A blast furnace condition fluctuation pre-judgment and multi-parameter coordinated attack and defense retreat regulation method

Abstract

The present application belongs to the technical field of blast furnace ironmaking, and specifically discloses a blast furnace condition fluctuation prediction and multi-parameter coordinated attack and defense retreat regulation method, comprising the following steps: first, collecting blast furnace gas flow, thermal state, raw materials and furnace condition state parameters, and setting early warning threshold and judgment logic; then, according to the parameter, determining the furnace condition fluctuation grade and coordinating the adjustment of load, charging, heat system, slag making system, air volume and blast dynamic energy according to the classification results; then, continuously monitoring the parameters in the furnace condition recovery, identifying and disposing the abnormalities such as hanging material, furnace cooling, slag skin falling off in real time; finally, counting the early warning and disposal effect, and updating the early warning threshold and regulation parameters. In view of the problems that the current furnace condition fluctuation and abnormal working condition prediction lags behind and the regulation is blind, thus leading to the easy deterioration of the furnace condition, the present application effectively solves the problems of blast furnace condition fluctuation prediction lag and blind regulation through multi-parameter coordinated prediction, attack and defense retreat classification regulation, real-time disposal of abnormal working conditions and whole-process closed-loop optimization, and improves the stability of blast furnace operation.

Description

Technical Field

[0001] This invention relates to the field of blast furnace ironmaking technology, specifically to a method for predicting blast furnace condition fluctuations and for multi-parameter coordinated offensive, defensive, and defensive control. Background Technology

[0002] Blast furnace smelting is a continuous production process involving high temperature, high pressure, and multiple coupled physicochemical reactions. The stability of the furnace conditions directly determines ironmaking capacity, energy consumption, and production safety. In actual production, fluctuations in raw material quality, deviations in operating parameters, enrichment of harmful elements (zinc, alkali metals), and the special smelting characteristics of vanadium-titanium ore and other special furnace materials can frequently cause furnace condition fluctuations and abnormal operating conditions, such as reduced blast volume, abnormal pressure difference, material collapse, suspended material, furnace cooling, and slag shedding. In severe cases, these can develop into serious accidents such as tuyeres and hearth accumulation, resulting in huge economic losses.

[0003] Current blast furnace condition management and abnormal operation handling typically suffer from the following problems: 1. Insufficient early warning and prediction capabilities: Traditional methods often rely on manual experience to identify furnace condition signs, using only a few single indicators such as pressure difference and molten iron temperature to determine trends, resulting in a false alarm rate as high as 30-40%. Furthermore, intervention is often reactive, only after production declines and indicators show obvious deterioration, with early warnings lagging by 2-4 hours, thus missing the optimal control window. 2. Lack of hierarchical and standardized control strategies: Handling furnace condition fluctuations and abnormal operation relies on experience-based "one-size-fits-all" approaches, easily leading to excessive blast reduction or blind load reduction, resulting in capacity and efficiency losses, or insufficient adjustments causing continued furnace condition fluctuations, which further deteriorate into serious accidents such as furnace cooling and tuyeres failure. 3. Poor parameter coordination: Control is often limited to adjusting blast volume, load, or thermal regime alone, neglecting the coordinated adaptation of parameters such as charging, distribution, blast energy, and slag composition. This creates a vicious cycle of single parameter adjustment → deterioration of other indicators → fluctuation rebound, resulting in low control efficiency. IV. Lack of adaptability to special working conditions: The early warning threshold and handling strategies have not been optimized for complex working conditions such as economical furnace charge, high zinc and high alkali, and vanadium-titanium ore smelting. Under these conditions, the incidence of abnormal furnace conditions is 30-35% higher than under normal conditions, and the monthly smooth operation index of the blast furnace is difficult to stabilize above 75.

