Sinter ore blending quality control risk evaluation method and system

Abstract

The present application belongs to the technical field of steel smelting, and relates to a sinter ore blending quality control risk evaluation method and system; the method comprises the following steps: calculating a blending scheme based on boundary constraint conditions and a raw material performance database and outputting prediction data; after preprocessing the prediction data, inputting the prediction data into a risk evaluation system to perform a predictive risk pre-evaluation, and obtaining a pre-evaluation comprehensive score; if the comprehensive score does not reach a preset qualified threshold, adjusting the blending ratio and cyclically iterating until the qualified threshold is reached, locking the scheme, and then performing sintering production; after stable production, collecting actual production data, substituting the actual production data into the risk evaluation system to perform a verification risk evaluation, and obtaining an actual comprehensive score; comparing the pre-evaluation comprehensive score with the actual comprehensive score, and optimizing the risk evaluation system according to a deviation result. The present application realizes pre-positioned prediction of blending, automatic iteration of unqualified products, verification feedback after production, and self-adaptive optimization of the risk evaluation system, and improves the scientificity and reliability of quality control.

Description

Technical Field

[0001] This invention belongs to the field of iron and steel smelting technology, specifically relating to a method and system for risk assessment of sinter blending quality control. Background Technology

[0002] In existing technologies, sintering ore blending and quality risk control mainly rely on the testing of finished ore after production. The significant limitations of this technology are: delayed risk assessment, as the ore blending stage cannot predict sintering and blast furnace risks, leading to raw material waste and production fluctuations; a disconnect between ore blending and evaluation, with no automatic correction for substandard evaluations, lacking an iterative closed loop; an incomplete evaluation system, focusing only on the chemical composition of the finished product while ignoring the performance of the mixture, fuels, and operating procedures, and lacking specific rules and quantitative scoring standards for substandard ore; and a missing data preprocessing stage, resulting in outliers and distorted evaluations. These are the shortcomings of existing technologies.

[0003] Therefore, there is an urgent need for a method that can perform online risk prediction and comprehensive performance quantification of ore blending schemes, automatically clean abnormal data, and correct ore blending strategies based on the evaluation results. Summary of the Invention

[0004] In a first aspect, embodiments of this application provide a method for risk assessment of sinter blending quality control, comprising the following steps: S1. Based on boundary constraints and the raw material performance database, calculate the blending ratio of raw materials to generate the current blending scheme; based on the current blending scheme, call the raw material performance database to perform physical property calculations and output predicted data; the current blending scheme includes at least the material ratio of each raw material; the predicted data includes at least the predicted chemical composition content and predicted process parameters. S2, Construct a risk assessment system; After preprocessing the predicted data, input it into the risk assessment system along with the current ore blending scheme to conduct a predictive risk pre-assessment and obtain a comprehensive pre-assessment score; S3, if the comprehensive score of the pre-evaluation does not reach the preset qualified threshold, then locate the risk indicators, trace the key raw materials, adjust the proportions according to the direction of exceeding the standard and the preset step size, and perform linkage compensation to generate a correction plan and new prediction data, and return to step S2 to perform predictive risk pre-evaluation again, iterating until the comprehensive score of the pre-evaluation reaches the preset qualified threshold; if the comprehensive score of the pre-evaluation reaches the preset qualified threshold, sintering production is carried out according to the prediction data and the current ore blending plan; S4. After production stabilizes, collect actual production data, substitute the actual production data into the risk assessment system described in step S2 to conduct a confirmatory risk assessment, and calculate the actual comprehensive score. S5 compares the preliminary comprehensive score with the actual comprehensive score, analyzes the deviation, and optimizes the predictive risk preliminary assessment process based on the deviation results.

[0005] Furthermore, the specific steps of step S1 are as follows: S11, Obtain boundary constraints, which include: raw material inventory data, sinter quality index requirements, raw material cost ceiling, harmful element control limits, and mixture process control range; S12, call the raw material performance database, take the boundary constraints as the solution objective, and take the raw material performance database as the calculation basis. Use the linear programming algorithm to solve the ratio of each raw material that satisfies all constraints and generate the current ore blending scheme; the database stores the chemical composition, loss on ignition, particle size distribution, price and harmful element content data of each raw material. S13. Based on the proportions of each raw material in the current ore blending scheme, and combined with the chemical composition and physical property data of each raw material in the raw material performance database, perform physical property calculations and output predicted data. The predicted data includes: predicted chemical component content and predicted process parameter values; The predicted chemical composition content includes at least: alkalinity value, TFe content, SiO2 content, MgO content, and Al2O3 content; the predicted process parameter values ​​include at least: mixture moisture value, average particle size value, and CaO / TFe molar ratio value.

[0006] Furthermore, the specific steps of step S2 are as follows: S21, adopts 3 The principle is to standardize the prediction data, remove outliers, and then normalize the removed data to obtain preprocessed prediction data. S22. A risk assessment system is constructed based on the sintering process principle. The risk assessment system includes five evaluation dimensions: chemical composition of sintered ore, performance of mixed materials, performance of fuel and auxiliary materials, operating system, and proportion of non-mainstream ore. The risk assessment system evaluates the predicted data and the current ore blending scheme based on the corresponding risk indicator standard values ​​and risk deviation thresholds pre-configured for the five evaluation dimensions. S23, input the preprocessed prediction data and the current ore blending scheme into the risk assessment system to obtain the risk level of each risk indicator in each dimension; the risk level includes: no risk level, low risk level or high risk level. S24. Based on the risk level of each dimension, a corresponding deduction coefficient is assigned to perform basic weighted scoring. Among them, the high risk level corresponds to the first deduction weight, the low risk level corresponds to the second deduction weight, and the no-risk level has a deduction of zero. The predicted values ​​of the risk indicators of each dimension are weighted and summed according to the corresponding deduction weight to obtain the deduction value. The deduction value is deducted with a maximum score of 100 points to obtain the basic score. S25, deduct points specifically for non-mainstream mineral types from the basic score to obtain the preliminary comprehensive score.

[0007] Furthermore, the specific steps of step S23 are as follows: The preprocessed predicted data and the current ore blending scheme are input into the risk assessment system. The data required for the risk indicators in each dimension are extracted and compared with the preset risk indicator standard values ​​to calculate the degree of deviation and obtain the indicator deviation value. The indicator deviation value is then compared with the risk deviation threshold. The risk deviation threshold includes: no risk deviation threshold, low risk deviation threshold, and high risk deviation threshold. If the deviation value of an indicator falls within the risk-free deviation threshold range, the indicator is determined to be at a risk-free level; if the deviation value falls within the low-risk deviation threshold range, it is determined to be at a low-risk level; if the deviation value falls within the high-risk deviation threshold range, it is determined to be at a high-risk level.

