Sintering ore blending method based on Monte Carlo algorithm

Abstract

The invention discloses a sintering ore blending method based on a Monte Carlo algorithm. The method comprises the following steps: a, obtaining all raw material components and price information for producing sintered ore; b, establishing a finished sinter chemical component prediction model and a sinter cost prediction model; c, raw material ratio constraint conditions and sinter chemical component constraint conditions are set; d, calculating a raw material ratio solution and sinter cost by adopting a Monte Carlo algorithm; and e, sorting the raw material proportioning solutions according to the sinter cost, and outputting the raw material proportioning solution with the lowest sinter cost and the corresponding finished product sinter chemical components to obtain an optimized ore proportioning scheme. According to the method, the finished product sinter chemical component prediction model and the sinter cost prediction model are established on the premise of fully considering the moisture and burning loss in the raw materials, and the raw material proportioning structure with the optimal cost is obtained by adopting the Monte Carlo algorithm, so that the production efficiency is improved, and the production cost of the sinter is remarkably reduced.

Description

Technical Field

[0001] The invention relates to a sintering ore blending method based on a Monte Carlo algorithm, and belongs to the technical field of metal smelting. Background Art

[0002] Sintering blending is a key process of uniformly mixing raw materials such as iron ore powder, flux, fuel, etc. according to a pre-defined ore ratio and generating sintered ore through a sintering process. At present, many companies rely on experience-based guidance to organize sintering production. Specifically, on the basis of mastering the quality of existing ore resources, they make adjustments based on the use effects of various types of ores in previous production and historical blending data, and finally formulate a planned blending with a relatively reasonable expected composition and cost. However, with the continuous fluctuations in the supply and price of imported iron ore powder, the variety of sintering raw materials has become increasingly diverse, the blending system has become increasingly complex, and its performance has become more unstable. These factors have significantly increased the difficulty of sintering blending.

[0003] In this context, the way of adjusting the ore blending structure by relying on people's subjective experience has been difficult to cope with the complex and changeable raw material characteristics and market environment. The invention patent with publication number CN115612833A discloses a method and system for screening iron ore sintering blending, including: presetting multiple blending schemes, wherein the proportion of iron ore species is set to more than two groups, and the proportion of flux and its corresponding fuel is set to more than two groups; extracting all the iron ore proportion groups in the preset blending scheme, and one group of flux and its corresponding fuel proportion group, combined into the first experimental group, and conducting the first assimilability temperature experiment; obtaining the iron ore proportion group with the lowest assimilability temperature in the first experimental group, and the proportion group of all flux and its corresponding fuel in the preset blending scheme, combined into the second experimental group, and conducting the second assimilability temperature experiment and sintering cup experiment; according to the effective sintering temperature area in the sintering cup experiment, screening the iron ore sintering blending scheme. The embodiment of the present invention adopts the effective sintering temperature area as an indicator, which is less affected by factors such as fuel, ventilation, air permeability, and water content, and is more accurate and reliable for guiding the determination of the sintering scheme. However, the existing methods cannot guarantee that the final ratio will reach the lowest cost under the premise of satisfying the rationality of chemical composition. This not only limits the production efficiency of the enterprise, but also hinders the realization of the enterprise's cost reduction and efficiency improvement goals to a certain extent. In addition, in the prior art, the prediction of the chemical composition of sintered ore is usually obtained by multiplying the component content of each type of ore powder by its ratio to obtain a weighted average value as the predicted composition of the mixed ore. However, this method does not fully consider the impact of moisture and burn loss in the raw materials on the final product, resulting in a certain deviation between the predicted results and actual production. Therefore, there is an urgent need for a scientific and systematic ore blending method to meet the challenges currently faced in sintering production. Summary of the invention

[0004] The purpose of the present invention is to provide a sintering ore blending method based on the Monte Carlo algorithm to address the shortcomings of the prior art, thereby ensuring that the chemical composition of the finished sintered ore meets the requirements, improving the production efficiency of the enterprise and reducing the production cost of the sintered ore.

