Intelligent distribution method for low-carbon blast furnace gas injection quantity

Abstract

The invention relates to a low-carbon blast furnace gas injection amount intelligent distribution method, and belongs to the technical field of metallurgical industry low-carbon blast furnace ironmaking process optimization, and the method comprises the following steps: S1, adopting a multi-dimensional furnace condition clustering model to divide high-quality furnace conditions; s2, dynamically calibrating the ideal point of the core index of the blast furnace; s3, calculating a standard Euclidean distance between the core index of the blast furnace and the ideal point P in the current state; s4, calculating an index weight factor by using a Bayesian optimization search algorithm; and S5, solving the multi-constraint gas injection quantity distribution model. According to the method, the blank of a low-carbon blast furnace gas injection quantity distribution method is filled, the balance relation between the furnace condition operation state of the blast furnace and the gas injection quantity is comprehensively considered, real-time dynamic optimization of the gas injection quantity under a blast furnace group collaborative production scene can be realized, and the gas injection quantity is reasonably distributed while the stable furnace condition of the blast furnace is ensured.

Description

Technical Field

[0001] This invention belongs to the technical field of low-carbon blast furnace ironmaking process optimization in the metallurgical industry, and relates to an intelligent allocation method for low-carbon blast furnace gas injection volume. Background Technology

[0002] Traditional blast furnace ironmaking technology, as the core process of steel production, has long relied on coke and pulverized coal as the main reducing agents and energy carriers. With the acceleration of global carbon neutrality, reducing the carbon footprint of blast furnace ironmaking has become a key challenge for the transformation and upgrading of the steel industry.

[0003] In the existing low-carbon blast furnace technology system, hydrogen-rich gas injection is widely recognized as the most feasible emission reduction path. This technology partially replaces traditional carbon-based reducing agents by injecting hydrogen-containing media such as coke oven gas, converter gas, natural gas, or pure hydrogen. Its emission reduction mechanism is mainly reflected in three aspects: first, the reduction reaction of hydrogen with iron ore only produces water, not carbon dioxide; second, the kinetic characteristics of hydrogen reduction reaction are better than those of carbon reduction; and third, the injected gas can optimize the blast furnace thermal regime and reduce the demand for carbon combustion.

[0004] However, in the actual scenario of multi-blast furnace collaborative production, the operation and management of the gas injection system faces numerous technical challenges. As a highly complex multiphase reactor, the blast furnace's gas injection capacity is constrained by dynamically changing furnace parameters, including fluctuations in the composition of the gas at the furnace top, migration of the softening zone, changes in permeability index, and heat load distribution. Different blast furnaces exhibit significant individual differences in their gas receiving capacity due to variations in volume, furnace age, and the degree of refractory material wear. Fluctuations in raw material conditions, such as ore grade, coke quality, and flux ratio, further affect the response characteristics of each blast furnace to the injected gas.

[0005] Currently, the gas distribution methods commonly used in steel enterprises rely primarily on the experience and judgment of operators. This manual decision-making model has significant limitations: first, it is difficult to capture changes in dynamic parameters of the blast furnace in a timely manner; second, it lacks a quantitative evaluation system, making it impossible to accurately balance the multi-objective optimization of carbon emission reduction and production efficiency; and third, it is difficult to coordinate the gas competition among multiple blast furnaces, often leading to a coexistence of gas shortage and gas surplus. More importantly, existing methods have not established mathematical models for injection volume and key process parameters, causing gas distribution to become disconnected from core process indicators such as blast furnace thermal state and reduction balance.

[0006] The direct consequence of this extensive management approach is low efficiency in the utilization of coal gas resources. Actual measurement data shows that there are excessive differences in coal gas utilization rates among different blast furnaces; the potential for carbon emission reduction has not been fully released, and the theoretical emission reduction effect is too high under the same total injection volume; it may even cause fluctuations in furnace conditions due to improper coal gas distribution, resulting in production losses or increased energy consumption. Especially when steel enterprises face complex operating conditions such as total coal gas resource constraints and multi-media coal gas mixed injection, the limitations of traditional manual distribution methods become more prominent.