[0008] Furthermore, in step S23, when the predicted data and the current ore blending scheme are input into the risk assessment system, index matching is performed using data tags as an index: The predicted chemical composition values ​​correspond to the chemical composition dimension of sintered ore, the predicted process parameter values ​​correspond to the performance dimension of the mixture, the performance dimension of the fuel and auxiliary materials, and the operating system dimension, and the raw material ratio values ​​in the current ore blending scheme correspond to the non-mainstream ore ratio dimension.

[0009] Furthermore, the specific deduction for non-mainstream mines in step S25 includes: S251, obtain the actual proportion of each non-mainstream mineral in the current ore blending scheme, wherein the non-mainstream minerals include at least Sierra Leone fines, laterite nickel ore, Malaysian fines and Indian fines; S252, based on the actual proportions of each non-mainstream mineral type, specific deductions will be implemented according to the preset deduction rules: Sierra Leone powder: Deduct 0.5 points for every 1% added; Laterite nickel ore: 0.5 points will be deducted for every 0.2% added; Malaysian rice noodles: Deduct 0.5 points for every 0.2% added; Indian powder: Deduct 0.2 points for every 1% added; S253, add up the deductions for each item to get the total deduction for non-mainstream mines; S254. Subtract the total deduction for the non-mainstream mine project from the base score to obtain the preliminary evaluation comprehensive score.

[0010] Furthermore, the specific steps of step S3 are as follows: S31, determine whether the pre-evaluation comprehensive score reaches the preset qualification threshold: If the comprehensive score of the pre-evaluation is greater than or equal to the preset qualified threshold, the current ore blending scheme is determined to be qualified, the current ore blending scheme and the corresponding prediction data are locked, sintering production is executed, and the process proceeds to step S4. If the overall score of the pre-evaluation is less than the preset qualified threshold, the current ore blending scheme is deemed unqualified, and S32 to S34 are executed. S32. Based on the risk level determination results of each dimension of risk indicators in step S2, locate the risk indicators that have not reached the risk-free level and their respective dimensions, and use correlation analysis to backtrack and determine the key raw materials that cause the risk indicators to exceed the standard. S33, determine the direction of increase or decrease of key raw materials based on the index deviation value in step S23: if the index deviation value is high, reduce the proportion of the raw material; if it is low, increase the proportion of the raw material; adjust in increments of 5% to 10% of the current proportion of the key raw material. After adjusting the key raw materials, the proportions of each raw material are recalculated to generate a revised ore blending scheme and output new forecast data. S34. Input the revised ore blending scheme and new prediction data into the risk assessment system of step S2, and conduct another predictive risk pre-assessment. Iterate in cycles until the comprehensive score of the pre-assessment reaches the preset qualified threshold.

[0011] Furthermore, the specific steps of step S4 are as follows: S41, organize sintering production according to the ore blending scheme locked in step S3. When the sintering production operating parameters have been running continuously and stably for a preset time, it is determined that the production has entered a stable state. S42, During the stable production phase, actual production data is collected according to a preset sampling cycle; the actual production data includes: actual chemical component content data, actual process parameter data, and actual material ratio data; the actual chemical component content data includes at least alkalinity, content, content, content, The content, the actual process parameter data includes at least the moisture content of the mixture, the thickness of the material layer, the vertical sintering speed, and the ignition temperature; S43, adopts 3 The principle is to preprocess the collected actual production data, remove outliers, and normalize the removed data to obtain preprocessed actual production data. S44. Substitute the preprocessed actual production data into the risk assessment system described in step S2, and perform confirmatory risk assessment and scoring according to steps S22 to S25 to obtain the actual comprehensive score.

[0012] Furthermore, the specific steps of step S5 are as follows: S51: Compare the preliminary comprehensive score with the actual comprehensive score and calculate the difference to obtain the comprehensive deviation value; analyze the comprehensive deviation value: if the comprehensive deviation value exceeds the preset deviation range, it is determined that there is a deviation in the risk assessment system, and S52 is executed to optimize the risk assessment system; if the comprehensive deviation value does not exceed the preset deviation range, the current risk assessment system parameters remain unchanged, and all the ore blending plan, forecast data, actual production data, preliminary comprehensive score and actual comprehensive score are archived for subsequent ore blending reference. S52, compare the deviation values ​​of the indicators for each dimension in step S23 with the actual deviation values ​​of the indicators for each dimension obtained by inputting the actual production data into the risk assessment system in step S44, calculate the risk score deviation value for each dimension, and locate one or more dimensions with the largest deviation values ​​for adjustment: For the chemical composition dimension of sintered ore, the deviation values ​​of each risk indicator are compared with the actual deviation values. Based on the direction and magnitude of the deviation of the individual risk indicator score, the preset risk deviation threshold of that risk indicator is adjusted accordingly: if the deviation value of a certain risk indicator is higher than the actual deviation value, the risk deviation threshold of that indicator is appropriately relaxed; if it is lower than the actual deviation value, the risk deviation threshold of that indicator is appropriately tightened. The adjustment range is determined based on the historical deviation statistical average. For the dimensions of mixture performance, fuel and auxiliary material performance, and operating procedures, the deduction weight of each dimension is adjusted according to the direction and magnitude of the risk score deviation: when the actual risk index score is lower than the predicted risk index score, the deduction weight of that dimension is increased; when the actual risk index score is higher than the predicted risk index score, the deduction weight of that dimension is decreased; the adjustment range is determined based on the ratio of the deviation of the score of that dimension to the overall deviation. For non-mainstream minerals, if a non-mainstream mineral type is used and the overall deviation value is related to the non-mainstream minerals, the unit deduction coefficient for that non-mainstream mineral type will be proportionally adjusted based on the ratio of the actual overall score to the pre-evaluation overall score. S53 will apply the revised risk deviation threshold, the deduction weights of each dimension, and the deduction coefficients of the non-mainstream mine project to the subsequent predictive risk pre-evaluation; and will enter the deviation data of this dual evaluation into the deviation database to drive subsequent cumulative statistical optimization.

