. In addition, the control unit 11 determines the blast furnace operating parameters corresponding to the preconditions by referring to the storage unit 12.

[0059] Step S120 : The control unit 11 inputs the data of the blast furnace operating parameters determined in step S110 to the learning model learned in step S100 .

[0060] Step S130: The control unit 11 infers the comprehensive costs of the blast furnace department and the energy department through the learning model based on the input data of the blast furnace operating parameters. In this inference process, the above-mentioned premise is used. Specifically, the learning model uses the rolling operation plan, the coal market situation, and the crude oil market situation, and infers the comprehensive costs of the blast furnace department and the energy department based on the input data of the blast furnace operating parameters.

[0061] Step S140 : The control unit 11 outputs the comprehensive cost estimated in step S130 via the output unit 15 .

[0062] Here, in the above, the output of the learning model is an example of comprehensive cost, but it is not limited to this. The learning model can also further infer the carbon dioxide emissions of the blast furnace department and the energy department based on the data of the input blast furnace operating parameters. The carbon dioxide emissions include the carbon dioxide emissions from the reducing materials, the carbon dioxide emissions from the purchased fuels, and the carbon dioxide emissions from the purchased electricity. In addition, the carbon dioxide emissions from the sold electricity are subtracted from the carbon dioxide emissions. That is, the carbon dioxide emissions are determined, for example, by the following mathematical formula (2).

[0063] [Mathematical formula 2]

[0064] CO2 emissions = CO2 emissions from reducing materials + CO2 emissions from purchased fuel + CO2 emissions from purchased electricity - CO2 emissions from sold electricity…(2)

[0065] In the case of inferring carbon dioxide emissions, the actual performance data in the above step S100 includes the actual performance of carbon dioxide emissions. That is, in step S100, the control unit 11 uses the actual performance data including the actual performance of blast furnace operating parameters, comprehensive costs, and carbon dioxide emissions as teaching data to learn the learning model. In addition, in step S130, the control unit 11 infers the comprehensive costs and carbon dioxide emissions of the blast furnace department and the energy department through the learning model based on the input data of the blast furnace operating parameters. Furthermore, in step S140, the control unit 11 outputs the inferred comprehensive costs and carbon dioxide emissions.

[0066] In addition, the comprehensive cost may also include the cost equivalent to the carbon dioxide emissions. The cost equivalent to the carbon dioxide emissions is determined by the carbon dioxide emissions and the unit price of carbon dioxide. In this case, the above-mentioned prerequisite includes the unit price of carbon dioxide. Hereinafter, the comprehensive cost including the cost equivalent to the carbon dioxide emissions is also referred to as the carbon assessment comprehensive cost. The carbon assessment comprehensive cost is determined, for example, by the following mathematical formula (3).

[0067] [Mathematical formula 3]

[0068] Carbon assessment comprehensive cost = comprehensive cost + carbon dioxide emissions × carbon dioxide unit price…(3)

[0069] In addition, the learning model is a model corresponding to an ironworks including a blast furnace department and an energy department and this is described, but is not limited to this. For example, the learning model may also be a combination of two models: a model corresponding to the blast furnace department (hereinafter also referred to as a blast furnace department model) and a model corresponding to the energy department (hereinafter also referred to as an energy department model). Furthermore, the learning model may also be a model obtained by combining models that further subdivide the blast furnace department model and the energy department model.

[0070] Reference Figure 3 An example of the structure of the learning model is shown. Figure 3The learning model shown includes a blast furnace department model 100 and an energy department model 200. The blast furnace department model 100 includes a blast furnace operation model 110 and a hot blast stove model 120. The blast furnace operation model 110 and the hot blast stove model 120 are models based on material heat balance of the blast furnace and the hot blast stove, respectively. The control unit 11 uses actual performance data as teaching data to train the blast furnace operation model 110 and the hot blast stove model 120. Blast furnace operation parameters based on preconditions are input to the blast furnace operation model 110 and the hot blast stove model 120. The preconditions include an iron-out schedule (iron-out amount, etc.). As described above, the control unit 11 determines the blast furnace operation parameters input to the model based on the iron-out schedule and the lookup table. The blast furnace operation model 110 and the hot blast stove model 120 output data required for the calculation of the comprehensive cost based on the input blast furnace operation parameters. The data is used as an input to the energy department model 200. The data required for the calculation of the comprehensive cost includes, for example, data related to B gas, which is a combustible gas produced as a by-product in the blast furnace. The data related to the B gas include the generation amount of the B gas and the heat per unit volume of the B gas.

