A method for improving the low temperature reduction disintegration index of iron ore sinter

Abstract

The present application relates to a kind of methods for improving iron ore sinter low temperature reduction pulverization index, belong to iron ore sinter processing technical field.The method includes: collecting finished iron ore sinter as target sample, obtaining the initial sinter low temperature reduction pulverization index of target sample, original chemical composition and its proportion;Hyperbolic space of sinter low temperature reduction pulverization index is constructed, and coarse proportioning matrix is calculated;According to coarse proportioning matrix, multi-dimensional variable solution group is calculated using genetic algorithm;According to preset gradient quantity, the optimal solution under each temperature gradient interval is filtered in multi-dimensional variable solution group, and temperature gradient solution group is generated comprehensively;According to temperature gradient solution group, the optimal sintering material proportioning scheme and the combination of optimal parameters under each temperature gradient interval are calculated, and the optimal sintering material proportioning scheme and the combination of optimal parameters are applied in low temperature reduction pulverization process.The present application improves control precision, and improves the improvement effect of iron ore sinter low temperature reduction pulverization index.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of iron ore sinter processing, and particularly relates to a method for improving low-temperature reduction and pulverization indexes of iron ore sinter. BACKGROUND

[0002] Currently, the quality of iron ore sinter is required not only in physical and chemical properties, but also in metallurgical properties. The degree of reduction of iron of the iron ore sinter is a basic metallurgical property, and the low-temperature reduction and pulverization property is an important metallurgical property. The low-temperature reduction and pulverization process is used in the metal reduction reaction process of the iron ore sinter, and the low-temperature reduction and pulverization indexes of the sinter reflect the reduction strength of the iron ore sinter in the upper part of the blast furnace. If the low-temperature reduction and pulverization index value is high, the permeability of the column of materials in the upper part of the blast furnace is good, the furnace top clogging can be reduced, the utilization rate of CO in the coal gas is increased, the smelting intensity is good, the coke ratio can be reduced, and the pig iron yield is high.

[0003] The traditional low-temperature reduction and pulverization process adjusts the low-temperature reduction and pulverization indexes of the sinter by adjusting the reaction temperature, the reaction time, the basicity or adding sintered materials as a combustion-supporting agent. However, these methods are often adjusted from only one direction, without considering the mutual relationship among the reaction temperature, the reaction time and the basicity, resulting in that the adjustment method of the low-temperature reduction and pulverization process of the sinter is unreasonable, the control precision is low, and the improvement effect of the low-temperature reduction and pulverization indexes of the sinter is poor. SUMMARY

[0004] To solve the above technical problems, the present application provides a method for improving the low-temperature reduction and pulverization indexes of the iron ore sinter. The technical scheme of the present application is as follows:

[0005] S1, collecting the finished iron ore sinter on the finished product belt conveyor, obtaining the target sample through screening and size reduction, and obtaining the initial low-temperature reduction and pulverization indexes of the sinter of the target sample;

[0006] S2, obtaining the original chemical composition and the proportion of the target sample, and constructing a hyperbolic space of the low-temperature reduction and pulverization indexes of the sinter in combination with all the preset sintered material compositions, and calculating the coarse proportioning matrix with the initial low-temperature reduction and pulverization indexes of the sinter as the lowest threshold value;

[0007] S3, taking the proportioning of the preset sintered material composition, the basicity, the reaction temperature and the reaction time as the multi-dimensional variables, taking the coarse proportioning matrix, the process parameters of the low-temperature reduction and pulverization process and the expected low-temperature reduction and pulverization indexes of the sinter as the constraint conditions, and calculating the multi-dimensional variable solution set by using the genetic algorithm;

[0008] S4, screening the optimal solution in each temperature gradient interval according to the preset gradient number, and comprehensively generating the temperature gradient solution set;

[0009] S5, calculating the optimal sintering material proportioning scheme and the optimal parameter combination in each temperature gradient interval according to the temperature gradient decomposition, wherein the optimal sintering material proportioning scheme comprises optimal sintering material component proportioning, and the optimal parameter comprises optimal basicity, optimal reaction temperature and optimal reaction time;

[0010] S6, applying the optimal sintering material proportioning scheme and the optimal parameter combination to the low-temperature reduction and pulverization process, and obtaining the sinter low-temperature reduction and pulverization index and performing feedback adjustment in the subsequent low-temperature reduction and pulverization process.

[0011] Preferably, the S2 comprises:

[0012] S21, obtaining the original chemical components and their proportions of the target sample according to the gas chromatography-mass spectrometry analysis method;

[0013] S22, constructing an original vector according to the original chemical component proportions of the target sample, and constructing a feasible proportioning space according to all preset sintering material components;

[0014] S23, inputting the original vector and the feasible proportioning space into the hyperbolic space of the sinter low-temperature reduction and pulverization index, taking the initial sinter low-temperature reduction and pulverization index as the minimum threshold, calculating the correlation strength of each preset sintering material component proportion and the sinter low-temperature reduction and pulverization index according to the hyperbolic space, and generating a feasible proportioning space solution according to the correlation strength of each preset sintering material component proportion;

[0015] S24, searching for a plurality of feasible proportioning schemes that meet the space constraint conditions in the feasible proportioning space solution through a numerical optimization algorithm, and constructing a coarse proportioning matrix from the plurality of feasible proportioning schemes, wherein each row of the coarse proportioning matrix represents a feasible proportioning scheme, and each column represents the proportion of a preset sintering material component in each feasible proportioning scheme.

