阅读说明：本技术 转炉的喷溅预知方法、转炉的操作方法及转炉的喷溅预知系统 (Method for predicting splash in converter, method for operating converter, and system for predicting splash in converter ) 是由 天野胜太 高桥幸雄 加濑宽人 菊池直树 于 2020-03-24 设计创作，主要内容包括：在转炉中的铁水的脱碳精炼中,无需将用于检知喷溅的传感器设置于转炉炉内或转炉炉内附近,且无时间延迟地预知喷溅的发生。本发明的转炉的喷溅预知方法是转炉的脱碳精炼中的喷溅的预知方法,所述转炉从顶吹喷枪向转炉内的铁水喷吹氧化性气体,或者进一步从底吹风口向转炉内的铁水吹入氧化性气体或非活性气体,对铁水进行脱碳精炼,从铁水制造钢水,其中,测定从所述转炉的炉口吹出的炉口燃烧火焰的发光光谱,算出测定的发光光谱的580～620nm的范围的波长中的发光强度,基于算出的发光强度的时序变化来预知喷溅的发生。(In decarburization refining of molten iron in a converter, it is not necessary to provide a sensor for detecting slopping in or near the converter, and occurrence of slopping is predicted without a time delay. A method for predicting splashing in a converter during decarburization refining of the converter, wherein an oxidizing gas is blown from a top-blowing lance into molten iron in the converter, or an oxidizing gas or an inert gas is further blown from a bottom-blowing tuyere into the molten iron in the converter to decarburize and refine the molten iron and produce molten steel from the molten iron, wherein an emission spectrum of a mouth combustion flame blown out from a mouth of the converter is measured, an emission intensity at a wavelength in a range of 580 to 620nm of the measured emission spectrum is calculated, and occurrence of splashing is predicted based on a time-series change in the calculated emission intensity.)

1. A method of predicting splashing in a converter for decarburization refining in a converter for producing molten steel from molten iron by blowing an oxidizing gas from a top-blowing lance into the molten iron in the converter or further blowing an oxidizing gas or an inert gas from a bottom-blowing tuyere into the molten iron in the converter to decarburize and refine the molten iron,

wherein an emission spectrum of a mouth combustion flame blown out from a mouth of the converter is measured, an emission intensity in a wavelength range of 580 to 620nm of the measured emission spectrum is calculated, and occurrence of the splash is predicted based on a time-series change of the calculated emission intensity.

2. The method for predicting the splashing of a converter according to claim 1,

and detecting the inflection point of the increase of the luminous intensity after the temporary reduction, and predicting the occurrence of the splashing by detecting the inflection point.

3. The method for predicting slopping of a converter according to claim 1 or 2,

the time-series change of the light emission intensity is obtained by moving average.

4. The method for predicting slopping of a converter according to claim 1 or 2,

the time-series change of the light emission intensity is obtained using an equation for determination based on a moving average.

5. The method for predicting the splashing of a converter according to claim 4,

the following expressions (1) to (3) are used as the expressions for the determination, and it is determined that the splash is generated when all of the expressions (1) to (3) are satisfied,

[ mathematical formula 1]

I(n，m₀)≥C₀…(1)

Wherein, I (n, m)₀) Is from the measuring point n-m₀Moving average of emission intensity indexes up to measurement point n (a.u.), I (n-L)₁,m₁) Is from the measurement point n-L₁-m₁To the measurement point n-L₁Moving average of the luminescence intensity indices so far (a.u.), I (n-2L.)₁,m₁) Is from the measurement point n-2L₁-m₁To the measurement point n-2L₁Moving average of the luminescence intensity indices so far (a.u.), I (n, m.)₂) Is from the measuring point n-m₂Moving average of emission intensity indexes up to measurement point n (a.u.), I (n-L)₂,m₂) Is from the measurement point n-L₂-m₂To the measurement point n-L₂Moving average of the luminous intensity indices so far (a.u.), C₀、C₁、C₂Is a threshold value for determination, C₀>0，C₂>0，C₁<C₂，L₁、L₂Is a constant and is an integer of 1 or more, m₀、m₁、m₂Is constant and is an integer of 0 or more.

