A method for intelligent prediction and response to blast furnace collapse

Abstract

This invention relates to a method for intelligent prediction and response to blast furnace collapse, comprising the steps of (1) calculating the average daily oxygen consumption per ton of iron charged in the blast furnace; (2) correcting the current oxygen consumption per ton of iron charged based on the current fuel ratio; and (3) calculating the difference X between the theoretical and actual iron charge at the current operating moment of the blast furnace. i Calculations are performed to determine the material feeding space inside the blast furnace in real time; (4) the collapse index η is calculated. i (5) The collapse risk is evaluated and the collapse index is assessed and corresponding measures are taken. The present invention obtains the collapse index by using the oxygen consumption index per ton of iron at the tuyere and the difference between the theoretical and actual iron content at the current working moment of the blast furnace. It evaluates the changes in the feeding space, tracks and evaluates the collapse risk in real time, realizes accurate prediction of blast furnace collapse, prevents collapse from occurring in advance, and reduces collapse accidents.

Description

Technical Field

[0001] This invention relates to the field of iron and steel smelting technology, and in particular to a method for intelligent prediction and response to blast furnace collapse. Background Technology

[0002] Blast furnace collapse is a common accident in blast furnace production. Collapse refers to the probe stopping and then suddenly falling. The main causes are excessive local gas flow resistance, local deterioration of the charge column permeability, forced changes in gas flow, and local gas flow velocity exceeding the charge fluidization velocity. Partially molten slag and iron are carried into the low-temperature zone by the gas flow and re-solidify, forming a "cavity". The normal descent condition of the charge is: the weight of the charge is greater than (the friction between the charge + the friction between the charge and the furnace wall + the supporting force of the gas). Therefore, changes in gas flow will lead to changes in the descent conditions of the charge, causing collapse accidents. Collapse accidents are very harmful to blast furnaces. In addition to inevitably causing a significant reduction in production and increased consumption, in severe cases, it will also lead to a series of consequences such as abnormal furnace conditions, deterioration of the hearth condition, thickening of the furnace wall, and burn-out of the tuyeres, posing a huge threat to the blast furnace.

[0003] Therefore, it is urgent to accurately predict material collapse accidents and take early control measures to eliminate material collapse in its infancy, adjust the abnormal gas distribution to normal as soon as possible, minimize losses, and avoid serious adverse consequences such as continuous material collapse, excessively deep material collapse leading to prolonged abnormal furnace conditions, and hearth freezing.

[0004] In the traditional ironmaking industry, the prediction of material collapse was based on the experience of operators. They could only make judgments a few minutes before the collapse occurred by observing the operation of the probe and fluctuations in air pressure. However, this experience-based judgment had poor accuracy and a short lead time. By the time the collapse was detected, it was already unavoidable or had already occurred, so the prediction was not very meaningful. According to field data, more than 95% of material collapse accidents were caused by abnormalities in the material feeding space index 1-3 hours before the collapse. Therefore, early prediction could solve more than 95% of material collapse accidents. Only a very small number of collapses were caused by a sudden deterioration in the resistance of the material column. Therefore, by establishing a tracking system between the theoretical and actual material feeding space of the blast furnace, the occurrence of material collapse can be effectively predicted.

[0005] Chinese patent CN110633657B discloses a method for predicting blast furnace collapse, including: acquiring a reference image; acquiring an image to be identified, performing image matching with the reference image, and confirming whether the acquired image to be identified shows a collapse trend; wherein, multiple features of the image before collapse are extracted for model training, and the training samples include the image to be identified, the reference image, the image contour clarity, the brightness information, and the matching degree. The image to be identified and the reference image are used as training inputs, and the image contour, brightness information, and matching degree are used as output reference values. The training samples are trained using a generative adversarial network, training the initial first convolutional layer, the initial second convolutional layer, and the initial generative adversarial network to obtain the trained first convolutional layer, the second convolutional layer, and the generative adversarial network, which can quickly and accurately analyze whether the current stage of blast furnace operation has a collapse trend; however, this patent relies on a large amount of data for model training based on feature extraction. Once the data is insufficient or due to the presence of interfering data, the training model is prone to getting stuck in a local loop, leading to model bias and inaccurate prediction. Summary of the Invention

[0006] This invention provides an intelligent prediction and response method for blast furnace collapse. It corrects the oxygen consumption per ton of iron at the tuyeres using blast furnace fuel consumption data, then accurately predicts the theoretical charge amount in real time based on the tuyeres' oxygen consumption per ton of iron. By comparing the theoretical and actual charge amounts, it assesses the size of voids in the blast furnace's internal charge column, thereby enabling real-time tracking and evaluation of the collapse probability. Combined with the collapse index, it formulates corresponding adjustments to blast furnace smelting operation parameters to achieve the goal of eliminating collapses in advance. This invention is the first to achieve the assessment of the size of voids in the blast furnace's internal charge column, solving the problems of other collapse prediction methods being limited by the accuracy of furnace top images and gas analyzers, resulting in poor prediction effects, short prediction times, and a lack of corresponding collapse prevention measures after prediction. This invention achieves accurate prediction of blast furnace collapses, preventing collapses in advance and reducing collapse accidents.

