Thermocouple arrangement method and system based on digital twin model of hearth

Abstract

The application provides a kind of hot thermocouple arrangement method and system based on furnace digital twin model, comprising: obtaining furnace digital twin model, and setting physical boundary conditions of multiple erosion types in furnace digital twin model, carrying out multi-working condition simulation to obtain high-precision temperature field distribution;Based on temperature field data, the temperature sequence of each candidate thermocouple arrangement scheme is extracted, and the inverse problem of furnace erosion is solved;By comparing the error of the inversion erosion line and the set erosion line, a multi-objective weighted comprehensive error evaluation function based on erosion probability is introduced, and finally the thermocouple arrangement is solved based on multi-attribute decision method.The application can effectively evaluate the rationality of thermocouple arrangement scheme under multiple erosion conditions by constructing furnace digital twin model and establishing erosion condition library, introducing multi-objective weighted evaluation system based on probability weight, solving the technical problems that traditional thermocouple arrangement relies on experience and is difficult to accurately capture local abnormal erosion.

Description

Technical Field

[0001] This invention relates to the field of blast furnace ironmaking technology, and in particular to a thermocouple arrangement method and system based on a digital twin model of the hearth. Background Technology

[0002] The blast furnace hearth and bottom, as the core part of the blast furnace ironmaking system, are a key limiting factor restricting the service life of the blast furnace. Currently, the main method for monitoring hearth erosion is to embed thermocouples inside refractory materials (such as refractories) and infer the erosion profile from the temperature data at the measuring points. Therefore, the rationality of the thermocouple arrangement directly determines the accuracy and timeliness of erosion monitoring, thus affecting the blast furnace's lifespan and production safety. However, existing blast furnace hearth thermocouple arrangement schemes face the following technical problems: (1) Design based on experience: The existing thermocouple layout of blast furnaces mainly relies on the engineering experience of designers and analog references, lacking quantitative and targeted optimization methods. However, designers usually refer to thermocouple configuration schemes or drawings of similar furnace types and make appropriate human adjustments. It is difficult to optimize the layout in terms of heat transfer for the specific size, masonry structure, refractory type and physical property parameters of the blast furnace. The lack of precise design of "one furnace, one policy" may lead to a mismatch between the spatial distribution of measuring points and heat transfer characteristics, resulting in low efficiency of temperature information utilization; insufficient measuring point layout in some areas, making it impossible to capture early key erosion information; and redundant measuring points in some areas, resulting in investment waste.

[0003] (2) One-sided data perception: Due to the significant three-dimensional characteristics of heat transfer within the furnace hearth, especially defects such as elephant foot erosion and rat hole-like local erosion, the disturbance characteristics of the temperature field are completely different. Since the temperature field response of different erosion types varies significantly, the requirements for thermocouple arrangement are different. Temperature data collected by the traditional uniform distribution method often cannot effectively characterize all erosion types, resulting in the furnace hearth erosion model back-inference results failing to accurately reflect or not reflecting the actual erosion situation inside the furnace. If the arrangement is optimized for a single erosion type or local area, the optimized scheme may not perform well for other erosion types. It is necessary to seek a balance between the monitoring needs of different erosion types, but there is a lack of effective multi-objective trade-off methods.

[0004] (3) Design verification is delayed: Existing technologies mostly focus on erosion diagnosis of in-service blast furnaces, and lack optimization of thermocouple arrangement schemes for the blast furnace design stage. The quality of thermocouple arrangement schemes can usually only be judged by data performance after the furnace is started. Once the furnace is built and put into operation, the position of the measuring points is difficult to adjust, resulting in long-term monitoring blind spots, which can only be supplemented in a post-event rescue when the blast furnace is shut down for major overhaul.

[0005] (4) Failure to consider the probability of erosion occurrence and unreasonable allocation of monitoring resources: The probability of occurrence of different erosion types varies significantly at different stages of furnace operation. For example, elephant foot erosion mostly occurs in the middle and late stages of furnace operation, and ring crack erosion is closely related to the start-up heating regime. Traditional methods do not take the probability of erosion occurrence into optimization considerations, which may lead to insufficient monitoring resources for high-probability erosion types, over-allocation of low-probability erosion types, and suboptimal overall monitoring efficiency throughout the furnace life cycle. Summary of the Invention

[0006] This invention provides a thermocouple arrangement method and system based on a furnace hearth digital twin model, in order to solve or alleviate the technical problems described above, such as the reliance on experience, conflicting schemes due to a single optimization objective, and failure to consider erosion probability in current thermocouple arrangement methods.

[0007] This invention provides a thermocouple arrangement method based on a furnace hearth digital twin model, comprising the following steps: acquiring a furnace hearth digital twin model and candidate thermocouple arrangement schemes; constructing an erosion simulation condition library based on the furnace hearth digital twin model, and establishing erosion probability models for various erosions at different stages of furnace service based on the erosion simulation condition library; wherein, the erosion simulation condition library includes at least one type of furnace hearth erosion condition, and the erosion probability model is used to assign furnace hearth erosion probability weights; performing numerical simulation based on the furnace hearth digital twin model, the candidate thermocouple arrangement schemes, and the erosion simulation condition library to solve for the three-dimensional structure of the furnace hearth and furnace bottom. The temperature field distribution is analyzed, and temperature data from measuring points are extracted based on the three-dimensional temperature field distribution of the furnace hearth and furnace bottom to form a virtual temperature measurement dataset. Simulated erosion lines for different erosion conditions are determined based on the virtual temperature measurement dataset, and these simulated erosion lines are compared with preset erosion lines. The comparison results are then correlated with the furnace hearth erosion probability weights to perform a comprehensive error calculation. The preset erosion lines include the erosion line contours set in the furnace hearth digital twin model by changing the physical boundaries of the refractory material at preset locations. When the calculated comprehensive error exceeds a preset threshold, the candidate thermocouple arrangement scheme is adjusted, and the adjusted thermocouple arrangement scheme is output.

[0008] In one embodiment of the present invention, the construction process of the digital twin model of the hearth includes: obtaining modeling parameters of the target blast furnace, wherein the modeling parameters include the geometric parameters of the hearth and bottom, the refractory material masonry structure, the physical property parameters of each layer of refractory material, the physical property parameters of molten iron, the physical property parameters of slag, the physical property parameters of the cooling medium, the physical property parameters of the solidified iron layer, and the configuration of the cooling system; wherein the geometric parameters of the hearth and bottom include at least one of the hearth diameter, hearth height, bottom thickness, dead iron layer depth, and thickness of each layer of refractory material, and the physical property parameters include the relationship between thermal conductivity and temperature, and the relationship between specific heat capacity and temperature. The cooling system configuration includes at least one of the following: cooling wall type, cooling water pipe arrangement, cooling water flow rate, and inlet temperature; the refractory material includes at least one of ceramic cups, ceramic pads, carbon bricks, graphite bricks, and ramming mixes; based on the geometric parameters of the hearth and furnace bottom, the refractory material masonry structure, the physical property parameters of each layer of refractory material, the physical property parameters of molten iron, the physical property parameters of slag, the physical property parameters of the cooling medium, the physical property parameters of the solidified iron layer, and the cooling system configuration, a digital twin model of the hearth is constructed using the finite element method or the finite volume method.