[0016] Preferably, in the S23, the correlation strength of each preset sintering material component proportion and the sinter low-temperature reduction and pulverization index is calculated according to the hyperbolic space through formula (1) and formula (2) with the initial sinter low-temperature reduction and pulverization index as the minimum threshold:

[0017] (1);

[0018] (2);

[0019] In formula (1), represents the correlation strength of the i-th preset sintering material component proportion and the sinter low-temperature reduction and pulverization index, c i represents the i-th preset sintering material component proportion, represents the initial sinter low-temperature reduction and pulverization index, represents the Sigmoid function; represents the hyperbolic distance of the i th preset sintering material component proportion and the low-temperature reduction pulverization index of the sinter, dist() represents a hyperbolic distance function;

[0020] In formula (2), arcosh() represents an inverse hyperbolic cosine function, s j represents the j th original chemical component proportion, represents an Euclidean distance function, m represents that the target sample has m original chemical components.

[0021] Preferably, in the S23, generating the feasible proportioning space solution according to the correlation strength of each preset sintering material component proportion comprises:

[0022] S231, combining each preset sintering material component and its proportion into a plurality of candidate proportioning schemes;

[0023] S232, if the sum of the correlation strengths of all preset sintering material component proportions in any candidate proportioning scheme is greater than the expected correlation strength, the candidate proportioning is filled into the feasible proportioning space as a feasible proportioning scheme;

[0024] S233, generating the feasible proportioning space solution according to all feasible proportioning schemes in the feasible proportioning space.

[0025] Preferably, the S24 comprises:

[0026] S241, taking the feasible proportioning scheme in the first row in the feasible proportioning space solution as an initial solution;

[0027] S242, taking the initial solution as an iteration starting point, iterating each feasible proportioning scheme in the feasible proportioning space solution, calculating the optimization degree of each iteration of the feasible proportioning scheme through the preset evaluation function, and updating the function parameters of the preset evaluation function according to the optimization degree of each iteration of the feasible proportioning scheme;

[0028] S243, if the optimization degree of any feasible proportioning scheme is greater than a preset optimization threshold, the feasible proportioning scheme is filled into the coarse proportioning matrix, and after iterating all feasible proportioning schemes in the feasible proportioning space solution, a coarse proportioning matrix is generated.

[0029] Preferably, the S3 comprises:

[0030] S31, taking the preset sintering material component proportion, the basicity, the reaction temperature and the reaction time as multi-dimensional variables, taking the coarse proportioning matrix, the process parameters of the low-temperature reduction pulverization process and the expected low-temperature reduction pulverization index of the sinter as constraint conditions, constructing a multi-dimensional objective function, and generating a plurality of original solution vectors according to the multi-dimensional objective function, and encoding each original solution vector into a gene group;

[0031] S32, calculate the fitness of each genome according to the fitness algorithm, and select the function with the fitness greater than a preset fitness as a parent from all the genomes, wherein each genome comprises a proportion of sintering material components and process parameters;

[0032] S33, cross and exchange the proportions of sintering material components and process parameters of all the genomes of the parents to obtain new offspring and their genomes, return to S32, and end the iteration when a preset iteration number is reached, and select the genome with the highest fitness in each iteration to generate a multi-dimensional variable solution group.

[0033] Preferably, the fitness of the kth genome in S32 is calculated according to the fitness algorithm by formula (3) :

[0034] (3);

[0035] In formula (3), f() represents a scalar function, w k represents the proportion of sintering material components of the kth genome, t k represents the process parameters of the kth genome, n represents the number of genomes, max() represents a maximization operation function, g k represents the racial constraint of the kth genome obtained according to the constraint condition.

[0036] Preferably, S4 comprises:

[0037] S41, normalize a preset temperature interval according to a preset number of gradients to obtain a preset number of temperature gradient intervals;

[0038] S42, for each temperature gradient interval, select all multi-dimensional variable solutions with the reaction temperature in the temperature gradient interval from the multi-dimensional variable solution group;

[0039] S43, calculate the comprehensive matching degree of each multi-dimensional variable solution in each temperature gradient interval, and select the optimal solution under each temperature gradient interval according to 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 comprehensively selecting the optimal solution under each temperature gradient interval.

[0040] Preferably, S43 comprises:

[0041] S431, calculate the dynamic threshold of each temperature gradient interval according to all multi-dimensional variable solutions in all temperature gradient intervals according to the normal distribution principle;

[0042] S432, calculate the matching degree of each multi-dimensional variable solution in each temperature gradient interval according to all multi-dimensional variable solutions in each temperature gradient interval;

[0043] S433, comparing the matching degree of each multi-dimensional variable solution of each temperature gradient interval with its dynamic threshold, for any temperature gradient interval, selecting the multi-dimensional variable solution with the maximum matching degree greater than its dynamic threshold in the temperature gradient interval as the optimal solution, and if there is no multi-dimensional variable solution or no multi-dimensional variable solution with the matching degree greater than the dynamic threshold in the temperature gradient interval, performing linear interpolation on the temperature gradient interval to obtain a filled multi-dimensional variable solution, taking the filled multi-dimensional variable solution as the optimal solution, and generating a temperature gradient solution group by synthesizing the optimal solutions under each temperature gradient interval.