[0013] Secondly, embodiments of this application also provide a sinter ore blending quality control risk assessment system, including: The ore blending scheme calculation module is used to calculate the blending ratio of raw materials based on boundary constraints and a raw material performance database, and generate the current ore blending scheme; based on the current ore blending scheme, it calls the raw material performance database to perform physical property calculations and outputs predicted data; the current ore blending scheme includes at least the material ratio of each raw material; the predicted data includes at least the predicted chemical composition content and predicted process parameters. The risk pre-assessment module is used to construct a risk assessment system. After preprocessing the predicted data, it is input into the risk assessment system along with the current ore blending scheme to conduct a predictive risk pre-assessment and obtain a comprehensive pre-assessment score. The proportioning iteration optimization module is used to locate risk indicators, trace key raw materials, adjust the proportions according to the direction of exceeding the standard and the preset step size, and perform linkage compensation when the comprehensive score of the pre-evaluation does not reach the preset qualified threshold. It generates a correction plan and new prediction data, and returns to the risk pre-evaluation module to perform predictive risk pre-evaluation again. The process is iterated until the comprehensive score of the pre-evaluation reaches the preset qualified threshold. If the comprehensive score of the pre-evaluation reaches the preset qualified threshold, sintering production is carried out according to the prediction data and the current ore blending plan. The on-site performance verification module is used to collect actual production data after production stabilizes, and then substitute the actual production data into the risk assessment system described in the risk pre-assessment module to conduct a confirmatory risk assessment and calculate the actual comprehensive score. The self-optimization module of the evaluation system is used to compare the pre-evaluation comprehensive score with the actual comprehensive score, analyze the deviation, and optimize the predictive risk pre-evaluation process based on the deviation results.

[0014] As can be seen from the above technical solutions, this application has the following advantages: By using predictive risk pre-assessment, sintering and blast furnace risks can be predicted at the ore blending stage, transforming risk assessment from post-event detection to pre-event prediction, avoiding raw material waste and production fluctuations, and solving the problem of delayed risk assessment. Through iterative processes, the ore blending scheme is first pre-evaluated. If the evaluation is unsatisfactory, the unsatisfactory items are automatically identified, the causes are located, and the scheme is corrected accordingly. Then, the evaluation is re-evaluated, and adjustments are made multiple times until it is satisfactory, thereby eliminating the defects of ore blending and evaluation being disconnected and lacking iterative feedback. With the help of a risk assessment system covering five dimensions and a special deduction mechanism for non-mainstream minerals, covering chemical composition, mixture performance, fuel and auxiliary material performance, operating procedures, and risk management of non-mainstream minerals, it makes up for the shortcomings of traditional methods, such as single evaluation dimensions and lack of special rules and quantitative scoring standards for inferior minerals. Through 3 The principle of data preprocessing eliminates outlier interference, avoiding the problem of outliers being mixed in and causing evaluation distortion due to the lack of data preprocessing, thus ensuring the objectivity and accuracy of the evaluation results. Furthermore, this application achieves full-process control from ore blending to production and then to the risk assessment system, improving the scientific nature and reliability of sinter quality control. Attached Figure Description

[0015] To more clearly illustrate the technical solution of this application, the accompanying drawings used in the description will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a flowchart of the risk assessment method for quality control of sintered ore blending provided by the present invention.

[0017] Figure 2 This is a schematic diagram of the risk assessment system for quality control of sintered ore blending provided by the present invention.

[0018] Among them, 1-ore blending scheme calculation module, 2-risk pre-evaluation module, 3-ratio iterative optimization module, 4-on-site performance verification module, and 5-evaluation system self-optimization module. Detailed Implementation

[0019] The various embodiments of this disclosure will be described more fully in the following detailed description of the specific steps involved in the risk assessment for sinter blending quality control. This disclosure may have various embodiments, and adjustments and changes may be made therein. However, it should be understood that there is no intention to limit the various embodiments of this disclosure to the specific embodiments disclosed herein, but rather this disclosure should be understood to cover all adjustments, equivalents, and / or alternatives falling within the spirit and scope of the various embodiments of this disclosure.

[0020] This embodiment provides a risk assessment method for the quality control of sintered ore blending. Through a dual evaluation approach of predictive pre-evaluation and verification evaluation, risk assessment is completed during the ore blending scheme formulation stage, avoiding the waste of raw materials caused by post-event testing. At the same time, by using quantitative scoring and a special deduction mechanism for non-mainstream ores, the accuracy of the evaluation and the adaptive capability of the model are improved.

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] Please see Figure 1 The diagram shows a flowchart of a risk assessment method for sinter blending quality control in a specific embodiment. The method includes the following steps: S1. Based on boundary constraints and the raw material performance database, calculate the blending ratio of raw materials to generate the current blending scheme; based on the current blending scheme, call the raw material performance database to perform physical property calculations and output predicted data; the current blending scheme includes at least the material ratio of each raw material; the predicted data includes at least the predicted chemical composition content and predicted process parameters. S2, Construct a risk assessment system; After preprocessing the predicted data, input it into the risk assessment system along with the current ore blending scheme to conduct a predictive risk pre-assessment and obtain a comprehensive pre-assessment score; S3, if the comprehensive score of the pre-evaluation does not reach the preset qualified threshold, then locate the risk indicators, trace the key raw materials, adjust the proportions according to the direction of exceeding the standard and the preset step size, and perform linkage compensation to generate a correction plan and new prediction data, and return to step S2 to perform predictive risk pre-evaluation again, iterating until the comprehensive score of the pre-evaluation reaches the preset qualified threshold; if the comprehensive score of the pre-evaluation reaches the preset qualified threshold, sintering production is carried out according to the prediction data and the current ore blending plan; S4. After production stabilizes, collect actual production data, substitute the actual production data into the risk assessment system described in step S2 to conduct a confirmatory risk assessment, and calculate the actual comprehensive score. S5 compares the preliminary comprehensive score with the actual comprehensive score, analyzes the deviation, and optimizes the predictive risk preliminary assessment process based on the deviation results.

[0023] This embodiment achieves a shift in risk management from passive post-event detection to proactive pre-judgment by employing a dual approach of pre-evaluation during ore blending and post-production verification evaluation, coupled with an automatic iterative correction mechanism for non-compliant schemes. This eliminates the shortcomings of ore blending and evaluation being separate processes. Furthermore, by utilizing quantitative scoring and special deduction rules for non-mainstream ores, it addresses the issues of vague evaluation standards and lack of control over substandard ores, avoiding raw material waste and production fluctuations caused by the commissioning of non-compliant schemes, and improving the accuracy of evaluation and the adaptive capability of the model.