[0005] The problem described in the present invention is solved by the following technical solution:

[0006] A sintering ore blending method based on Monte Carlo algorithm, the method comprising the following steps:

[0007] a. Obtain all raw material composition and price information for sinter production;

[0008] b. Establish a finished sintered ore chemical composition prediction model and a sintered ore cost prediction model based on raw material composition and price information. The finished sintered ore chemical composition prediction model is:

[0009]

[0010] In the formula, C TFe represents the TFe content of sintered ore; C CaO Represents the CaO content of sintered ore; Represents the SiO2 content of sintered ore; C MgO Represents the MgO content of sintered ore; represents the Al2O3 content of sintered ore; R represents the basicity of sintered ore; n is the number of raw materials used to produce sintered ore; X i is the ratio of the i-th raw material, in %; H2O(i) represents the H2O content corresponding to the i-th raw material, in %; LOI(i) represents the burnout corresponding to the i-th raw material, in %; TFe(i) represents the TFe content of the i-th raw material, in %; CaO(i) represents the CaO content of the i-th raw material, in %; SiO2(i) represents the SiO2 content of the i-th raw material, in %; MgO(i) represents the MgO content of the i-th raw material, in %; Al2O3(i) represents the Al2O3 content of the i-th raw material, in %; Csur is the total residual of raw materials, in %; the calculation formula for the total residual of raw materials is:

[0011]

[0012] The sinter cost prediction model is:

[0013]

[0014] Where, F is the cost of sintered ore; price(i) represents the price corresponding to the i-th raw material, in yuan / ton;

[0015] c. Set the constraints of each raw material ratio and the chemical composition of the sintered ore, wherein the constraints of each raw material ratio include the equation constraint and the upper and lower limits of each raw material ratio. The equation constraint is that the sum of the raw material ratios is 100%; the chemical composition constraint of the sintered ore includes the upper and lower limits of the contents of various chemical components in the sintered ore;

[0016] d. Monte Carlo algorithm is used to calculate the raw material ratio solution and sintering cost;

[0017] e. Sort the raw material ratio solutions according to the size of the sintered ore cost, and output the raw material ratio solution with the lowest sintered ore cost and its corresponding finished sintered ore chemical composition to obtain the optimized ore ratio solution.

[0018] The above-mentioned sintering ore blending method based on the Monte Carlo algorithm uses the following specific steps to calculate the raw material blending solution and the sintering ore cost using the Monte Carlo algorithm:

[0019] ① Set the calculation step size and number of iterations for each raw material ratio;

[0020] ② Initialize the storage list of raw material ratio solution and sintering ore cost;

[0021] ③ Randomly generate the ratio value X of the first n-1 kinds of raw materials i , i = 1, 2, 3, ..., n-1, the ratio of each raw material is an integer multiple of the calculation step length, and the random generation formula for the ratio of the first n-1 raw materials is:

[0022]

[0023] Wherein, UB(i) is the upper limit of the ratio of the i-th raw material, in %; LB(i) is the lower limit of the ratio of the i-th raw material, in %; S t is the set calculation step size; rand is a random number between 0 and 1;

[0024] ④ Calculate the ratio value X of the nth raw material according to the equality constraint conditions n :

[0025]

[0026] A raw material ratio solution is formed by the ratio values ​​of n kinds of raw materials;

[0027] ⑤ Input the raw material ratio solution into the sinter ore chemical composition prediction model to calculate the corresponding finished sinter ore chemical composition;

[0028] ⑥ Verify the calculated chemical composition of the finished sintered ore to determine whether it meets all the constraints of the chemical composition of the sintered ore. If it does not meet the constraints of the chemical composition of the sintered ore, the raw material ratio solution is discarded; if it meets the constraints of the chemical composition of the sintered ore, the raw material ratio solution is input into the sintered ore cost prediction model to calculate the corresponding sintered ore cost, and the raw material ratio solution and its sintered ore cost are recorded together in the storage list of raw material ratio solution and sintered ore cost;

[0029] ⑦ Determine whether the number of calculations currently completed has reached the set number of iterations. If not, return to step ③; if reached, output the storage list of raw material ratio solutions and sintered ore costs.