[0044] All the optional technical solutions described above can be combined arbitrarily, and the present application does not describe the structures after one-by-one combination in detail.

[0045] Through the above-mentioned scheme, the beneficial effects of the present application are as follows:

[0046] By constructing a hyperbolic space of a sinter low temperature reduction pulverization index according to the original chemical composition of a target sample of iron ore sinter and all preset sinter material compositions, a coarse proportioning matrix is calculated; the preset sinter material composition proportion, the alkalinity, the temperature and the time are taken as multi-dimensional variables, and the coarse proportioning matrix, the process parameters and the preset sinter low temperature reduction pulverization index are taken as constraint conditions to calculate a multi-dimensional variable solution group; the multi-dimensional variable solution group is filtered according to a preset gradient number to obtain the optimal solution under each temperature gradient interval, and a temperature gradient solution group is generated by synthesis, the combination of the optimal sinter material proportioning scheme and the optimal parameters is calculated according to the temperature gradient solution group, and the original sinter low temperature reduction pulverization scheme is adjusted, thereby providing a method for improving the sinter low temperature reduction pulverization index of iron ore, which comprehensively considers the process parameters and the sinter material composition proportion, introduces the concept of temperature gradient, and comprehensively considers the mutual relationship of the reaction temperature, the reaction time and the alkalinity, thereby providing a reasonable adjustment method for the low temperature reduction pulverization index based on multi-dimensional variables, and the control is combined with the temperature gradient, thereby improving the control precision and increasing the improvement effect of the sinter low temperature reduction pulverization index of iron ore.

[0047] The above description is only a summary of the technical solutions of the present application, in order to more clearly understand the technical means of the present application, and the content of the specification can be implemented as follows. BRIEF DESCRIPTION OF DRAWINGS

[0048] Figure 1 is a flowchart of a method for improving the sinter low temperature reduction pulverization index of iron ore provided by an embodiment of the present application.

[0049] Figure 2 is a schematic diagram of the generation of a child gene group by crossing and exchanging a parent gene group in the genetic algorithm of an embodiment of the present application. DETAILED DESCRIPTION

[0050] The specific embodiments of the present application are described in further detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present application but are not intended to limit the scope of the present application.

[0051] In combination with the above, as shown in the method for improving the low-temperature reduction disintegration index of iron ore sintered ore provided by the embodiments of the present application can be realized through the following steps S1 to S6: Figure 1

[0052] S1, collecting finished iron ore sintered ore on a finished product belt conveyor, obtaining a target sample through screening and size reduction, and obtaining the initial low-temperature reduction disintegration index of the target sample.

[0053] Specifically, when collecting the finished iron ore sintered ore, attention should be paid to the following: the material should not be taken on the base material belt conveyor, and the finished iron ore sintered ore should be taken on the finished product belt conveyor. After screening the finished iron ore sintered ore with a size of +12.5 mm and a weight of 15 kg, the finished iron ore sintered ore is poured into a medium crushing crusher for crushing. The 10-12.5 mm size fraction is obtained through screening, and the sample amount of not less than 3 kg is obtained through uniform mixing and size reduction as the target sample. Then, the target sample is sent to the pre-iron laboratory for detection to obtain the initial low-temperature reduction disintegration index of the target sample.

[0054] S2, obtaining the original chemical composition and its proportion of the target sample, and constructing a hyperbolic space of the low-temperature reduction disintegration index of the sintered ore in combination with all the preset sintering material compositions, and calculating the coarse proportioning matrix with the initial low-temperature reduction disintegration index of the sintered ore as the lowest threshold.

[0055] Specifically, the fundamental reason for the low-temperature reduction disintegration of the sintered ore is that the phase transition temperature of Fe2O3 to Fe3O4 plays an important role in the formation of regenerated Fe2O3. Any composition that can increase the phase transition temperature of Fe2O3 to Fe3O4 is conducive to the generation of regenerated Fe2O3, such as TiO2, Al2O3, K2O, Na2O, etc. Any composition that can reduce the phase transition temperature of Fe2O3 to Fe3O4 is not conducive to the generation of regenerated Fe2O3, such as CaO, SiO2, MgO, FeO, etc. Through the scatter plot, the linear correlation method is used to obtain the main factors affecting the low-temperature reduction disintegration index of the sintered ore: the low-temperature reduction disintegration index of the sintered ore has a relatively significant correlation with FeO, TiO2, Al2O3, K2O, Na2O, and has a certain influence on CaO, SiO2, MgO, but the linear correlation is not strong. Therefore, in the embodiments of the present application, the preset sintering material compositions are selected as TiO2, Al2O3, K2O, and Na2O.

[0056] In one specific embodiment, the S2 includes:

[0057] ​S21, obtaining the original chemical components and their proportions of the target sample according to a gas chromatography-mass spectrometry analysis method.

[0058] In the specific implementation of this step, first, the target sample is separated by a gas chromatograph (GC), and the separated components are detected according to different retention times to obtain different groups. Then, all the groups are put into a mass spectrometer (MS), and the mass spectrum of the target sample is generated through ionization and mass analysis of the mass spectrometer. Different ion peaks in the mass spectrum correspond to different components in the target sample, and qualitative analysis is performed by comparison with a standard spectral library to obtain the original chemical components of the target sample. At the same time, 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 components and their proportions of the target sample are obtained.