[0007] From a technical perspective, the core of the problem lies in the lack of three key capabilities in existing systems: first, the ability to monitor the process status of multiple blast furnaces in real time; second, the ability to model the dynamic relationship between gas injection rate and process parameters; and third, the computational capability to make rapid optimization decisions under multi-objective and multi-constraint conditions. These deficiencies make it difficult for steel companies to translate theoretical emission reduction potential into actual emission reduction performance when implementing low-carbon blast furnace technology, becoming a significant technical bottleneck restricting the industry's green transformation. Summary of the Invention

[0008] In view of this, the purpose of the present invention is to provide a method for intelligent allocation of low-carbon blast furnace gas injection volume.

[0009] To achieve the above objectives, the present invention provides the following technical solution:

[0010] A method for intelligent allocation of low-carbon blast furnace gas injection volume includes the following steps:

[0011] S1: High-quality furnace conditions are classified using a multi-dimensional furnace condition clustering model;

[0012] S2: Dynamically calibrate the ideal point of the core indicators of the blast furnace;

[0013] S3: Calculate the standard Euclidean distance between the core indicators of the blast furnace under the current state and the ideal point P;

[0014] S4: Calculate the index weight factors using the Bayesian optimization search algorithm;

[0015] S5: Solve the multi-constraint gas injection quantity allocation model.

[0016] Furthermore, the multi-dimensional furnace condition clustering model described in step S1 for classifying high-quality furnace conditions specifically includes: selecting the economic and status indicators that the enterprise is most concerned about, using a clustering algorithm to perform cluster analysis on historical data, classifying historical furnace conditions according to the clustering results, and classifying high-quality furnace conditions.

[0017] Furthermore, the dynamic calibration of the ideal point of the core indicators of the blast furnace in step S2 specifically includes: selecting the core indicators of the blast furnace, back-calculating the distribution of the core indicators of the high-quality furnace conditions based on the high-quality furnace conditions divided in step 1, and selecting the value corresponding to the position with the maximum distribution density of the core indicators to form a multi-dimensional ideal point P.

[0018] Furthermore, in step S3, the standard Euclidean distance between the core indicators of the blast furnace under the current state and the ideal point P is calculated, and the indicators are corrected by adding a weighting factor λ:

[0019]

[0020] Where x i Indices i and y represent the current state. i Indices i and s representing the ideal point i λ represents the standard deviation of index i. i This represents the weighting factor of index i.

[0021] Furthermore, step S4, which involves using a Bayesian optimization search algorithm to calculate the indicator weight factors, specifically includes: setting the R² value of the weighted standard Euclidean distance fitted to the economic indicator as the objective function, and searching for the weight combination that maximizes the objective function.

[0022] Furthermore, the solution to the multi-constraint gas injection quantity allocation model described in step S5 specifically includes:

[0023] The gas injection rate for each blast furnace is calculated using a programming model, with the objective function being:

[0024] min∑(d i *x i )

[0025] Where d i x is the current weighted standard Euclidean distance for blast furnace i. i The amount of hydrogen-rich gas allocated to blast furnace number i;

[0026] The constraints include:

[0027] Total quantity constraint: x1+x2+…+xn=X, where X represents the total available amount of injected gas;

[0028] Safety boundary: xi_min≤xi≤xi_max, that is, setting the upper and lower limits of blast furnace gas injection volume;

[0029] Air volume constraint: Q_min≤Q-μ(xi-x)≤Q_max, that is, setting upper and lower limits for blast furnace air volume;

[0030] Where x represents the current gas injection volume of the blast furnace, and μ represents the amount of air volume reduction caused by a unit gas injection volume;

[0031] The ideal combustion temperature is: T_min≤T-η(xi-x)≤T_max, that is, setting the upper and lower limits of the ideal combustion temperature of the blast furnace;

[0032] Where x represents the current amount of gas injected into the blast furnace, and η represents the change in combustion temperature caused by a unit amount of gas injected.