[0024] In one embodiment of the present invention, based on step S1, the following will provide a possible embodiment and describe its specific implementation in a non-limiting manner.

[0025] The specific steps of step S1 are as follows: S11, Obtain boundary constraints, which include: raw material inventory data, sinter quality index requirements, raw material cost ceiling, harmful element control limits, and mixture process control range; For example, the quality requirements for the sintered ore are: basicity 1.95~2.05, TFe≥53.5%, SiO2 5.57%~5.95%; the upper limit of raw material cost is: Indian powder ≤2%; the control limit of harmful elements is: alkali metals ≤0.20%; the process control range of the mixture is: moisture 7.5%~8.5%; S12, call the raw material performance database, take the boundary constraints as the solution objective, and take the raw material performance database as the calculation basis. Use the linear programming algorithm to solve the ratio of each raw material that satisfies all constraints and generate the current ore blending scheme; the database stores the chemical composition, loss on ignition, particle size distribution, price and harmful element content data of each raw material. For example, the initial ore blending scheme generated after solving the problem using a linear programming algorithm is as follows: Mac coal powder 23.7%, superfine powder 11.2%, coarse SSFT coal 9.5%, Indian coal powder 2.0%, quicklime 12.0%, sintered anthracite 4.5%, and cold return ore 18.0%. S13. Based on the proportions of each raw material in the current ore blending scheme, and combined with the chemical composition and physical property data of each raw material in the raw material performance database, perform physical property calculations and output predicted data. The predicted data includes: predicted chemical component content and predicted process parameter values; The predicted chemical composition content includes at least: alkalinity value, TFe content, SiO2 content, MgO content, and Al2O3 content; the predicted process parameter values ​​include at least: mixture moisture value, average particle size value, and CaO / TFe molar ratio value. For example, the output predicted data includes: alkalinity 1.98, 5.90%, TFe 53.8%, MgO 1.95%, 1.85%, moisture content of the mixture is 8.1%, and the average particle size is 3.2 mm.

[0026] It should be noted that steps S11 to S13 upgrade manual experience-based ore blending to intelligent quantitative algorithm-driven calculation. By using linear programming, the optimal blending ratio that satisfies multi-dimensional boundary constraints is solved in one go, avoiding the shortcomings of traditional trial-and-error methods, such as low efficiency and easy deviation from the constraint boundaries. The introduction of the raw material performance database ensures that the blending ratio calculation results are always based on real physical property data, providing accurate and complete data input for subsequent predictive risk pre-evaluation, and laying the data foundation for full-process cyclical management and control.

[0027] In one embodiment of the present invention, based on step S2, the following will provide a possible embodiment and describe its specific implementation in a non-limiting manner.

[0028] The specific steps of step S2 are as follows: S21, adopts 3 The principle is to standardize the prediction data, remove outliers, and then normalize the removed data to obtain preprocessed prediction data. S22. A risk assessment system is constructed based on the sintering process principle. The risk assessment system includes five evaluation dimensions: chemical composition of sintered ore, performance of mixed materials, performance of fuel and auxiliary materials, operating system, and proportion of non-mainstream ore. The risk assessment system evaluates the predicted data and the current ore blending scheme based on the corresponding risk indicator standard values ​​and risk deviation thresholds pre-configured for the five evaluation dimensions. S23, input the preprocessed prediction data and the current ore blending scheme into the risk assessment system to obtain the risk level of each risk indicator in each dimension; the risk level includes: no risk level, low risk level or high risk level. In step S23, when the predicted data and the current ore blending scheme are input into the risk assessment system, index matching is performed using data tags as an index: The predicted chemical composition values ​​correspond to the chemical composition dimension of sintered ore, the predicted process parameter values ​​correspond to the performance dimension of mixture, the performance dimension of fuel and auxiliary materials, and the operating system dimension, and the raw material ratio values ​​in the current ore blending scheme correspond to the non-mainstream ore ratio dimension. The specific steps of step S23 are as follows: The preprocessed predicted data and the current ore blending scheme are input into the risk assessment system. The data required for the risk indicators in each dimension are extracted and compared with the preset risk indicator standard values ​​to calculate the degree of deviation and obtain the indicator deviation value. The indicator deviation value is then compared with the risk deviation threshold. The risk deviation threshold includes: no risk deviation threshold, low risk deviation threshold, and high risk deviation threshold. If the deviation value of an indicator falls within the risk-free deviation threshold range, the indicator is determined to be at the risk-free level; if the deviation value of an indicator falls within the low-risk deviation threshold range, it is determined to be at the low-risk level; if the deviation value of an indicator falls within the high-risk deviation threshold range, it is determined to be at the high-risk level. For example, the preset standards and risk range thresholds for the chemical composition dimension of sinter in the risk assessment system are as follows: Basicity standard value 2.00, preset risk deviation thresholds are ±0.05 for no risk, ±0.05~0.10 for low risk, and exceeding ±0.10 for high risk; TFe standard value 54.0%, risk deviation thresholds are ±0.3% for no risk, ±0.3%~0.6% for low risk, and exceeding ±0.6% for high risk; SiO2 content standard value 5.70%, risk deviation thresholds are ±0.15% for no risk, ±0.15%~0.30% for low risk, and exceeding ±0.30% for high risk. Therefore, the predicted risk index values ​​for the chemical composition dimension of sinter in this embodiment include: basicity deviation 0.02, TFe content deviation 0.2%, ... The content deviation is 0.20%, therefore the alkalinity and TFe content are both classified as risk-free. The content is at a low risk level; The preset standards and risk deviation thresholds for the performance dimension of the mixture in the risk assessment system are as follows: the standard value of the mixture moisture is 8.0%, the risk deviation threshold is within ±0.3% as no risk, ±0.3%~0.5% as low risk, and exceeding ±0.5% as high risk; therefore, the risk deviation thresholds for the performance dimension of the mixture in this embodiment include: a predicted moisture deviation of ±0.1% is judged as no risk level; In both the operational procedure dimension and the fuel and auxiliary material performance dimension, all process parameters are within the preset standard range and are all determined to be at a risk-free level.