[0059] S22, constructing an original vector according to the original chemical component proportions of the target sample, and constructing a feasible proportioning space according to all the preset sintering material components.

[0060] For example, if the preset sintering material components include A, B, C, and D, a feasible proportioning space containing these four preset sintering material components is constructed according to A, B, C, and D, and the proportions of the four preset sintering material components are filled into the feasible proportioning space. The feasible proportioning space after filling the proportions of the preset sintering material components can generate different preset sintering material component proportions based on the four preset sintering material components and their proportions.

[0061] S23, inputting the original vector and the feasible proportioning space into a hyperbolic space of the sinter low-temperature reduction disintegration index, taking the initial sinter low-temperature reduction disintegration index as the lowest threshold, calculating the correlation strength of each preset sintering material component proportion and the sinter low-temperature reduction disintegration index according to the hyperbolic space, and generating a feasible proportioning space solution according to the correlation strength of each preset sintering material component proportion.

[0062] In a specific embodiment, in S23, the initial sinter low-temperature reduction disintegration index is taken as the lowest threshold, and the correlation strength of each preset sintering material component proportion and the sinter low-temperature reduction disintegration index is calculated according to the hyperbolic space through formulas (1) and (2):

[0063] (1);

[0064] (2);

[0065] In formula (1), represents the correlation strength of the i-th preset sintering material component proportion and the sinter low-temperature reduction disintegration index, c i represents the i-th preset sintering material component proportion, represents the initial sinter low-temperature reduction disintegration index, represents a Sigmoid function; represents the hyperbolic distance between the ith preset sintering material composition ratio and the sinter low temperature reduction disintegration index, dist() represents a hyperbolic distance function;

[0066] In formula (2), arcosh() represents an inverse hyperbolic cosine function, s j represents the jth original chemical composition ratio, represents an Euclidean distance function, m represents that the target sample has m original chemical compositions.

[0067] Specifically, the hyperbolic distance in this embodiment 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 into the interval of (0, 1) through the Sigmoid function, and if the hyperbolic distance of any preset sintering material composition ratio is close to r0, the correlation strength is close to 0; this relationship shows that the closer the preset sintering material composition ratio is to the minimum threshold value, the smaller the correlation strength between the preset sintering material composition ratio and the sinter low temperature reduction disintegration index. Formula (2) maps the original chemical composition ratio and each preset sintering material composition ratio into the hyperbolic space, and the smaller the hyperbolic distance, the closer the overall structure of the preset sintering material composition ratio c i to the original chemical composition ratio s j , and the more reasonable the sinter low temperature reduction disintegration index obtained according to the preset sintering material composition ratio.

[0068] In a specific embodiment, in the S23, generating a feasible proportioning space solution according to the correlation strength of each preset sintering material composition ratio comprises:

[0069] S231, combining each preset sintering material composition and its ratio into a plurality of candidate proportioning schemes.

[0070] Specifically, the sum of all preset sintering material composition ratios of each candidate proportioning scheme is 100%.

[0071] S232, if the sum of the correlation strengths of all preset sintering material composition ratios in any candidate proportioning scheme is greater than the expected correlation strength, the candidate proportioning scheme is filled into the feasible proportioning space as a feasible proportioning scheme.

[0072] S233, generating a feasible proportioning space solution according to all feasible proportioning schemes in the feasible proportioning space.

[0073] Specifically, each row in the feasible proportioning space solution represents a feasible proportioning scheme, and all feasible proportioning schemes are filled into the feasible proportioning space according to their preset sintering material compositions and proportions to obtain the feasible proportioning space solution. The expected correlation strength is a threshold value of the correlation strength obtained according to experience.

[0074] S24, searching for a plurality of feasible proportioning schemes satisfying the space constraint conditions in the feasible proportioning space solution by a numerical optimization algorithm, and constructing the plurality of feasible proportioning schemes into a coarse proportioning matrix, wherein each row of the coarse proportioning matrix represents a feasible proportioning scheme, and each column indicates the proportion of a preset sintering material component in each feasible proportioning scheme.

[0075] Specifically, the space constraint conditions include a preset sintering material component proportion range and a minimum correlation strength between the preset sintering material component proportion and the low-temperature reduction pulverization index of the sinter.

[0076] In a specific embodiment, the S24 includes:

[0077] S241, taking the feasible proportioning scheme in the first row in the feasible proportioning space solution as an initial solution.

[0078] Specifically, the ranking of all feasible proportioning schemes in the feasible proportioning space solution has no precedence, and the feasible proportioning scheme in the first row is selected as the initial solution for the convenience of subsequent iterations.

[0079] S242, taking the initial solution as the iteration starting point, iterating each feasible proportioning scheme in the feasible proportioning space solution, calculating the optimization degree of each iteration of the feasible proportioning scheme by a preset evaluation function, and updating the function parameters of the preset evaluation function according to the optimization degree of each iteration of the feasible proportioning scheme.

[0080] Wherein, the function parameters of the preset evaluation function include the preset weight of each preset sintering material component, the space constraint condition and the preset sintering material component ideal proportion, the preset evaluation function is composed of the function parameters, and in the embodiment of the application, the preset evaluation function can be realized by minimizing the objective function, which adjusts the ideal proportion of each preset sintering material component according to the optimization degree of each iteration of the feasible proportioning scheme.