[0071] The energy department model 200 includes a utility balance logic 210. The control unit 11 uses the actual performance data as teaching data to train the utility balance logic 210. The utility balance logic 210 outputs the comprehensive cost and the carbon dioxide emission based on the input prerequisites and the data related to B gas. The prerequisites include the rolling operation plan, the coal market conditions, and the crude oil market conditions. That is, the energy department model 200 uses the prerequisites and the output of the blast furnace department model 100 as inputs and outputs the comprehensive cost and the carbon dioxide emission.

[0072] (Operation Example 2 of the Operation Support Device 10)

[0073] Next, operation example 2 of the operation support device 10 according to the present embodiment will be described. Figure 4 1 is a flowchart showing an example of an operation support method performed by the operation support device 10 according to the present embodiment. Here, an example of an operation in which the operation support device 10 outputs a plurality of operation data candidates (hereinafter also referred to as candidate operation data) based on the comprehensive cost inferred by the learning model is described. Figure 3 The processing in step S100 uses the trained learning model.

[0074] Step S210: The control unit 11 of the operation support device 10 determines the representative value of the blast furnace operation parameter (hereinafter referred to as the basic condition) based on the precondition. The precondition is set, for example, by user input. The precondition includes the iron output (or iron output ratio), the coke unit price, the rolling operation plan, the coal market conditions, and the crude oil market conditions. Specifically, the control unit 11 determines the basic conditions, for example, using a lookup table corresponding to the iron output and the basic conditions. In this case, the storage unit 12 stores the lookup table. In addition, the control unit 11 determines the basic conditions corresponding to the preconditions by referring to the storage unit 12. The basic conditions include the coke ratio, the fine carbon ratio, the supply air humidity, the supply air temperature, and the supply air oxygen concentration. In addition, at least a part of the basic conditions can also be set by manual input by the user.

[0075] Step S220: The control unit 11 inputs the data of the blast furnace operating parameters within the specified range into the learning model based on the basic conditions. That is, the control unit 11 uses the five operating factors, namely, the coke ratio, the fine carbon ratio, the supply air humidity, the supply air temperature and the supply air oxygen concentration, contained in the basic conditions as operating factors and inputs the data of the blast furnace operating parameters within the specified range into the learning model. The specified range is determined for each of the five operating factors. Specifically, the lower limit of the specified range of the coke ratio is 280kg / t, and the upper limit is 450kg / t. In addition, the lower limit of the specified range of the fine carbon ratio is 80kg / t, and the upper limit is 260kg / t. In addition, the lower limit of the specified range of the supply air humidity is 5g / Nm 3 , the upper limit is 70g / Nm 3 . In addition, the lower limit of the prescribed range of the supply air temperature is 900°C and the upper limit is 1200°C. In addition, the lower limit of the prescribed range of the supply air oxygen concentration is 0% and the upper limit is 10%. These prescribed ranges are determined based on the upper and lower limits of the actual operating performance under the steady state. In addition, the prescribed range can also be determined based on the constraints on equipment management. For example, from the perspective of stably maintaining the function of the hot air furnace, the prescribed range of the supply air temperature is determined to be, for example, the above-mentioned range.

[0076] In addition, the control unit 11 inputs the combination of all operating factors in the specified range into the learning model based on the grid specified within the specified range (the scale within the specified range). For example, with respect to the supply air temperature, the specified range is set to 900°C to 1200°C. In addition, the grid can be specified as 50°C. When the specified range and the grid are specified in this way, 900°C, 950°C, ... 1200°C are input into the learning model as the supply air temperature. In other words, the control unit 11 performs a grid search for the five operating factors using the specified range of each operating factor (for example, the settable upper and lower limits of each operating factor) as the search range. The size of the grid is appropriately determined according to the computational load involved in the grid search.

[0077] Step S230: The control unit 11 estimates the total costs of the blast furnace department and the energy department corresponding to each input input in step S220 through the learning model.

[0078] Step S240: The control unit 11 outputs a plurality of candidate operation data through the output unit 15 based on the comprehensive cost inferred in step S230. In other words, the control unit 11 outputs a plurality of candidate operation data by searching with the comprehensive cost as the objective function. The plurality of candidate operation data are, for example, operation data within the upper 20 in sequence starting from the operation data with the low comprehensive cost inferred corresponding to each input. The operation data includes the coke ratio, the fine carbon ratio, the supply air humidity, the supply air temperature, and the supply air oxygen concentration. That is, the user can easily grasp the operation data within the upper 20 with the low comprehensive cost. In addition, the user can select the operation data that can be actually used from the operation data.