[0009] In one embodiment of the present invention, the process of determining simulated erosion lines for different erosion conditions based on the virtual temperature measurement dataset includes: setting boundary conditions for the simulation model according to the preset erosion lines, and calculating the model temperature field based on the simulation model boundary conditions and the thermal conductivity of the refractory material; extracting the predicted temperature value at the corresponding position of the thermocouple on the calculation section from the model temperature field, and comparing it with the temperature data at the corresponding thermocouple measurement point in the virtual temperature measurement dataset to calculate the relative temperature error; when the relative temperature error is higher than the preset temperature threshold, calculating the moving direction and step size of the erosion node using the steepest descent method with a penalty function, and controlling the moving distance of the erosion node each time it moves inward or outward, until all erosion nodes complete one movement and a new target temperature isotherm position is obtained; wherein, the background mesh is kept fixed when the erosion node moves, and only the erosion boundary position is updated; performing iterative calculation of the temperature field, and comparing the erosion degree of each position point on the target temperature isotherm position obtained after the iteration convergence with the historical erosion line, selecting the position point with the most severe erosion, and using the selected position point with the most severe erosion to update the historical erosion line to obtain the simulated erosion line.

[0010] In one embodiment of the present invention, the process of comparing the simulated erosion line with a preset erosion line and correlating the comparison result with the furnace hearth erosion probability weight to perform comprehensive error calculation includes: comparing the simulated erosion line with the preset erosion line, and determining the average profile deviation, local extreme value deviation, and heat flux density deviation based on the corresponding comparison results; constructing a comprehensive error evaluation function based on the average profile deviation, the local extreme value deviation, the heat flux density deviation, and the furnace hearth erosion probability weight, and performing comprehensive error calculation through the comprehensive error evaluation function; wherein, the comprehensive error evaluation function is: In the formula, E 综合 This is the comprehensive error evaluation function. n This represents the total number of erosion conditions in the erosion simulation condition library. l 1 represents the weighting coefficient for the average profile deviation. l 2 represents the weighting coefficient for local extremum deviation. l 3 is the weighting coefficient for the heat flux density deviation, and l 1+ l 2+ l 3=1, w i For the first i The probability weights of hearth erosion for different erosion types e i,轮廓 For the first i Average profile deviation for each type of erosion. e i,极值 For the first i Local extreme deviations for various erosion types e i,热流 For the first i Heat flux density deviation for different erosion types.

[0011] In one embodiment of the present invention, the process of obtaining the candidate thermocouple arrangement scheme includes: using a parameterization method to generate the candidate thermocouple arrangement scheme based on decision variables, wherein the decision variables include the number of circumferential cross-sections, the number of radial measuring points, the number of axial layers, and the arrangement pattern, wherein the arrangement pattern includes at least one of uniform arrangement, gradient arrangement, and asymmetric densification arrangement; wherein, the uniform arrangement is in which thermocouples are uniformly distributed circumferentially, radially, and / or axially in the furnace hearth and bottom; the gradient arrangement is based on the gradient change of the thermal conductivity of the refractory material, with denser arrangement in areas where the thermal conductivity is lower than a preset thermal conductivity, and sparse arrangement in areas where the thermal conductivity is higher than a preset thermal conductivity, and the spacing between adjacent measuring points is inversely proportional to the thermal conductivity; the asymmetric densification arrangement is to locally densify in the target area and maintain a conventional arrangement in non-target areas, wherein the target area includes below the taphole and the center of the furnace bottom.

[0012] In one embodiment of the present invention, the process of adjusting the candidate thermocouple arrangement scheme includes: adjusting the candidate thermocouple arrangement scheme according to at least one of a densification strategy, a migration strategy, and a layering strategy; wherein, the densification strategy is used to increase the number of thermocouples and shorten the thermocouple spacing according to the error contribution; the migration strategy is used to migrate the thermocouples to the target area and optimize the spatial distribution of measuring points; the layering strategy is used to add thermocouples according to the interface of thermal conductivity change and improve the ability to capture interlayer temperature gradients; the target area includes the area below the taphole and the center of the furnace bottom.

[0013] In one embodiment of the present invention, the process of adjusting the candidate thermocouple arrangement scheme further includes: solving for the Pareto optimal solution set with the minimum number of thermocouples, the minimum comprehensive error, and the most uniform error distribution as the optimization objectives; and adjusting the candidate thermocouple arrangement scheme according to the number of thermocouples, comprehensive error, and error distribution corresponding to the Pareto optimal solution set; wherein the optimization variables corresponding to the optimization objectives include thermocouple position coordinates, number of circumferential sections, number of radial measuring points, number of axial layers, arrangement density, and measuring point combination method; and / or, analyzing the heat flux density gradient distribution based on the furnace hearth digital twin model, determining the regions where the heat flux density change exceeds a preset density threshold and the regions where the three-dimensional heat transfer effect exceeds a preset heat transfer threshold based on the heat flux density gradient distribution, and densifying the thermocouple arrangement in the corresponding regions.

[0014] In one embodiment of the present invention, the erosion simulation condition library includes at least one type of hearth erosion condition among rat hole erosion, elephant foot erosion, pot bottom erosion and ring crack erosion, and each type of hearth erosion is divided into multiple levels or degrees according to the erosion depth or range.

[0015] This invention also provides a thermocouple arrangement system based on a furnace hearth digital twin model. The system includes: a furnace hearth digital twin model module for constructing a furnace hearth digital twin model; a thermocouple arrangement scheme module for generating candidate thermocouple arrangement schemes; an erosion simulation condition library module for constructing an erosion simulation condition library based on the furnace hearth digital twin model and establishing erosion probability models for various erosions at different stages of furnace service based on the erosion simulation condition library; wherein the erosion simulation condition library includes at least one type of furnace hearth erosion condition, and the erosion probability model is used to assign furnace hearth erosion probability weights; and an erosion scenario simulation module for performing numerical simulations based on the furnace hearth digital twin model, the candidate thermocouple arrangement schemes, and the erosion simulation condition library to solve for furnace hearth erosion probability. The system includes: a three-dimensional temperature field distribution at the furnace bottom; a temperature data extraction module for extracting temperature data from measuring points based on the three-dimensional temperature field distribution at the furnace bottom to form a virtual temperature measurement dataset; a simulated erosion line module for determining simulated erosion lines for different erosion conditions based on the virtual temperature measurement dataset; a scheme evaluation and optimization module for comparing the simulated erosion lines with preset erosion lines and correlating the comparison results with the furnace bottom erosion probability weights to perform comprehensive error calculation; and adjusting the candidate thermocouple arrangement scheme when the calculated comprehensive error exceeds a preset threshold. The preset erosion lines include erosion line contours set in the furnace bottom digital twin model by changing the physical boundaries of the refractory material at preset locations. A result output module is used to output the adjusted thermocouple arrangement scheme.

[0016] The present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the thermocouple arrangement method based on the furnace hearth digital twin model described in any one of the above.

[0017] The beneficial effects of this invention are as follows: This invention proposes a thermocouple arrangement method and system based on a digital twin model of the furnace hearth. By constructing a digital twin model of the furnace hearth and establishing an erosion condition database, and introducing a multi-objective weighted evaluation system based on probability weights, it can effectively evaluate the rationality of thermocouple arrangement schemes under various erosion conditions. This solves the technical problem that traditional thermocouple arrangement relies on experience and is difficult to accurately capture local abnormal erosion. It realizes scientific and quantitative decision-making for thermocouple arrangement schemes, which can improve the longevity guarantee capability and intelligent operation and maintenance level of blast furnaces, extend the service life of blast furnaces, and is applicable to the design and construction of new blast furnaces and the intelligent transformation of in-service blast furnaces, realizing an engineering and technological leap from "experience-based placement" to "scientific design".