[0030] In the sintering ore matching method based on the Monte Carlo algorithm, the calculation step length S t Set to 0.1 or 1.

[0031] The above-mentioned sintering ore blending method based on the Monte Carlo algorithm should verify the integrity of the data after obtaining all the raw material composition and price information for producing sintered ore. The data that need to be verified for integrity include the material composition information and price information data of iron-containing materials, fuels, and fluxes.

[0032] The above-mentioned sintering ore blending method based on the Monte Carlo algorithm should perform data rationality verification on the set raw material ratio constraints. The specific verification method is: the sum of the lower limit values ​​of all raw material ratios should be less than or equal to 100%, and the sum of the upper limit values ​​of the ratios should be greater than or equal to 100%.

[0033] Beneficial Effects

[0034] The present invention establishes a finished sintered ore chemical composition prediction model and a sintered ore cost prediction model under the premise of fully considering the moisture content and burnt loss in the raw materials, and adopts the Monte Carlo algorithm to obtain the raw material ratio structure with the optimal cost, which not only improves the production efficiency but also significantly reduces the production cost of the sintered ore, and has good practical application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The present invention will be further described below in conjunction with the accompanying drawings.

[0036] Figure 1 is a flow chart of the present invention;

[0037] Figure 2 It is a flow chart for calculating the raw material ratio solution and sintering ore cost using the Monte Carlo algorithm.

[0038] The symbols in this article are: C TFe represents the TFe content of sintered ore; C CaO Represents the CaO content of sintered ore; Represents the SiO2 content of sintered ore; CMgO Represents the MgO content of sintered ore; represents the Al2O3 content of sintered ore; R represents the basicity of sintered ore; n is the number of raw materials used to produce sintered ore; X i is the ratio of the i-th raw material, in %; H2O(i) represents the H2O content corresponding to the i-th raw material, in %; LOI(i) represents the burn loss corresponding to the i-th raw material, in %; TFe(i) represents the TFe content of the i-th raw material, in %; CaO(i) represents the CaO content of the i-th raw material, in %; SiO2(i) represents the SiO2 content of the i-th raw material, in %; MgO(i) represents the MgO content of the i-th raw material, in %; Al2O3(i) represents the Al2O3 content of the i-th raw material, in %; Csur is the total residual raw material, in %; F is the sintered ore cost; price(i) represents the price corresponding to the i-th raw material, in yuan; UB(i) is the upper limit of the i-th raw material ratio, in %; LB(i) is the lower limit of the i-th raw material ratio, in %; S t is the set calculation step size; rand is a random number between 0 and 1. DETAILED DESCRIPTION

[0039] The present invention provides a sintering ore matching method based on Monte Carlo algorithm, which can accurately and effectively calculate an optimized ore matching scheme that meets constraint conditions to meet the needs of actual smelting process and improve the efficiency of blast furnace ironmaking.

[0040] The present invention comprises the following steps:

[0041] a. Obtain all raw material components and price information for sinter production, and perform data integrity verification. All raw material components for sinter production include TFe, CaO, MgO, SiO2, Al2O3, H2O, and burnout. The data that need to be integrity verified include the material composition information and price information data of iron-containing materials, fuels, and fluxes.

[0042] b. Establish a finished sintered ore chemical composition prediction model and a sintered ore cost prediction model based on raw material composition and price information to calculate the finished sintered ore chemical composition and sintered ore cost.