[0004] In summary, it is necessary to design a method for predicting blast furnace condition fluctuations and coordinating multi-parameter attack, defense, and retreat control to address the problem that under the current technology, the prediction of blast furnace condition fluctuations and abnormal operating conditions is lagging and the control is blind, which leads to the easy deterioration of the furnace condition and causes serious accidents such as hanging materials and furnace cooling. Summary of the Invention

[0005] The purpose of this invention is to provide a method for predicting blast furnace condition fluctuations and for multi-parameter coordinated offensive, defensive, and defensive control, in order to solve the problem that in the existing blast furnace smelting process, the prediction of furnace condition fluctuations and abnormal operating conditions is lagging and the control is blind, which leads to the easy deterioration of the furnace condition and causes serious accidents such as hanging materials and furnace cooling.

[0006] To achieve the above objectives, the basic solution provided by this invention is: a method for predicting blast furnace condition fluctuations and coordinating multi-parameter attack, defense, and retreat control, comprising the following steps: S1: First, collect the gas flow parameters, thermal parameters, raw material parameters and furnace condition parameters of the blast furnace in real time, and set the early warning threshold and early warning judgment logic at the same time. S2: Next, the current blast furnace condition fluctuation level is determined according to the early warning judgment logic set in S1. Then, the blast furnace load, charging system, thermal system, slagging system, air volume and blast energy are adjusted in coordination according to the blast furnace condition fluctuation level. S3: Then, during the process of adjusting the blast furnace parameters to restore the furnace condition, continuously monitor the furnace operating parameters and determine in real time whether abnormal conditions such as hanging materials, furnace cooling, or slag skin falling off occur. S4: Finally, calculate the accuracy rate of early warning and the success rate of handling, and update the early warning threshold and control parameters based on the accuracy rate of early warning and the success rate of handling.

[0007] The beneficial effects of the present invention are as follows: (1) By constructing a multi-dimensional index system of airflow, hot state, raw materials and furnace condition, the present invention can identify signs of furnace condition fluctuation 2-4 hours in advance and automatically determine the warning level by quantifying the threshold, which solves the problem of lagging and high misjudgment of traditional manual judgment and greatly reduces the probability of furnace condition fluctuation escalating into serious accidents such as furnace cooling and tuyeres; (2) By matching the three-level strategies of "offensive" prevention, "defensive" furnace stabilization and "retreat" furnace rescue according to the degree of furnace condition deterioration, the five major systems of blast furnace load, charging, hot state, slag making and air volume are linked in sync, avoiding blind adjustment in a "one-size-fits-all" manner, thereby minimizing capacity loss and coke ratio increase while rapidly restoring the furnace condition; (3) A closed loop is formed by real-time monitoring, dynamic upgrading, effect verification and threshold optimization of the control process; at the same time, the prediction threshold and control parameters are flexibly adjusted according to the characteristics of raw materials and the load of harmful elements, so that it can adapt to the complex working conditions of high zinc and high alkali and vanadium titanium ore, and improve the accuracy of adaptation to different working conditions.

[0008] Option 2, an optimized version of the basic option, includes the following parameters in S1: airflow shrinkage rate, pressure difference fluctuation rate, and gas utilization rate; hot parameters: molten iron temperature, molten iron Si content, and furnace top temperature; raw material parameters: coke M40, coke CSR, and sinter RDI; and furnace condition parameters: probe difference, tuyeres temperature, and number of collapses. The collected parameters cover all dimensions of airflow, hot state, raw materials, and furnace condition, providing a more comprehensive basis for prediction.

[0009] Option 3, an optimized version of the basic option, sets the following warning thresholds in S1: airflow reduction rate ≥3% / h, pressure difference fluctuation rate ≥5% / h, gas utilization rate decrease ≥2% / h, hot metal temperature fluctuation ≥30℃ / furnace, hot metal Si content 0.8% < 0.3%, coke M40 decrease ≥2%, coke CSR decrease ≥3%, sinter RDI +3.15mm < 65%, probe difference > 0.8m, and material collapse ≥ 2 times / shift. When any two of the above warning thresholds are triggered, the current furnace condition is determined to be slightly fluctuating; when any three to four of the above warning thresholds are triggered, the current furnace condition is determined to be moderately fluctuating; and when any five or more of the above warning thresholds are triggered, the current furnace condition is determined to be severely fluctuating. By setting unified and quantifiable warning thresholds and establishing three-level fluctuation judgment rules (slight, moderate, and severe), standardized classification of furnace condition fluctuations is achieved.