[0018] Furthermore, this invention establishes a complete technical chain from digital twin modeling, composite erosion condition simulation, probabilistic assessment to multi-objective optimization, providing blast furnace design and operation personnel with an operable quantitative decision-making tool. This enhances the scientific rigor and standardization of thermocouple placement design, guiding thermocouple placement schemes during the design phase. Moreover, by introducing a probabilistic weighted comprehensive error evaluation function, this invention quantifies the monitoring requirements of different erosion types into comparable and weighted numerical indicators, avoiding the "one-sided approach" caused by single optimization, promoting the design of globally optimal thermocouple placement schemes, and resolving prediction conflicts among multiple erosion types. Simultaneously, this invention reduces redundant measuring points through optimization, decreasing the number of thermocouples while maintaining monitoring accuracy, thus reducing system complexity and maintenance workload. The introduction of a dynamic evolution mechanism for erosion occurrence probability ensures that the optimized scheme covers the entire furnace operation process, not just a specific moment, guaranteeing basic monitoring capabilities for all types of erosion, improving robustness, and reducing maintenance costs. Attached Figure Description

[0019] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention. It is obvious that the drawings described below are merely some embodiments of the invention, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0020] In the attached diagram: Figure 1 This is a flowchart illustrating a thermocouple arrangement method based on a furnace hearth digital twin model, as provided in one embodiment of the present invention. Figure 2 This is a flowchart illustrating a thermocouple arrangement method based on a furnace hearth digital twin model, provided in another embodiment of the present invention. Figure 3 The local temperature field distribution of the furnace hearth is obtained by simulating the furnace hearth under the rat hole erosion condition provided in one embodiment of the present invention. Figure 4 This is a schematic diagram of the process for determining simulated erosion lines provided in one embodiment of the present invention; Figure 5 This is a schematic diagram of the process for determining simulated erosion lines provided in another embodiment of the present invention; Figure 6 This is a schematic diagram of the hardware structure of a thermocouple arrangement system based on a digital twin model of the furnace hearth, provided in one embodiment of the present invention. Detailed Implementation

[0021] 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. Various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0022] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. 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.

[0023] 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.

[0024] Figure 1 A flowchart illustrating a thermocouple arrangement method based on a digital twin model of a furnace hearth is shown. Specifically, in an exemplary embodiment, as... Figure 1 As shown, this embodiment provides a thermocouple arrangement method based on a digital twin model of a furnace hearth, including the following steps: S110, obtain the digital twin model of the furnace hearth and candidate thermocouple arrangement schemes.

[0025] In some examples, the process of constructing a digital twin model of the hearth may include: obtaining modeling parameters of the target blast furnace, including hearth and bottom geometric parameters, refractory material masonry structure, physical property parameters of each layer of refractory material, physical property parameters of molten iron, physical property parameters of slag, physical property parameters of cooling medium, physical property parameters of solidified iron layer, and cooling system configuration; wherein, the hearth and bottom geometric parameters include at least one of hearth diameter, hearth height, bottom thickness, dead iron layer depth, and thickness of each layer of refractory material; the physical property parameters include the relationship between thermal conductivity and temperature, the relationship between specific heat capacity and temperature, and / or the relationship between density and temperature; the cooling system configuration includes at least one of cooling wall type, cooling water pipe arrangement, cooling water flow rate, and inlet temperature; the refractory material includes at least one of ceramic cups, ceramic pads, carbon bricks, graphite bricks, and ramming mix. Based on the geometric parameters of the hearth and furnace bottom, the refractory material masonry structure, the physical properties of each layer of refractory material, the physical properties of molten iron, the physical properties of slag, the physical properties of the cooling medium, the physical properties of the solidified iron layer, and the configuration of the cooling system, a digital twin model of the hearth is constructed using the finite element method or the finite volume method. In some examples, the finite element method or the finite volume method can be used to construct a three-dimensional heat transfer digital twin model of the hearth and furnace bottom that considers latent heat of solidification and unsteady heat transfer.

[0026] In some examples, the process of obtaining candidate thermocouple arrangement schemes may include: using a parametric method to generate candidate thermocouple arrangement schemes based on decision variables, including the number of circumferential profiles, the number of radial measuring points, the number of axial layers, and the arrangement pattern. The arrangement pattern includes at least one of uniform arrangement, gradient arrangement, and asymmetric dense arrangement. The uniform arrangement involves thermocouples being evenly distributed circumferentially, radially, and / or axially in the furnace hearth and bottom. The gradient arrangement, based on the gradient change in the thermal conductivity of the refractory material, involves denser arrangement in areas with thermal conductivity lower than a preset thermal conductivity and sparser arrangement in areas with thermal conductivity higher than a preset thermal conductivity, with the spacing between adjacent measuring points inversely proportional to the thermal conductivity. The asymmetric dense arrangement involves localized dense arrangement in the target area and maintaining a conventional arrangement in non-target areas. The target area includes the area below the taphole and the center of the furnace bottom. In some examples, the target area may also include sidewall corners, the main scouring zone of the circulating flow opposite the taphole, historical abnormal temperature rises, and areas of heat flux peak evolution, where abnormal temperature rises refer to temperature increases exceeding a set temperature range.

[0027] S120, an erosion simulation condition library is constructed based on the digital twin model of the furnace hearth, and erosion occurrence probability models for various erosion types at different stages of furnace service are established according to the erosion simulation condition library. The erosion simulation condition library includes at least one type of furnace hearth erosion condition, and the erosion occurrence probability models are used to assign probabilistic weights to the furnace hearth erosion. In some examples, the erosion simulation condition library includes at least one type of furnace hearth erosion condition among rat-hole erosion, elephant-foot erosion, pot-bottom erosion, and ring-crack erosion. Each type of furnace hearth erosion is divided into multiple levels or degrees according to erosion depth or range. For example, each type of furnace hearth erosion is divided into three levels or degrees—light, moderate, and severe—according to erosion depth or range, forming a set of erosion scenarios.

[0028] S130. Numerical simulation is performed based on the digital twin model of the furnace hearth, candidate thermocouple arrangement schemes, and erosion simulation case library to solve the three-dimensional temperature field distribution of the furnace hearth and furnace bottom. Based on the three-dimensional temperature field distribution of the furnace hearth and furnace bottom, temperature data of the measuring points are extracted to form a virtual temperature measurement dataset.

[0029] S140, based on the virtual temperature measurement dataset, simulated erosion lines for different erosion conditions are determined, and the simulated erosion lines are compared with preset erosion lines. The comparison results are then correlated with the furnace hearth erosion probability weights to perform comprehensive error calculation. The preset erosion lines include erosion line contours set in the furnace hearth digital twin model by changing the physical boundaries of the refractory material at preset locations.