[0043] The chemical composition prediction model of the finished sintered ore is used to calculate the chemical composition of the finished sintered ore, including the TFe content C of the sintered ore. TFe , CaO content of sintered ore C CaO , SiO2 content of sintered ore Sintered ore MgO content C MgO , Al2O3 content of sintered ore The calculation formula of sinter basicity R is:

[0044]

[0045] Where n is the number of raw materials used to produce sintered ore, including flux and fuel; X i is the ratio of the i-th raw material, in %; Csur is the total residual raw material, in %; H2O(i) represents the H2O content corresponding to the i-th raw material, in %; LOI(i) represents the burnout corresponding to the i-th raw material, in %. TFe(i) represents the TFe content of the i-th raw material, in %; CaO(i) represents the CaO content of the i-th raw material, in %; SiO2(i) represents the SiO2 content of the i-th raw material, in %; MgO(i) represents the MgO content of the i-th raw material, in %; Al2O3(i) represents the Al2O3 content of the i-th raw material, in %;

[0046] The calculation formula for the total residual raw material is:

[0047]

[0048] The sinter cost prediction model is used to calculate the sinter cost, including the calculation formula of sinter cost:

[0049]

[0050] Where F is the cost of sintered ore; price(i) represents the price corresponding to the i-th raw material, in yuan.

[0051] c. Set constraints on the ratio of each raw material and the chemical composition of the sintered ore, and perform data rationality verification.

[0052] The raw material ratio constraint conditions include the equation constraint conditions and the upper and lower limits of the raw material ratios. The equation constraint conditions are: the sum of the raw material ratios is 100%; the sinter ore chemical composition constraint conditions include the upper and lower limits of the contents of various chemical components in the sinter ore.

[0053] The constraints of each raw material ratio are defined according to the constraints of the chemical composition of the sintered ore to ensure that the calculated ore blending scheme meets the chemical composition requirements. The equality constraint is used to ensure that the sum of all raw material ratios is 100% to meet the proportion balance in the production process. The upper and lower limits of each raw material ratio are set according to the actual production needs.

[0054] Data rationality check: The ratio constraints of each raw material should be able to make the total ratio reach 100%, that is, the sum of the lower limit values ​​of all raw material ratios should be less than or equal to 100%, and the sum of the upper limit values ​​of the ratios should be greater than or equal to 100%.

[0055] d. The Monte Carlo algorithm is used to calculate the raw material ratio solution and the sintering cost, including the following steps:

[0056] ① Set the calculation step size and number of iterations for each raw material ratio;

[0057] ② Initialize the storage list of raw material ratio solution and sintering ore cost;

[0058] ③ Randomly generate the ratio value X of the first n-1 kinds of raw materials i , i = 1, 2, 3, ..., n-1, ensuring that the ratio of each raw material is an integer multiple of the calculation step length. The formula for randomly generating the ratio of the first n-1 raw materials is:

[0059]

[0060] Wherein, UB(i) is the upper limit of the ratio of the i-th raw material, in %; LB(i) is the lower limit of the ratio of the i-th raw material, in %; S t is the set calculation step size, generally set to 0.1 or 1; rand is a random number between 0 and 1.

[0061] ④ Calculate the ratio value X of the nth raw material according to the equality constraint conditions n , to ensure that the total ratio of n kinds of raw materials is 100%, the calculation formula is:

[0062]

[0063] A raw material ratio solution is formed by the ratio values ​​of n kinds of raw materials;

[0064] ⑤ Input the raw material ratio solution into the sinter ore chemical composition prediction model to calculate the corresponding finished sinter ore chemical composition;

[0065] ⑥ Verify the calculated chemical composition of the finished sintered ore to determine whether it meets all the constraints of the chemical composition of the sintered ore. If it does not meet the constraints of the chemical composition of the sintered ore, the raw material ratio solution is discarded; if it meets the constraints of the chemical composition of the sintered ore, the raw material ratio solution is input into the sintered ore cost prediction model to calculate the corresponding sintered ore cost, and the qualified raw material ratio solution and its sintered ore cost are recorded together in the storage list of raw material ratio solution and sintered ore cost;

[0066] ⑦ Determine whether the number of calculations currently completed has reached the set number of iterations. If not, return to step ③; if reached, output the storage list of raw material ratio solutions and sintered ore costs.