[0010] Option 4, an optimal choice of Option 3, involves adjusting the blast furnace charging regime, thermal regime, and blast furnace load in S2 if the current furnace condition is determined to be slightly fluctuating. Specifically, the ore-coke ratio is first increased by 0.05-0.1, followed by increasing the number of coke rings by one and lowering the charging angle by 0.5-1° during charging. Then, the blast temperature is increased by 10-15°C and the oxygen enrichment rate is increased by 0.2-0.3%. By strengthening the central airflow and increasing the hearth heat reserve, fluctuations are suppressed at the source, and the blast furnace load is finely adjusted to stabilize the furnace condition while avoiding economic losses.

[0011] Option 5, an optimal choice from Option 4, involves adjusting the blast furnace load, charging regime, thermal regime, and slagging regime in S2 if the current furnace condition is determined to be moderately fluctuating. Specifically, the ore-coke ratio is first reduced by 0.1-0.2, while maintaining the current blast volume or reducing it by 50-100 Nm³. 3 The charge rate was adjusted to 0.5 min, then the charging mode was changed to C333222 or O3321, and the charge line was raised by 0.1 m. Next, the blast temperature was increased by 20-30℃ and the coke ratio was increased by 5-8 kg / t. Then, a batch of clean coke was added to the blast furnace, increasing the binary basicity of the slag by 0.03-0.05 and the MgO content to 9.0%. The coordinated adjustment of the four systems of blast furnace load, charging, hot state, and slag formation made the control more efficient, quickly stabilized the furnace condition, and prevented further deterioration of the furnace condition.

[0012] Option 6, an optimal choice from Option 5, involves adjusting the blast furnace load, blast volume, thermal regime, and charging regime in S2 if the current furnace condition is determined to be severely fluctuating. Specifically, the ore-coke ratio is first reduced by 0.3-0.5 to 3.5-3.8, while the blast volume is simultaneously reduced by 200-300 Nm³. 3The blast furnace is operated at a speed of 500-800 J / s, followed by an increase in blast temperature of 30-40℃ and coke ratio of 10-15 kg / t. Then, 2-3 batches of clean coke are added to the blast furnace, and pulverized coal injection is stopped or significantly reduced. Next, the material line is raised by 0.2-0.3 m, the material distribution angle is increased by 1-1.5°, and the ore batch weight is reduced by 5%. This method can forcibly stabilize the furnace conditions, effectively preventing severe fluctuations in furnace conditions from escalating into critical abnormalities or catastrophic accidents, thereby curbing material collapse and hanging.

[0013] Option 7, an optimal choice from Option 6, states that in S2, if the furnace condition cannot be stably controlled and deteriorates significantly after adjustments based on slight, moderate, and severe fluctuations, the ore-coke ratio should be reduced to 3.2-3.5, and the blast volume reduced by 300-400 Nm³. 3 At the same time, 4-5 batches of clean coke are added to the blast furnace, and acid washing is started to reduce the slag basicity to 1.05-1.08. By reducing the load, significantly reducing the blast, and adding more clean coke, the blast furnace is forcibly kept in place to prevent major accidents such as furnace cooling and hearth freezing.

[0014] Option 8, an optimal choice from the basic option, in S3, triggers a material suspension warning when the pressure difference > 175 kPa, the air permeability index < 4.2, the probe length difference > 0.8 m, and the duration is ≥ 5 minutes. In this case, the air volume is reduced by 300-500 Nm. 3 Simultaneously reduce the ore-to-coke ratio to 3.5-3.8, then raise the feed line by 0.3m, and add 3-4 batches of clean coke to the blast furnace while stopping pulverized coal injection. When the pressure difference is <165kPa, increase the feed rate to 50Nm. 3 The blast volume is gradually restored at a rate of / min. When the molten iron temperature is <1470℃, the Si content in the molten iron is <0.3%, and the blast temperature utilization rate is <90%, a furnace cooling warning is triggered. At this time, the blast temperature is increased by 40-60℃, the coke ratio is increased by 20-25 kg / t, the oxygen enrichment rate is increased by 0.8-1.0%, and 5-6 batches of clean coke are added to the blast furnace. At the same time, the ore-coke ratio is reduced to below 3.2, and the vanadium-titanium ore blending ratio is reduced to <5%, so that the molten iron temperature is ≥1490℃. When the blast furnace top temperature rises sharply by ≥50℃, the blast furnace edge temperature fluctuates by ≥30℃, and the pressure difference drops sharply by ≥6kPa, a slag detachment warning is triggered. At this time, the material line is lowered by 0.15-0.2m, the number of coke rings at the edge is increased by 1-2 rings, the ore batch weight is reduced by 3-5%, and the blast temperature is increased by 20-30℃. Real-time monitoring of blast furnace operating parameters during the control process prevents the expansion of abnormal operating conditions, thereby significantly reducing production risks and capacity losses.

