Method for improving low-temperature reduction degradation index of iron ore sinter

Abstract

The invention relates to a method for improving the low-temperature reduction degradation index of iron ore sinter, and belongs to the technical field of iron ore sinter processing. The method comprises the following steps: collecting finished iron ore sinter as a target sample, and obtaining initial sinter low-temperature reduction degradation indexes, original chemical components and proportions thereof of the target sample; constructing a hyperbolic space of sintering ore low-temperature reduction degradation indexes, and calculating a coarse matching matrix; calculating a multi-dimensional variable solution group by adopting a genetic algorithm according to the rough matching matrix; screening an optimal solution in each temperature gradient interval from the multi-dimensional variable solution group according to a preset gradient number, and comprehensively generating a temperature gradient solution group; and calculating the combination of the optimal sintering material proportioning scheme and the optimal parameters in each temperature gradient interval according to the temperature gradient solution, and applying the combination of the optimal sintering material proportioning scheme and the optimal parameters to the low-temperature reduction degradation process. According to the method, the control precision is improved, and the improvement effect of the low-temperature reduction degradation index of the iron ore sinter is improved.

Description

Technical Field

[0001] This invention relates to the field of iron ore sinter processing technology, and in particular to a method for improving the low-temperature reduction pulverization index of iron ore sinter. Background Technology

[0002] Currently, the quality requirements for iron ore sinter not only include physicochemical properties but also metallurgical properties. The reducibility of iron ore sinter to iron is a fundamental metallurgical property, while low-temperature reduction pulverization is an important one. The low-temperature reduction pulverization process is used in the metal reduction reaction of iron ore sinter. The low-temperature reduction pulverization index of the sinter reflects the reduction intensity of the iron ore sinter in the upper part of the blast furnace. A high low-temperature reduction pulverization index value results in good permeability of the upper burden column in the furnace, reducing furnace body caking; increased CO utilization in the gas; better smelting intensity; lower coke ratio; and higher pig iron production.

[0003] Traditional low-temperature reduction pulverization processes adjust the low-temperature reduction pulverization index of sinter by adjusting the reaction temperature, reaction time, alkalinity, or by adding sintering materials as combustion aids. However, these methods often only adjust one aspect and do not consider the relationship between reaction temperature, reaction time, and alkalinity. This results in unreasonable adjustment methods for the low-temperature reduction pulverization process of sintering and low control precision, leading to poor improvement in the low-temperature reduction pulverization index of sintering. Summary of the Invention

[0004] To address the aforementioned technical problems, this invention provides a method for improving the low-temperature reduction pulverization index of iron ore sinter. The technical solution of this invention is as follows: S1, collect finished iron ore sinter on the finished product belt conveyor, obtain the target sample after screening and reduction, and obtain the initial low temperature reduction pulverization index of the target sample sinter. S2, obtain the original chemical composition and proportion of the target sample, and construct the hyperbolic space of the low-temperature reduction pulverization index of sinter by combining all the preset sintering material components, and calculate the coarse ratio matrix with the initial low-temperature reduction pulverization index of sintering as the lowest threshold. S3 uses the preset sintering material composition ratio, basicity, reaction temperature and reaction time as multi-dimensional variables, and the coarse ratio matrix, the process parameters of the low-temperature reduction pulverization process and the expected low-temperature reduction pulverization index of the sintering ore as constraints. The genetic algorithm is used to calculate the solution of the multi-dimensional variables. S4: The multi-dimensional variable solution group is sorted according to the preset number of gradients to select the optimal solution in each temperature gradient interval, and the temperature gradient solution group is generated in combination. S5. Calculate the optimal sintering material batching scheme and optimal parameter combination for each temperature gradient interval based on the temperature gradient ungrouping. The optimal sintering material batching scheme includes the optimal sintering material composition ratio, and the optimal parameters include the optimal basicity, optimal reaction temperature, and optimal reaction time. S6 applies the optimal sinter batching scheme and optimal parameter combination to the low-temperature reduction pulverization process, and obtains the low-temperature reduction pulverization index of sinter as needed in the subsequent low-temperature reduction pulverization process and performs feedback adjustment.

[0005] Preferably, S2 includes: S21. Obtain the original chemical composition and proportion of the target sample using gas chromatography-mass spectrometry. S22, construct the original vector based on the original chemical composition ratio of the target sample, and construct the feasible ratio space based on all preset sintering material components; S23. Input the original vector and feasible proportion space into the hyperbolic space of the low-temperature reduction pulverization index of sinter. Use the initial low-temperature reduction pulverization index of sinter as the minimum threshold. Calculate the correlation strength between each preset sinter component ratio and the low-temperature reduction pulverization index of sinter based on the hyperbolic space. Generate a feasible proportion space solution based on the correlation strength of each preset sinter component ratio. S24. A numerical optimization algorithm is used to search for multiple feasible proportioning schemes that satisfy the spatial constraints in the feasible proportioning space. The multiple feasible proportioning schemes are then constructed into a coarse proportioning matrix. In the coarse proportioning matrix, each row represents a feasible proportioning scheme, and each column represents the proportion of a preset sintering material component in each feasible proportioning scheme.