[0079] Here, in the above, the example in which the output of the learning model is the comprehensive cost is described, but it is not limited to this. The learning model can also further infer the carbon dioxide emissions of the blast furnace department and the energy department based on the data of the input blast furnace operation parameters. The carbon dioxide emissions are expressed by the above mathematical formula (2). At this time, in the above steps S220 to S240, the control unit 11 outputs multiple candidate operation data by searching the comprehensive cost and carbon dioxide emissions as the objective function. The candidate operation data at this time is the operation data within the upper 20 based on the order determined based on the comprehensive cost and carbon dioxide emissions inferred corresponding to each input. The order determined based on the comprehensive cost and carbon dioxide emissions is appropriately determined by the evaluation value based on the comprehensive cost and carbon dioxide emissions. For example, or the learning model can also infer the carbon evaluation comprehensive cost. The carbon evaluation comprehensive cost is determined, for example, by the above mathematical formula (3). At this time, in the above steps S220 to S240, the control unit 11 outputs multiple candidate operation data by searching the carbon evaluation comprehensive cost as the objective function. The candidate operation data in this case are, for example, operation data ranked within the top 20 in order from the operation data with the lowest carbon evaluation comprehensive cost estimated corresponding to each input.

[0080] As described above, according to the present embodiment, the operation support device 10 inputs data of blast furnace operation parameters that have an impact on the energy sector into the learning model to output comprehensive costs, etc. In addition, the learning model is trained based on actual performance data. Therefore, according to the present embodiment, the accuracy of inference of comprehensive costs, etc. in the iron and steel plant can be improved. In addition, according to the present embodiment, the operation support device 10 inputs data of blast furnace operation parameters within a specified range to infer comprehensive costs, etc. corresponding to each input and outputs a plurality of candidate operation data. Therefore, a plurality of candidate operation data that are preferred from the perspective of comprehensive costs, etc. can be output from the data of blast furnace operation parameters within a specified range. In this way, according to the present embodiment, the technology for supporting operations in the iron and steel plant can be improved.

[0081] Here, there are various operating conditions that affect the overall cost of the blast furnace department and the energy department. For example, in the blast furnace, the reducing material conditions such as the coke ratio and the fine powder carbon ratio, which are closely related to the B gas generation conditions, and the air supply conditions such as the air humidity, the air supply temperature, and the air supply oxygen concentration also have a great influence on the overall cost and the carbon dioxide emission. According to this embodiment, since these reducing material conditions and air supply conditions are considered as variables, more accurate cost management can be performed, and a more executable operation plan can be prepared.

[0082] Figure 5 This is the result of a case study when different operation data of 6 models were experimentally operated under the same premise. The blast furnace operating parameters of the 6 cases were all thermally equivalent conditions, and were set in the form of a combination of multiple parameter changes equivalent to a coke ratio of 2kg / t. Figure 5 In the case study results, the cost advantages of the blast furnace department and the energy department are shown in bar graphs. In addition, the cost advantage is the value obtained by multiplying the cost by -1. In addition, the comprehensive cost advantage obtained by adding up the cost advantages of the blast furnace department and the energy department is shown in a line graph. The comprehensive cost advantage is the sum of the cost advantages of the blast furnace department and the energy department. Figure 5 As shown, in each of the blast furnace department and the energy department, the case with the greatest cost advantage is Case VI in the blast furnace department and Case IV in the energy department. On the other hand, the case with the greatest overall cost advantage is Case V. In this way, the case with the greatest cost advantage in the blast furnace department, the case with the greatest cost advantage in the energy department, and the case with the greatest overall cost advantage are not necessarily the same. That is, it is necessary to determine the blast furnace parameters based not only on the individual costs of the blast furnace department and the energy department, but also on the comprehensive costs. By inferring the comprehensive cost using the operation support device 10 involved in this embodiment, the overall optimization of the iron and steel plant can be performed.

