Integrated burdening method for ironmaking with configurable process flow and process prediction model

Abstract

The application provides an integrated burdening method of ironmaking with configurable process flow and process prediction model, and comprises the following steps: process prediction model setting, process flow setting, raw fuel setting and process requirement setting; a process prediction model is established according to the process prediction model setting content; a flow model is established according to the process prediction model, the process flow setting content and the raw fuel setting content; an integrated burdening model is established according to the process constraints obtained from the flow model and the process requirement setting content; and the integrated burdening model is solved, which belongs to a constraint nonlinear programming model, so that the integrated burdening result is obtained. The application provides an integrated burdening scheme including four links of model setting, flow modeling, mathematical programming modeling and mathematical programming solving, and can realize flexible setting of different process paths and adjustment of the process prediction model.

Description

Technical Field

[0001] This invention relates to the field of metallurgical ironmaking, specifically to an integrated ironmaking batching method with configurable process flow and process prediction models. Background Technology

[0002] The batching process in the ironmaking system of steel enterprises involves multiple steps, including mixing, sintering, coking, pelletizing, pulverized coal injection, and blast furnace. The raw material and fuel costs involved account for approximately 90% of the total cost, and are diverse, complex in composition, and subject to significant price fluctuations. Traditionally, the batching in the ironmaking system has been confined to within each step, achieving only locally optimal batching. Therefore, considering the interrelationships between steps from a holistic perspective of the ironmaking system and implementing integrated batching is of great significance for steel enterprises to reduce costs and increase efficiency.

[0003] Some enterprises have explored and implemented integrated ironmaking batching methods. Patent document CN108154295A proposes an optimized ore blending method for sintering-pelletizing-ironmaking linkage, establishing calculation modules for sintering, pelletizing, and blast furnace batching, setting constraints and solving the batching model to achieve simultaneous optimization of sintering, pelletizing, and blast furnace batching. Patent document CN114091871A proposes an ore blending method for sintering-blast furnace, which, through sintering batching constraints, a sinter quality prediction model, and the establishment of an optimization model linking sintering and blast furnace batching, ultimately obtains iron ore allocation and procurement plans. Patent document CN112699613B proposes a multi-objective integrated batching optimization method for sintering-pelletizing-blast furnace. It uses neural networks to establish a predictive model of the performance indicators of sinter and pellets, obtains parameter constraint intervals based on clustering methods and expert experience, transforms multiple objectives such as production increase benefits, coke saving benefits, and ore blending cost optimization benefits into single objectives, and uses a genetic algorithm to solve for the integrated optimal batching scheme of blast furnace, sintering and pelletizing.

[0004] The existing achievements generally involve three processes: first, modeling the predictive models for processes such as sintering and blast furnace; second, modeling the integrated batching mathematical programming model after the process path is determined; and third, the solution algorithm for the integrated batching model. This approach has two drawbacks: first, it cannot set different process paths; and second, the process prediction model cannot be adjusted. This makes the integrated batching model inflexible during use. When the process path changes, such as from mixing-sintering-pelletizing-blast furnace to mixing-sintering-blast furnace, the integrated batching model becomes unusable. When steel enterprises across different bases have multiple process paths to choose from, it is impossible to use the integrated batching model to select the lowest-cost path. Furthermore, when multiple equipment options exist for a certain process path, such as two coke ovens (6m and 4m) in the coking process, changes to the equipment will cause the predictive model to no longer match the actual situation, rendering the integrated batching model infeasible.

[0005] Therefore, increasing the flexibility in the process of use has become one of the technical problems that urgently need to be solved in the field of integrated ironmaking batching. Summary of the Invention

[0006] To address the shortcomings of existing technologies, the purpose of this invention is to provide an integrated ironmaking batching method with configurable process flow and procedure prediction models.

[0007] According to the present invention, an integrated ironmaking batching method with a configurable process flow and process prediction model includes:

[0008] Model setting steps: Set up the process prediction model, process flow, raw materials and fuels, and process requirements; establish the process prediction model based on the settings of the process prediction model.