6. The method for predicting the splashing of a converter according to claim 5,

c for determining the threshold value of the determination in the expressions (1) to (3) is determined by using 1 or more of the transition of the luminous intensity in oxygen converting, the exhaust gas flow rate, the exhaust gas composition, the oxygen supply rate from the top-blowing lance, and the lance height of the top-blowing lance₀、C₁、C₂。

7. The method for predicting the splashing of a converter according to claim 5,

c of the threshold value for determination in expressions (1) to (3) is determined by machine learning using 1 or more of the transition of luminous intensity in oxygen converting, the exhaust gas flow rate, the exhaust gas composition, the oxygen supply rate from the top-blowing lance, and the lance height of the top-blowing lance₀、C₁、C₂。

8. A method of operating a converter that manufactures molten steel from molten iron,

wherein, when it is judged that the splash is generated by the method for predicting the splash of the converter according to any one of claims 1 to 7,

at the time point when it is determined that the splash occurs, 1 or 2 or more of the adjustment of the flow rate of the oxidizing gas blown from the top-blowing lance, the adjustment of the lance height of the top-blowing lance, the adjustment of the height position of the movable hood, the adjustment of the flow rate of the oxidizing gas or the inert gas blown from the bottom-blowing tuyere, and the input of the sedative material are performed.

9. A splash prediction system for a converter for producing molten steel from molten iron by blowing an oxidizing gas from a top-blowing lance into the molten iron in the converter or further blowing an oxidizing gas or an inert gas from a bottom-blowing tuyere into the molten iron in the converter to decarburize and refine the molten iron,

wherein the splash forecasting system has:

a spectral camera disposed around the converter and configured to photograph a combustion flame at a furnace mouth from a gap between the converter and the movable hood; and

and an image analysis device for recording the image data sent from the spectral camera in a manner of being capable of being taken out, calculating the luminous intensity in the wavelength range of 580-620 nm of the luminous spectrum of the image data, and predicting the occurrence of the splash based on the calculated time-series change of the luminous intensity.

10. The system for predicting splashing of a converter according to claim 9,

the image analysis device further includes a control computer that transmits a control signal to change an operation condition based on data input from the image analysis device.

11. The system for predicting slopping of a converter according to claim 9 or 10,

the image analysis device detects the inflection point at which the luminous intensity is temporarily reduced and then is increased, and the occurrence of the splash is predicted by detecting the inflection point.

12. The system for predicting slopping of a converter according to any one of claims 9 to 11,

the image analysis device obtains the time-series change of the light emission intensity by using an equation for determination based on a moving average.

13. The system for sputter forecasting of a converter according to claim 12,

the following expressions (1) to (3) are used as the expressions for the determination, and it is determined that the splash is generated when all of the expressions (1) to (3) are satisfied,

[ mathematical formula 2]

I(n，m₀)≥C₀…(1)

Wherein, I (n, m)₀) Is from the measuring point n-m₀Moving average of emission intensity indexes up to measurement point n (a.u.), I (n-L)₁,m₁) Is from the measurement point n-L₁-m₁To the measurement point n-L₁Moving average of the luminescence intensity indices so far (a.u.), I (n-2L.)₁,m₁) Is from the measurement point n-2L₁-m₁To the measurement point n-2L₁Moving average of the luminescence intensity indices so far (a.u.), I (n, m.)₂) Is from the measuring point n-m₂Moving average of emission intensity indexes up to measurement point n (a.u.), I (n-L)₂,m₂) Is from the measurement point n-L₂-m₂To the measurement point n-L₂Moving average of the luminous intensity indices so far (a.u.), C₀、C₁、C₂Is a threshold value for determination, C₀>0，C₂>0，C₁<C₂，L₁、L₂Is a constant and is an integer of 1 or more, m₀、m₁、m₂Is constant and is an integer of 0 or more.