[0030] In some examples, the process of determining simulated erosion lines for different erosion conditions based on virtual temperature measurement datasets may include: setting boundary conditions for the simulation model according to preset erosion lines, and calculating the model temperature field based on the simulation model boundary conditions and the thermal conductivity of the refractory material; extracting the predicted temperature values ​​at the corresponding thermocouple locations on the calculation cross-section from the model temperature field, and comparing them with the temperature data at the corresponding thermocouple measurement points in the virtual temperature measurement dataset to calculate the relative temperature error; when the relative temperature error is higher than a preset temperature threshold, the simulated erosion line is solved using the inverse problem of heat transfer, calculating the direction and step size of erosion node movement using the steepest descent method with a penalty function, and controlling the movement distance of the erosion nodes each time they move inward or outward, until all erosion nodes have completed one movement and a new target temperature isotherm position is obtained; wherein, the inversion process uses background mesh technology, keeping the background mesh fixed when the erosion nodes move, and only updating the erosion boundary positions to avoid computational redundancy caused by mesh reconstruction. The temperature field is iteratively calculated, and the erosion degree of each point on the target temperature isotherm obtained after the iteration converges is compared with that of the historical erosion line. The point with the most severe erosion is selected, and the historical erosion line is updated using the selected point with the most severe erosion to obtain the simulated erosion line. The target temperature can be selected or set according to the actual situation, and there is no specific numerical limit here. For example, it can be 1150℃.

[0031] In some examples, the process of comparing simulated erosion lines with preset erosion lines and correlating the comparison results with the hearth erosion probability weights to calculate the comprehensive error may include: comparing simulated erosion lines with preset erosion lines, and determining the average profile deviation, local extreme value deviation, and heat flux density deviation based on the corresponding comparison results; constructing a comprehensive error evaluation function based on the average profile deviation, local extreme value deviation, heat flux density deviation, and hearth erosion probability weights, and calculating the comprehensive error using the comprehensive error evaluation function; wherein, the comprehensive error evaluation function is: In the formula, E 综合 This is the comprehensive error evaluation function. n This represents the total number of erosion scenarios in the erosion simulation scenario library. l 1 represents the weighting coefficient for the average profile deviation. l 2 represents the weighting coefficient for local extremum deviation. l 3 is the weighting coefficient for the heat flux density deviation, and l 1+ l 2+ l 3=1, w i For the first i The probability weights of hearth erosion for different erosion types e i,轮廓 For the first i Average profile deviation for each type of erosion. e i,极值 For the first i Local extreme deviations for various erosion types e i,热流 For the first i Heat flux density deviation for different erosion types.

[0032] S150: When the calculated comprehensive error exceeds the preset threshold, adjust the candidate thermocouple arrangement scheme and output the adjusted thermocouple arrangement scheme.

[0033] In some examples, the process of adjusting the candidate thermocouple arrangement scheme may include: adjusting the candidate thermocouple arrangement scheme according to at least one of a densification strategy, a migration strategy, and a layering strategy; wherein, the densification strategy is used to increase the number of thermocouples and shorten the thermocouple spacing according to the error contribution; the migration strategy is used to migrate thermocouples to the target area and optimize the spatial distribution of measuring points; the layering strategy is used to add thermocouples according to the interface of thermal conductivity change and improve the ability to capture interlayer temperature gradients; the target area includes the area below the taphole and the center of the furnace bottom.

[0034] In some examples, the process of adjusting the candidate thermocouple arrangement scheme may also include: solving for the Pareto optimal solution set with the optimization objectives of minimizing the number of thermocouples, minimizing the overall error, and achieving the most uniform error distribution; and adjusting the candidate thermocouple arrangement scheme according to the number of thermocouples, overall error, and error distribution corresponding to the Pareto optimal solution set; wherein the optimization variables corresponding to the optimization objectives include thermocouple position coordinates, number of circumferential profiles, number of radial measuring points, number of axial layers, arrangement density, and measuring point combination method; and / or, analyzing the heat flux density gradient distribution based on the furnace cylinder digital twin model, determining regions where the heat flux density change exceeds a preset density threshold (i.e., regions with drastic heat flux density changes) and regions where the three-dimensional heat transfer effect exceeds a preset heat transfer threshold (i.e., regions with significant three-dimensional heat transfer effect) based on the heat flux density gradient distribution, and densifying the thermocouple arrangement in the corresponding regions.

[0035] like Figure 2 As shown, in another exemplary embodiment of the present invention, a thermocouple arrangement method based on a furnace hearth digital twin model is provided, comprising the following steps: S1: Obtain the geometric parameters of the hearth and bottom of the target blast furnace, the refractory material masonry structure, the physical properties of each layer of refractory material, the physical properties of molten iron, the physical properties of slag, the physical properties of the cooling medium, the physical properties of the solidified iron layer, and the configuration of the cooling system. Using the finite element method or finite volume method, construct a three-dimensional heat transfer digital twin model of the hearth and bottom, considering latent heat of solidification and unsteady heat transfer. The geometric parameters of the hearth and bottom include the hearth diameter, height, bottom thickness, dead iron layer depth, and the thickness of each refractory layer; the physical properties include the relationship between thermal conductivity and temperature, the relationship between specific heat capacity and temperature, and / or the relationship between density and temperature; the cooling system configuration includes the type of cooling wall, the arrangement of cooling water pipes, the cooling water flow rate, and the inlet temperature; the refractory material includes at least one of ceramic cups, ceramic pads, carbon bricks, graphite bricks, and ramming mixes.

[0036] S2: Based on the blast furnace erosion mechanism, the erosion line contour is set in the digital twin model of the hearth by changing the physical boundary of the refractory material in specific parts. Different erosion conditions are preset to construct a simulation condition library covering different types and degrees of erosion. The erosion types include: rat-hole erosion, elephant-foot erosion, pot-bottom erosion, and ring-crack erosion. Each erosion type is divided into three levels—light, moderate, and severe—based on erosion depth or range, forming a set of erosion scenarios. Therefore, in some examples, the erosion condition library can contain at least 12 erosion scenarios (4 types × 3 degrees), and can be used individually or in combination to form composite erosion conditions. In some examples, the range of light erosion is set to 10%–20% of the refractory material thickness, moderate erosion to 20%–40% of the refractory material thickness, and severe erosion to 40%–60% of the refractory material thickness. In some examples, rat-hole erosion is simulated by setting a specified physical boundary of carbon bricks inward in a local area of ​​the furnace hearth sidewall below the taphole in the digital twin model of the furnace hearth, thus simulating the localized point-like deep erosion caused by rat-hole erosion; elephant-foot erosion is simulated by setting an erosion boundary with a certain angle and depth transition in the corner area of ​​the furnace bottom in the digital twin model of the furnace hearth, thus simulating the sudden temperature change in the corner area caused by elephant-foot erosion; pot bottom erosion is simulated by setting a concave curved boundary in the central area of ​​the furnace bottom in the digital twin model of the furnace hearth, thus simulating the downward movement of the isothermal line in the central area caused by pot bottom erosion; and ring crack erosion is simulated by setting an annular air gap or thinning layer at a specified elevation on the sidewall of the furnace hearth in the digital twin model of the furnace hearth, thus simulating the temperature step change on both sides of the crack caused by ring crack erosion.