[0081] S243, if the optimization degree of any feasible proportioning scheme is greater than a preset optimization threshold, the feasible proportioning scheme is filled into the coarse proportioning matrix, and after iterating all feasible proportioning schemes in the feasible proportioning space solution, the coarse proportioning matrix is generated.

[0082] Wherein, the preset optimization threshold is an optimization threshold set according to a large amount of historical experimental data.

[0083] S3, taking the preset sintering material component proportion, the basicity, the reaction temperature and the reaction time as multi-dimensional variables, taking the coarse proportioning matrix, the process parameters of the low-temperature reduction pulverization process and the expected low-temperature reduction pulverization index of the sinter as constraint conditions, and using a genetic algorithm to calculate a multi-dimensional variable solution group.

[0084] According to the national standard, the low-temperature reduction disintegration index of the sinter is expected to be not less than 65%.

[0085] Specifically, one of the fundamental reasons for the low-temperature reduction disintegration of the sinter is that the volume expands when Fe2O3 is reduced to Fe3O4 or FeO, stress is released, and crack propagation is intensified to cause disintegration. The content of CaO in the low-temperature reduction disintegration process represents the basicity in the process parameters. For example, if the basicity of the sinter is increased from 1.83±0.05 to 1.95±0.05 by adding CaO in the low-temperature reduction disintegration process, the low-temperature reduction disintegration index of the sinter is finally greatly improved, so the basicity in the process parameters has a great influence on the low-temperature reduction disintegration index of the sinter. Therefore, the basicity is taken as a variable in the embodiment of the present application.

[0086] In addition, the second fundamental reason for the low-temperature reduction disintegration of the sinter is that the lattice transformation occurs when the regenerated skeleton-like hematite is transformed from α-Fe2O3 to γ-Fe2O3, internal stress is generated, and the sinter is broken and disintegrated during the extrusion and collision process. The main reason for the formation of skeleton Fe2O3 is high sintering temperature and fast cooling speed. Therefore, the reaction temperature and reaction time in the process parameters are one of the important factors affecting the low-temperature reduction disintegration index of the sinter, so the reaction temperature and reaction time are also taken as variables in the embodiment of the present application.

[0087] In a specific embodiment, the S3 comprises:

[0088] S31, taking the preset sintering material component proportion ratio, basicity, reaction temperature and reaction time as multi-dimensional variables, taking the coarse proportioning matrix, the process parameters of the low-temperature reduction disintegration process and the expected low-temperature reduction disintegration index of the sinter as constraint conditions, constructing a multi-dimensional objective function, and generating a plurality of original solution vectors according to the multi-dimensional objective function, and encoding each original solution vector into a gene group.

[0089] The multi-dimensional objective function is a kind of objective function constructed by taking the preset sintering material component proportion ratio, basicity, reaction temperature and reaction time as multi-dimensional variables, and taking the coarse proportioning matrix, the process parameters of the low-temperature reduction disintegration process and the expected low-temperature reduction disintegration index of the sinter as constraint conditions, aiming to meet the constraint conditions at the same time to obtain a plurality of original solution vectors.

[0090] For example, assuming that an original solution vector includes: preset sintering material component proportion ratio: [0.3, 0.5, 0.2], basicity: 1.5, reaction temperature: 550℃, reaction time 50min, then after real number coding of the original solution vector, the gene group is represented as [(0.3, 0.5, 0.2), (1.5, 550, 50)].

[0091] S32, calculate the fitness of each genome according to the fitness algorithm, and select the function with the fitness greater than a preset fitness in all genomes as a parent, wherein each genome comprises a proportion of sintering material components and a process parameter.

[0092] Specifically, the fitness of the kth genome in the S32 is calculated according to the fitness algorithm by formula (3) :

[0093] (3).

[0094] In formula (3), f() represents a scalar function, w k represents the proportion of sintering material components of the kth genome, t k represents the process parameter of the kth genome, n represents the number of genomes, max() represents a maximization operation function, g k represents the racial constraint of the kth genome obtained according to the constraint condition.

[0095] Specifically, in formula (3), the scalar function is a target term, which is used to quantify the performance of the genome in the entire population; is a constraint penalty term, which is used to dynamically penalize all genomes that violate the constraint and reduce their performance in the entire population.

[0096] S33, cross and exchange the proportion of sintering material components and the process parameter of all genomes of the parent to obtain new offspring and their genomes, return to S32, and end the iteration when the preset iteration number is iterated, and select the highest fitness genome in each iteration to generate a multi-dimensional variable solution set.

[0097] For example, the process of cross and exchanging genomes of the parent 1 and the parent 2 to generate the offspring 1 and the offspring 2 is as shown in Figure 2 In addition, the preset iteration number is generally 600-1000 times.

[0098] S4, screen the optimal solution in each temperature gradient interval according to a preset gradient number, and comprehensively generate a temperature gradient solution set;

[0099] In one specific embodiment, the S4 comprises:

[0100] S41, normalize a preset temperature interval according to a preset gradient number to obtain N temperature gradient intervals.

[0101] Specifically, assuming that the preset temperature interval is [T1, T2] and the preset gradient number is N, then the temperature range of the jth temperature gradient interval after normalization is: .

[0102] S42, for each temperature gradient interval, screening all multi-dimensional variable solutions with the reaction temperature within the temperature gradient interval from the multi-dimensional variable solution group.