14. The system for sputter forecasting of a converter according to claim 13,

the image analysis device is provided with a machine learning model which determines C of the threshold value for determination in expressions (1) to (3) by machine learning using 1 or more of the transition of the emission intensity in oxygen converting, the exhaust gas flow rate, the exhaust gas component, the oxygen supply speed from the top-blowing lance, and the lance height of the top-blowing lance₀、C₁、C₂。

15. The system for sputter forecasting of a converter according to claim 13,

the oxygen lance control system further comprises a machine learning computer for determining the threshold value C of the determination in the expressions (1) to (3) by machine learning using 1 or more of the transition of the emission intensity in oxygen blowing, the exhaust gas flow rate, the exhaust gas component, the oxygen supply rate from the top-blowing lance, and the lance height of the top-blowing lance₀、C₁、C₂The machine learning model of (1).

Technical Field

The present invention relates to a method and a system for predicting occurrence of splash (discharge of slag and molten iron from a furnace) in decarburization refining of molten iron in a converter. Further, the present invention relates to a method for operating a converter for oxygen-converting molten iron while preventing occurrence of slopping.

Background

Molten iron discharged from a blast furnace is charged into a converter, and an oxidizing gas (oxygen) is supplied to the molten iron charged into the converter from a top-blowing lance or a bottom-blowing tuyere, whereby the molten iron is decarburized and refined in the converter, and molten steel is smelted from the molten iron. In this converter, an oxidizing gas (referred to as "oxygen blowing") is blown into the converter to slag the flux to form slag, and impurity elements (such as P, Si) contained in the molten iron are removed from the slag. However, when the flux is sufficiently melted, the generated slag foams (foams), and so-called "slopping" may occur in which slag and molten iron (molten iron or molten steel) in the furnace are suddenly discharged from the furnace mouth to the outside of the furnace during oxygen blowing.

In particular, when a large amount of an iron oxide source (iron ore, mill scale, etc.) is charged into the furnace or during the soft blowing operation, it is said that the amount of oxygen stored in the slag (FeO amount) increases, and a decarburization reaction (C + O → CO) occurs explosively at the interface between the slag and the molten iron (molten iron or molten steel), and a large amount of CO gas is generated, so that splashing occurs.

The splash disturbs the molten steel components to reduce the tapping yield, and causes various problems such as an increase in the decarburization refining time, a reduction in the gas recovery rate in the OG facility (non-combustion type exhaust gas treatment facility), a reduction in the working environment, and a failure of peripheral equipment. Therefore, various methods for predicting splashing have been proposed.

For example, patent document 1 proposes the following sputtering prediction method: the vibration of the top-blowing lance is measured by a vibration sensor provided in the top-blowing lance, and the ratio of a signal larger than a predetermined amplitude set value among the measured vibration signals is calculated within a predetermined time period.

Patent document 2 proposes the following refining method: the method includes projecting microwaves to a slag surface in the converter, capturing the microwaves reflected from the slag surface, calculating the frequency of a mixture of the projected waves and the reflected waves and/or the microwave reflectance at the slag surface, detecting the slag surface and the slag state based on the calculated values, and controlling the influence elements so as to maintain the slag surface and the slag state in a predetermined reference state.

Patent document 3 proposes the following converter refining method: in a converter exhaust gas treatment apparatus for cooling, dedusting and recovering exhaust gas generated from a converter, the slag condition is determined based on information detected by an acoustics meter, an exhaust gas composition analysis and a dust concentration meter, and based on the determination result, the lance height, the oxygen flow rate, the top-bottom blowing ratio or the amount of auxiliary raw material to be charged are controlled in order to suppress the occurrence of splashing and splashing.

Documents of the prior art

Patent document

Patent document 1: japanese laid-open patent publication No. 6-248321

Patent document 2: japanese laid-open patent publication No. 59-41409

Patent document 3: japanese laid-open patent publication No. 6-256832

Disclosure of Invention

Problems to be solved by the invention

However, the above-described conventional techniques have the following problems.