[0037] S3: Combining historical data and damage survey data of similar blast furnaces, erosion statistics, and expert experience, establish probability models for various erosion types at different stages of furnace operation to assign probability weights to each erosion condition. In some examples, the probability model can consider factors such as furnace age, operating conditions, and raw material conditions to determine the probability weights of each erosion type, with the probability weights dynamically adjusted with furnace age. By introducing a dynamic evolution mechanism for erosion occurrence probability, the probability weights of each erosion type change dynamically with the increase of furnace life. Specifically, the dynamic evolution mechanism can be implemented by dividing the furnace life into three stages: early (0-3 years), middle (3-7 years), and late (over 7 years). Within each stage, the probability weights of elephant-foot erosion, rat-hole erosion, bottom erosion, and ring crack erosion are assigned according to the cumulative distribution function of statistical data on damage of similar blast furnaces, ensuring that the sum of the probability weights of all erosion types within each stage is 1. As an example, in the blast furnace design stage, considering that elephant-foot erosion is a common erosion type in the middle and late stages of furnace operation, a probability weight for elephant-foot erosion is set. w 象脚The value is 0.35~0.45; considering that the probability of bottom erosion increases in the later stages of furnace operation, a probability weight for bottom erosion is set. w 锅底 The value is set to 0.25~0.35; considering the suddenness and low probability of rat burrow erosion, a probability weight for rat burrow erosion is set. w 鼠洞 The value is 0.15~0.25; considering that ring crack erosion is related to refractory material quality and heating operation regime, in the blast furnace design stage, it is assumed that the refractory material quality meets the requirements, and a ring crack erosion probability weight is set. w 环裂 The values ​​are 0.10 to 0.15. It should be noted that the above probability weights are dynamically adjusted with the furnace age. In the early stages of furnace service (less than 3 years), the weight of bottom erosion is reduced, while the weight of annular crack erosion is increased; in the later stages of furnace service (greater than or equal to 7 years), the weights of elephant foot erosion and bottom erosion are increased. Therefore, the error term corresponding to high-probability erosion types has a larger weight in the objective function, ensuring that optimization resources are tilted towards high-risk scenarios.

[0038] S4: Based on the structural characteristics of the furnace hearth and bottom and the principles of heat transfer, multiple candidate thermocouple arrangement schemes are generated. For various erosion conditions and candidate arrangement schemes, numerical simulations are performed using a digital twin model of the furnace hearth to solve the three-dimensional temperature field distribution of the furnace hearth and bottom, extracting the temperature data of the corresponding measuring points to form a virtual temperature measurement dataset. In some examples, candidate thermocouple arrangement schemes can be generated using a parametric method. The corresponding decision variables include the number of circumferential profiles, the number of radial measuring points, the number of axial layers, and the arrangement pattern. The arrangement patterns include uniform arrangement, gradient arrangement, and asymmetric densification arrangement. Each arrangement scheme includes a set of arrangement angles for each thermocouple in the circumferential direction, a set of arrangement depths in the radial direction, and a set of arrangement heights in the axial direction, generating virtual measuring point coordinates for each arrangement scheme. Among them, the uniform arrangement involves thermocouples being evenly distributed circumferentially, radially, and / or axially in the hearth and bottom of the furnace; the gradient arrangement, based on the gradient change in the thermal conductivity of the refractory material, involves denser arrangement in areas with low thermal conductivity and sparser arrangement in areas with high thermal conductivity; the asymmetric dense arrangement involves localized denser arrangement in erosion-sensitive areas (such as the area below the taphole, the center of the furnace bottom, etc.), while maintaining a conventional arrangement in other areas. For each erosion condition and each candidate arrangement scheme, numerical simulation can be performed using a digital twin model of the hearth to solve the three-dimensional temperature field distribution of the hearth and bottom of the furnace. The coordinates of virtual measuring points for each arrangement scheme are mapped to the temperature field corresponding to each erosion condition, and the temperature data of the corresponding measuring points are extracted to form a virtual temperature measurement dataset for different conditions under that arrangement scheme. Figure 3 As shown, this is a simulation of the local temperature field distribution of the furnace hearth under a rat-hole-like erosion condition with an erosion depth of 600 mm, using a digital twin model.

[0039] S5: For each thermocouple arrangement scheme and its corresponding virtual temperature measurement data under various operating conditions, inverse heat transfer problems are used to calculate and reverse-engineer the simulated erosion lines under each operating condition. In some examples, the steepest descent method with a penalty function can be used to calculate the direction and step size of the erosion node movement. To avoid instability in the calculation process caused by excessively large step sizes, the movement distance of the erosion node each time it moves inward or outward can be controlled. The furnace hearth digital twin model uses background mesh technology, keeping the background mesh fixed when the erosion node moves, only updating the erosion boundary position, avoiding redundant mesh reconstruction operations, and improving calculation efficiency while ensuring calculation accuracy. The reverse-engineering process of the simulated erosion line can be as follows: Figure 4 and Figure 5 shown. Specifically, S51: Set the boundary conditions of the simulation model according to the preset erosion line, and solve the heat transfer problem according to the boundary conditions of the simulation model and the thermal conductivity of the refractory material to calculate the temperature field of the model. S52: Capture the predicted temperature value at the corresponding position of the thermocouple on the calculation section, compare it with the corresponding temperature data in the virtual temperature measurement dataset, and calculate the relative error; S53: When the relative error is higher than the set threshold, control the erosion nodes to move inward or outward. After all erosion nodes have completed one movement, they will obtain a new 1150℃ isotherm position. This process is for solving the inverse heat transfer problem. S54: Perform temperature field calculations again, iterating repeatedly until the iteration exit condition is met, and obtain the final 1150℃ isotherm position. Simultaneously, compare the degree of erosion at each point on the 1150℃ isotherm with the historical erosion lines, select the points with the most severe erosion, and update the erosion lines. S55: Outputs information such as the simulated isotherm distribution inside the furnace hearth, the simulated 1150℃ isotherm, and the simulated erosion lines.

[0040] S6: Compare the simulated erosion line with the pre-set erosion line. For each working condition, calculate the average profile deviation, local extreme value deviation, and heat flux density deviation. Construct a probability-weighted comprehensive error evaluation function and calculate the comprehensive monitoring index for each candidate layout scheme. The comprehensive error evaluation function is as follows: In the formula, E 综合 This is the comprehensive error evaluation function. n This represents the total number of erosion scenarios in the erosion simulation scenario library. l 1 represents the weighting coefficient for the average profile deviation. l 2 represents the weighting coefficient for local extremum deviation. l 3 is the weighting coefficient for the heat flux density deviation, and l 1+ l 2+ l 3=1, wi For the first i The probability weights of hearth erosion for different erosion types e i,轮廓 For the first i Average profile deviation for each type of erosion. e i,极值 For the first i Local extreme deviations for various erosion types e i,热流 For the first i The heat flux density deviation varies depending on the type of erosion. In some examples, the average profile deviation is calculated by uniformly selecting along the erosion line. M (Using positive integers) feature points, calculate the root mean square of the radial distance between the simulated erosion line and the set erosion line at each feature point; the local extreme value deviation is calculated by identifying the location and region of the maximum deviation between the set erosion line and the simulated erosion line, and calculating the distance deviation between the simulated erosion line and the set erosion line within that region; the heat flux density deviation is calculated by selecting [a specific feature point] on the cooling wall surface. K Using (positive integer) monitoring points, the relative error between the heat flux density obtained from the simulated erosion line inversion and the heat flux density calculated based on the set erosion line is calculated. Therefore, by introducing a probability-weighted comprehensive error evaluation function, the optimization process can not pursue the ultimate accuracy under a single operating condition, but rather maximize the overall monitoring efficiency based on the probability of different erosion risks throughout the entire furnace life cycle. This resolves potential conflicts in the optimal arrangement scheme under different erosion types and improves the engineering robustness of thermocouple placement.