[0067] The calculation step size is set according to the calculation accuracy requirements, and the number of iterations is set by the user according to the convergence speed of the Monte Carlo algorithm and the accuracy of the expected results.

[0068] e. Sort the raw material ratio solutions according to the size of the sintered ore cost, and output the raw material ratio solution with the lowest sintered ore cost and its corresponding finished sintered ore chemical composition to obtain the optimized ore ratio solution.

[0069] Example 1

[0070] This embodiment describes in detail the sintering ore blending method proposed by the present invention based on actual production data of a sintering plant, including the following steps:

[0071] a. Obtain all raw material composition and price information for producing sintered ore, and perform data integrity verification. Currently, a sintering plant uses 18 kinds of raw materials to produce finished sintered ore, including iron ore powder, flux and fuel. The raw material composition and price information are shown in Table 1:

[0072] Table 1 Raw material composition and price information

[0073]

[0074] b. Establish a finished sintered ore chemical composition prediction model and a sintered ore cost prediction model based on raw material composition and price information to calculate the finished sintered ore chemical composition and sintered ore cost.

[0075] The chemical composition of the finished sintered ore includes the TFe content of the sintered ore, C TFe , CaO content of sintered ore C CaO , SiO2 content of sintered ore Sintered ore MgO content C MgO , Al2O3 content of sintered ore The basicity R of sintered ore is calculated as follows:

[0076]

[0077] Total residual C sur The calculation formula is:

[0078]

[0079] The formula for calculating the cost of sintered ore is:

[0080]

[0081] In the formula, X iis the ratio of the i-th raw material, in %; Csur is the total residual raw material, in %; H2O(i) represents the H2O content corresponding to the i-th raw material, in %; LOI(i) represents the burn-out corresponding to the i-th raw material, in %. TFe(i) represents the TFe content of the i-th raw material, in %; CaO(i) represents the CaO content of the i-th raw material, in %; SiO2(i) represents the SiO2 content of the i-th raw material, in %; MgO(i) represents the MgO content of the i-th raw material, in %; Al2O3(i) represents the Al2O3 content of the i-th raw material, in %; price(i) represents the price corresponding to the i-th raw material, in yuan.

[0082] c. Set constraints on the ratio of each raw material and the chemical composition of the sintered ore, and perform data rationality verification.

[0083] The upper and lower limits of the raw material ratios are shown in Table 2, and the constraints on the chemical composition of the sintered ore are shown in Table 3.

[0084] Table 2 Upper and lower limits of each raw material ratio

[0085]

[0086] Table 3 Constraints on chemical composition of sintered ore

[0087] Chemical composition of sintered ore Lower limit Upper limit TFe / % 55 57 CaO / % 10 11 SiO2 / % 5 5.7 Al2O3 / % 2 2.6 MgO / % 2 2.8 Alkalinity R 1.91 1.92

[0088] d. The Monte Carlo algorithm is used to calculate the raw material ratio solution and the sintering cost, including the following steps:

[0089] ① Set the calculation step size and number of iterations for each raw material ratio; the chemical composition of the sintered ore calculated by the sintered ore chemical composition prediction model must meet the chemical composition constraints of the sintered ore, that is, the upper and lower limits of the contents of various chemical components set in advance. The sum of all raw material ratios must be equal to 100%; the upper and lower limits of each raw material ratio specified in step c constitute the boundary conditions of each raw material ratio. Based on the actual material bin unloading control capability on site, the calculation step size of each raw material ratio is set to 0.1. At the same time, in order to ensure the optimization of the calculation efficiency, the preset number of iterations is 1000 times.