[0015] Option 9, an optimal choice from the basic option, in S4, if used for high zinc and high alkali conditions (i.e., when Zn load ≥ 800 g / t and alkali load > 5 kg / t), the warning threshold is adjusted to trigger an alarm when the pressure difference fluctuation ≥ 7 kPa. During regulation, the blast energy is increased by 500 J / s, the ore-coke ratio is reduced by 0.05-0.1, and a simple rejection operation is performed simultaneously. If used for vanadium-titanium ore smelting (i.e., when TiO2 load ≥ 12 kg / t), the alarm threshold for abnormal furnace cooling is adjusted to molten iron temperature < 1480℃. During regulation, the blast temperature is increased by 10-15℃, the Si+Ti content is controlled at 0.4-0.55%, and the single adjustment range of the ore-coke ratio is controlled to ≤ 0.1. During "retreat" regulation, the MgO content in the slag is increased to 9.5-10%, and 0.3-0.5% fluorite is added to the furnace charge. By monitoring blast furnace operating parameters in real time and statistically analyzing the accuracy and success rate of blast furnace early warnings, and updating the early warning thresholds, the adaptability to different operating conditions is improved. Attached Figure Description

[0016] Figure 1 This is a flowchart of the present invention, which describes a method for predicting blast furnace condition fluctuations and for multi-parameter coordinated attack, defense, and retreat control. Detailed Implementation