[0083] Figure 6These are the results of case studies when operations were experimentally performed under different preconditions. Figure 6 All blast furnace operating parameters in the six cases are thermally equivalent conditions, and are set in the form of a combination of multiple parameter changes equivalent to a coke ratio of 2 kg / t. The A and B rolling operation plans respectively represent the case where all the main rolling mills are operated and the case where all the main rolling mills are not operated. The normal and high coke unit price and crude oil market conditions are appropriately set to the range of normal unit price and high unit price. In addition, Figure 6 In the table, the first and second units in terms of comprehensive cost advantage are highlighted. Figure 6 As shown, it can be seen that the cases with the greatest cost advantage are all different under the six preconditions. In other words, when the preconditions change, the comprehensive costs in the blast furnace department and the energy department need to be re-evaluated. According to the operation support device 10 involved in this embodiment, since the comprehensive cost can be easily estimated, even when the preconditions change, the recalculation related to the optimization of the entire ironworks can be carried out in a timely manner.

[0084] In addition, in step S240, the plurality of candidate operation data are operation data of the top 20 digits based on the comprehensive cost and this is explained, but not limited to this. The comprehensive cost involved in the plurality of candidate operation data may also be operation data that meets a specified benchmark. For example, the specified benchmark may be a benchmark based on the ranking, or it may be that the comprehensive cost is lower than a specified threshold. The specified threshold may be a fixed value, or it may be determined based on the average value of the comprehensive cost corresponding to the input searched by the grid search, etc.

[0085] In addition, multiple candidate operation data may also be operation data that meets the specified constraints. In other words, the constraints of the blast furnace department and the energy department in operation or equipment management may be set. When the candidate operation data conflicts with the constraints, they may be removed during sorting. The constraints include the front end temperature of the tuyere, the air supply volume, etc. For example, the lower limit and upper limit of the front end temperature of the tuyere may be specified as 2250°C and 2320°C, respectively. In addition, for example, the lower limit and upper limit of the air supply volume may be specified as 6500Nm, respectively. 3 / min、7400Nm 3 / min. In addition, the constraint condition may include the minimum heavy oil usage in the power generation equipment. The plurality of candidate operation data may be only data satisfying these constraint conditions.

[0086] Furthermore, the control unit 11 may highlight and output candidate operation data having the same or similar operation conditions as the actual performance data among the plurality of candidate operation data output by the output unit 15. In this way, the user can more easily select actually available operation data from the candidate operation data.

[0087] In addition, in the present embodiment, an example is shown in which the learning model outputs any one of the comprehensive cost, the comprehensive cost and the carbon dioxide emission, or the carbon evaluation comprehensive cost, but it is not limited thereto. The learning model may also output the carbon dioxide emission. In other words, the operation support device 10 may also infer and output the carbon dioxide emission only without inferring the comprehensive cost. In this case, in step S130 of the above-mentioned action example 1, the control unit 11 infers the carbon dioxide emission of the blast furnace department and the energy department through the learning model based on the data of the input blast furnace operation parameters. Then, in step S140, the control unit outputs the inferred carbon dioxide emission through the output unit 15. In addition, in step S230 of the above-mentioned action example 2, the control unit 11 infers the carbon dioxide emission of the blast furnace department and the energy department corresponding to each input input through the learning model. Then, in step S240, the control unit 11 outputs a plurality of candidate operation data through the output unit 15 based on the inferred carbon dioxide emission. In other words, in this case, the control unit 11 outputs a plurality of candidate operation data by searching the carbon dioxide emission as the objective function.

[0088] In addition, in the present embodiment, an example is described in which the operation support device 10 is provided with an output unit 15 and the result calculated by the control unit 11 of the operation support device 10 is outputted through the output unit 15, but it is not limited to this. For example, the technology involved in the present embodiment can also be executed in the form of SaaS (Software as a Service). In this case, the operation support device 10 functions as a cloud server. The operation support device 10 outputs the calculation result through another display device different from the operation support device 10. The display device is provided with a communication unit, and receives the calculation result from the operation support device 10 via the communication unit to display information such as the comprehensive cost. In other words, the display device displays the comprehensive cost of the blast furnace department and the energy department inferred by the learning model involved in the ironworks based on the data of the blast furnace operation parameters that have an impact on the energy department by the control unit 11 of the operation support device 10. Alternatively, the display device displays the carbon dioxide emissions of the blast furnace department and the energy department inferred by the learning model involved in the ironworks based on the data of the blast furnace operation parameters that have an impact on the energy department by the control unit 11 of the operation support device 10. Note that any method may be used to request the operation support device 10 for the calculation result. For example, the display device may include an input unit that receives input from the user, and transmit a request for the calculation result to the operation support device 10 based on the input received by the input unit.