[0033] The beneficial effects of this invention are as follows: The low-carbon blast furnace gas injection quantity allocation model and method provided by this invention fills the gap in the market for low-carbon blast furnace gas injection quantity allocation methods. It comprehensively considers the balance between the blast furnace operating status and the gas injection quantity, ensuring that blast furnaces with better furnace conditions are allocated more gas injection quantity, minimizing fluctuations in blast furnace conditions, and ensuring that the blast furnace still operates at a high quality while reducing carbon emissions.

[0034] Other advantages, objectives, and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination, or may be learned from practice of the invention. The objectives and other advantages of the invention can be realized and obtained through the following description. Attached Figure Description

[0035] To make the objectives, technical solutions, and advantages of the present invention clearer, the preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings, wherein:

[0036] Figure 1 This is a flowchart of the low-carbon blast furnace gas injection quantity allocation method of the present invention;

[0037] Figure 2 The following diagrams show the results of the multi-dimensional clustering model for classifying furnace conditions in Embodiment 1 of the present invention: (a) shows the clustering results for 1 furnace, (b) shows the clustering results for 2 furnaces, and (c) shows the clustering results for 3 furnaces.

[0038] Figure 3 This is a density distribution diagram of the core indicators of high-quality furnace conditions in Embodiment 1 of the present invention, where (a) is wind speed, (b) is blast kinetic energy, (c) is top pressure, (d) is oxygen enrichment, (e) is top temperature, and (f) is soft water heat load. Detailed Implementation

[0039] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0040] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0041] In the following description, numerous details are explored to provide a more thorough explanation of embodiments of the invention. However, it will be apparent to those skilled in the art that embodiments of the invention may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the invention.

[0042] Example 1

[0043] like Figure 1 As shown, the present invention provides a method for distributing low-carbon blast furnace gas injection, comprising the following steps:

[0044] Step 1: Use a multi-dimensional furnace condition clustering model to classify high-quality furnace conditions;

[0045] Step 2: Dynamic calibration of the ideal point of core blast furnace indicators;

[0046] Step 3: Calculate the weighted standard Euclidean distance;

[0047] Step 4: Bayesian optimization of search metric weight factors;

[0048] Step 5: Solve the multi-constraint gas injection quantity allocation model.

[0049] This example collects six months of production data from three blast furnaces of a steel company, removes abnormal data such as those from shutdown periods and blast furnace maintenance periods, and fills in missing values ​​for preliminary preprocessing.

[0050] Step 1: Perform two-dimensional clustering using two economic indicators, production output and fuel ratio, as clustering features. Set the cluster centers to 4. In this example, the K-Means clustering algorithm is used to cluster all furnace conditions into 4 classes based on production output and fuel ratio. The clustering results for the 3 furnaces are shown below. Figure 2 As shown in (a)-(c), based on the clustering results, the triangles in the figure represent high-quality furnace conditions, with the highest output and the lowest fuel ratio.

[0051] Step 2: Select six core indicators: wind speed, blower kinetic energy, top pressure, oxygen enrichment, top temperature, and soft water heat load. Figure 3Tables (a)-(f) show the distribution of the six core indicators for a high-quality furnace condition in Furnace 1. The maximum distribution density of each indicator is selected to form the ideal point. The ideal point status indicators for the three furnaces are shown in Table 1. To ensure the timeliness of the ideal points, the ideal point status indicators are dynamically updated using the production data of the previous six months as the time window.

[0052] Table 1

[0053] wind speed Blowering kinetic energy Top pressure oxygen content top temperature Soft water heat load 1 furnace 255.36 12863 233.40 28288 128.87 55.27 2 furnaces 266.39 13878 230.28 27438 112.23 83.14 3 furnaces 258.88 13168 227.11 28762 125.19 60.26

[0054] Step 3: Bayesian optimization finds the value that minimizes the objective function by establishing a replacement function based on past evaluation results of the objective function. In this example, the R² value fitted between the weighted standard Euclidean distance and the output is used as the objective function to search for the combination of weight factors that maximizes the R² value. The weight factors for the three furnaces are calculated as shown in Table 2.