[0006] Preferably, in step S23, the initial low-temperature reduction pulverization index of the sinter is used as the minimum threshold, and the correlation strength between the component ratio of each preset sinter and the low-temperature reduction pulverization index of the sinter is calculated according to hyperbolic space using formulas (1) and (2): (1); (2); In formula (1), c represents the correlation strength between the i-th preset sinter composition ratio and the low-temperature reduction pulverization index of the sinter. i This represents the component ratio of the i-th preset sintering material. This indicates the initial low-temperature reduction and pulverization index of the sinter. Represents the Sigmoid function; Let represent the hyperbolic distance between the i-th preset sinter composition ratio and the low-temperature reduction pulverization index of the sinter, and let dist() represent the hyperbolic distance function; In formula (2), arcosh() represents the inverse hyperbolic cosine function, and s j This represents the proportion of the j-th original chemical component. Let m represent the Euclidean distance function, and m represent the m original chemical components of the target sample.

[0007] Preferably, in step S23, generating a feasible sizing space solution based on the correlation strength of each preset sintering material component ratio includes: S231, combine each preset sintering material component and its proportion into multiple candidate proportion schemes; S232, if the sum of the correlation strengths of all preset sintering material component ratios in any candidate ratio scheme is greater than the expected correlation strength, then the candidate ratio is filled into the feasible ratio space as a feasible ratio scheme. S233, generate a feasible matching space solution based on all feasible matching schemes in the feasible matching space.

[0008] Preferably, S24 includes: S241, take the feasible matching scheme in the first row of the feasible matching space as the initial solution; S242, taking the initial solution as the starting point of the iteration, iterates each feasible ratio scheme in the feasible ratio space solution, calculates the optimization degree of the feasible ratio scheme in each iteration through the preset evaluation function, and updates the function parameters of the preset evaluation function according to the optimization degree of the feasible ratio scheme in each iteration. S243, if the optimization degree of any feasible ratio scheme is greater than the preset optimization threshold, then the feasible ratio scheme is filled into the coarse ratio matrix. After iterating through all feasible ratio schemes in the feasible ratio space solution, the coarse ratio matrix is ​​generated.

[0009] Preferably, S3 includes: S31 uses the preset sintering material composition ratio, basicity, reaction temperature and reaction time as multi-dimensional variables, and the coarse ratio matrix, the process parameters of the low-temperature reduction pulverization process and the expected low-temperature reduction pulverization index of the sintering ore as constraints to construct a multi-dimensional objective function, and generates multiple original solution vectors based on the multi-dimensional objective function, and encodes each original solution vector into a genome; S32, calculate the fitness of each genome according to the fitness algorithm, and select the function with a fitness greater than the preset fitness among all genomes as the parent generation. Each genome includes the sintering material composition ratio and process parameters. S33, cross-interchange the sintering material composition ratio and process parameters of all the genomes of the parent generation to obtain the new offspring and its genome, return to S32, and end the iteration when the preset number of iterations is completed, and select the genome with the highest fitness in each iteration to generate multi-dimensional variable ungrouping.

[0010] Preferably, in step S32, the fitness of the k-th genome is calculated using formula (3) according to the fitness algorithm. : (3); In formula (3), f() represents a scalar function, and wk t represents the sintering composition ratio of the k-th genome. k Let g represent the process parameters of the k-th genome, n represent the number of genomes, max() represent the maximization operation function, and g k This represents the racial constraint of the k-th genome obtained based on the constraint conditions.

[0011] Preferably, S4 includes: S41, normalize the preset temperature range according to the preset number of gradients to obtain the preset number of temperature gradient ranges. S42, For each temperature gradient interval, select all multi-dimensional variable solutions for the reaction temperature within the temperature gradient interval from the multi-dimensional variable solution set; S43, calculate the comprehensive matching degree of each multi-dimensional variable solution in each temperature gradient interval, and filter the optimal solution in each temperature gradient interval based on the comprehensive matching degree of each multi-dimensional variable solution in each temperature gradient interval and the dynamic threshold of the corresponding temperature gradient interval, and generate a temperature gradient solution group by combining the optimal solutions in each temperature gradient interval.

[0012] Preferably, S43 includes: S431, calculate the dynamic threshold of each temperature gradient interval based on the normal distribution principle of all multidimensional variable solutions in all temperature gradient intervals; S432, calculate the matching degree of each multidimensional variable solution in each temperature gradient interval based on all multidimensional variable solutions in each temperature gradient interval; S433, compare the matching degree of each multidimensional variable solution in each temperature gradient interval with its dynamic threshold. For any temperature gradient interval, select the multidimensional variable solution in the temperature gradient interval whose matching degree is greater than its dynamic threshold and whose matching degree is the maximum value as the optimal solution. If there is no multidimensional variable solution in the temperature gradient interval or no multidimensional variable solution whose matching degree is greater than the dynamic threshold, then perform linear interpolation on the temperature gradient interval to obtain a filled multidimensional variable solution. Take the filled multidimensional variable solution as the optimal solution, and generate a temperature gradient solution group by combining the optimal solutions under each temperature gradient interval.

[0013] All of the above-mentioned optional technical solutions can be combined arbitrarily, and the present invention will not provide a detailed description of the structure after each combination.