[0029] S24, assign corresponding deduction coefficients to each dimension of risk level for basic weighted scoring; S25, deduct points specifically for non-mainstream mineral types from the basic score to obtain the preliminary comprehensive score; The specific steps of step S25 are as follows: S251, obtain the actual proportion of each non-mainstream mineral in the current ore blending scheme, wherein the non-mainstream minerals include at least Sierra Leone fines, laterite nickel ore, Malaysian fines and Indian fines; S252, based on the actual proportions of each non-mainstream mineral type, specific deductions will be implemented according to the preset deduction rules: Sierra Leone powder: Deduct 0.5 points for every 1% added; Laterite nickel ore: 0.5 points will be deducted for every 0.2% added; Malaysian rice noodles: Deduct 0.5 points for every 0.2% added; Indian powder: Deduct 0.2 points for every 1% added; S253, add up the deductions for each item to get the total deduction for non-mainstream mines; S254, the preliminary evaluation comprehensive score is obtained by subtracting the total deduction of the non-mainstream mine special project from the base score; For example, the basic weight allocation for each dimension is as follows: sintered ore chemical composition dimension: 50 points; mixture performance dimension: 20 points; fuel and auxiliary material performance dimension: 10 points; operating system dimension: 10 points; non-mainstream ore ratio dimension: 10 points; risk level deduction coefficient is as follows: high risk level deducts 60%~80% of the full score of this dimension, low risk level deducts 20%~40% of the full score of this dimension, and no risk level deducts zero. For example, As a low-risk level, 29.2% of the full score for the chemical composition dimension is deducted, i.e., 50 × 29.2% = 14.6 points. All other dimensions are at a risk-free level with zero deductions. The base score is 100 - 14.6 = 85.4 points. In the special deduction for non-mainstream minerals, the proportion of Indian powder is 2.0%, and 0.2 points are deducted for every 1% added, resulting in a special deduction of 0.4 points. The final pre-evaluation comprehensive score is 85.4 - 0.4 = 85 points.

[0030] It should be noted that steps S21 to S25 are performed through 3 The principle involves preprocessing the predicted data to eliminate abnormal interference at the input end, reduce evaluation errors, and ensure the objectivity and accuracy of subsequent risk assessment. Five dimensions of evaluation indicators comprehensively cover all risk sources, from raw material properties and fuel / auxiliary material compatibility to operating procedures, filling the blind spot of traditional methods that only focus on the chemical composition of finished products. Risk interval threshold comparison and risk level determination quantify and standardize the evaluation logic. The composite algorithm of basic weighted scoring and special deduction for non-mainstream minerals ensures the balance of mainstream indicator evaluation while establishing a special control channel for non-mainstream inferior minerals, making the risk assessment system universally applicable to different raw material structures and providing clear indicator positioning basis for the precise iteration of subsequent unqualified solutions.

[0031] In one embodiment of the present invention, based on step S3, the following will provide a possible embodiment and describe its specific implementation in a non-limiting manner.

[0032] The specific steps of step S3 are as follows: S31, determine whether the pre-evaluation comprehensive score reaches the preset passing threshold of 85 points: If the comprehensive score of the pre-evaluation is greater than or equal to the preset qualified threshold, the current ore blending scheme is determined to be qualified, the current ore blending scheme and the corresponding prediction data are locked, sintering production is executed, and the process proceeds to step S4. If the overall score of the pre-evaluation is less than the preset qualified threshold, the current ore blending scheme is deemed unqualified, and S32 to S34 are executed. S32. Based on the risk level determination results of each dimension of risk indicators in step S2, locate the risk indicators that have not reached the risk-free level and their respective dimensions, and use correlation analysis to backtrack and determine the key raw materials that cause the risk indicators to exceed the standard. S33, determine the direction of increase or decrease of key raw materials based on the index deviation value in step S23: if the index deviation value is high, reduce the proportion of the raw material; if it is low, increase the proportion of the raw material; adjust in increments of 5% to 10% of the current proportion of the key raw material. After adjusting the key raw materials, the proportions of each raw material are recalculated to generate a revised ore blending scheme and output new forecast data. S34. Input the revised ore blending scheme and new prediction data into the risk assessment system of step S2, and conduct a predictive risk pre-assessment again. Iterate until the comprehensive score of the pre-assessment reaches the preset qualified threshold. For example, if the comprehensive score of the pre-evaluation is greater than or equal to the preset qualified threshold of 85 points, the current ore blending scheme is determined to be qualified, the current ore blending scheme and the corresponding predicted data are locked, sintering production is executed, and the process proceeds to step S4.

[0033] It should be noted that steps S31 to S34, through a dual evaluation approach of pre-evaluation followed by iterative correction and verification evaluation, eliminate the bottleneck of passive and delayed problem detection in traditional ore blending models. By tracing risk indicators and raw material properties bidirectionally, precise identification of specific raw materials requiring adjustment is achieved, avoiding the inefficiency of blind trial-and-error. The dual evaluation approach ensures that any ore blending scheme has passed risk verification before being implemented in production, preventing unqualified schemes from entering the production process and avoiding raw material waste and production fluctuations. The targeted adjustment strategy retains the proportion structure of qualified indicators, implementing precise correction only for problematic raw materials, enabling rapid convergence of the iterative process and improving the efficiency and accuracy of ore blending scheme optimization.

[0034] In one embodiment of the present invention, based on step S4, the following will provide a possible embodiment and describe its specific implementation in a non-limiting manner.

[0035] The specific steps of step S4 are as follows: S41, organize sintering production according to the ore blending scheme locked in step S3. When the sintering production operating parameters have been running continuously and stably for a preset time, it is determined that the production has entered a stable state. For example, in this embodiment, the production parameters are set as follows: moisture content of the mixture 8.0%, material layer thickness 900 mm, vertical sintering speed 20 mm / min, and ignition temperature 1100℃; the materials are accurately proportioned according to the ratio, and the metering error is controlled within ±0.1%; the entire production process of mixing, granulation, sintering, crushing, and screening is completed according to the set process parameters; after the sintering production conditions have been running continuously and stably for 4 hours, the production is determined to have entered a stable state. S42, During the stable production phase, actual production data is collected according to a preset sampling cycle; the actual production data includes: actual chemical composition data, actual process parameter data, and actual material ratio data. For example, the actual chemical composition data includes: alkalinity 2.02, 5.85%, TFe content 53.6%, MgO content 1.98%, Content 1.80%; The actual process parameters include: mixture moisture 8.0%, material layer thickness 902mm, vertical sintering speed 19.8mm / min, ignition temperature 1105℃; The actual material ratios include: Mac powder 23.5%, super-fine powder 11.0%, coarse SSFT 9.6%, Indian powder 2.0%, quicklime 12.2%, sintered anthracite 4.5%, cold return ore 18.2%; S43, adopts 3 The principle is to preprocess the collected actual production data, remove outliers, and normalize the removed data to obtain preprocessed actual production data. S44. Substitute the preprocessed actual production data into the risk assessment system described in step S2, and perform confirmatory risk assessment and scoring according to steps S22 to S25 to obtain an actual comprehensive score of 88 points.