[0017] The present invention will be further described in detail below through specific embodiments: Example like Figure 1 As shown: A method for predicting blast furnace condition fluctuations and coordinating multi-parameter offensive, defensive, and defensive control includes the following steps: S1: Construct a multi-dimensional fluctuation prediction index system: First, use infrared thermometers, pressure sensors, and online spectrometers deployed on the blast furnace to collect airflow parameters, thermal parameters, raw material parameters, and furnace condition parameters in real time at a frequency of ≥1 time / minute. The temperature error measured by the infrared thermometer is ≤3℃, and the pressure error measured by the pressure sensor is ≤0.5kPa. Specifically, this includes: airflow shrinkage rate, pressure difference fluctuation rate, gas utilization rate, molten iron temperature, molten iron Si content, furnace top temperature, coke M40, coke CSR, sinter RDI, probe difference, tuyeres temperature, and number of collapses. Then, set early warning thresholds, specifically: airflow shrinkage rate ≥ The following are considered minor fluctuations in blast furnace conditions during smelting: 3% / h, differential pressure fluctuation rate ≥5% / h, gas utilization rate decrease ≥2% / h, molten iron temperature fluctuation ≥30℃ / furnace, molten iron Si content <0.8% <0.3%, coke M40 decrease ≥2%, coke CSR decrease ≥3%, sinter RDI+3.15mm <65%, probe difference >0.8m, and material collapse ≥2 times / shift. S2: Blast Furnace Condition Fluctuation Classification and "Attack-Defense-Retreat" Classified Control: Following the warning judgment logic set in S1, the current blast furnace condition fluctuation level is determined and categorized into mild, moderate, and severe fluctuations. Then, based on the degree of fluctuation, the blast furnace load, charging regime, thermal regime, slagging regime, blast volume, and blast energy are adjusted accordingly. If any 1-2 warning thresholds in S1 are triggered, and the remaining indicators are stable with no material collapse or hanging, the current furnace condition is determined to be a mild fluctuation. At this time, the blast furnace smooth operation index is 70-80. The blast furnace charging regime, thermal regime, and blast furnace load are fine-tuned according to the "offensive" control strategy. Specifically: First, the ore-coke ratio is increased by 0.05-0.1. Then, during charging, the number of coke rings is increased by 1 ring while the charging angle is decreased by 0.5°-1°. Next, the blast temperature is increased by 10-15℃ and the oxygen enrichment rate is increased by 0.2-0.3%. If any 3-4 warning thresholds in S1 are triggered, and 1-2 instances of material collapse occur simultaneously with a pressure differential fluctuation of 8-10 kPa, the current furnace condition is determined to be moderately fluctuating. At this time, the blast furnace smooth operation index is 65-70. Adjustments are made to the blast furnace load, charging system, thermal system, and slagging system according to the "defensive" control strategy. Specifically: first, the ore-coke ratio is reduced by 0.1-0.2, and the current blast volume is maintained or reduced by 50-100 Nm³. 3The charge rate was increased by 0.1m, followed by adjusting the charging mode to C333222 and O3321, raising the charge line by 0.1m, increasing the blast temperature by 20-30℃ and the coke ratio by 5-8kg / t, then adding one batch of clean coke to the blast furnace, increasing the slag binary basicity by 0.03-0.05, and raising the MgO content to 9.0%. If any 5 or more of the warning thresholds in S1 are triggered, and there are ≥3 instances of suspended or collapsed material per shift, and the blast volume reduction rate is ≥8%, then the current furnace condition is determined to be severely fluctuating. At this time, the blast furnace smooth operation index is 60-65. The blast furnace load, blast volume, thermal regime, and charging regime are adjusted according to the "retreat" control. Specifically, the ore-coke ratio is first reduced by 0.3-0.5, so that the ore-coke ratio drops to 3.5-3.8, and the blast volume is reduced by 200-300 Nm³. 3 Increase the blast furnace speed and blast energy by 500-800 J / s, then increase the blast temperature by 30-40℃ and the coke ratio by 10-15 kg / t. Next, add 2-3 batches of clean coke to the blast furnace and stop or significantly reduce pulverized coal injection. Then, raise the material line by 0.2-0.3 m, increase the material distribution angle by 1-1.5°, and reduce the ore batch weight by 5%. If the blast furnace running index is <60 and cannot be stably controlled through "attack-defense-retreat" regulation, then reduce the ore-coke ratio to 3.2-3.5 and decrease the blast volume by 300-400 Nm³. 3 At the same time, 4-5 batches of clean coke are added to the blast furnace, and pickling is started to remove impurities and reduce the slag basicity to 1.05-1.08. Under mild fluctuation conditions, after parameter adjustment, the predictive indicators are monitored every 2 hours. Under moderate and severe fluctuation conditions, after parameter adjustment, the predictive indicators are monitored every 30 minutes. If the fluctuation indicators recover to ≥50% within 4-6 hours after parameter adjustment, the current control strategy is maintained. If the indicators do not improve or worsen, the control level is upgraded (e.g., the "defensive" control is changed to "retreating" control), and the adjustment range is increased by 50%. If the indicators recover to ≥80%, the parameters are gradually pulled back. S3: Emergency Handling of Abnormal Operating Conditions: During the process of adjusting blast furnace parameters to restore furnace conditions, continuously monitor furnace operating parameters and determine in real time whether abnormal operating conditions such as material suspension, furnace cooling, or slag skin detachment occur; when the pressure difference > 175 kPa, permeability index < 4.2, probe difference > 0.8 m, and the duration is ≥ 5 minutes, a material suspension warning is triggered. At this time, the blast volume is reduced by 300-500 Nm³. 3 Simultaneously reduce the ore-to-coke ratio to 3.5-3.8, then raise the feed line by 0.3m, and add 3-4 batches of clean coke to the blast furnace while stopping pulverized coal injection. When the pressure difference is <165kPa, increase the feed rate to 50Nm. 3The blast volume is gradually restored at a rate of / min. When the molten iron temperature is <1470℃, the Si content in the molten iron is <0.3%, and the blast temperature utilization rate is <90%, a furnace cooling warning is triggered. At this time, the blast temperature is increased by 40-60℃, the coke ratio is increased by 20-25kg / t, the oxygen enrichment rate is increased by 0.8-1.0%, and 5-6 batches of clean coke are added to the blast furnace. At the same time, the ore-coke ratio is reduced to below 3.2, and the vanadium-titanium ore blending ratio is reduced to <5%, so that the molten iron temperature is ≥1490℃. When the blast furnace top temperature rises sharply by ≥50℃, the blast furnace edge temperature fluctuates by ≥30℃, and the pressure difference drops sharply by ≥6kPa, a slag skin detachment warning is triggered. At this time, the material line is lowered by 0.15-0.2m, the number of coke rings at the edge is increased by 1-2 rings, the ore batch weight is reduced by 3-5%, and the blast temperature is increased by 20-30℃ to stabilize the airflow and heat state and avoid material collapse.