[0014] By means of the above solution, the beneficial effects of the present invention are as follows: By constructing a hyperbolic space for the low-temperature reduction pulverization index of sintered ore based on the original chemical composition of the target sample and all preset sintering material components, a coarse proportioning matrix is ​​calculated. Using the preset sintering material composition, basicity, temperature, and time as multi-dimensional variables, and the coarse proportioning matrix, process parameters, and the preset low-temperature reduction pulverization index of sintered ore as constraints, a multi-dimensional variable solution is calculated. The multi-dimensional variable solution is then filtered according to the preset number of gradients to select the optimal solution for each temperature gradient interval, and a comprehensive temperature gradient solution is generated. Based on the temperature gradient solution, the optimal sintering material batching scheme and the optimal parameter combination are calculated, adjusting the original low-temperature reduction pulverization scheme of sintered ore. This method provides a way to improve the low-temperature reduction pulverization index of iron ore sintered ore. This method comprehensively considers process parameters, sintering material composition, introduces the concept of temperature gradient, and comprehensively considers the relationship between reaction temperature, reaction time, and basicity. It provides a reasonable adjustment method for the low-temperature reduction pulverization index based on multi-dimensional variables. Furthermore, by combining temperature gradient control, the control accuracy is improved, and the improvement effect on the low-temperature reduction pulverization index of iron ore sintered ore is enhanced.

[0015] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, the preferred embodiments of the present invention are described in detail below with reference to the accompanying drawings. Attached Figure Description

[0016] Figure 1 This is a flowchart of a method for improving the low-temperature reduction pulverization index of iron ore sinter provided in an embodiment of the present invention.

[0017] Figure 2 This is a schematic diagram of the generation of offspring genomes through crossover of parent genomes in the genetic algorithm of this invention. Detailed Implementation

[0018] The specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and are not intended to limit the scope of the invention.

[0019] In summary, such as Figure 1 As shown in the embodiment of the present invention, a method for improving the low-temperature reduction pulverization index of iron ore sinter can be achieved through the following steps S1 to S6: S1. Collect finished iron ore sinter on the finished product conveyor belt, obtain the target sample after screening and reduction, and obtain the initial low-temperature reduction pulverization index of the target sample sinter.

[0020] Specifically, when collecting finished iron ore sinter, the following precautions should be taken: Material must not be taken from the bottom-laying conveyor belt; finished iron ore sinter must be taken from the finished product conveyor belt. After screening out finished iron ore sinter with a size of +12.5mm and a weight of 15kg, it is poured into a medium crusher for crushing. The resulting particle size is 10–12.5mm after screening. After mixing and reduction, a sample weight of no less than 3kg is obtained as the target sample. The target sample is then sent to a pre-iron testing laboratory to determine the initial low-temperature reduction pulverization index of the target sinter.

[0021] S2, obtain the original chemical composition and proportion of the target sample, and construct the hyperbolic space of the low-temperature reduction pulverization index of sinter by combining all the preset sintering material components, and calculate the coarse proportion matrix with the initial low-temperature reduction pulverization index of sintering as the lowest threshold.

[0022] Specifically, the fundamental reason for the low-temperature reduction pulverization of sinter is that the phase transition temperature of Fe2O3 to Fe3O4 plays a crucial role in the formation of regenerated Fe2O3. Components that can increase the phase transition temperature of Fe2O3 to Fe3O4, such as TiO2, Al2O3, K2O, and Na2O, are conducive to the formation of regenerated Fe2O3; conversely, components that can decrease the phase transition temperature of Fe2O3 to Fe3O4, such as CaO, SiO2, MgO, and FeO, are detrimental to the formation of regenerated Fe2O3. Using a scatter plot and linear correlation method, the main factors affecting the low-temperature reduction pulverization index of sinter are: the low-temperature reduction pulverization index of sinter is significantly correlated with FeO, TiO2, Al2O3, K2O, and Na2O, and has some influence with CaO, SiO2, and MgO, but the linear correlation is not strong. Therefore, in this embodiment of the invention, the preset sinter composition is TiO2, Al2O3, K2O, and Na2O.

[0023] In one specific embodiment, S2 includes: S21. Obtain the original chemical composition and proportion of the target sample using gas chromatography-mass spectrometry.

[0024] In practice, this step involves first separating the target sample using gas chromatography (GC). The separated components are detected at different retention times to obtain different groups. Then, all groups are placed in a mass spectrometer (MS). Through ionization and mass analysis, a mass spectrum of the target sample is generated. Different ion peaks in the mass spectrum correspond to different components in the target sample. By comparing with a standard spectral library, qualitative analysis is performed to obtain the original chemical composition of the target sample. Simultaneously, the signal intensity of the mass spectrum is used for quantitative analysis to determine the relative content of each original chemical component in the target sample. Thus, the original chemical composition and proportions of the target sample are obtained.

[0025] S22, construct the original vector based on the original chemical composition ratio of the target sample, and construct the feasible proportion space based on all preset sintering material components.