[0009] Process modeling steps: Establish a process model based on the process prediction model, process flow settings, and raw material settings;

[0010] Mathematical programming modeling steps: Based on the process model and process requirements, set the process constraints to establish an integrated batching model;

[0011] The mathematical programming solution steps are as follows: Solve the integrated batching model, which belongs to the constrained nonlinear programming model, to obtain the integrated batching result.

[0012] Preferably, in the model setting step:

[0013] The process prediction model settings include the process prediction model and its empirical parameters for the performance indicators, price, and output of the materials produced in each process. There are multiple process prediction models to choose from or to configure parameters for each process.

[0014] The process flow settings include the process name, output material name, and next process name for each step involved in the integrated batching process.

[0015] The raw material and fuel settings include the material number, material name, material category name, price, and test values ​​of each performance indicator for each process.

[0016] The process requirements include the names and upper and lower limits of the raw materials and fuels used in this process, the names and upper and lower limits of the performance indicators of the mixed materials used in this process, and the names and upper and lower limits of the performance indicators of the materials produced in this process.

[0017] Preferably, in the process modeling step:

[0018] The process model is a set of quintuples, each quintuple corresponding to a process. The first position of the quintuple is the set of input materials, the second position is the process name, the third position is the process prediction model set for that process, the fourth position is the sequence of input material ratio variables, and the fifth position is the set of output materials.

[0019] Preferably, the set of input materials includes the set of input materials for this process in the raw material and fuel settings, and also includes the set of output materials for this process as the next process in the process flow settings.

[0020] The output material set is the set of output materials for this process in the process flow setting. The sequence of input material ratio variables and the information of each material in the input material set are input into the process prediction model to obtain the expressions for the performance index, price and output of each output material.

[0021] Preferably, in the mathematical programming modeling step:

[0022] The optimization objective of the integrated batching model is to minimize the cost per ton of iron. The cost per ton of iron is the ratio of the weighted sum of the prices of the input materials in the blast furnace process under the current batching variable sequence to the iron production in the output materials of the blast furnace process.

[0023] For each five-tuple in the process model, based on the names and upper and lower limits of the raw material and fuel categories set according to the process requirements, the first type of constraint expression is the sum of the material ratio variables belonging to that category in the input material set; based on the names and upper and lower limits of the performance indicators of the input mixture set set according to the process requirements, the second type of constraint expression is the weighted sum of the performance indicators of the input material set under the current ratio variable sequence; based on the names and upper and lower limits of the performance indicators of the output materials set according to the process requirements, the third type of constraint expression is the expression of that indicator in the output material set.

[0024] According to the present invention, an integrated ironmaking batching system with a configurable process flow and operation prediction model includes:

[0025] Model setting: Set up the process prediction model, process flow, raw materials and fuels, and process requirements; establish the process prediction model based on the settings of the process prediction model.

[0026] Process modeling model: Establish a process model based on the process prediction model, process flow settings, and raw material settings;

[0027] Mathematical programming modeling model: Based on the process model and process requirements, an integrated batching model is established to define the process constraints.

[0028] Mathematical programming solution model: Solve the integrated batching model. This type of model belongs to constrained nonlinear programming model and obtains the integrated batching result.

[0029] Preferably, in the model setting model:

[0030] The process prediction model settings include the process prediction model and its empirical parameters for the performance indicators, price, and output of the materials produced in each process. There are multiple process prediction models to choose from or to configure parameters for each process.

[0031] The process flow settings include the process name, output material name, and next process name for each step involved in the integrated batching process.

[0032] The raw material and fuel settings include the material number, material name, material category name, price, and test values ​​of each performance indicator for each process.

[0033] The process requirements include the names and upper and lower limits of the raw materials and fuels used in this process, the names and upper and lower limits of the performance indicators of the mixed materials used in this process, and the names and upper and lower limits of the performance indicators of the materials produced in this process.

[0034] Preferably, in the process modeling model:

[0035] The process model is a set of quintuples, each quintuple corresponding to a process. The first position of the quintuple is the set of input materials, the second position is the process name, the third position is the process prediction model set for that process, the fourth position is the sequence of input material ratio variables, and the fifth position is the set of output materials.