[0007] To achieve the above objectives, the present invention employs the following technical solution:

[0008] A method for intelligent prediction and response to blast furnace collapse includes the following steps:

[0009] (1) Calculate the average daily oxygen consumption per ton of iron charged in the blast furnace;

[0010] (2) Adjust the current oxygen consumption per ton of iron fed based on the current fuel ratio;

[0011] (3) The difference X between the theoretical and actual iron feeding at the current working moment of the blast furnace. i Calculations are performed to determine the material feeding space inside the blast furnace in real time;

[0012] (4) Regarding the collapse index η i The following formula is used to perform real-time calculations and assess the risk of material collapse:

[0013] η i =X i / TFe 批 (1)

[0014] Where, η i The collapse index of the blast furnace at time i is dimensionless.

[0015] TFe 批 This represents the theoretical iron content of each batch of ore currently being processed in the blast furnace, expressed in tons per batch.

[0016] (5) Assess the material collapse index and take corresponding measures.

[0017] When η i When the value is ≤1.8, there is no risk of material collapse in the blast furnace charging space, and no measures are required.

[0018] When 1.8 < η i When the pressure is ≤2.0, there are voids in the blast furnace charging space, which will reduce the operating limit pressure difference by 3 to 5 kPa;

[0019] When 2.0 < η i When the pressure difference is ≤2.5, voids exist in the blast furnace charging space, and the probability of material collapse is >80%. Measures include reducing the operating limit pressure difference by 5 kPa, reducing the operating blast pressure by 10–20 kPa, and reducing the blast volume by 200–300 m³ / h. 3 / min, coke load reduces ore batch weight by 2-4t;

[0020] When η i When the pressure is greater than 2.5 kPa, voids exist in the blast furnace charging space, and the probability of material collapse is greater than 90%. To mitigate this, the operating limit pressure difference should be reduced by 5–10 kPa, the operating blast pressure by 20–30 kPa, and the blast volume by 300–400 m³ / h. 3 / min, the coke load is reduced by 2-4t of ore batch weight.

[0021] Furthermore, the average daily oxygen consumption per ton of iron charged in the blast furnace is calculated from the average blast volume, total oxygen content, and theoretical iron output of the previous 24-hour working days, as shown in the following formula:

[0022] V o2 =(q*0.21*1440+v) / m (2)

[0023] Among them, V o2 This represents the oxygen consumption per ton of iron charged per day in the blast furnace, expressed in m³. 3 / t,

[0024] q represents the average blast furnace air volume over 24 hours, in m³ / s.3 / min,

[0025] v represents the total oxygen enrichment in the blast furnace over 24 hours, in cubic meters (m³). 3 ,

[0026] m represents the theoretical iron content in the blast furnace charge batch over 24 hours, expressed in tons (t).

[0027] Furthermore, the formula for correcting the current oxygen consumption per ton of iron fed is as follows:

[0028] V o2 '=V o2 +0.933*K Φ *(C) K *(K2-K1)+0.933*(C) M *(M2-M1) (3)

[0029] Among them, V o2 'This is the corrected oxygen consumption per ton of iron charged under the current fuel ratio of the blast furnace, in m³.' 3 / t,

[0030] K Φ This refers to the combustion rate of coke before the tuyeres in the blast furnace, a dimensionless value ranging from 0.60% to 0.75%.

[0031] (C) K The carbon content of coke is fixed, with no dimensionless unit.

[0032] (C) M The carbon content of the pulverized coal is fixed, and the unit is dimensionless.

[0033] K1 represents the blast furnace coke ratio over the previous working day (24 hours), expressed in kg / t.

[0034] K2 represents the cumulative coke ratio in the blast furnace at the current moment, in kg / t.

[0035] M1 represents the blast furnace pulverized coal injection ratio for the previous working day (24 hours), expressed in kg / t.

[0036] M2 represents the cumulative pulverized coal injection ratio of the blast furnace at the current moment, in kg / t.

[0037] Furthermore, the formula for the difference between the theoretical and actual iron quantity fed into the blast furnace at the current operating moment is as follows:

[0038]

[0039] Among them, X i This represents the difference between the theoretical iron content and the actual iron content of the blast furnace charge from the start of blast furnace operation to time i on that day, expressed in tons (t).