[0026] For example, if the preset sintering material composition includes A, B, C, and D, then a feasible proportioning space containing these four preset sintering material compositions is constructed based on A, B, C, and D, and the proportions of these four preset sintering material compositions are filled into the feasible proportioning space. The feasible proportioning space after filling in the proportions of the preset sintering material compositions can generate different preset sintering material composition proportions based on these four preset sintering material compositions and their proportions.

[0027] S23. Input the original vector and feasible proportion space into the hyperbolic space of the low-temperature reduction pulverization index of sinter. Use the initial low-temperature reduction pulverization index of sinter as the minimum threshold. Calculate the correlation strength between each preset sinter component ratio and the low-temperature reduction pulverization index of sinter based on the hyperbolic space. Generate a feasible proportion space solution based on the correlation strength of each preset sinter component ratio.

[0028] In a specific embodiment, in S23, the initial low-temperature reduction pulverization index of the sinter is used as the minimum threshold, and the correlation strength between each preset sintering material composition ratio and the low-temperature reduction pulverization index of the sinter is calculated according to hyperbolic space using formulas (1) and (2): (1); (2); In formula (1), c represents the correlation strength between the i-th preset sinter composition ratio and the low-temperature reduction pulverization index of the sinter. i This represents the component ratio of the i-th preset sintering material. This indicates the initial low-temperature reduction and pulverization index of the sinter. Represents the Sigmoid function; Let represent the hyperbolic distance between the i-th preset sinter composition ratio and the low-temperature reduction pulverization index of the sinter, and let dist() represent the hyperbolic distance function; In formula (2), arcosh() represents the inverse hyperbolic cosine function, and s j This represents the proportion of the j-th original chemical component. Let m represent the Euclidean distance function, and m represent the m original chemical components of the target sample.

[0029] Specifically, in this embodiment, the hyperbolic distance represents the sum of the single distances between any preset sintering material composition ratio and each original chemical composition ratio in the hyperbolic space. Formula (1) maps the hyperbolic distance to the interval (0,1) using the Sigmoid function. If the hyperbolic distance of any preset sintering material composition ratio is close to r0, the correlation strength is close to 0. This relationship indicates that the closer the preset sintering material composition ratio is to the minimum threshold, the weaker the correlation strength between the preset sintering material composition ratio and the low-temperature reduction pulverization index of the sinter. Formula (2) maps the original chemical composition ratio and each preset sintering material composition ratio to the hyperbolic space. The smaller the hyperbolic distance, the stronger the correlation between the preset sintering material composition ratio c. i Ratio of original chemical components s j The closer the overall structure is, the more reasonable the low-temperature reduction pulverization index of the sinter obtained according to the preset sintering material composition ratio is.

[0030] In a specific embodiment, step S23, generating a feasible sizing space solution based on the correlation strength of each preset sintering material component ratio, includes: S231, combine each preset sintering material component and its proportion into multiple candidate proportion schemes.

[0031] Specifically, the sum of the proportions of all preset sintering materials in each candidate formulation is 100%.

[0032] S232, if the sum of the correlation strengths of all preset sintering material component ratios in any candidate ratio scheme is greater than the expected correlation strength, then the candidate ratio is filled into the feasible ratio space as a feasible ratio scheme.

[0033] S233, generate a feasible matching space solution based on all feasible matching schemes in the feasible matching space.

[0034] Specifically, each row in the feasible mix design space represents a feasible mix design scheme. All feasible mix designs are filled into the feasible mix design space according to their preset sintering material composition and proportion, thus obtaining the feasible mix design space solution. The expected correlation strength is a threshold value for correlation strength obtained based on empirical values.

[0035] S24. A numerical optimization algorithm is used to search for multiple feasible proportioning schemes that satisfy the spatial constraints in the feasible proportioning space. The multiple feasible proportioning schemes are then constructed into a coarse proportioning matrix. In the coarse proportioning matrix, each row represents a feasible proportioning scheme, and each column represents the proportion of a preset sintering material component in each feasible proportioning scheme.

[0036] Specifically, the spatial constraints include the preset range of sinter composition ratios and the minimum correlation strength between the preset sinter composition ratios and the low-temperature reduction pulverization index of the sinter.

[0037] In one specific embodiment, S24 includes: S241, take the feasible matching scheme in the first row of the feasible matching space as the initial solution.

[0038] Specifically, there is no order in which all feasible matching schemes in the feasible matching space are ordered. The feasible matching scheme in the first row is selected as the initial solution to facilitate subsequent iterations.

[0039] S242, taking the initial solution as the starting point of the iteration, iterates each feasible ratio scheme in the feasible ratio space solution, calculates the optimization degree of the feasible ratio scheme in each iteration through the preset evaluation function, and updates the function parameters of the preset evaluation function according to the optimization degree of the feasible ratio scheme in each iteration.

[0040] The preset evaluation function includes preset weights, spatial constraints, and ideal proportions of each preset sintering material component as its function parameters. The preset evaluation function is composed of function parameters. In this embodiment of the invention, the preset evaluation function can be implemented by minimizing the objective function, which adjusts the ideal proportions of each preset sintering material component according to the optimization degree of the feasible proportioning scheme in each iteration.

[0041] S243, if the optimization degree of any feasible ratio scheme is greater than the preset optimization threshold, then the feasible ratio scheme is filled into the coarse ratio matrix. After iterating through all feasible ratio schemes in the feasible ratio space solution, the coarse ratio matrix is ​​generated.