[0036] Preferably, the set of input materials includes the set of input materials for this process in the raw material and fuel settings, and also includes the set of output materials for this process as the next process in the process flow settings.

[0037] The output material set is the set of output materials for this process in the process flow setting. The sequence of input material ratio variables and the information of each material in the input material set are input into the process prediction model to obtain the expressions for the performance index, price and output of each output material.

[0038] Preferably, in the mathematical programming modeling model:

[0039] The optimization objective of the integrated batching model is to minimize the cost per ton of iron. The cost per ton of iron is the ratio of the weighted sum of the prices of the input materials in the blast furnace process under the current batching variable sequence to the iron production in the output materials of the blast furnace process.

[0040] For each five-tuple in the process model, based on the names and upper and lower limits of the raw material and fuel categories set according to the process requirements, the first type of constraint expression is the sum of the material ratio variables belonging to that category in the input material set; based on the names and upper and lower limits of the performance indicators of the input mixture set set according to the process requirements, the second type of constraint expression is the weighted sum of the performance indicators of the input material set under the current ratio variable sequence; based on the names and upper and lower limits of the performance indicators of the output materials set according to the process requirements, the third type of constraint expression is the expression of that indicator in the output material set.

[0041] Compared with the prior art, the present invention has the following beneficial effects:

[0042] This invention proposes an integrated batching scheme that includes four stages: model setting, process modeling, mathematical programming modeling, and mathematical programming solution. It can flexibly set different process paths and adjust the process prediction model. Attached Figure Description

[0043] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0044] Figure 1 This is a schematic diagram illustrating the principle of the present invention. Detailed Implementation

[0045] The present invention will now be described in detail with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the present invention, but do not limit the invention in any way. It should be noted that those skilled in the art can make several changes and improvements without departing from the concept of the present invention. These all fall within the protection scope of the present invention.

[0046] To address the lack of flexibility in existing integrated ironmaking batching models, this invention proposes an integrated ironmaking batching method with configurable process flow and operation prediction models. This method comprises four steps: model setting, process modeling, mathematical programming modeling, and mathematical programming solution. It allows for flexible setting of different process paths and adjustment of the operation prediction model. These steps are described below.

[0047] This invention allows for the configuration of both the process flow and the process prediction model, flexibly meeting the needs of actual ironmaking production and thus creating different process paths under different configurations. It is particularly suitable for steel enterprises to optimize material allocation and processes by comprehensively considering the relationships between various processes from a holistic perspective.

[0048] Model setting steps: This involves setting up the process modeling steps and mathematical programming modeling steps required for the integrated batching model, including four setting stages: process flow setting, process requirement setting, raw material and fuel setting, and process prediction model setting. The process flow setting includes the process name, output material name, and next process name for each process involved in the integrated batching; for example, the blending-sintering-blast furnace process flow is set as: (blending, blending ore, sintering), (sintering, sintered ore, blast furnace), (blast furnace, {molten iron, slag}). The process requirement setting includes the name and upper / lower limits of the raw material and fuel categories input in this process, the name and upper / lower limits of the performance indicators of the mixed materials input in this process, and the name and upper / lower limits of the performance indicators of the output materials in this process; for example, the name and upper / lower limits of the raw material and fuel categories input in the blending process are set as: (category name: fine ore, upper / lower limits: [30,80]), (category name: magnetic separation, upper / lower limits: [15,30]), ... The performance index names and upper and lower limits of the mixed materials input in the blending process are set as follows: (Performance index name: SiO2, upper and lower limits: [4.8, 6]), (Performance index name: TFe, upper and lower limits: [52, 62]), ... The raw material and fuel settings are used to provide the material number, material name, material category name, price, and test values ​​of each performance index of the raw materials and fuels input in each process; for example, the raw materials and fuels settings for the blending process are as follows: (Material number: k001, Material name: 77 refined, Material category: magnetic separation ore, Price: 789, TFe: 57.232, SiO2: 7.496, S: 0.117), (Material number: k002, Material name: Newman powder, Material category: fine ore, Price: 942, TFe: 62.4, SiO2: 2.35, S: 0.093), ... The process prediction model parameter settings include prediction models and their empirical parameters for the performance indicators, prices, and output of materials produced in each process. Each process should have multiple prediction models to choose from or have their parameters configured. The prediction models can be established using methods based on element balance, empirical formulas, or data-based methods, which will not be elaborated here. For example, the prediction model for the sintering process is based on the element balance method, that is, the SiO2 content of the sinter is predicted by the weighted sum of the SiO2 content of the input materials and their proportions, denoted as ModelSinter_balance.