[0055] Table 2

[0056] wind speed Blowering kinetic energy Top pressure oxygen content top temperature Soft water heat load 1 furnace 0.15 0.13 0.08 0.35 0.18 0.11 2 furnaces 0.22 0.21 0.03 0.16 0.23 0.16 3 furnaces 0.21 0.23 0.01 0.28 0.15 0.12

[0057] Step 4: Select the blast furnace status indicators for the current period. This example uses 12 hours. Assume that the current blast velocity, blast energy, top pressure, oxygen enrichment, top temperature, and soft water heat load of blast furnace 1 over the past 12 hours are 257.1, 13064, 238.27, 28362, 141.09, and 60.36, respectively. The standard deviations of these indicators over the past 6 months are 4.16, 531, 5.55, 8.86, 1544, and 10.78, respectively. Therefore, the weighted standard Euclidean distance of blast furnace 1 from the ideal point is...

[0058]

[0059] Step 5: Assume that the weighted standard Euclidean distances for boilers 1, 2, and 3 are calculated to be 0.69, 1.12, and 0.51 respectively, and the total gas volume is 18000 m³. 3 / h. Different compositions and properties of the injected gas, along with varying blast furnace volumes, lead to the injection of different hydrogen-rich gases, resulting in variations in blast furnace airflow, combustion temperature, and the amount of gas injected into the furnace belly. These variations require theoretical calculations or data analysis based on actual production conditions, depending on the specific circumstances. This example assumes a 1m³ increase in the gas injection rate. 3 / h, air volume decreases by 0.015m³ 3 / min, the combustion temperature drops by 0.005℃, and the current gas injection rate of the three furnaces is 0.

[0060] but:

[0061] The objective function is: min(0.69x1+1.12x2+0.51x3)

[0062] Constraint set:

[0063] Total constraint: x1 + x2 + ... + xn = 18000

[0064] Safety boundary constraints: 0≤xi≤9000 (In this example, the maximum injection rate per furnace is set at this time, m) 3 / h)

[0065] Airflow constraint: 5200≤Q-0.015xi≤5500 (This example sets upper and lower limits for airflow, in meters) 3 / min)

[0066] Theoretical ignition temperature: 2350≤T-0.005xi≤2450 (This example sets the upper and lower limits of the theoretical ignition temperature in °C)

[0067] Table 1 shows the results of five gas distributions using the multi-constraint pulverized coal injection distribution model.

[0068] Table 3

[0069]

[0070] Example 2:

[0071] An electronic device, comprising a memory and a processor;

[0072] The memory is used to store computer programs;

[0073] The processor is configured to implement the method described in Embodiment 1 when executing the computer program.

[0074] Example 3:

[0075] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in Embodiment 1.

[0076] Example 4:

[0077] A computer program product includes a computer program that, when executed by a processor, implements the method described in Example 1.

[0078] In the above embodiments, the reference to "this embodiment" in the specification indicates that a specific feature, structure, or characteristic described in connection with the embodiment is included in at least some embodiments, but not necessarily all embodiments. Multiple appearances of "this embodiment" do not necessarily refer to the same embodiment.

[0079] In the above embodiments, the reference to "this embodiment" in the specification indicates that a specific feature, structure, or characteristic described in connection with the embodiment is included in at least some embodiments, but not necessarily all embodiments. Multiple appearances of "this embodiment" do not necessarily refer to the same embodiment.

[0080] In the above embodiments, although the invention has been described in conjunction with specific embodiments thereof, many substitutions, modifications, and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory structures (e.g., dynamic RAM (DRAM)) may be used with the embodiments discussed. The embodiments of the invention are intended to cover all such substitutions, modifications, and variations falling within the broad scope of the appended claims.

[0081] As will be understood by those skilled in the art, the computer-readable storage medium described in this embodiment allows for the implementation of all or part of the steps in the above method embodiments by computer program-related hardware. The aforementioned computer program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0082] The electronic terminal provided in this embodiment includes a processor, a memory, a transceiver, and a communication interface. The memory and the communication interface are connected to the processor and the transceiver and complete communication between them. The memory is used to store computer programs, the communication interface is used to perform communication, and the processor and the transceiver are used to run the computer programs, so that the electronic terminal performs the steps of the above method.

[0083] In this embodiment, the memory may include random access memory (RAM) and may also include non-volatile memory, such as at least one disk storage device.