[0042] The preset optimization threshold is an optimization threshold set based on a large amount of historical experimental data.

[0043] S3 uses the preset sintering material composition ratio, basicity, reaction temperature and reaction time as multi-dimensional variables, and the coarse ratio matrix, the process parameters of the low-temperature reduction pulverization process and the expected low-temperature reduction pulverization index of the sintered ore as constraints. The genetic algorithm is used to calculate the solution of the multi-dimensional variables.

[0044] According to national standards, the low-temperature reduction pulverization index of sintered ore is expected to be no less than 65%.

[0045] Specifically, one of the fundamental reasons for the low-temperature reduction pulverization of sinter is the volume expansion during the reduction of Fe2O3 to Fe3O4 or FeO, which releases stress and exacerbates crack propagation, leading to pulverization. The CaO content in the low-temperature reduction pulverization process represents the basicity in the process parameters. For example, if adding CaO to the low-temperature reduction pulverization process increases the basicity of the sinter from 1.83±0.05 to 1.95±0.05, it significantly improves the low-temperature reduction pulverization index of the sinter. Therefore, the basicity in the process parameters has a significant impact on the low-temperature reduction pulverization index of the sinter. Thus, in this embodiment of the invention, basicity is treated as a variable.

[0046] Furthermore, the second fundamental reason for the low-temperature reduction pulverization of sintered ore is the lattice transformation of regenerated skeletal hematite from α-Fe₂O₃ to γ-Fe₂O₃, generating internal stress. During the compression and collision process, the sintered ore fragments and pulverizes. The main reason for the formation of skeletal Fe₂O₃ is the high sintering temperature and rapid cooling rate. Therefore, the reaction temperature and reaction time in the process parameters are important factors affecting the low-temperature reduction pulverization index of sintered ore. Thus, in this embodiment of the invention, reaction temperature and reaction time are also included as variables.

[0047] In one specific embodiment, S3 includes: S31 uses the preset sintering material composition ratio, basicity, reaction temperature and reaction time as multi-dimensional variables, and the coarse ratio matrix, the process parameters of the low-temperature reduction pulverization process and the expected low-temperature reduction pulverization index of the sintered ore as constraints to construct a multi-dimensional objective function, and generates multiple original solution vectors based on the multi-dimensional objective function, and encodes each original solution vector into a genome.

[0048] The multidimensional objective function is an objective function constructed by using the preset sintering material composition ratio, basicity, reaction temperature and reaction time as multidimensional variables, and the coarse ratio matrix, the process parameters of the low-temperature reduction pulverization process and the desired low-temperature reduction pulverization index of the sinter as constraints. It aims to obtain multiple original solution vectors by simultaneously satisfying the constraints.

[0049] For example, suppose an original solution vector includes: a preset sintering material composition ratio of [0.3, 0.5, 0.2], an alkalinity of 1.5, a reaction temperature of 550℃, and a reaction time of 50 min. Then, after real number encoding of the original solution vector, the genome is represented as [(0.3, 0.5, 0.2), (1.5, 550, 50)].

[0050] S32, calculate the fitness of each genome according to the fitness algorithm, and select the function with a fitness greater than the preset fitness among all genomes as the parent generation. Each genome includes the sintering material composition ratio and process parameters.

[0051] Specifically, in S32, the fitness of the k-th genome is calculated using formula (3) according to the fitness algorithm. : (3); In formula (3), f() represents a scalar function, and w k t represents the sintering composition ratio of the k-th genome. k Let g represent the process parameters of the k-th genome, n represent the number of genomes, max() represent the maximization operation function, and g k This represents the racial constraint of the k-th genome obtained based on the constraint conditions.

[0052] Specifically, in formula (3), the scalar function is the objective term, used to quantify the performance of the genome in the entire population; This is a constraint penalty term used to dynamically penalize all genomes that violate the constraint, reducing their performance across the entire population.

[0053] S33, cross-interchange the sintering material composition ratio and process parameters of all the genomes of the parent generation to obtain the new offspring and its genome, return to S32, and end the iteration when the preset number of iterations is completed, and select the genome with the highest fitness in each iteration to generate multi-dimensional variable ungrouping.

[0054] For example, the process of crossover between parent generation 1 and parent generation 2 to generate offspring generation 1 and offspring generation 2 is as follows: Figure 2 As shown. Additionally, the preset number of iterations is typically 600-1000.

[0055] S4: The multi-dimensional variable solution group is sorted according to the preset number of gradients to select the optimal solution in each temperature gradient interval, and the temperature gradient solution group is generated in combination. In one specific embodiment, S4 includes: S41, normalize the preset temperature range according to the preset number of gradients to obtain the preset number of temperature gradient ranges.

[0056] Specifically, assuming the preset temperature range is [T1, T2] and the preset number of gradients is N, then the normalized temperature range of the j-th temperature gradient interval is: .

[0057] S42, for each temperature gradient interval, select all multidimensional variable solutions for the reaction temperature within the temperature gradient interval from the multidimensional variable solution set.

[0058] Specifically, a multi-dimensional variable solution includes four variables: the preset sintering material composition ratio, basicity, reaction temperature, and reaction time.