[0049] Process modeling steps: A process model is established based on the defined process flow and raw material / fuel settings. The process model is a set of quintuples, each corresponding to a process step. The first position of each quintuple represents the set of input materials, the second position the process name, the third position the prediction model set for that process, the fourth position the sequence of input material ratio variables, and the fifth position the set of output materials. The input material set includes the input materials for that process in the raw material / fuel settings, as well as the output materials set for the next process in the process flow settings. The output material set is the set of output materials for that process in the process flow settings. By inputting the sequence of input material ratio variables and the information of each material in the input material set into the prediction model, expressions for the performance indicators, price, and output of each output material can be obtained. For example, the first position of the quintuple corresponding to the blending process is: inputBlend={(material number:k001, material name:77 fine, material category: magnetic separation, price:789, TFe:57.232, SiO2:7.496, S:0.117), (material number:k002, material name: Newman powder, material category: powdered ore, price:942, TFe:62.4, SiO2:2.35, S:0.093),…}, the second position is: blending process, the third position is: ModelSinter_balance, and the fourth position is: ratioBlend=(x1,x2,…,xn), where n is the cardinality of the input material set. The fifth position is: ModelSinter_balance(inputBlend, ratioBlend) = (Material Name: Blended Ore, Material Category: , Price: 789×x1+942×x2+,…, TFe: x1×57.232+x2×62.4+,… SiO2: x1×7.496+x2×2.35+,…, S: x1×0.117+x2×0.093+,…).

[0050] Mathematical programming modeling steps: Establish the objective function and constraint inequalities. The optimization objective of the integrated batching model is to minimize the cost per ton of iron, which is the ratio of the weighted sum of the prices of the input materials in the blast furnace process under the current ratio variable sequence to the iron production in the output materials of the blast furnace process. For the five-tuple of each process in the process model, based on the names and upper and lower limits of the raw material and fuel categories set by the process requirements, the first type of constraint expression is the sum of the ratio variables of the input materials belonging to that category; based on the names and upper and lower limits of the performance indicators of the input mixture set by the process requirements, the second type of constraint expression is the weighted sum of the performance indicators of the input materials under the current ratio variable sequence; based on the names and upper and lower limits of the performance indicators of the output materials set by the process requirements, the third type of constraint expression is the expression of that indicator in the output materials.

[0051] Mathematical programming solution steps: Solve the integrated batching model. This type of model belongs to constrained nonlinear programming model and can be solved using intelligent optimization algorithms such as numerical optimization algorithms and genetic algorithms, which will not be elaborated here.

[0052] This invention also provides an integrated ironmaking batching system with configurable process flow and process prediction models. This integrated ironmaking batching system can be implemented by executing the steps of the integrated ironmaking batching method with configurable process flow and process prediction models. That is, the integrated ironmaking batching method with configurable process flow and process prediction models can be understood as a preferred embodiment of the integrated ironmaking batching system with configurable process flow and process prediction models. According to the integrated ironmaking batching system with configurable process flow and process prediction models provided by this invention, it includes:

[0053] Model setting: Set up the process prediction model, process flow, raw materials and fuels, and process requirements; establish the process prediction model based on the settings of the process prediction model.

[0054] Process modeling model: Establish a process model based on the process prediction model, process flow settings, and raw material settings;

[0055] Mathematical programming modeling model: Based on the process model and process requirements, an integrated batching model is established to define the process constraints.