[0018] S4: Finally, calculate the early warning accuracy rate and handling success rate weekly, and update the early warning threshold and control parameters based on these rates; specifically: if the false alarm rate of a certain parameter is ≥20%, adjust the early warning threshold for that parameter by ±5%; if the handling success rate of an abnormal operating condition is <70%, adjust the handling parameter by ±20% (e.g., if the original air volume reduction was 300Nm). 3 / min, after adjustment it becomes 240Nm 3 / min or 360Nm 3 / min); If used in high zinc and high alkali conditions, i.e. when Zn load ≥ 800 g / t and alkali load > 5 kg / t, the warning threshold is adjusted to: when the pressure difference fluctuation ≥ 7 kPa, the warning is triggered. In response, the blast energy is increased by 500 J / s, the ore-coke ratio is reduced by 0.05-0.1, and a simple rejection operation is performed simultaneously. If used in vanadium-titanium ore smelting, i.e. when TiO2 load ≥ 12 kg / t, the warning threshold for abnormal furnace cooling is adjusted to molten iron temperature < 1480℃. In response, the blast temperature is increased by 10-15℃, the Si+Ti content is controlled at 0.4-0.55%, and the single adjustment range of the ore-coke ratio is controlled to ≤ 0.1. In response to the "reduction" control, the MgO content in the slag is increased to 9.5-10%, and 0.3-0.5% fluorite is added to the furnace charge.

[0019] The above methods enable the identification of furnace condition fluctuations 2-4 hours in advance, reducing the incidence of mild furnace condition fluctuations developing into moderate fluctuations by 80% and the probability of severe furnace condition fluctuations escalating into abnormalities by 90%. The recovery time for moderate furnace condition fluctuations is shortened to 6-8 hours, a 50% reduction compared to traditional methods, and the recovery time for severe furnace condition fluctuations is shortened to 12-16 hours. This avoids long-term deterioration of the furnace condition, keeps the monthly blast furnace smooth operation index stable above 78, reduces the number of hanging and collapsing events by 85%, and eliminates serious accidents such as furnace cooling and tuyeres. Ultimately, the capacity loss rate caused by furnace condition fluctuations is reduced to below 3%, and the coke ratio is reduced by 18-22 kg / t.

[0020] In summary, by setting quantitative early warning thresholds and grading judgment rules, the furnace conditions are divided into four levels: mild, moderate, severe fluctuations, and critical abnormalities. A three-level differentiated control strategy of "offensive," "defensive," and "retreat" is formed. While implementing coordinated adjustments to blast furnace load, charging, thermal regime, slagging regime, blast volume, and blast energy, three abnormal conditions—suspended material, furnace cooling, and slag shedding—are monitored in real time throughout the process, and rapid emergency response is provided. A complete closed loop is formed through dynamic optimization and effect verification. In addition, the early warning thresholds and control parameters are dynamically updated for special conditions such as high zinc and high alkali and vanadium-titanium ore smelting, thereby improving the adaptability accuracy for different conditions.