[0103] Specifically, one multi-dimensional variable solution includes four variables of preset sintering material ingredient proportion, basicity, reaction temperature and reaction time.

[0104] S43, calculating the comprehensive matching degree of each multi-dimensional variable solution in each temperature gradient interval, and screening the optimal solution in each temperature gradient interval according to 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 generating a temperature gradient solution group by comprehensively generating the optimal solution in each temperature gradient interval.

[0105] Specifically, the single-variable matching degree of each multi-dimensional variable in each multi-dimensional variable solution in each temperature gradient interval is calculated using the Euclidean distance, and finally the sum of the single-variable matching degrees of all multi-dimensional variables is taken as the comprehensive matching degree of each multi-dimensional variable solution. Wherein, for the preset sintering material ingredient proportion in the multi-dimensional variable solution, the single-variable matching degree of each preset sintering material ingredient proportion is calculated by weighting and adding the preset ingredient weight and the preset sintering material ingredient proportion, and the preset ingredient weight of each preset sintering material ingredient is determined by experience.

[0106] In one specific embodiment, the S43 comprises:

[0107] S431, calculating the dynamic threshold of each temperature gradient interval according to the normal distribution principle and all multi-dimensional variable solutions in all temperature gradient intervals.

[0108] In specific implementation, the single-variable dynamic threshold of each multi-dimensional variable in each multi-dimensional variable solution in each temperature gradient interval is calculated according to the normal distribution principle and all multi-dimensional variable solutions in each temperature gradient interval, and the dynamic threshold of each temperature gradient interval is the sum of the single-variable thresholds of each multi-dimensional variable in the temperature gradient interval. For example, if the mean value of the preset sintering material ingredient proportion in a temperature gradient interval is 0.85, the standard deviation is 0.05, and the constant value is 1, then the single-variable threshold of the preset sintering material ingredient proportion in the temperature gradient interval is 0.85+1×0.05=0.9.

[0109] Still taking the above example as an example, the single-variable threshold of the preset sintering material ingredient proportion in the temperature gradient interval is 0.9, the single-variable threshold of the basicity is 0.1, the single-variable threshold of the reaction temperature is 0.5, and the single-variable threshold of the reaction time is 0.6, then the dynamic threshold of the temperature gradient interval is 0.9+0.1+0.5+0.6=2.1.

[0110] S432, calculating the matching degree of each multi-dimensional variable solution in each temperature gradient interval according to all multi-dimensional variable solutions in each temperature gradient interval.

[0111] S433, comparing the matching degree of each multi-dimensional variable solution in each temperature gradient interval with its dynamic threshold, for any temperature gradient interval, selecting the multi-dimensional variable solution with the largest matching degree greater than its dynamic threshold in the temperature gradient interval as the optimal solution, and if there is no multi-dimensional variable solution or no multi-dimensional variable solution with the matching degree greater than the dynamic threshold in the temperature gradient interval, performing linear interpolation on the temperature gradient interval to obtain a filled multi-dimensional variable solution, taking the filled multi-dimensional variable solution as the optimal solution, and generating a temperature gradient solution set by synthesizing the optimal solutions in each temperature gradient interval.

[0112] Specifically, for any temperature gradient interval without multi-dimensional variable solution or without multi-dimensional variable solution with the matching degree greater than the dynamic threshold, calculating the single variable linear interpolation result of each multi-dimensional variable in the temperature gradient interval, and combining the single variable linear interpolation results of all multi-dimensional variables into the filled multi-dimensional variable solution of the temperature gradient interval. Wherein, for a certain temperature gradient interval without multi-dimensional variable solution or without multi-dimensional variable solution with the matching degree greater than the dynamic threshold, the embodiment of the present application calculates the mean value of each multi-dimensional variable on both sides of the temperature gradient interval as the single variable linear interpolation result in the temperature gradient interval.

[0113] S5, calculating the optimal sintering material batching scheme and the combination of optimal parameters in each temperature gradient interval according to the temperature gradient solution set, wherein the optimal sintering material batching scheme includes the optimal sintering material component ratio, and the optimal parameters include the optimal basicity, the optimal reaction temperature and the optimal reaction time.

[0114] Specifically, according to the numerical relationship of the optimal reaction temperature and the optimal reaction time, the sinter can be ensured to be fully sintered while avoiding the deterioration of mineral structure and the decline of reduction performance caused by excessively high temperature and excessively long time.

[0115] S6, applying the combination of the optimal sintering material batching scheme and the optimal parameters to the low-temperature reduction and pulverization process, and obtaining the sinter low-temperature reduction and pulverization index in the subsequent low-temperature reduction and pulverization process and performing feedback adjustment.

[0116] Specifically, by applying the optimal sintering material batching scheme and the optimal parameters, the low-temperature reduction disintegration index of the sinter is significantly improved. The test shows that the method makes the low-temperature reduction disintegration index of the sinter reach more than 70%, effectively reducing the fragmentation and disintegration phenomenon of the sinter in the low-temperature reduction process. Moreover, due to the improvement of the low-temperature reduction disintegration index and the optimization of the quality of the sinter, the permeability of the material column in the blast furnace is significantly improved, the gas distribution is more uniform, the furnace condition is more stable, the problems such as deterioration of the permeability of the furnace charge, fluctuation or abnormality of the furnace condition caused by sinter disintegration are reduced, the risk of production accidents is reduced, and the production efficiency and safety of the blast furnace are improved.