[0041] S7: Determine if the overall error is less than a preset threshold. If not, adjust the thermocouple arrangement and repeat steps S4 to S7 until the requirements are met. Specifically, the overall error can be determined. E 综合 Is it less than the preset threshold? e ,like E 综合 ≤ e If so, it is considered that the current thermocouple arrangement scheme can meet the monitoring needs of various erosions with the expected probability-weighted accuracy; if E 综合 > eBased on the error and heat flux density gradient sensitivity distribution under various operating conditions, the thermocouple arrangement scheme is automatically or manually adjusted through optimization strategies to generate a new thermocouple deployment scheme. Steps S4-S7 are repeated until the error requirements are met. In some examples, optimization strategies include: densification strategy, migration strategy, and layering strategy. Among them, the densification strategy increases the number of thermocouples and shortens the thermocouple spacing in areas with large errors, and prioritizes the placement of new measuring points at nodes where the temperature gradient direction changes drastically. The migration strategy moves thermocouples to erosion-sensitive areas, optimizing the spatial distribution of measuring points, ensuring that the distance between the migrated measuring point and the original position does not exceed 50% of the original spacing. The layering strategy adds thermocouples at interfaces with abrupt changes in thermal conductivity to improve the ability to capture temperature gradients between layers. Two to four thermocouples are added at each interface with abrupt changes in thermal conductivity, and they are placed at different radial depths on both sides of the interface. In other examples, as an improvement, the thermocouple arrangement is adjusted according to the gradient densification principle: based on the heat flux density gradient distribution analyzed from the furnace hearth digital twin model, the thermocouple arrangement is densified in areas with drastic heat flux density changes and significant heat transfer effects (such as the center of the furnace bottom, below the taphole, and other target areas); in brick layers with low thermal conductivity, the spacing between thermocouples should be smaller than in brick layers with high thermal conductivity. The optimization algorithm aims to minimize the number of thermocouples, minimize the overall error, and achieve the most uniform error distribution. A multi-objective genetic algorithm (Non-dominated Sorting Genetic Algorithm II, NSGA-II) is used to solve for the Pareto optimal solution set. Optimization variables include thermocouple position coordinates, number of circumferential profiles, number of radial measuring points, number of axial layers, arrangement density, and measuring point combination method.

[0042] S8: Combining constraints and investment budget, a multi-attribute decision-making method is used to select the final implementation scheme, outputting the optimized thermocouple layout scheme. Specifically, based on the Pareto front solution set, and combining structural and process constraints and investment budget, the TOPSIS (Technique for Order Preference by Similarity to an IdealSolution) multi-attribute decision-making method can be used to select the final implementation scheme; the optimized thermocouple layout drawings, measuring point numbers, and installation process requirements are output. In some examples, structural constraints include that thermocouples must not be placed in cooling water pipes, steel structure frames, or iron tapping and slag tapping channels; process constraints include that the spacing between thermocouples in the same cross-section should not be less than 100mm to avoid drilling stress concentration; spare thermocouples should be set at key measuring points, and maintainability requirements should be met for measuring points in important areas; investment budget cost constraints include that the total number of thermocouples should not exceed the budget limit, and the comprehensive cost of a single thermocouple includes installation, commissioning, and maintenance costs.

[0043] In summary, this invention proposes a thermocouple arrangement method based on a digital twin model of the furnace hearth. A three-dimensional heat transfer digital twin model of the furnace hearth and furnace bottom is constructed based on the actual blast furnace geometric parameters, refractory material properties, and operating conditions. Multiple physical boundary conditions for typical erosion types are set in the digital twin model, and multi-condition simulations are conducted to obtain a high-precision temperature field distribution. Temperature sequences at measurement points for each candidate thermocouple arrangement scheme are extracted based on the temperature field data, and the inverse problem of furnace hearth erosion is solved. By comparing the error between the inverted erosion line and the set erosion line, a multi-objective weighted comprehensive error evaluation function based on the probability of erosion occurrence is introduced, and the thermocouple arrangement is iteratively solved based on a multi-attribute decision-making method. Therefore, this method, by constructing a digital twin model of the furnace hearth and establishing an erosion condition library, and introducing a multi-objective weighted evaluation system based on probability weights, can effectively assess the rationality of thermocouple placement schemes under various erosion conditions. It solves the technical problem of traditional thermocouple placement relying on experience and being difficult to accurately capture local abnormal erosion, achieving scientific and quantitative decision-making for thermocouple placement schemes. This can improve the longevity guarantee capability and intelligent operation and maintenance level of blast furnaces, extending their service life. It is applicable to the design and construction of new blast furnaces and the intelligent transformation of in-service blast furnaces, realizing an engineering technology leap from "experience-based placement" to "scientific design." Furthermore, by establishing a complete technical chain from digital twin modeling, composite erosion condition simulation, probabilistic assessment to multi-objective optimization, this method provides blast furnace design and operation personnel with an operable quantitative decision-making tool, improving the scientificity and standardization of thermocouple placement design and providing guidance for thermocouple placement schemes during the design phase. Furthermore, this method introduces a probability-weighted comprehensive error evaluation function to quantify the monitoring requirements of different erosion types into comparable and weighted numerical indicators, avoiding the "one-sided approach" caused by single optimization, and promoting the design of a globally optimal thermocouple arrangement scheme to resolve the prediction conflict problem of multiple erosion types. Simultaneously, this method reduces redundant measuring points through optimization, thereby reducing the number of thermocouples to a certain extent while ensuring monitoring accuracy, thus lowering system complexity and maintenance workload. The introduction of a dynamic evolution mechanism for the probability of erosion occurrence ensures that the optimized scheme is not only targeted at a specific moment but covers the entire furnace operation process, guaranteeing that the optimized scheme has basic monitoring capabilities for all types of erosion, improving robustness and reducing maintenance costs.

[0044] In an exemplary embodiment of the present invention, such as Figure 6 As shown, a thermocouple arrangement system based on a digital twin model of a furnace hearth is provided, comprising: The furnace hearth digital twin model module is used to build a digital twin model of the furnace hearth. Thermocouple arrangement scheme module, used to generate candidate thermocouple arrangement schemes; The erosion simulation condition library module is used to construct an erosion simulation condition library based on the digital twin model of the hearth, and to establish erosion probability models for various types of erosion at different stages of furnace service based on the erosion simulation condition library; wherein, the erosion simulation condition library includes at least one type of hearth erosion condition, and the erosion probability model is used to assign hearth erosion probability weights. The erosion scenario simulation module is used to perform numerical simulations based on the furnace hearth digital twin model, candidate thermocouple arrangement schemes, and erosion simulation case library to solve the three-dimensional temperature field distribution of the furnace hearth and furnace bottom. The temperature data extraction module is used to extract temperature data from measuring points based on the three-dimensional temperature field distribution of the furnace hearth and furnace bottom, forming a virtual temperature measurement dataset. The simulated erosion line module is used to determine simulated erosion lines for different erosion conditions based on virtual temperature measurement datasets. The scheme evaluation and optimization module is used to compare the simulated erosion line with the preset erosion line, and associate the comparison result with the furnace hearth erosion probability weight to perform comprehensive error calculation; and, when the calculated comprehensive error is greater than the preset threshold, to adjust the candidate thermocouple arrangement scheme; wherein, the preset erosion line includes the erosion line contour set in the furnace hearth digital twin model by changing the physical boundary of the refractory material of the preset part. The result output module is used to output the adjusted thermocouple arrangement scheme.