[0084] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0085] This invention can be used in a wide range of general-purpose or special-purpose computing system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices, etc.

[0086] This invention can be described in the general context of computer-executable instructions, such as program modules, that are executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This invention can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0087] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for intelligent allocation of low-carbon blast furnace gas injection volume, characterized in that: Includes the following steps: S1: High-quality furnace conditions are classified using a multi-dimensional furnace condition clustering model; S2: Dynamically calibrate the ideal point of the core indicators of the blast furnace; S3: Calculate the standard Euclidean distance between the core indicators of the blast furnace under the current state and the ideal point P; S4: Calculate the index weight factors using the Bayesian optimization search algorithm; S5: Solve the multi-constraint gas injection quantity allocation model.

2. The intelligent allocation method for low-carbon blast furnace gas injection volume according to claim 1, characterized in that: Step S1, which describes the use of a multi-dimensional furnace condition clustering model to classify high-quality furnace conditions, specifically includes: selecting the economic and status indicators that the enterprise is most concerned about, using a clustering algorithm to perform cluster analysis on historical data, classifying historical furnace conditions according to the clustering results, and classifying high-quality furnace conditions.

3. The intelligent allocation method for low-carbon blast furnace gas injection volume according to claim 1, characterized in that: Step S2, which describes the dynamic calibration of the ideal point of the core indicators of the blast furnace, specifically includes: selecting the core indicators of the blast furnace, calculating the distribution of the core indicators of the high-quality furnace conditions based on the high-quality furnace conditions divided in Step 1, and selecting the values ​​corresponding to the positions with the highest distribution density of the core indicators to form a multi-dimensional ideal point P.

4. The intelligent allocation method for low-carbon blast furnace gas injection volume according to claim 1, characterized in that: In step S3, the standard Euclidean distance between the core indicators of the blast furnace under the current state and the ideal point P is calculated, and the indicators are corrected by adding a weighting factor λ. Where x i Indices i and y represent the current state. i Indices i and s representing the ideal point i λ represents the standard deviation of index i. i This represents the weighting factor of index i.

5. The intelligent allocation method for low-carbon blast furnace gas injection volume according to claim 1, characterized in that: Step S4, which involves using a Bayesian optimization search algorithm to calculate the indicator weight factors, specifically includes: setting the R² value of the weighted standard Euclidean distance fitted to the economic indicator as the objective function, and searching for the weight combination that maximizes the objective function.

6. The intelligent allocation method for low-carbon blast furnace gas injection volume according to claim 1, characterized in that: Step S5, which involves solving the multi-constraint gas injection quantity allocation model, specifically includes: The gas injection rate for each blast furnace is calculated using a programming model, with the objective function being: min∑(d i *x i ) Where d i x is the current weighted standard Euclidean distance for blast furnace i. i The amount of hydrogen-rich gas allocated to blast furnace number i; The constraints include: Total quantity constraint: x1+x2+…+xn=X, where X represents the total available amount of injected gas; Safety boundary: xi_min≤xi≤xi_max, that is, setting the upper and lower limits of blast furnace gas injection volume; Air volume constraint: Q_min≤Q-μ(xi-x)≤Q_max, that is, setting upper and lower limits for blast furnace air volume; Where x represents the current gas injection volume of the blast furnace, and μ represents the amount of air volume reduction caused by a unit gas injection volume; The ideal combustion temperature is: T_min≤T-η(xi-x)≤T_max, that is, setting the upper and lower limits of the ideal combustion temperature of the blast furnace; Where x represents the current amount of gas injected into the blast furnace, and η represents the change in combustion temperature caused by a unit amount of gas injected.

7. An electronic device, characterized in that, Including memory and processor; The memory is used to store computer programs; The processor is configured to implement the intelligent allocation method for low-carbon blast furnace gas injection as described in any one of claims 1-6 when executing the computer program.

8. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the intelligent allocation method for low-carbon blast furnace gas injection as described in any one of claims 1-6.

9. A computer program product, characterized in that: It includes a computer program that, when executed by a processor, implements the intelligent allocation method for low-carbon blast furnace gas injection as described in any one of claims 1-6.

Images