[0117] In addition, in the actual application process, a specific area is selected on the sintering production line, small-scale trial production is carried out according to the combination of the optimal sintering material batching scheme and the optimal parameters, a large amount of actual production data is collected, the combination of the optimal sintering material batching scheme and the optimal parameters is further verified and optimized, the optimized sintering material batching scheme and the optimized parameter combination are obtained, and the stability and operability in the actual industrial environment are focused on. The optimized sintering material batching scheme and the optimized parameter combination are applied to the whole production line, the low-temperature reduction disintegration index of the sinter, the utilization coefficient of the blast furnace, the coke ratio and other operation indexes are continuously tracked and monitored, the technical scheme is adjusted in time according to the feedback information, and long-term stable good effects are ensured.

[0118] According to the above embodiment, the present application obtains the coarse proportioning matrix by constructing the hyperbolic space of the low-temperature reduction disintegration index of the sinter according to the original chemical composition of the target sample and the proportion thereof and all preset sintering material compositions, fully considers the influence degree of the proportioning of the preset sintering material compositions on the low-temperature reduction disintegration index of the sinter, and provides a data basis for subsequent multi-dimensional adjustment of the low-temperature reduction disintegration index of the sinter. Further, the preset sintering material composition proportion, the alkalinity, the reaction temperature and the reaction time are taken as multi-dimensional variables, the multi-dimensional variable solution in each temperature gradient interval is obtained as the combination of the optimal sintering material batching scheme and the optimal parameters by using the genetic algorithm and the temperature gradient screening method, and the combination of the optimal sintering material batching scheme and the optimal parameters is applied to actual production. By comprehensively considering the process parameters and the proportioning of the sintering material compositions, introducing the concept of the temperature gradient, and comprehensively considering the mutual relationship of the reaction temperature, the reaction time and the alkalinity, a reasonable adjustment method based on multi-dimensional variables is provided for the low-temperature reduction disintegration index, and the control precision is improved by combining the temperature gradient, and the improvement effect of the low-temperature reduction disintegration index of the iron ore sinter is increased.

[0119] The above only describes the preferred embodiments of the present application and is not used to limit the present application. It should be noted that, for ordinary skilled persons in the technical field, several improvements and modifications can be made without departing from the technical principles of the present application, and these improvements and modifications should be considered as the protection scope of the present application.

Claims

1. A method of improving the low temperature reduction disintegration index of iron ore sinter, characterised in that, The method comprises the following steps: S1, collecting finished iron ore sinter on a finished product belt conveyor, obtaining a target sample after screening and sub-division, and obtaining an initial sinter low-temperature reduction index of the target sample; S2, obtaining the original chemical composition and proportion of the target sample, and constructing a hyperbolic space of the sinter low-temperature reduction index combined with all preset sinter material compositions, and calculating a coarse proportioning matrix with the initial sinter low-temperature reduction index as the lowest threshold value; S3, taking the preset sinter material composition proportion, the basicity, the reaction temperature and the reaction time as multi-dimensional variables, taking the coarse proportioning matrix, the process parameters of the low-temperature reduction process and the expected sinter low-temperature reduction index as constraint conditions, and calculating a multi-dimensional variable solution set by using a genetic algorithm; S4, screening the optimal solution in each temperature gradient interval according to a preset gradient number, and comprehensively generating a temperature gradient solution set; S5, calculating the optimal sinter material proportioning scheme and the optimal parameter combination in each temperature gradient interval according to the temperature gradient solution set, wherein the optimal sinter material proportioning scheme comprises an optimal sinter material composition proportion, and the optimal parameters comprise an optimal basicity, an optimal reaction temperature and an optimal reaction time; S6, applying the optimal sinter material proportioning scheme and the optimal parameter combination to the low-temperature reduction process, and obtaining the sinter low-temperature reduction index in the subsequent low-temperature reduction process as needed and performing feedback adjustment.

2. A method of improving low temperature reduction degradation index of iron ore sinter as claimed in claim 1 wherein, The S2 comprises the following steps: S21, obtaining the original chemical composition and proportion of the target sample by using a gas chromatography-mass spectrometry analysis method; S22, constructing an original vector according to the original chemical composition proportion of the target sample, and constructing a feasible proportioning space according to all preset sinter material compositions; S23, inputting the original vector and the feasible proportioning space into the hyperbolic space of the sinter low-temperature reduction index, taking the initial sinter low-temperature reduction index as the lowest threshold value, calculating the correlation intensity of each preset sinter material composition proportion and the sinter low-temperature reduction index according to the hyperbolic space, and generating a feasible proportioning space solution according to the correlation intensity of each preset sinter material composition proportion; S24, searching for a plurality of feasible proportioning schemes that meet the space constraint conditions in the feasible proportioning space solution by using a numerical optimization algorithm, and constructing the plurality of feasible proportioning schemes into a coarse proportioning matrix, wherein each row in the coarse proportioning matrix represents a feasible proportioning scheme, and each column represents the proportion of one preset sinter material composition in each feasible proportioning scheme.