[0056] Mathematical programming solution model: Solve the integrated batching model. This type of model belongs to constrained nonlinear programming model and obtains the integrated batching result.

[0057] In the model setting model:

[0058] The process prediction model settings include the process prediction model and its empirical parameters for the performance indicators, price, and output of the materials produced in each process. There are multiple process prediction models to choose from or to configure parameters for each process.

[0059] The process flow settings include the process name, output material name, and next process name for each step involved in the integrated batching process.

[0060] The raw material and fuel settings include the material number, material name, material category name, price, and test values ​​of each performance indicator for each process.

[0061] The process requirements include the names and upper and lower limits of the raw materials and fuels used in this process, the names and upper and lower limits of the performance indicators of the mixed materials used in this process, and the names and upper and lower limits of the performance indicators of the materials produced in this process.

[0062] In the process modeling model:

[0063] The process model is a set of quintuples, each quintuple corresponding to a process. The first position of the quintuple is the set of input materials, the second position is the process name, the third position is the process prediction model set for that process, the fourth position is the sequence of input material ratio variables, and the fifth position is the set of output materials.

[0064] The input material set includes the input material set for this process in the raw material and fuel settings, as well as the output material set for this process as the next process in the process flow settings.

[0065] The output material set is the set of output materials for this process in the process flow setting. The sequence of input material ratio variables and the information of each material in the input material set are input into the process prediction model to obtain the expressions for the performance index, price and output of each output material.

[0066] In the mathematical programming model:

[0067] The optimization objective of the integrated batching model is to minimize the cost per ton of iron. The cost per ton of iron is the ratio of the weighted sum of the prices of the input materials in the blast furnace process under the current batching variable sequence to the iron production in the output materials of the blast furnace process.

[0068] For each five-tuple in the process model, based on the names and upper and lower limits of the raw material and fuel categories set according to the process requirements, the first type of constraint expression is the sum of the material ratio variables belonging to that category in the input material set; based on the names and upper and lower limits of the performance indicators of the input mixture set set according to the process requirements, the second type of constraint expression is the weighted sum of the performance indicators of the input material set under the current ratio variable sequence; based on the names and upper and lower limits of the performance indicators of the output materials set according to the process requirements, the third type of constraint expression is the expression of that indicator in the output material set.

[0069] Those skilled in the art will understand that, in addition to implementing the system, apparatus, and their modules provided by this invention in purely computer-readable program code, the same program can be implemented in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers by logically programming the method steps. Therefore, the system, apparatus, and their modules provided by this invention can be considered a hardware component, and the modules included therein for implementing various programs can also be considered structures within the hardware component; alternatively, modules for implementing various functions can be considered both software programs implementing the method and structures within the hardware component.

[0070] Specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art can make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. Unless otherwise specified, the embodiments and features described in this application can be arbitrarily combined with each other.

Claims

1. A method for integrated ironmaking batching with a configurable process flow and procedure prediction model, characterized in that, include: Model setting steps: Set up the process prediction model, process flow, raw materials and fuels, and process requirements; Establish a process prediction model based on the settings of the process prediction model; Process modeling steps: Establish a process model based on the process prediction model, process flow settings, and raw material settings; Mathematical programming modeling steps: Based on the process model and process requirements, set the process constraints to establish an integrated batching model; The steps for solving mathematical programming problems are as follows: Solve the integrated batching model, which is a constrained nonlinear programming model, to obtain the integrated batching results; In the process modeling step: The process model is a set of quintuples, each quintuple corresponding to a process. The first position of the quintuple is the set of input materials, the second position is the process name, the third position is the process prediction model set for that process, the fourth position is the sequence of input material ratio variables, and the fifth position is the set of output materials. The set of input materials includes the set of input materials for this process in the raw material and fuel settings, as well as the set of output materials for this process as the next process in the process flow settings; The output material set is the set of output materials for this process in the process flow setting. The sequence of input material ratio variables and the information of each material in the input material set are input into the process prediction model to obtain the expressions for the performance index, price and output of each output material. In the mathematical programming modeling steps described above: The optimization objective of the integrated batching model is to minimize the cost per ton of iron. The cost per ton of iron is the ratio of the weighted sum of the prices of the input materials in the blast furnace process under the current batching variable sequence to the iron production in the output materials of the blast furnace process. For each five-tuple in the process model, based on the names and upper and lower limits of the raw material and fuel categories set according to the process requirements, the first type of constraint expression is the sum of the material ratio variables belonging to that category in the input material set; based on the names and upper and lower limits of the performance indicators of the input mixture set set according to the process requirements, the second type of constraint expression is the weighted sum of the performance indicators of the input material set under the current ratio variable sequence; based on the names and upper and lower limits of the performance indicators of the output materials set according to the process requirements, the third type of constraint expression is the expression of that indicator in the output material set.