[0059] S43, calculate the comprehensive matching degree of each multi-dimensional variable solution in each temperature gradient interval, and filter the optimal solution in each temperature gradient interval based on the comprehensive matching degree of each multi-dimensional variable solution in each temperature gradient interval and the dynamic threshold of the corresponding temperature gradient interval, and generate a temperature gradient solution group by combining the optimal solutions in each temperature gradient interval.

[0060] Specifically, Euclidean distance is used to calculate the univariate matching degree between each multidimensional variable of each multidimensional variable solution and the standard value in each temperature gradient interval. Finally, the sum of the univariate matching degrees of all multidimensional variables is taken as the comprehensive matching degree of each multidimensional variable solution. Among them, for the preset sintering material composition ratio in the multidimensional variable solution, the univariate matching degree of each preset sintering material composition ratio is obtained by weighting the preset component weights with the preset sintering material composition ratio. The preset component weights of each preset sintering material composition are determined empirically.

[0061] In one specific embodiment, S43 includes: S431, calculate the dynamic threshold for each temperature gradient interval based on the normal distribution principle of all multidimensional variable solutions in all temperature gradient intervals.

[0062] In practice, this step calculates the univariate dynamic threshold of each multidimensional variable in each multidimensional variable solution within each temperature gradient interval, based on the normal distribution principle and all multidimensional variable solutions for each temperature gradient interval. The dynamic threshold for each temperature gradient interval is the sum of the univariate thresholds of each of its multidimensional variables. For example, if the preset sintering material composition ratio for a certain temperature gradient interval has a mean of 0.85, a standard deviation of 0.05, and a constant value of 1, then the univariate threshold for the preset sintering material composition ratio for that temperature gradient interval is: 0.85 + 1 × 0.05 = 0.9.

[0063] Taking the above example again, if the single-variable threshold of the preset sintering material composition ratio in the temperature gradient range is 0.9, the basicity is 0.1, the reaction temperature is 0.5, and the reaction time is 0.6, then the dynamic threshold of the temperature gradient range is 0.9+0.1+0.5+0.6=2.1.

[0064] S432, calculate the matching degree of each multidimensional variable solution for each temperature gradient interval based on all multidimensional variable solutions for each temperature gradient interval.

[0065] S433, compare the matching degree of each multidimensional variable solution in each temperature gradient interval with its dynamic threshold. For any temperature gradient interval, select the multidimensional variable solution in the temperature gradient interval whose matching degree is greater than its dynamic threshold and whose matching degree is the maximum value as the optimal solution. If there is no multidimensional variable solution in the temperature gradient interval or no multidimensional variable solution whose matching degree is greater than the dynamic threshold, then perform linear interpolation on the temperature gradient interval to obtain a filled multidimensional variable solution. Take the filled multidimensional variable solution as the optimal solution, and generate a temperature gradient solution group by combining the optimal solutions under each temperature gradient interval.

[0066] Specifically, for any temperature gradient interval where there is no multidimensional variable solution or no multidimensional variable solution with a matching degree greater than the dynamic threshold, the univariate linear interpolation result of each multidimensional variable in the temperature gradient interval is calculated, and the univariate linear interpolation results of all multidimensional variables are combined to form the filled multidimensional variable solution of the temperature gradient interval. Specifically, for a temperature gradient interval where there is no multidimensional variable solution or no multidimensional variable solution with a matching degree greater than the dynamic threshold, this embodiment of the invention calculates the mean of each multidimensional variable on both sides of the temperature gradient interval as the univariate linear interpolation result in that temperature gradient interval.

[0067] S5. Calculate the optimal sintering material batching scheme and optimal parameter combination for each temperature gradient interval based on the temperature gradient ungrouping. The optimal sintering material batching scheme includes the optimal sintering material composition ratio, and the optimal parameters include the optimal basicity, optimal reaction temperature, and optimal reaction time.

[0068] Specifically, based on the numerical relationship between the optimal reaction temperature and the optimal reaction time, it is possible to ensure that the sintered ore is fully sintered while avoiding the deterioration of mineral structure and the decline in reducibility caused by excessively high temperature and excessively long reaction time.

[0069] S6 applies the optimal sinter batching scheme and optimal parameter combination to the low-temperature reduction pulverization process, and obtains the low-temperature reduction pulverization index of sinter as needed in the subsequent low-temperature reduction pulverization process and performs feedback adjustment.

[0070] Specifically, by applying the optimal sintering material batching scheme and optimal parameters, the low-temperature reduction pulverization index of sinter is significantly improved. Experiments show that this method can achieve a low-temperature reduction pulverization index of sinter of over 70%, effectively reducing the fragmentation and pulverization of sinter during the low-temperature reduction process. Furthermore, due to the improvement in the low-temperature reduction pulverization index and the optimization of the quality of sinter, the permeability of the blast furnace charge is significantly improved, the gas distribution is more uniform, and the furnace condition is more stable. This reduces problems such as deterioration of charge permeability, furnace condition fluctuations or abnormalities caused by sinter pulverization, lowers the risk of production accidents, and improves the production efficiency and safety of the blast furnace.