[0036] It should be noted that steps S41-S44, through the collection of actual production data and verification risk assessment, achieve substantial verification of the ore blending scheme from pre-judgment compliance to actual achievement, eliminating the blind spot of disconnect between scheme design and actual effect in the traditional model. Based on stable operating condition judgment and periodic sampling strategy, the actual data is ensured to have statistical representativeness and a steady-state basis for the process, avoiding interference from transient fluctuations on the evaluation results. Substituting the actual production data back into the unified risk assessment system for isomorphic comparison provides quantifiable and comparable data basis for subsequent deviation tracing and optimization of the risk assessment system.

[0037] In one embodiment of the present invention, based on step S5, a possible embodiment will be given below, and its specific implementation will be described in a non-limiting manner.

[0038] The specific steps of step S5 are as follows: S51, compare the preliminary comprehensive score with the actual comprehensive score, and calculate the comprehensive deviation value by taking the difference; analyze the comprehensive deviation value: if the comprehensive deviation value exceeds the preset deviation range, it is determined that there is a deviation in the risk assessment system, and S52 is executed to optimize the risk assessment system; if the comprehensive deviation value does not exceed the preset deviation range, the current risk assessment system parameters remain unchanged, and all the ore blending plan, prediction data, actual production data, preliminary comprehensive score and actual comprehensive score are archived. For example, the preliminary evaluation score of 85 points is compared with the actual comprehensive score of 88 points, and the difference is calculated to obtain a comprehensive deviation value of 3 points. The comprehensive deviation value is analyzed: the comprehensive deviation value does not exceed the preset deviation range of ±5 points, the parameters of the current risk assessment system remain unchanged, and the ore blending plan, prediction data, actual production data, preliminary evaluation comprehensive score and actual comprehensive score are all archived for reference in subsequent ore blending. S52, compare the deviation values ​​of the indicators for each dimension in step S23 with the actual deviation values ​​of the indicators for each dimension obtained by inputting the actual production data into the risk assessment system in step S44, calculate the risk score deviation value for each dimension, and locate one or more dimensions with the largest deviation values ​​for adjustment: For the chemical composition dimension of sintered ore, the deviation values ​​of each risk indicator are compared with the actual deviation values. Based on the direction and magnitude of the deviation of the individual risk indicator score, the preset risk deviation threshold of that risk indicator is adjusted accordingly: if the deviation value of a certain risk indicator is higher than the actual deviation value, the risk deviation threshold of that indicator is appropriately relaxed; if it is lower than the actual deviation value, the risk deviation threshold of that indicator is appropriately tightened. The adjustment range is determined based on the historical deviation statistical average. For the performance dimensions of the mixture, the performance dimensions of the fuel and auxiliary materials, and the operating procedures, the deduction weight of each dimension is adjusted according to the direction and magnitude of the risk score deviation value of each dimension. The adjustment range is determined based on the ratio of the score deviation value of that dimension to the overall deviation value. For non-mainstream minerals, if a non-mainstream mineral type is used and the overall deviation value is related to the non-mainstream minerals, the unit deduction coefficient for that non-mainstream mineral type will be proportionally adjusted based on the ratio of the actual overall score to the pre-evaluation overall score. S53, the revised risk deviation threshold, the deduction weights of each dimension, and the deduction coefficients of the non-mainstream mine project will be applied to the subsequent predictive risk pre-evaluation; and the deviation data of this dual evaluation will be entered into the deviation database.

[0039] It should be noted that step S5 establishes an adaptive optimization mechanism for the parameters of the risk assessment system through a two-way deviation comparison analysis between the pre-evaluation and the actual evaluation. When the deviation exceeds the limit, differentiated correction strategies are implemented for the chemical composition dimension, the mixture / fuel / auxiliary material / operating system dimension, and the non-mainstream mine-specific deduction items—threshold expansion and contraction, dynamic adjustment of weights, and proportional correction of deduction coefficients, so that the risk assessment system continuously approaches the actual production law during continuous use; when the deviation does not exceed the limit, the data is archived and accumulated to provide historical sample support for subsequent statistical optimization, thereby improving the long-term adaptability of the risk assessment system to different raw material conditions and production conditions.

[0040] like Figure 2As shown, the following are embodiments of the sinter blending quality control risk assessment system provided in this disclosure. This system and the sinter blending quality control risk assessment methods in the above embodiments belong to the same inventive concept. For details not described in detail in the embodiments of the sinter blending quality control risk assessment system, please refer to the embodiments of the above sinter blending quality control risk assessment methods.

[0041] The system includes: The ore blending scheme calculation module 1 is used to calculate the blending ratio of raw materials based on boundary constraints and raw material performance database, and generate the current ore blending scheme; based on the current ore blending scheme, it calls the raw material performance database to perform physical property calculation and outputs predicted data; the current ore blending scheme includes at least the material ratio of each raw material; the predicted data includes at least the predicted chemical composition content and predicted process parameters. Risk pre-assessment module 2 is used to construct a risk assessment system; after preprocessing the prediction data, it is input into the risk assessment system along with the current ore blending scheme to conduct a predictive risk pre-assessment and obtain a comprehensive pre-assessment score; The proportioning iteration optimization module 3 is used to locate risk indicators, trace key raw materials, adjust the proportioning according to the direction of exceeding the standard and the preset step size, and perform linkage compensation when the comprehensive score of the pre-evaluation does not reach the preset qualified threshold. It generates a correction plan and new prediction data, and returns to the risk pre-evaluation module 2 to perform predictive risk pre-evaluation again. The cycle iterates until the comprehensive score of the pre-evaluation reaches the preset qualified threshold. If the comprehensive score of the pre-evaluation reaches the preset qualified threshold, sintering production is carried out according to the prediction data and the current ore blending plan. The on-site performance verification module 4 is used to collect actual production data after production stabilizes, and then substitute the actual production data into the risk assessment system described in the risk pre-assessment module 2 to conduct a confirmatory risk assessment and calculate the actual comprehensive score. The self-optimization module 5 of the evaluation system is used to compare the pre-evaluation comprehensive score with the actual comprehensive score, analyze the comparison deviation, and optimize the predictive risk pre-evaluation process based on the deviation results.