2. The integrated ironmaking batching method with configurable process flow and process prediction model according to claim 1, characterized in that, In the model setting step: The process prediction model settings include the process prediction model and its empirical parameters for the performance indicators, price, and output of the materials produced in each process. There are multiple process prediction models to choose from or to configure parameters for each process. The process flow settings include the process name, output material name, and next process name for each step involved in the integrated batching process. The raw material and fuel settings include the material number, material name, material category name, price, and test values ​​of each performance indicator for each process. The process requirements include the names and upper and lower limits of the raw materials and fuels used in this process, the names and upper and lower limits of the performance indicators of the mixed materials used in this process, and the names and upper and lower limits of the performance indicators of the materials produced in this process.

3. A configurable integrated ironmaking batching system with process flow and operation prediction models, characterized in that, include: Model setting: Set up the process prediction model, process flow, raw materials and fuels, and process requirements; Establish a process prediction model based on the settings of the process prediction model; Process modeling model: Establish a process model based on the process prediction model, process flow settings, and raw material settings; Mathematical programming modeling model: Based on the process model and process requirements, an integrated batching model is established to define the process constraints. Mathematical programming solution model: Solving the integrated batching model, this type of model belongs to constrained nonlinear programming model, and obtaining the integrated batching result; In the process modeling model: The process model is a set of quintuples, each quintuple corresponding to a process. The first position of the quintuple is the set of input materials, the second position is the process name, the third position is the process prediction model set for that process, the fourth position is the sequence of input material ratio variables, and the fifth position is the set of output materials. The set of input materials includes the set of input materials for this process in the raw material and fuel settings, as well as the set of output materials for this process as the next process in the process flow settings; The output material set is the set of output materials for this process in the process flow setting. The sequence of input material ratio variables and the information of each material in the input material set are input into the process prediction model to obtain the expressions for the performance index, price and output of each output material. In the mathematical programming model: The optimization objective of the integrated batching model is to minimize the cost per ton of iron. The cost per ton of iron is the ratio of the weighted sum of the prices of the input materials in the blast furnace process under the current batching variable sequence to the iron production in the output materials of the blast furnace process. For each five-tuple in the process model, based on the names and upper and lower limits of the raw material and fuel categories set according to the process requirements, the first type of constraint expression is the sum of the material ratio variables belonging to that category in the input material set; based on the names and upper and lower limits of the performance indicators of the input mixture set set according to the process requirements, the second type of constraint expression is the weighted sum of the performance indicators of the input material set under the current ratio variable sequence; based on the names and upper and lower limits of the performance indicators of the output materials set according to the process requirements, the third type of constraint expression is the expression of that indicator in the output material set.

4. The configurable integrated ironmaking batching system with process flow and process prediction model according to claim 3, characterized in that, In the model setting model: The process prediction model settings include the process prediction model and its empirical parameters for the performance indicators, price, and output of the materials produced in each process. There are multiple process prediction models to choose from or to configure parameters for each process. The process flow settings include the process name, output material name, and next process name for each step involved in the integrated batching process. The raw material and fuel settings include the material number, material name, material category name, price, and test values ​​of each performance indicator for each process. The process requirements include the names and upper and lower limits of the raw materials and fuels used in this process, the names and upper and lower limits of the performance indicators of the mixed materials used in this process, and the names and upper and lower limits of the performance indicators of the materials produced in this process.

Images