[0090] ② Initialize the storage list of raw material ratio solutions and sintered ore costs to facilitate recording and management of raw material ratio solutions and sintered ore costs in the subsequent optimization process.

[0091] ③ Based on the set boundary conditions, randomly generate the ratio values ​​of the first 17 raw materials and ensure that these values ​​are integer multiples of 0.1. The calculation formula is as follows:

[0092]

[0093] Wherein, UB(i) is the upper limit of the ratio of the i-th raw material, in %, i = 1, 2, 3, ..., 17; LB(i) is the lower limit of the ratio of the i-th raw material, in %, S t is the set calculation step size, generally set to 0.1 or 1; rand is a random number between 0 and 1.

[0094] ④ Calculate the ratio value X of the 18th raw material according to the equality constraint conditions 18 To ensure that the sum of all ratios is equal to 100%, the calculation formula is:

[0095]

[0096] ⑤ Input the raw material ratio solution into the sinter ore chemical composition prediction model to calculate the corresponding finished sinter ore chemical composition;

[0097] ⑥ Verify the calculated chemical composition of the finished sintered ore to determine whether it meets all the constraints of the chemical composition of the sintered ore. If it does not meet the constraints of the chemical composition of the sintered ore, the raw material ratio solution is discarded; if it meets the constraints of the chemical composition of the sintered ore, the raw material ratio solution is input into the sintered ore cost prediction model to calculate the corresponding sintered ore cost, and the qualified raw material ratio solution and its sintered ore cost are recorded together in the storage list of raw material ratio solution and sintered ore cost;

[0098] ⑦ Determine whether the number of calculations currently completed has reached the set number of iterations. If not, return to step ③; if reached, output the storage list of raw material ratio solutions and sintered ore costs.

[0099] e. Sort the raw material ratio solutions according to the size of the sintered ore cost, and output the raw material ratio solution with the lowest sintered ore cost and its corresponding finished sintered ore chemical composition to obtain the optimized ore ratio solution.

[0100] The optimized ore blending scheme is compared with the benchmark ore blending scheme calculated by the empirical method. The comparison of raw material proportions is shown in Table 4, and the comparison of chemical composition of finished sintered ore is shown in Table 5.

[0101] Table 4 Comparison of raw material ratios

[0102]

[0103] Table 5 Comparison of chemical composition and cost of finished sintered ore

[0104] Chemical composition and cost of sintered ore Benchmark ore allocation plan Optimized ore blending plan TFe / % 55 55.2 CaO / % 10.79 10.73 SiO2 / % 5.65 5.62 Al2O3 / % 2.66 2.62 MgO / % 2.53 2.51 Alkalinity R 1.91 1.91 Sinter cost 889.08 871.18

[0105] The present invention establishes a chemical composition prediction model for finished sintered ore and a sintered ore cost prediction model based on the chemical composition information and price information of raw materials. Under the condition of fully considering the moisture and burnout in the raw materials, the raw material ratio and the sintered ore composition constraints, the Monte Carlo algorithm is used to accurately calculate the chemical composition and sintered ore cost of the finished sintered ore, thereby obtaining the raw material ratio structure with the best cost. This method not only improves production efficiency, but also significantly reduces production costs, and has good practical application value.