[0071] Furthermore, in practical applications, specific areas on the sintering production line are selected for small-scale trial production using the optimal sintering material batching scheme and optimal parameter combination. Extensive actual production data is collected to further verify and optimize the optimal sintering material batching scheme and optimal parameter combination, resulting in an optimized sintering material batching scheme and optimized parameter combination. Emphasis is placed on the stability and operability under actual industrial conditions. The optimized sintering material batching scheme and optimized parameter combination are then promoted and applied throughout the entire production line. Continuous monitoring of various operating indicators such as the low-temperature reduction pulverization index of the sinter, the blast furnace utilization coefficient, and the coke ratio is maintained. Based on feedback information, the technical scheme is fine-tuned in a timely manner to ensure long-term stable and effective results.

[0072] According to the above embodiments, this invention constructs a hyperbolic space of the low-temperature reduction pulverization index of sintered ore based on the original chemical composition and proportion of the target sample and all preset sintering material components to obtain a coarse proportioning matrix. This fully considers the influence of the preset sintering material composition ratio on the low-temperature reduction pulverization index of sintered ore, providing a data foundation for subsequent multi-dimensional adjustment of the low-temperature reduction pulverization index of sintered ore. Furthermore, using the preset sintering material composition ratio, basicity, reaction temperature, and reaction time as multi-dimensional variables, a genetic algorithm and temperature gradient screening method are used to obtain the multi-dimensional variable solutions under each temperature gradient interval as the optimal sintering material batching scheme and optimal parameter combination. The optimal sintering material batching scheme and optimal parameter combination are then applied to actual production. By comprehensively considering process parameters, sintering material composition ratio, introducing the concept of temperature gradient, and comprehensively considering the relationship between reaction temperature, reaction time, and basicity, a reasonable adjustment method based on multi-dimensional variables is provided for the low-temperature reduction pulverization index. Furthermore, by combining temperature gradient control, the control accuracy is improved, and the improvement effect of the low-temperature reduction pulverization index of iron ore sintered ore is enhanced.

[0073] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for improving the low-temperature reduction pulverization index of iron ore sinter, characterized in that, include: S1, collect finished iron ore sinter on the finished product belt conveyor, obtain the target sample after screening and reduction, and obtain the initial low temperature reduction pulverization index of the target sample sinter. S2, obtain the original chemical composition and proportion of the target sample, and construct the hyperbolic space of the low-temperature reduction pulverization index of sinter by combining all the preset sintering material components, and calculate the coarse ratio matrix with the initial low-temperature reduction pulverization index of sintering as the lowest threshold. S3 uses the preset sintering material composition ratio, basicity, reaction temperature and reaction time as multi-dimensional variables, and the coarse ratio matrix, the process parameters of the low-temperature reduction pulverization process and the expected low-temperature reduction pulverization index of the sintering ore as constraints. The genetic algorithm is used to calculate the solution of the multi-dimensional variables. S4: The multi-dimensional variable solution group is sorted according to the preset number of gradients to select the optimal solution in each temperature gradient interval, and the temperature gradient solution group is generated in combination. S5. Calculate the optimal sintering material batching scheme and optimal parameter combination for each temperature gradient interval based on the temperature gradient ungrouping. The optimal sintering material batching scheme includes the optimal sintering material composition ratio, and the optimal parameters include the optimal basicity, optimal reaction temperature, and optimal reaction time. S6 applies the optimal sinter batching scheme and optimal parameter combination to the low-temperature reduction pulverization process, and obtains the low-temperature reduction pulverization index of sinter as needed in the subsequent low-temperature reduction pulverization process and performs feedback adjustment.

2. The method for improving the low-temperature reduction pulverization index of iron ore sinter according to claim 1, characterized in that, S2 includes: S21. Obtain the original chemical composition and proportion of the target sample using gas chromatography-mass spectrometry. S22, construct the original vector based on the original chemical composition ratio of the target sample, and construct the feasible ratio space based on all preset sintering material components; S23. Input the original vector and feasible proportion space into the hyperbolic space of the low-temperature reduction pulverization index of sinter. Use the initial low-temperature reduction pulverization index of sinter as the minimum threshold. Calculate the correlation strength between each preset sinter component ratio and the low-temperature reduction pulverization index of sinter based on the hyperbolic space. Generate a feasible proportion space solution based on the correlation strength of each preset sinter component ratio. S24. A numerical optimization algorithm is used to search for multiple feasible proportioning schemes that satisfy the spatial constraints in the feasible proportioning space. The multiple feasible proportioning schemes are then constructed into a coarse proportioning matrix. In the coarse proportioning matrix, each row represents a feasible proportioning scheme, and each column represents the proportion of a preset sintering material component in each feasible proportioning scheme.