[0045] It is understood that the thermocouple arrangement system based on the furnace hearth digital twin model provided in the above embodiments and the thermocouple arrangement method based on the furnace hearth digital twin model provided in the above embodiments belong to the same concept. The specific operation of the thermocouple arrangement method based on the furnace hearth digital twin model has been described in detail in the above method embodiments and will not be repeated here. In practical applications, the thermocouple arrangement system based on the furnace hearth digital twin model provided in the above embodiments can allocate the above functions to different functional modules as needed. That is, the internal structure of the thermocouple arrangement system based on the furnace hearth digital twin model is divided into different functional modules, and then all or part of the functions of the corresponding functional modules are implemented through the thermocouple arrangement method based on the furnace hearth digital twin model described in the above embodiments. For example, the relevant execution process of step S110 can be used to implement all or part of the functions of the furnace hearth digital twin model module and the thermocouple arrangement scheme module; the relevant execution process of step S120 can be used to implement all or part of the functions of the erosion simulation working condition library module; the relevant execution process of step S130 can be used to implement all or part of the functions of the erosion scene simulation module and the temperature data extraction module; the relevant execution process of step S140 can be used to implement all or part of the functions of the simulated erosion line module and the scheme evaluation and optimization module; and the relevant execution process of step S150 can be used to implement all the functions of the result output module. For specific implementation processes, please refer to the above embodiments, and no specific limitations are imposed here.

[0046] Therefore, this invention proposes a thermocouple arrangement system based on a digital twin model of the furnace hearth. A three-dimensional heat transfer digital twin model of the furnace hearth and furnace bottom is constructed based on the actual blast furnace geometric parameters, refractory material properties, and operating conditions. Multiple physical boundary conditions for typical erosion types are set in the digital twin model of the furnace hearth, and multi-condition simulations are conducted to obtain a high-precision temperature field distribution. Based on the temperature field data, the temperature sequences of the measuring points for each candidate thermocouple arrangement scheme are extracted, and the inverse problem of furnace hearth erosion is solved. By comparing the error between the inverted erosion line and the set erosion line, a multi-objective weighted comprehensive error evaluation function based on the probability of erosion occurrence is introduced, and the thermocouple arrangement is iteratively solved based on a multi-attribute decision-making method. Therefore, this system, by constructing a digital twin model of the furnace hearth and establishing an erosion condition database, and introducing a multi-objective weighted evaluation system based on probability weights, can effectively assess the rationality of thermocouple placement schemes under various erosion conditions. It solves the technical problem of traditional thermocouple placement relying on experience and being difficult to accurately capture local abnormal erosion, achieving scientific and quantitative decision-making for thermocouple placement schemes. This can improve the longevity guarantee capability and intelligent operation and maintenance level of blast furnaces, extending their service life. It is applicable to the design and construction of new blast furnaces and the intelligent transformation of in-service blast furnaces, realizing an engineering technology leap from "experience-based placement" to "scientific design." Furthermore, by establishing a complete technical chain from digital twin modeling, composite erosion condition simulation, probabilistic assessment to multi-objective optimization, this system provides blast furnace design and operation personnel with an operable quantitative decision-making tool, improving the scientificity and standardization of thermocouple placement design and providing guidance for thermocouple placement schemes during the design phase. Furthermore, this system introduces a probability-weighted comprehensive error evaluation function to quantify the monitoring requirements of different erosion types into comparable and weighted numerical indicators, avoiding the "one-sided approach" caused by single optimization, promoting the design of a globally optimal thermocouple arrangement scheme, and resolving the prediction conflict problem of multiple erosion types. Simultaneously, this system reduces redundant measuring points through optimization, decreasing the number of thermocouples while ensuring monitoring accuracy, thereby reducing system complexity and maintenance workload. The introduction of a dynamic evolution mechanism for erosion occurrence probability ensures that the optimized scheme is not only targeted at a specific moment but covers the entire furnace operation process, guaranteeing that the optimized scheme has basic monitoring capabilities for all types of erosion, improving robustness, and reducing maintenance costs.

[0047] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in the present invention should still be covered by the claims of the present invention.

Claims

1. A thermocouple arrangement method based on a digital twin model of a furnace hearth, characterized in that, The method includes: Obtain the digital twin model of the furnace hearth and candidate thermocouple arrangement schemes; Based on the digital twin model of the hearth, an erosion simulation condition library is constructed, and based on the erosion simulation condition library, erosion occurrence probability models for various types of erosion at different stages of furnace service are established; wherein, the erosion simulation condition library includes at least one type of hearth erosion condition, and the erosion occurrence probability model is used to allocate hearth erosion probability weights. Numerical simulation is performed based on the digital twin model of the furnace hearth, the candidate thermocouple arrangement scheme, and the erosion simulation condition library to solve the three-dimensional temperature field distribution of the furnace hearth and furnace bottom. Based on the three-dimensional temperature field distribution of the furnace hearth and furnace bottom, temperature data of the measuring points are extracted to form a virtual temperature measurement dataset. Based on the virtual temperature measurement dataset, simulated erosion lines for different erosion conditions are determined, and the simulated erosion lines are compared with preset erosion lines. The comparison results are then correlated with the furnace hearth erosion probability weights to perform comprehensive error calculations. The preset erosion lines include the erosion line contours set in the furnace hearth digital twin model by changing the physical boundaries of the refractory material in preset locations. When the calculated overall error exceeds a preset threshold, the candidate thermocouple arrangement scheme is adjusted, and the adjusted thermocouple arrangement scheme is output.

2. The thermocouple arrangement method based on a digital twin model of a furnace hearth according to claim 1, characterized in that, The process of constructing the digital twin model of the furnace hearth includes: The modeling parameters for the target blast furnace are obtained, including the geometric parameters of the hearth and bottom, the refractory material masonry structure, the physical properties of each refractory layer, the physical properties of molten iron, the physical properties of slag, the physical properties of the cooling medium, the physical properties of the solidified iron layer, and the configuration of the cooling system. The geometric parameters of the hearth and bottom include at least one of the following: hearth diameter, hearth height, bottom thickness, dead iron layer depth, and the thickness of each refractory layer. The physical properties include the relationship between thermal conductivity and temperature, the relationship between specific heat capacity and temperature, and / or the relationship between density and temperature. The configuration of the cooling system includes at least one of the following: cooling wall type, cooling water pipe arrangement, cooling water flow rate, and inlet temperature. The refractory materials include at least one of the following: ceramic cups, ceramic pads, carbon bricks, graphite bricks, and ramming mixes. Based on the geometric parameters of the hearth and furnace bottom, the refractory material masonry structure, the physical properties of each layer of refractory material, the physical properties of molten iron, the physical properties of slag, the physical properties of the cooling medium, the physical properties of the solidified iron layer, and the configuration of the cooling system, a digital twin model of the hearth is constructed using the finite element method or the finite volume method.