[0021] The above descriptions are merely embodiments of the present invention, and common knowledge regarding specific structures and characteristics is not elaborated upon here. It should be noted that those skilled in the art can make various modifications and improvements without departing from the structure of the present invention, and these should also be considered within the scope of protection of the present invention. These modifications and improvements will not affect the effectiveness of the present invention or the practicality of the patent. The scope of protection claimed in this application should be determined by the content of its claims, and the specific embodiments described in the specification can be used to interpret the content of the claims.

Claims

1. A method for predicting blast furnace condition fluctuations and coordinating multi-parameter offensive, defensive, and defensive control, characterized in that, Includes the following steps: S1: First, collect the gas flow parameters, thermal parameters, raw material parameters and furnace condition parameters of the blast furnace in real time, and set the early warning threshold and early warning judgment logic at the same time. S2: Next, the current blast furnace condition fluctuation level is determined according to the early warning judgment logic set in S1. Then, the blast furnace load, charging system, thermal system, slagging system, air volume and blast energy are adjusted in coordination according to the blast furnace condition fluctuation level. S3: Then, during the process of adjusting the blast furnace parameters to restore the furnace condition, continuously monitor the furnace operating parameters and determine in real time whether abnormal conditions such as hanging materials, furnace cooling, or slag skin falling off occur. S4: Finally, calculate the accuracy rate of early warning and the success rate of handling, and update the early warning threshold and control parameters based on the accuracy rate of early warning and the success rate of handling.

2. The method for predicting blast furnace condition fluctuations and coordinating multi-parameter attack, defense, and retreat control according to claim 1, characterized in that, In S1, the airflow parameters include: air volume shrinkage rate, pressure difference fluctuation rate, and gas utilization rate; the hot parameters include: molten iron temperature, molten iron Si content, and furnace top temperature; the raw material parameters include: coke M40, coke CSR, and sinter RDI; and the furnace condition parameters include: probe difference, tuyer temperature, and number of collapses.

3. The method for predicting blast furnace condition fluctuations and coordinating multi-parameter attack, defense, and retreat control according to claim 1, characterized in that, In S1, the set warning thresholds are: air volume shrinkage rate ≥3% / h, pressure difference fluctuation rate ≥5% / h, gas utilization rate decrease ≥2% / h, molten iron temperature fluctuation ≥30℃ / furnace, molten iron Si content 0.8% < 0.3%, coke M40 decrease ≥2%, coke CSR decrease ≥3%, sinter RDI+3.15mm < 65%, probe difference > 0.8m, and material collapse ≥ 2 times / shift. When any 2 of the above warning thresholds are triggered, the current furnace condition is determined to be slightly fluctuating; when any 3-4 of the above warning thresholds are triggered, the current furnace condition is determined to be moderately fluctuating; when any 5 or more of the above warning thresholds are triggered, the current furnace condition is determined to be severely fluctuating.

4. The method for predicting blast furnace condition fluctuations and coordinating multi-parameter attack, defense, and retreat control according to claim 3, characterized in that, In S2, if the current furnace condition is determined to be slightly fluctuating, the blast furnace charging system, thermal system and blast furnace load are finely adjusted. Specifically, the ore-coke ratio is increased by 0.05-0.1, then the number of coke rings is increased by 1 ring during charging while the charging angle is lowered by 0.5-1°, and then the blast temperature is increased by 10-15℃ and the oxygen enrichment rate is increased by 0.2-0.3%.

5. The method for predicting blast furnace condition fluctuations and coordinating multi-parameter attack, defense, and retreat control according to claim 4, characterized in that, In S2, if the current furnace condition is determined to be moderately fluctuating, adjustments are made to the blast furnace load, charging system, thermal system, and slagging system. Specifically, the ore-coke ratio is first reduced by 0.1-0.2, and the current blast volume is maintained or reduced by 50-100 Nm³. 3 / min, then adjust the feeding mode to C333222, O3321, and raise the material line by 0.1m. Then increase the blast temperature by 20-30℃ and the coke ratio by 5-8kg / t. Then add a batch of clean coke to the blast furnace, increase the binary basicity of the slag by 0.03-0.05, and increase the MgO content to 9.0%.