Claims

1. A sintering ore blending method based on Monte Carlo algorithm, characterized in that: The method comprises the following steps: a. Obtain all raw material composition and price information for sinter production; b. Establish a finished sintered ore chemical composition prediction model and a sintered ore cost prediction model based on raw material composition and price information. The finished sintered ore chemical composition prediction model is: In the formula, C TFe represents the TFe content of sintered ore; C CaO Represents the CaO content of sintered ore; Represents the SiO2 content of sintered ore; C MgO Represents the MgO content of sintered ore; represents the Al2O3 content of sintered ore; R represents the basicity of sintered ore; n is the number of raw materials used to produce sintered ore; X i is the ratio of the i-th raw material, in %; H2O(i) represents the H2O content corresponding to the i-th raw material, in %; LOI(i) represents the burnout corresponding to the i-th raw material, in %; TFe(i) represents the TFe content of the i-th raw material, in %; CaO(i) represents the CaO content of the i-th raw material, in %; SiO2(i) represents the SiO2 content of the i-th raw material, in %; MgO(i) represents the MgO content of the i-th raw material, in %; Al2O3(i) represents the Al2O3 content of the i-th raw material, in %; Csur is the total residual of raw materials, in %; the calculation formula for the total residual of raw materials is: The sinter cost prediction model is: Where F is the cost of sintered ore; price(i) represents the price corresponding to the i-th raw material, in yuan; c. Set the constraints of each raw material ratio and the chemical composition of the sintered ore, wherein the constraints of each raw material ratio include the equation constraint and the upper and lower limits of each raw material ratio. The equation constraint is that the sum of the raw material ratios is 100%; the chemical composition constraint of the sintered ore includes the upper and lower limits of the contents of various chemical components in the sintered ore; d. Monte Carlo algorithm is used to calculate the raw material ratio solution and sintering cost; e. Sort the raw material ratio solutions according to the size of the sintered ore cost, and output the raw material ratio solution with the lowest sintered ore cost and its corresponding finished sintered ore chemical composition to obtain the optimized ore ratio solution.

2. The sintering ore blending method based on Monte Carlo algorithm according to claim 1 is characterized in that: The specific steps of using the Monte Carlo algorithm to calculate the raw material ratio solution and sintering cost are as follows: ① Set the calculation step size and number of iterations for each raw material ratio; ② Initialize the storage list of raw material ratio solution and sintering ore cost; ③ Randomly generate the ratio value X of the first n-1 kinds of raw materials i , i = 1, 2, 3, ..., n-1, the ratio of each raw material is an integer multiple of the calculation step length, and the random generation formula for the ratio of the first n-1 raw materials is: Wherein, UB(i) is the upper limit of the ratio of the i-th raw material, in %; LB(i) is the lower limit of the ratio of the i-th raw material, in %; S t is the set calculation step size; rand is a random number between 0 and 1; ④ Calculate the ratio value X of the nth raw material according to the equality constraint conditions n : A raw material ratio solution is formed by the ratio values ​​of n kinds of raw materials; ⑤ Input the raw material ratio solution into the sinter ore chemical composition prediction model to calculate the corresponding finished sinter ore chemical composition; ⑥ Verify the calculated chemical composition of the finished sintered ore to determine whether it meets all the constraints of the chemical composition of the sintered ore. If it does not meet the constraints of the chemical composition of the sintered ore, the raw material ratio solution is discarded; If the chemical composition constraint of the sintered ore is met, the raw material ratio solution is input into the sintered ore cost prediction model to calculate the corresponding sintered ore cost, and the raw material ratio solution and its sintered ore cost are recorded together in the storage list of raw material ratio solution and sintered ore cost; ⑦ Determine whether the number of calculations currently completed has reached the set number of iterations. If not, return to step ③; if reached, output the storage list of raw material ratio solutions and sintered ore costs.

3. The sintering ore blending method based on Monte Carlo algorithm according to claim 1 is characterized in that: The calculation step size S t Set to 0.1 or 1.

4. The sintering ore blending method based on Monte Carlo algorithm according to claim 1 is characterized in that: After obtaining all raw material composition and price information for the production of sintered ore, the integrity of the data should be verified. The data that need to be verified for integrity include the material composition information and price information data of iron-containing materials, fuels, and fluxes.

5. The sintering ore blending method based on Monte Carlo algorithm according to claim 1 is characterized in that: Data rationality verification should be performed on the set raw material ratio constraints. The specific verification method is: the sum of the lower limit values ​​of all raw material ratios should be less than or equal to 100%, and the sum of the upper limit values ​​of the ratios should be greater than or equal to 100%.

Images