3. The thermocouple arrangement method based on a digital twin model of a furnace hearth according to claim 1, characterized in that, The process of determining simulated erosion lines for different erosion conditions based on the virtual temperature measurement dataset includes: Set the boundary conditions of the simulation model according to the preset erosion line, and calculate the temperature field of the model according to the boundary conditions of the simulation model and the thermal conductivity of the refractory material. The predicted temperature value at the corresponding thermocouple position on the calculation section is extracted from the temperature field of the model and compared with the temperature data at the corresponding thermocouple measuring point in the virtual temperature measurement dataset to calculate the relative temperature error. When the relative temperature error is higher than the preset temperature threshold, the direction and step size of the erosion node movement are calculated by the steepest descent method with a penalty function, and the movement distance of the erosion node is controlled each time it moves inward or outward until all erosion nodes have completed one movement and a new target temperature isotherm position is obtained; wherein, the background mesh is kept fixed when the erosion node moves, and only the position of the erosion boundary is updated. The temperature field is iteratively calculated, and the erosion degree of each point on the target temperature isotherm obtained after the iteration converges is compared with the historical erosion line. The point with the most severe erosion is selected, and the historical erosion line is updated using the selected point with the most severe erosion to obtain the simulated erosion line.

4. The thermocouple arrangement method based on a digital twin model of a furnace hearth according to claim 1 or 3, characterized in that, The process of comparing the simulated erosion line with the preset erosion line and correlating the comparison result with the furnace hearth erosion probability weight to perform comprehensive error calculation includes: The simulated erosion line is compared with the preset erosion line, and the average profile deviation, local extreme value deviation and heat flux density deviation are determined based on the corresponding comparison results. Based on the average profile deviation, the local extreme value deviation, the heat flux density deviation, and the furnace hearth erosion probability weight, a comprehensive error evaluation function is constructed, and the comprehensive error is calculated using the comprehensive error evaluation function; wherein, the comprehensive error evaluation function is: ； In the formula, E 综合 This is the comprehensive error evaluation function. n This represents the total number of erosion conditions in the erosion simulation condition library. λ 1 represents the weighting coefficient for the average profile deviation. λ 2 represents the weighting coefficient for local extremum deviation. λ 3 is the weighting coefficient for the heat flux density deviation, and λ 1+ λ 2+ λ 3=1, w i For the first i The probability weights of hearth erosion for different erosion types e i,轮廓 For the first i Average profile deviation for each type of erosion e i,极值 For the first i Local extreme deviations for various erosion types e i,热流 For the first i Heat flux density deviation for different erosion types.

5. The thermocouple arrangement method based on a digital twin model of a furnace hearth according to claim 1, characterized in that, The process of obtaining the candidate thermocouple arrangement scheme includes: using a parameterization method to generate the candidate thermocouple arrangement scheme based on decision variables, the decision variables including the number of circumferential profiles, the number of radial measuring points, the number of axial layers, and the arrangement pattern, the arrangement pattern including at least one of uniform arrangement, gradient arrangement, and asymmetric densification arrangement; wherein, the uniform arrangement is that the thermocouples are uniformly distributed in the circumferential, radial, and / or axial directions of the furnace hearth and bottom; the gradient arrangement is based on the gradient change of the thermal conductivity of the refractory material, with denser arrangement in areas where the thermal conductivity is lower than the preset thermal conductivity, and sparse arrangement in areas where the thermal conductivity is higher than the preset thermal conductivity, and the spacing between adjacent measuring points is inversely proportional to the thermal conductivity; the asymmetric densification type is to locally densify in the target area and maintain a conventional arrangement in non-target areas, the target area including below the taphole and the center of the furnace bottom.

6. The thermocouple arrangement method based on a digital twin model of a furnace hearth according to claim 1, characterized in that, The process of adjusting the candidate thermocouple arrangement scheme includes: adjusting the candidate thermocouple arrangement scheme according to at least one of the densification strategy, migration strategy, and layering strategy; wherein, the densification strategy is used to increase the number of thermocouples and shorten the thermocouple spacing according to the error contribution; the migration strategy is used to migrate the thermocouples to the target area and optimize the spatial distribution of measuring points; the layering strategy is used to add thermocouples according to the interface of thermal conductivity change and improve the ability to capture interlayer temperature gradients; the target area includes the area below the taphole and the center of the furnace bottom.

7. The thermocouple arrangement method based on a digital twin model of a furnace hearth according to claim 1 or 6, characterized in that, The process of adjusting the candidate thermocouple arrangement scheme further includes: solving for the Pareto optimal solution set with the minimum number of thermocouples, the minimum comprehensive error, and the most uniform error distribution as the optimization objectives; and adjusting the candidate thermocouple arrangement scheme according to the number of thermocouples, comprehensive error, and error distribution corresponding to the Pareto optimal solution set; wherein, the optimization variables corresponding to the optimization objectives include thermocouple position coordinates, number of circumferential sections, number of radial measuring points, number of axial layers, arrangement density, and measuring point combination method; and / or, analyzing the heat flux density gradient distribution based on the furnace hearth digital twin model, determining the regions where the heat flux density change exceeds a preset density threshold and the regions where the three-dimensional heat transfer effect exceeds a preset heat transfer threshold based on the heat flux density gradient distribution, and densifying the thermocouple arrangement in the corresponding regions.

8. The thermocouple arrangement method based on a digital twin model of a furnace hearth according to claim 1, characterized in that, The erosion simulation case library includes at least one type of hearth erosion case among rat hole erosion, elephant foot erosion, pot bottom erosion and ring crack erosion. Each type of hearth erosion is divided into multiple levels or degrees according to the erosion depth or range.

9. A thermocouple arrangement system based on a digital twin model of a furnace hearth, characterized in that, The system includes: The furnace hearth digital twin model module is used to build a digital twin model of the furnace hearth. Thermocouple arrangement scheme module, used to generate candidate thermocouple arrangement schemes; The erosion simulation condition library module is used to construct an erosion simulation condition library based on the digital twin model of the hearth, and to establish erosion probability models for various types of erosion at different stages of furnace service based on the erosion simulation condition library; wherein, the erosion simulation condition library includes at least one type of hearth erosion condition, and the erosion probability model is used to assign hearth erosion probability weights. The erosion scenario simulation module is used to perform numerical simulation based on the digital twin model of the furnace hearth, the candidate thermocouple arrangement scheme and the erosion simulation condition library, and solve the three-dimensional temperature field distribution of the furnace hearth and furnace bottom. The temperature data extraction module is used to extract temperature data from measuring points based on the three-dimensional temperature field distribution of the furnace hearth and furnace bottom, forming a virtual temperature measurement dataset. The simulated erosion line module is used to determine simulated erosion lines for different erosion conditions based on the virtual temperature measurement dataset. The scheme evaluation and optimization module is used to compare the simulated erosion line with the preset erosion line, and associate the comparison result with the furnace hearth erosion probability weight to perform comprehensive error calculation; and, when the calculated comprehensive error is greater than the preset threshold, to adjust the candidate thermocouple arrangement scheme; wherein, the preset erosion line includes the erosion line contour set in the furnace hearth digital twin model by changing the physical boundary of the refractory material of the preset part; The result output module is used to output the adjusted thermocouple arrangement scheme.

Images