[0040] Xi-1 This represents the difference between the theoretical and actual iron content of the blast furnace feed from the start of blast furnace operation to 24:00 on the day before time i, expressed in tons (t). It is equivalent to the value of X at 24:00 on the previous working day. i ,

[0041] Q i The total blast volume of the blast furnace up to time i, in m³ / s. 3 ,

[0042] V i The total oxygen enrichment of the blast furnace up to time i, in m³. 3 ,

[0043] m i This represents the theoretical amount of iron in the blast furnace charge at time i, expressed in tons.

[0044] Compared with the prior art, the beneficial effects of the present invention are:

[0045] 1) Correct the oxygen consumption per ton of iron at the tuyeres by using blast furnace fuel consumption data, then accurately predict the theoretical charge amount of the blast furnace in real time by using the oxygen consumption per ton of iron at the tuyeres, and finally evaluate the size of the voids in the charge column inside the blast furnace by comparing the theoretical charge amount with the actual charge amount.

[0046] 2) Based on the magnitude of the blast furnace collapse index, formulate corresponding methods for adjusting blast furnace smelting operation parameters to achieve the goal of eliminating collapse in advance, realize accurate prediction of blast furnace collapse, prevent collapse from occurring in advance, and reduce collapse accidents. Attached Figure Description

[0047] Figure 1 This is a schematic diagram showing the theoretical cumulative iron content, the actual cumulative iron content, and the collapse index for the day. Detailed Implementation

[0048] The specific embodiments of the present invention will be further described below with reference to the accompanying drawings:

[0049] For example, see the measured data from a steel mill on July 1, 2022. Figure 1 A diagram showing the theoretical cumulative iron content, the actual cumulative iron content, and the collapse index for the day.

[0050] This invention discloses an intelligent prediction and response method for blast furnace collapse, comprising the following steps:

[0051] (1) The daily average oxygen consumption per ton of iron charged in the blast furnace was calculated, based on the 24-hour average blast volume of the blast furnace on previous working days, q = 5034.6 m³ / h. 3 / min, total oxygen enrichment in blast furnace over 24 hours v = 359821m³ 3 The theoretical iron content in the blast furnace charge batch over 24 hours is m = 7107.8t, and the average daily oxygen consumption per ton of iron charged in the blast furnace is V.o2 The calculation is as follows:

[0052] V o2 =(q*0.21*1440+v) / m=264.8m 3 / t (5)

[0053] (2) The current oxygen consumption per ton of iron charged is corrected based on the current fuel ratio. The oxygen consumption per ton of iron charged on the 21st is estimated based on the coke ratio and coal ratio of the blast furnace at various times on the 21st. The coke ratio is based on the cumulative coke ratio fed into the blast furnace up to that time, and the coal ratio is based on the cumulative amount of pulverized coal injected into the blast furnace up to that time / the cumulative theoretical amount of iron. Taking the data at 21:44, one minute before the collapse, as an example, the parameters are as follows:

[0054] From 0:00 to 21:44 on the 21st, the cumulative coke ratio K2 of a blast furnace was 328 kg / t, the pulverized coal injection ratio M2 of the blast furnace over 24 hours on the 21st was 196 kg / t, the average daily coke ratio K1 of the 20th was 324 kg / t, the pulverized coal injection ratio M1 of the blast furnace over 24 hours on the 20th was 193 kg / t, and the coke combustion rate K at the tuyeres of the blast furnace was... Φ The fixed carbon content (C) of coke over 21 days was 65%. K The fixed carbon content (C) of pulverized coal is 85%. M The oxygen consumption per ton of iron charged in the blast furnace is 75% based on the following formula, calculated from 21:00 to 21:44.

[0055]

[0056] (3) The difference X between the theoretical and actual iron feeding at the current working moment of the blast furnace. i Calculations were performed to assess the real-time material feeding space inside the blast furnace. As of 21:44 on the 21st, the average airflow calculated by the blast furnace computer system was 4962.4 m³ / h. 3 / min, then the cumulative air volume is: Q i =4962.4 * 1304 = 6470969m 3 Computer collects total oxygen V i =311661m 3 This value can also be calculated from the hourly average oxygen content, specifically the difference X between the theoretical iron content and the actual iron content of the iron fed on the 20th. i-1 = -12.5t, theoretical iron content m in the blast furnace charge at time i. i =6043.6t,

[0057]

[0058] (4) Regarding the collapse index η i Perform real-time calculations and assess the risk of material collapse, TFe批 The theoretical iron content (TFe) of each batch of ore in the blast furnace at present. 批 =44.1,