That is, patent document 1 uses a vibration sensor, and patent document 2 uses a microwave oven slag level gauge, and in the spatter prediction method using these sensors, it is necessary to provide the sensors in the converter or in the vicinity of the inside of the converter. Since the installed sensor is exposed to molten steel, slag, exhaust gas, and the like, which vigorously flows in the converter during oxygen blowing, there is a problem in durability and continuous operability of the equipment.

Patent document 3 has a problem that measurement is performed in an exhaust pipe of a converter exhaust gas treatment facility, and the exhaust pipe can be operated relatively stably because the ambient temperature is low, but the amount of time for which exhaust gas moves to a measurement position of the exhaust pipe is delayed at the measurement timing.

The present invention has been made in view of the above circumstances, and the object thereof is as follows. Provided are a prediction method and a prediction system for predicting the occurrence of slopping without a time delay without installing a sensor for detecting slopping in or near a converter furnace in decarburization refining of molten iron in the converter. Further, the object is to provide a method for operating a converter which performs oxygen blowing while preventing occurrence of slopping.

Means for solving the problems

The gist of the present invention for solving the above problems is as follows.

[1] A method of predicting splashing in a converter for decarburization refining of the converter for producing molten steel from molten iron by blowing an oxidizing gas from a top-blowing lance into the molten iron in the converter or further blowing an oxidizing gas or an inert gas from a bottom-blowing tuyere into the molten iron in the converter to decarburize and refine the molten iron,

measuring a luminescence spectrum of a mouth combustion flame blown out from a mouth of the converter,

calculating the emission intensity in the wavelength range of 580-620 nm of the measured emission spectrum,

the occurrence of sputtering is predicted based on the calculated time-series change in the emission intensity.

[2] According to the method for predicting the occurrence of slopping in a converter described in the above [1], the inflection point at which the luminous intensity temporarily decreases and then increases is detected, and the occurrence of slopping is predicted by detecting the inflection point.

[3] According to the method for predicting the splash in the converter according to the above [1] or [2], the time-series change in the emission intensity is obtained by a moving average.

[4] According to the method for predicting the splash of the converter according to the above [1] or [2], the time-series change of the emission intensity is obtained using an equation for determination based on a moving average.

[5] According to the method for predicting the occurrence of the slopping in the converter described in [4], the following expressions (1) to (3) are used as the expression for the determination, and it is determined that the slopping occurs when all of the expressions (1) to (3) are satisfied.

[ mathematical formula 1]

I(n，m₀)≥G₀…(1)

Wherein, I (n, m)₀) Is from the measuring point n-m₀Moving average of emission intensity indexes up to measurement point n (a.u.), I (n-L)₁,m₁) Is from the measurement point n-L₁-m₁To the measurement point n-L₁Moving average of the luminescence intensity indices so far (a.u.), I (n-2L.)₁,m₁) Is from the measurement point n-2L₁-m₁To the measurement point n-2L₁Moving average of the luminescence intensity indices so far (a.u.), I (n, m.)₂) Is from the measuring point n-m₂Moving average of emission intensity indexes up to measurement point n (a.u.), I (n-L)₂,m₂) Is from the measurement point n-L₂-m₂To the measurement point n-L₂Moving average of the luminous intensity indices so far (a.u.), C₀、C₁、C₂Is a threshold value for determination, C₀>0、C₂>0、C₁<C₂，L₁、L₂Is constant and is 1 or moreInteger of (1), m₀、m₁、m₂Is constant and is an integer of 0 or more.