3. A method of improving low temperature reduction degradation index of iron ore sinter as claimed in claim 2 wherein, In the S23, the correlation intensity of each preset sinter material composition proportion and the sinter low-temperature reduction index is calculated according to the hyperbolic space by using formula (1) and formula (2) with the initial sinter low-temperature reduction index as the lowest threshold value: （1）； （2）； In formula (1), represents the correlation strength of the i-th preset sinter material composition ratio and the low-temperature reduction pulverization index of the sinter, c i represents the i-th preset sinter material composition ratio, represents the low-temperature reduction pulverization index of the initial sinter, represents a Sigmoid function; represents the hyperbolic distance of the i-th preset sinter material composition ratio and the low-temperature reduction pulverization index of the sinter, dist() represents a hyperbolic distance function; In formula (2), arcosh( ) represents an inverse hyperbolic cosine function, s j represents the jth original chemical component ratio, represents a Euclidean distance function, and m represents that the target sample has m original chemical components.

4. A method of improving the low temperature reduction disintegration index of iron ore sinter according to claim 2 or 3, characterised in that, In the S23, the feasible proportioning space solution is generated according to the correlation intensity of each preset sinter material composition proportion, which comprises the following steps: S231, combining each preset sinter material composition and its proportion into a plurality of candidate proportioning schemes; S232, if the sum of the correlation intensities of all preset sinter material composition proportions in any candidate proportioning scheme is greater than the expected correlation intensity, the candidate proportioning scheme is filled into the feasible proportioning space as a feasible proportioning scheme; S233, generating a feasible proportioning space solution according to all feasible proportioning schemes in the feasible proportioning space.

5. A method of improving low temperature reduction disintegration index of iron ore sinter as claimed in claim 4 wherein, The S24 comprises: S241, taking the feasible proportioning scheme in the first row of the feasible proportioning space solution as an initial solution; S242, taking the initial solution as an iteration starting point, iterating each feasible proportioning scheme in the feasible proportioning space solution, calculating the optimization degree of each iteration of the feasible proportioning scheme by a preset evaluation function, and updating the function parameters of the preset evaluation function according to the optimization degree of each iteration of the feasible proportioning scheme; S243, if the optimization degree of any feasible proportioning scheme is greater than a preset optimization threshold, filling the feasible proportioning scheme into the coarse proportioning matrix, and generating the coarse proportioning matrix after iterating all feasible proportioning schemes in the feasible proportioning space solution.

6. A method of improving low temperature reduction degradation index of iron ore sinter as claimed in claim 1 wherein, The S3 comprises: S31, taking the preset sintering material component proportioning, basicity, reaction temperature and reaction time as multi-dimensional variables, taking the coarse proportioning matrix, the process parameters of the low-temperature reduction and pulverization process and the expected sinter low-temperature reduction and pulverization index as constraint conditions, constructing a multi-dimensional objective function, and generating a plurality of original solution vectors according to the multi-dimensional objective function, and encoding each original solution vector into a gene group; S32, calculating the fitness of each gene group according to the fitness algorithm, and selecting the functions with fitness greater than a preset fitness from all gene groups as parents, wherein each gene group comprises sintering material component proportioning and process parameters; S33, crossing and exchanging the sintering material component proportioning and process parameters of all gene groups of the parents to obtain new offspring and their gene groups, returning to S32 until the iteration is completed for a preset iteration number, and selecting the gene group with the highest fitness in each iteration to generate a multi-dimensional variable solution group.

7. A method of improving low temperature reduction disintegration index of iron ore sinter as claimed in claim 6 wherein, The fitness of the kth genome is calculated according to the fitness algorithm by formula (3) in S32 : （3）； In Equation (3), f() represents a scalar function, w k represents the composition ratio of the sintered material of the kth genome, t k represents the process parameter of the kth genome, n represents the number of genomes, max() represents a maximization operation function, g k represents the race constraint of the kth genome obtained according to the constraint condition.

8. A method of improving low temperature reduction degradation index of iron ore sinter as claimed in claim 1 wherein, The S4 comprises: S41, normalizing the preset temperature interval according to a preset gradient number to obtain a preset gradient number of temperature gradient intervals; S42, for each temperature gradient interval, selecting all multi-dimensional variable solutions with the reaction temperature in the temperature gradient interval from the multi-dimensional variable solution group; S43, calculating the comprehensive matching degree of each multi-dimensional variable solution in each temperature gradient interval, and selecting the optimal solution under each temperature gradient interval according to 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 generating a temperature gradient solution group by comprehensively selecting the optimal solution under each temperature gradient interval.

9. A method of improving low temperature reduction disintegration index of iron ore sinter as claimed in claim 8 wherein, The S43 comprises: S431, calculating the dynamic threshold of each temperature gradient interval according to all multi-dimensional variable solutions in all temperature gradient intervals by adopting the normal distribution principle; S432, calculating the matching degree of each multi-dimensional variable solution in each temperature gradient interval according to all multi-dimensional variable solutions in each temperature gradient interval; S433, comparing the matching degree of each multi-dimensional variable solution of each temperature gradient interval with its dynamic threshold value, for any temperature gradient interval, selecting the multi-dimensional variable solution with the maximum matching degree greater than its dynamic threshold value in the temperature gradient interval as the optimal solution, and if there is no multi-dimensional variable solution or no multi-dimensional variable solution with the matching degree greater than the dynamic threshold value in the temperature gradient interval, performing linear interpolation on the temperature gradient interval to obtain a filled multi-dimensional variable solution, taking the filled multi-dimensional variable solution as the optimal solution, and generating a temperature gradient solution set by comprehensively considering the optimal solution under each temperature gradient interval.

Images