[0042] This embodiment first performs a predictive risk pre-evaluation after generating the ore blending scheme. If it fails to meet the standard, it automatically iterates and corrects until it meets the standard before production can begin. After production stabilizes, a verification evaluation is performed using actual data. Then, the pre-evaluation and actual evaluation results are compared and analyzed to optimize the risk evaluation system in reverse. This achieves proactive risk control, precise ore blending schemes, and continuous self-optimization of the evaluation model. It avoids the passivity of post-event detection and eliminates the problem of the disconnect between scheme design and actual effect.

[0043] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for risk assessment of sinter blending quality control, characterized in that, Includes the following steps: S1, based on boundary constraints and raw material performance database, calculates the blending ratio of raw materials and generates the current blending scheme; Based on the current ore blending scheme, the raw material performance database is called to perform physical property calculations and output predicted data; the current ore blending scheme includes at least the material ratio of each raw material; the predicted data includes at least the predicted chemical composition content and predicted process parameters. S2, Construct a risk assessment system; After preprocessing the predicted data, input it into the risk assessment system along with the current ore blending scheme to conduct a predictive risk pre-assessment and obtain a comprehensive pre-assessment score; S3. If the overall score of the pre-evaluation does not reach the preset qualified threshold, locate the risk indicators, trace the key raw materials, adjust the proportions according to the direction of exceeding the standard and the preset step size, and carry out linkage compensation to generate a correction plan and new prediction data. Then return to step S2 to conduct a predictive risk pre-evaluation again, and iterate until the overall score of the pre-evaluation reaches the preset qualified threshold. If the comprehensive score of the pre-evaluation reaches the preset qualified threshold, sintering production will be carried out according to the predicted data and the current ore blending scheme. S4. After production stabilizes, collect actual production data, substitute the actual production data into the risk assessment system described in step S2 to conduct a confirmatory risk assessment, and calculate the actual comprehensive score. S5 compares the preliminary comprehensive score with the actual comprehensive score, analyzes the deviation, and optimizes the predictive risk preliminary assessment process based on the deviation results.

2. The risk assessment method for sinter blending quality control according to claim 1, characterized in that, Step S1 specifically includes: S11, Obtain boundary constraints, which include: raw material inventory data, sinter quality index requirements, raw material cost ceiling, harmful element control limits, and mixture process control range; S12, call the raw material performance database, take the boundary constraints as the solution objective and the raw material performance database as the calculation basis, and use the linear programming algorithm to solve the proportion of each raw material that satisfies all constraints to generate the current ore blending scheme; S13. Based on the proportions of each raw material in the current ore blending scheme, and combined with the chemical composition and physical property data of each raw material in the raw material performance database, perform physical property calculations and output predicted data. The predicted data includes: predicted chemical component content and predicted process parameter values; The predicted chemical composition content includes at least: alkalinity value, TFe content, SiO2 content, MgO content, and Al2O3 content; the predicted process parameter values ​​include at least: mixture moisture value, average particle size value, and CaO / TFe molar ratio value.

3. The risk assessment method for sinter blending quality control according to claim 1, characterized in that, Step S2 specifically includes: S21, adopts 3 The principle is to standardize the prediction data, remove outliers, and then normalize the removed data to obtain preprocessed prediction data. S22. A risk assessment system is constructed based on the sintering process principle. The risk assessment system includes five evaluation dimensions: chemical composition of sintered ore, performance of mixed materials, performance of fuel and auxiliary materials, operating system, and proportion of non-mainstream ore. The risk assessment system evaluates the predicted data and the current ore blending scheme based on the corresponding risk indicator standard values ​​and risk deviation thresholds pre-configured for the five evaluation dimensions. S23, input the preprocessed prediction data and the current ore blending scheme into the risk assessment system to obtain the risk level of each risk indicator in each dimension; the risk level includes: no risk level, low risk level or high risk level. S24. Based on the risk level of each dimension, a corresponding deduction coefficient is assigned to perform basic weighted scoring. High risk level corresponds to the first deduction weight, low risk level corresponds to the second deduction weight, and no risk level has a deduction of zero. The deviation values ​​of each dimension indicator are weighted and summed according to the corresponding deduction weight to obtain the deduction value. The deduction value is deducted out of a maximum of 100 points to obtain the basic score. S25, deduct points specifically for non-mainstream mineral types from the basic score to obtain the preliminary comprehensive score.

4. The risk assessment method for sinter blending quality control according to claim 3, characterized in that, Step S23 specifically includes: The preprocessed predicted data and the current ore blending scheme are input into the risk assessment system. The data required for the risk indicators in each dimension are extracted and compared with the preset risk indicator standard values ​​to calculate the degree of deviation and obtain the indicator deviation value. The indicator deviation value is then compared with the risk deviation threshold. The risk deviation threshold includes: no risk deviation threshold, low risk deviation threshold, and high risk deviation threshold. If the deviation value of an indicator falls within the risk-free deviation threshold range, the indicator is determined to be at a risk-free level; if the deviation value falls within the low-risk deviation threshold range, it is determined to be at a low-risk level; if the deviation value falls within the high-risk deviation threshold range, it is determined to be at a high-risk level.

5. The risk assessment method for sinter blending quality control according to claim 3, characterized in that, The specific deduction for non-mainstream mines in step S25 includes: S251, obtain the actual proportion of each non-mainstream mineral in the current ore blending scheme, wherein the non-mainstream minerals include at least Sierra Leone fines, laterite nickel ore, Malaysian fines and Indian fines; S252, based on the actual proportions of each non-mainstream mineral type, specific deductions will be implemented according to the preset deduction rules: Sierra Leone powder: Deduct 0.5 points for every 1% added; Laterite nickel ore: 0.5 points will be deducted for every 0.2% added; Malaysian rice noodles: Deduct 0.5 points for every 0.2% added; Indian powder: Deduct 0.2 points for every 1% added; S253, add up the deductions for each item to get the total deduction for non-mainstream mines; S254. Subtract the total deduction for the non-mainstream mine project from the base score to obtain the preliminary evaluation comprehensive score.