6. The method for predicting blast furnace condition fluctuations and coordinating multi-parameter attack, defense, and retreat control according to claim 5, characterized in that, In S2, if the current furnace condition is determined to be severely fluctuating, adjustments are made to the blast furnace load, blast volume, thermal regime, and charging regime. Specifically, the ore-coke ratio is first reduced by 0.3-0.5 to 3.5-3.8, while the blast volume is reduced by 200-300 Nm³. 3 Increase the blast furnace speed by 500-800 J / s, then increase the blast temperature by 30-40℃ and the coke ratio by 10-15 kg / t. Next, add 2-3 batches of clean coke to the blast furnace and stop or significantly reduce pulverized coal injection. Then, raise the material line by 0.2-0.3m, increase the material distribution angle by 1-1.5°, and reduce the ore batch weight by 5%.

7. The method for predicting blast furnace condition fluctuations and coordinating multi-parameter attack, defense, and retreat control according to claim 6, characterized in that, In S2, if the furnace condition cannot be stably controlled and deteriorates significantly after adjustments based on slight, moderate, and severe fluctuations, the ore-coke ratio should be reduced to 3.2-3.5, and the blast volume reduced by 300-400 Nm³. 3 At the same time, 4-5 batches of clean coke are added to the blast furnace, and acid washing is started to remove impurities and reduce the slag basicity to 1.05-1.

08.

8. The method for predicting blast furnace condition fluctuations and coordinating multi-parameter attack, defense, and retreat control according to claim 1, characterized in that, In S3, a material suspension warning is triggered when the pressure difference is greater than 175 kPa, the air permeability index is less than 4.2, the probe difference is greater than 0.8 m, and the duration is greater than or equal to 5 minutes. In this case, the air volume is reduced by 300-500 Nm. 3 Simultaneously reduce the ore-to-coke ratio to 3.5-3.8, then raise the feed line by 0.3m, and add 3-4 batches of clean coke to the blast furnace while stopping pulverized coal injection. When the pressure difference is <165kPa, increase the feed rate to 50Nm. 3 The blast volume is gradually restored at a rate of / min. When the molten iron temperature is <1470℃, the Si content in the molten iron is <0.3%, and the blast temperature utilization rate is <90%, a furnace cooling warning is triggered. At this time, the blast temperature is increased by 40-60℃, the coke ratio is increased by 20-25kg / t, the oxygen enrichment rate is increased by 0.8-1.0%, and 5-6 batches of clean coke are added to the blast furnace. At the same time, the ore-coke ratio is reduced to below 3.2, and the vanadium-titanium ore blending ratio is reduced to <5%, so that the molten iron temperature is ≥1490℃. When the blast furnace top temperature rises sharply by ≥50℃, the blast furnace edge temperature fluctuates by ≥30℃, and the pressure difference drops sharply by ≥6kPa, a slag skin detachment warning is triggered. At this time, the material line is lowered by 0.15-0.2m, the number of coke rings at the edge is increased by 1-2 rings, the ore batch weight is reduced by 3-5%, and the blast temperature is increased by 20-30℃.

9. The method for predicting blast furnace condition fluctuations and coordinating multi-parameter attack, defense, and retreat control according to claim 1, characterized in that, In S4, if used for high zinc and high alkali conditions, i.e., when the Zn load is ≥800g / t and the alkali load is >5kg / t, the warning threshold is adjusted to trigger a warning when the pressure difference fluctuation is ≥7kPa. During regulation, the blast energy is increased by 500J / s, the ore-coke ratio is reduced by 0.05-0.1, and a simple rejection operation is performed simultaneously. If used for vanadium-titanium ore smelting, i.e., when the TiO2 load is ≥12kg / t, the warning threshold for abnormal furnace cooling is adjusted to the molten iron temperature <1480℃. During regulation, the blast temperature is increased by 10-15℃, the Si+Ti content is controlled at 0.4-0.55%, and the single adjustment range of the ore-coke ratio is controlled to ≤0.

1. During "retreat" regulation, the MgO content in the slag is increased to 9.5-10%, and 0.3-0.5% fluorite is added to the furnace charge.

Images