[0089] In addition, the operation support device 10 is a device for outputting blast furnace operating parameters, wherein the blast furnace operating parameters are used to optimize the comprehensive cost and carbon dioxide emissions obtained by combining the blast furnace department and the energy department. This embodiment is used by the blast furnace operator to study and implement changes in the blast furnace operating parameters based on this, and is therefore different from optimizing the supply and demand of gas, steam, electricity, etc. only in a specific department such as the energy department.

[0090] Although the present disclosure is described based on the drawings and embodiments, it should be noted that it is easy for those skilled in the art to make various modifications and variations based on the present disclosure. Therefore, it should be noted that these modifications and variations are included in the scope of the present disclosure. For example, the functions included in each means or each step can be reconfigured in a logically non-contradictory manner, and multiple means or steps can be combined into one or divided.

[0091] Description of reference numerals:

[0092] 10…operation support device; 11…control unit; 12…storage unit; 13…communication unit; 14…input unit; 15…output unit; 100…blast furnace department model; 110…blast furnace operation model; 120…hot blast stove model; 200…energy department model; 210…utility balancing logic.

Claims

1. An operation support method, used in an ironworks having a blast furnace department and an energy department, executed by an information processing device, The operation support method is characterized in that: inputting data on blast furnace operating parameters having an impact on the energy sector into a learning model related to the ironworks, Based on the input data of the blast furnace operating parameters, the comprehensive costs of the blast furnace department and the energy department are inferred by the learning model, Output the inferred comprehensive cost, The learning model is a model trained based on actual performance data including actual performance of blast furnace operating parameters and comprehensive costs.

2. The operation support method according to claim 1, characterized in that: The learning model further infers the carbon dioxide emissions of the blast furnace department and the energy department based on the input data of the blast furnace operating parameters, and The inferred carbon dioxide emission is output.

3. The operation support method according to claim 1 or 2, characterized in that: The comprehensive cost includes costs equivalent to carbon dioxide emissions.

4. The operation support method according to any one of claims 1 to 3, characterized in that: Blast furnace operating parameters having an impact on the energy sector include reduced material parameters and air supply parameters.

5. The operation support method according to claim 4, characterized in that: The reducing material parameters include coke ratio and fine carbon ratio. The air supply parameters include air supply humidity, air supply temperature and air supply oxygen concentration.

6. The operation support method according to claim 5, characterized in that: The reducing material parameters include the composition and carbon content of the coke.

7. The operation support method according to any one of claims 1 to 5, characterized in that: Inputting data of a specified range of blast furnace operating parameters into the learning model to infer the comprehensive costs of the blast furnace department and the energy department corresponding to each input, and A plurality of candidate operating data are output based on the inferred comprehensive cost of the energy department.

8. The operation support method according to claim 7, characterized in that: The comprehensive costs involved in the plurality of candidate operation data meet a specified benchmark.

9. The operation support method according to claim 7 or 8, characterized in that: The plurality of candidate operation data satisfy prescribed constraints.

10. An operation support device, comprising a control unit, and being an operation support device in an ironworks having a blast furnace department and an energy department, The operation support device is characterized in that The control unit inputs data of blast furnace operating parameters having an impact on the energy department into the learning model related to the ironworks, Based on the input data of the blast furnace operating parameters, the comprehensive costs of the blast furnace department and the energy department are inferred by the learning model, and Output the inferred comprehensive cost, The learning model is a model trained using actual performance data including actual performance of blast furnace operating parameters and comprehensive costs as teaching data.

11. A display device for displaying the output of an operation support device having a control unit in an ironworks having a blast furnace department and an energy department, The display device is characterized in that The display device displays the comprehensive costs of the blast furnace department and the energy department estimated by the control unit of the operation support device through the learning model involved in the ironworks based on the data of the blast furnace operation parameters that have an impact on the energy department, The learning model is a model trained using actual performance data including actual performance of blast furnace operating parameters and comprehensive costs as teaching data.

12. An operation support program executed by an information processing device in an ironworks having a blast furnace department and an energy department, The operation support program is characterized in that Make the computer execute the following processing, namely: inputting data on blast furnace operating parameters having an impact on the energy sector into a learning model related to the ironworks; Inferring the comprehensive costs of the blast furnace department and the energy department through the learning model based on the input data of the blast furnace operating parameters; and Output the inferred comprehensive cost, The learning model is a model trained using actual performance data including actual performance of blast furnace operating parameters and comprehensive costs as teaching data.