3. The method for improving the low-temperature reduction pulverization index of iron ore sinter according to claim 2, characterized in that, In S23, the initial low-temperature reduction pulverization index of the sinter is used as the minimum threshold. Based on hyperbolic space, the correlation strength between the component ratio of each preset sinter and the low-temperature reduction pulverization index of the sinter is calculated using formulas (1) and (2): （1）； （2）； In formula (1), c represents the correlation strength between the i-th preset sinter composition ratio and the low-temperature reduction pulverization index of the sinter. i This represents the component ratio of the i-th preset sintering material. This indicates the initial low-temperature reduction and pulverization index of the sinter. Represents the Sigmoid function; Let represent the hyperbolic distance between the i-th preset sinter composition ratio and the low-temperature reduction pulverization index of the sinter, and let dist() represent the hyperbolic distance function; In formula (2), arcosh() represents the inverse hyperbolic cosine function, and s j This represents the proportion of the j-th original chemical component. Let m represent the Euclidean distance function, and m represent the m original chemical components of the target sample.

4. A method for improving the low-temperature reduction pulverization index of iron ore sinter according to claim 2 or 3, characterized in that, In step S23, a feasible sizing space solution is generated based on the correlation strength of each preset sintering material component ratio, including: S231, combine each preset sintering material component and its proportion into multiple candidate proportion schemes; S232, if the sum of the correlation strengths of all preset sintering material component ratios in any candidate ratio scheme is greater than the expected correlation strength, then the candidate ratio is filled into the feasible ratio space as a feasible ratio scheme. S233, generate a feasible matching space solution based on all feasible matching schemes in the feasible matching space.

5. The method for improving the low-temperature reduction pulverization index of iron ore sinter according to claim 4, characterized in that, S24 includes: S241, take the feasible matching scheme in the first row of the feasible matching space as the initial solution; S242, taking the initial solution as the starting point of the iteration, iterates each feasible ratio scheme in the feasible ratio space solution, calculates the optimization degree of the feasible ratio scheme in each iteration through the preset evaluation function, and updates the function parameters of the preset evaluation function according to the optimization degree of the feasible ratio scheme in each iteration. S243, if the optimization degree of any feasible ratio scheme is greater than the preset optimization threshold, then the feasible ratio scheme is filled into the coarse ratio matrix. After iterating through all feasible ratio schemes in the feasible ratio space solution, the coarse ratio matrix is ​​generated.

6. The method for improving the low-temperature reduction pulverization index of iron ore sinter according to claim 1, characterized in that, S3 includes: S31 uses the preset sintering material composition ratio, basicity, reaction temperature and reaction time as multi-dimensional variables, and the coarse ratio matrix, the process parameters of the low-temperature reduction pulverization process and the expected low-temperature reduction pulverization index of the sintering ore as constraints to construct a multi-dimensional objective function, and generates multiple original solution vectors based on the multi-dimensional objective function, and encodes each original solution vector into a genome; S32, calculate the fitness of each genome according to the fitness algorithm, and select the function with a fitness greater than the preset fitness among all genomes as the parent generation. Each genome includes the sintering material composition ratio and process parameters. S33, cross-interchange the sintering material composition ratio and process parameters of all the genomes of the parent generation to obtain the new offspring and its genome, return to S32, and end the iteration when the preset number of iterations is completed, and select the genome with the highest fitness in each iteration to generate multi-dimensional variable ungrouping.

7. The method for improving the low-temperature reduction pulverization index of iron ore sinter according to claim 6, characterized in that, In S32, the fitness of the k-th genome is calculated using formula (3) according to the fitness algorithm. : （3）； In formula (3), f() represents a scalar function, and w k t represents the sintering composition ratio of the k-th genome. k Let g represent the process parameters of the k-th genome, n represent the number of genomes, max() represent the maximization operation function, and g k This represents the racial constraint of the k-th genome obtained based on the constraint conditions.

8. The method for improving the low-temperature reduction pulverization index of iron ore sinter according to claim 1, characterized in that, S4 includes: S41, normalize the preset temperature range according to the preset number of gradients to obtain the preset number of temperature gradient ranges. S42, For each temperature gradient interval, select all multi-dimensional variable solutions for the reaction temperature within the temperature gradient interval from the multi-dimensional variable solution set; S43, calculate the comprehensive matching degree of each multi-dimensional variable solution in each temperature gradient interval, and filter the optimal solution in each temperature gradient interval based on the comprehensive matching degree of each multi-dimensional variable solution in each temperature gradient interval and the dynamic threshold of the corresponding temperature gradient interval, and generate a temperature gradient solution group by combining the optimal solutions in each temperature gradient interval.

9. A method for improving the low-temperature reduction pulverization index of iron ore sinter according to claim 8, characterized in that, S43 includes: S431, calculate the dynamic threshold of each temperature gradient interval based on the normal distribution principle of all multidimensional variable solutions in all temperature gradient intervals; S432, calculate the matching degree of each multidimensional variable solution in each temperature gradient interval based on all multidimensional variable solutions in each temperature gradient interval; S433, compare the matching degree of each multidimensional variable solution in each temperature gradient interval with its dynamic threshold. For any temperature gradient interval, select the multidimensional variable solution in the temperature gradient interval whose matching degree is greater than its dynamic threshold and whose matching degree is the maximum value as the optimal solution. If there is no multidimensional variable solution in the temperature gradient interval or no multidimensional variable solution whose matching degree is greater than the dynamic threshold, then perform linear interpolation on the temperature gradient interval to obtain a filled multidimensional variable solution. Take the filled multidimensional variable solution as the optimal solution, and generate a temperature gradient solution group by combining the optimal solutions under each temperature gradient interval.

Images