6. The risk assessment method for sinter blending quality control according to claim 1, characterized in that, Step S3 specifically includes: S31, determine whether the pre-evaluation comprehensive score reaches the preset qualification threshold: If the comprehensive score of the pre-evaluation is greater than or equal to the preset qualified threshold, the current ore blending scheme is deemed qualified, the current ore blending scheme and the corresponding prediction data are locked, sintering production is executed, and step S4 is entered; if the comprehensive score of the pre-evaluation is less than the preset qualified threshold, the current ore blending scheme is deemed unqualified, and steps S32 to S34 are executed. S32. Based on the risk level determination results of each dimension of risk indicators in step S2, locate the risk indicators that have not reached the risk-free level and their respective dimensions, and use correlation analysis to backtrack and determine the key raw materials that cause the risk indicators to exceed the standard. S33, determine the direction of increase or decrease of key raw materials based on the index deviation value in step S23: if the index deviation value is high, reduce the proportion of the raw material; if it is low, increase the proportion of the raw material; adjust in increments of 5% to 10% of the current proportion of the key raw material. After adjusting the key raw materials, the proportions of each raw material are recalculated to generate a revised ore blending scheme and output new forecast data. S34. Input the revised ore blending scheme and new prediction data into the risk assessment system of step S2, and conduct another predictive risk pre-assessment. Iterate in cycles until the comprehensive score of the pre-assessment reaches the preset qualified threshold.

7. The risk assessment method for sinter blending quality control according to claim 1, characterized in that, Step S4 specifically includes: S41, organize sintering production according to the ore blending scheme locked in step S3. When the sintering production operating parameters have been running continuously and stably for a preset time, it is determined that the production has entered a stable state. S42, During the stable production phase, actual production data is collected according to a preset sampling cycle; the actual production data includes: actual chemical component content data, actual process parameter data, and actual material ratio data; the actual chemical component content data includes at least alkalinity, content, content, content, The content, the actual process parameter data includes at least the moisture content of the mixture, the thickness of the material layer, the vertical sintering speed, and the ignition temperature; S43, adopts 3 The principle is to preprocess the collected actual production data, remove outliers, and normalize the removed data to obtain preprocessed actual production data. S44. Substitute the preprocessed actual production data into the risk assessment system described in step S2, and perform confirmatory risk assessment and scoring according to steps S22 to S25 to obtain the actual comprehensive score.

8. The risk assessment method for sinter blending quality control according to claim 3, characterized in that, In step S23, when the predicted data and the current ore blending scheme are input into the risk assessment system, index matching is performed using data tags as an index: The predicted chemical composition values ​​correspond to the chemical composition dimension of sintered ore, the predicted process parameter values ​​correspond to the performance dimension of the mixture, the performance dimension of the fuel and auxiliary materials, and the operating system dimension, and the raw material ratio values ​​in the current ore blending scheme correspond to the non-mainstream ore ratio dimension.

9. The risk assessment method for sinter blending quality control according to claim 4, characterized in that, Step S5 specifically includes: S51, compare the preliminary comprehensive score with the actual comprehensive score, and calculate the comprehensive deviation value by taking the difference; analyze the comprehensive deviation value: if the comprehensive deviation value exceeds the preset deviation range, it is determined that there is a deviation in the risk assessment system, and S52 is executed to optimize the risk assessment system; if the comprehensive deviation value does not exceed the preset deviation range, the current risk assessment system parameters remain unchanged, and all the ore blending plan, prediction data, actual production data, preliminary comprehensive score and actual comprehensive score are archived. S52, compare the deviation values ​​of the indicators for each dimension in step S23 with the actual deviation values ​​of the indicators for each dimension obtained by inputting the actual production data into the risk assessment system in step S44, calculate the risk score deviation value for each dimension, and locate one or more dimensions with the largest deviation values ​​for adjustment: For the chemical composition dimension of sintered ore, the deviation values ​​of each risk indicator are compared with the actual deviation values. Based on the direction and magnitude of the deviation of the individual risk indicator score, the preset risk deviation threshold of the risk indicator is adjusted accordingly. The adjustment range is determined based on the historical deviation statistical average. For the performance dimensions of the mixture, the performance dimensions of the fuel and auxiliary materials, and the operating procedures, the deduction weight of each dimension is adjusted according to the direction and magnitude of the risk score deviation value of each dimension. The adjustment range is determined based on the ratio of the score deviation value of that dimension to the overall deviation value. For non-mainstream minerals, if a non-mainstream mineral type is used and the overall deviation value is related to the non-mainstream minerals, the unit deduction coefficient for that non-mainstream mineral type will be proportionally adjusted based on the ratio of the actual overall score to the pre-evaluation overall score. S53, the revised risk deviation threshold, the deduction weights of each dimension, and the deduction coefficients of the non-mainstream mine project will be applied to the subsequent predictive risk pre-evaluation; and the deviation data of this dual evaluation will be entered into the deviation database.

10. A risk assessment system for quality control of sintered ore blending, characterized in that, The system is used to implement the risk assessment method for quality control of sintered ore blending as described in any one of claims 1 to 9; The system includes: The ore blending scheme calculation module is used to calculate the blending ratio of raw materials based on boundary constraints and a raw material performance database, and generate the current ore blending scheme; based on the current ore blending scheme, it calls the raw material performance database to perform physical property calculations and outputs predicted data; the current ore blending scheme includes at least the material ratio of each raw material; the predicted data includes at least the predicted chemical composition content and predicted process parameters. The risk pre-assessment module is used to construct a risk assessment system. After preprocessing the predicted data, it is input into the risk assessment system along with the current ore blending scheme to conduct a predictive risk pre-assessment and obtain a comprehensive pre-assessment score. The proportioning iteration optimization module is used to locate risk indicators, trace key raw materials, adjust the proportions according to the direction of exceeding the standard and the preset step size, and perform linkage compensation when the comprehensive score of the pre-evaluation does not reach the preset qualified threshold. It generates a correction plan and new prediction data, and returns to the risk pre-evaluation module to perform predictive risk pre-evaluation again. The process is iterated until the comprehensive score of the pre-evaluation reaches the preset qualified threshold. If the comprehensive score of the pre-evaluation reaches the preset qualified threshold, sintering production is carried out according to the prediction data and the current ore blending plan. The on-site performance verification module is used to collect actual production data after production stabilizes, and then substitute the actual production data into the risk assessment system described in the risk pre-assessment module to conduct a confirmatory risk assessment and calculate the actual comprehensive score. The self-optimization module of the evaluation system is used to compare the pre-evaluation comprehensive score with the actual comprehensive score, analyze the deviation, and optimize the predictive risk pre-evaluation process based on the deviation results.

Images