[0059] η i =X i / TFe 批 =156.5 / 44.1=3.54 (8)

[0060] Since 3.54 is greater than 2.5, a collapse of the blast furnace was inevitable at 21:44 on the 21st if no remedial measures were taken in advance. Similarly, the collapse index at various times was calculated, and the real-time collapse index of the blast furnace was tracked. This showed that at 19:00, the collapse index of the blast furnace satisfied 1.8 < η. i The blast furnace was predicted to collapse 2.44 minutes in advance when the index was ≤2.0. By 21:00, the collapse index had reached 2.6. Therefore, the operating limit pressure difference should be reduced by 5 kPa at 19:00. The collapse index should be monitored in real time and measures should be taken to prevent collapse.

[0061] Table 1

[0062]

[0063] The above embodiments are implemented based on the technical solution of the present invention, providing detailed implementation methods and specific operation processes. However, the scope of protection of the present invention is not limited to the above embodiments. Unless otherwise specified, the methods used in the above embodiments are conventional methods.

Claims

1. A method for intelligent prediction and response to blast furnace collapse, characterized in that, Includes the following steps: (1) Calculate the average daily oxygen consumption per ton of iron charged in the blast furnace; The average daily oxygen consumption per ton of iron charged into the blast furnace is calculated from the average blast volume, total oxygen content, and theoretical iron output of the previous 24-hour working days, as shown in the following formula: ； in, This represents the oxygen consumption per ton of iron charged per day in the blast furnace, expressed in m³. 3 / t, q represents the average blast furnace air volume over 24 hours, in m³ / s. 3 / min, v represents the total oxygen enrichment in the blast furnace over 24 hours, in cubic meters (m³). 3 , m represents the theoretical iron content in the blast furnace charge batch over 24 hours, in tons. (2) Adjust the current oxygen consumption per ton of iron fed based on the current fuel ratio; ； in, This is the corrected oxygen consumption per ton of iron charged under the current fuel ratio conditions of the blast furnace, in m³. 3 / t, This refers to the combustion rate of coke before the tuyeres in the blast furnace, a dimensionless value ranging from 0.60% to 0.75%. The carbon content of coke is fixed, with no dimensionless unit. The carbon content of the pulverized coal is fixed, with dimensionless units. The ratio of blast furnace coke in the previous 24 hours is expressed in kg / t. This represents the cumulative coke ratio of the blast furnace at the current moment, in kg / t. The pulverized coal injection ratio for the blast furnace over the previous 24-hour period is expressed in kg / t. This represents the cumulative pulverized coal injection ratio of the blast furnace at the current moment, in kg / t. (3) Difference between the theoretical and actual iron feeding at the current working moment of the blast furnace Calculations are performed to determine the material feeding space inside the blast furnace in real time; The formula for the difference between the theoretical and actual iron feeding at the current operating moment of the blast furnace is as follows: ； in, This represents the difference between the theoretical iron content and the actual iron content of the blast furnace charge from the start of blast furnace operation to time i on that day, expressed in tons (t). This represents the difference between the theoretical and actual iron content of the blast furnace feed from the start of blast furnace operation to 24:00 on the day before time i, expressed in tons (t). , The total blast volume of the blast furnace up to time i, in m³ / s. 3 , The total oxygen enrichment of the blast furnace up to time i, in m³. 3 , This represents the theoretical amount of iron in the blast furnace charge at time i, expressed in tons. (4) Collapse index The formula for performing real-time calculations and assessing the risk of material collapse is as follows: ； in, The collapse index of the blast furnace at time i is dimensionless. This represents the theoretical iron content of each batch of ore currently being processed in the blast furnace, expressed in tons per batch. (5) Assess the material collapse index and take corresponding measures. when At that time, there was no risk of material collapse in the blast furnace charging space, and no measures were required; when At that time, there were voids in the blast furnace charging space, which reduced the operating limit pressure difference by 3 to 5 kPa; when At that time, if there are voids in the blast furnace charging space, the probability of material collapse is >80%. To mitigate this, measures should be taken such as reducing the operating limit pressure difference by 5 kPa, reducing the operating blast pressure by 10–20 kPa, and reducing the blast volume by 200–300 m³ / h. 3 / min, coke load reduces ore batch weight by 2-4t; when When voids exist in the blast furnace charging space, the probability of material collapse is >90%. To mitigate this, reduce the operating limit pressure difference by 5–10 kPa, the operating blast pressure by 20–30 kPa, and the blast volume by 300–400 m³ / h. 3 / min, the coke load is reduced by 2-4t of ore batch weight.

Images