[6]According to [5] above]The method for predicting the splash in a converter determines C, which is a threshold value for determination in expressions (1) to (3), by using 1 or more of the transition of luminous intensity in oxygen converting, the exhaust gas flow rate, the exhaust gas composition, the oxygen supply rate from a top-blowing lance, and the lance height of the top-blowing lance₀、C₁、C₂。

[7]According to [5] above]The method for predicting the splash in a converter determines C, which is a threshold value for determination in expressions (1) to (3), by machine learning using 1 or more of a transition of luminous intensity in oxygen converting, an exhaust gas flow rate, an exhaust gas component, an oxygen supply rate from a top-blowing lance, and a lance height of the top-blowing lance₀、C₁、C₂。

[8] A method of operating a converter that manufactures molten steel from molten iron,

wherein, when it is determined that the splash is generated by the method for predicting the splash of the converter according to any one of the above [1] to [7],

at the time point when it is determined that the splash occurs, 1 or 2 or more of the adjustment of the flow rate of the oxidizing gas blown from the top-blowing lance, the adjustment of the lance height of the top-blowing lance, the adjustment of the height position of the movable hood, the adjustment of the flow rate of the oxidizing gas or the inert gas blown from the bottom-blowing tuyere, and the input of the sedative material are performed.

[9] A splash prediction system for a converter for producing molten steel from molten iron by blowing an oxidizing gas from a top-blowing lance into the molten iron in the converter or further blowing an oxidizing gas or an inert gas from a bottom-blowing tuyere into the molten iron in the converter to decarburize and refine the molten iron,

wherein the splash forecasting system has:

a spectral camera disposed around the converter and configured to photograph a combustion flame at a furnace mouth from a gap between the converter and the movable hood; and

an image analysis device records the image data sent from the spectral camera in a manner of being capable of being taken out, calculates the luminous intensity in the wavelength range of 580-620 nm of the luminous spectrum of the image data, and predicts the occurrence of the splash based on the time-series change of the calculated luminous intensity.

[10] The system for predicting splashing in a converter according to the above [9], further comprising a control computer for transmitting a control signal to change an operation condition based on data input from the image analysis device.

[11] The system for predicting slopping in a converter according to claim 9 or 10, wherein said image analysis device detects an inflection point at which said emission intensity changes to increase after temporarily decreasing, and predicts occurrence of slopping by detecting said inflection point.

[12] According to the system for predicting slopping in a converter described in any one of the above [9] to [11], the image analysis device obtains the time-series change in the emission intensity by using an equation for determination based on a moving average.

[13] According to the system for predicting the occurrence of slopping in a converter as described in [12], the following expressions (1) to (3) are used as the expression for the determination, and it is determined that slopping occurs when all of the expressions (1) to (3) are satisfied.

[ mathematical formula 2]

I(n，m₀)≥C₀…(1)

Wherein, I (n, m)₀) Is from the measuring point n-m₀Moving average of emission intensity indexes up to measurement point n (a.u.), I (n-L)₁,m₁) Is from the measurement point n-L₁-m₁To the measurement point n-L₁Moving average of the luminescence intensity indices so far (a.u.), I (n-2L.)₁,m₁) Is from the sideFixed point n-2L₁-m₁To the measurement point n-2L₁Moving average of the luminescence intensity indices so far (a.u.), I (n, m.)₂) Is from the measuring point n-m₂Moving average of emission intensity indexes up to measurement point n (a.u.), I (n-L)₂,m₂) Is from the measurement point n-L₂-m₂To the measurement point n-L₂Moving average of the luminous intensity indices so far (a.u.), C₀、C₁、C₂Is a threshold value for determination, C₀>0、C₂>0、C₁<C₂，L₁、L₂Is a constant and is an integer of 1 or more, m₀、m₁、m₂Is constant and is an integer of 0 or more.

[14]According to the above [13]]The system for predicting the splash in a converter comprises a machine learning model for determining the threshold value C for determination in expressions (1) to (3) by machine learning using 1 or more of the transition of the emission intensity in oxygen converting, the exhaust gas flow rate, the exhaust gas composition, the oxygen supply rate from the top-blowing lance, and the lance height of the top-blowing lance₀、C₁、C₂。

[15]According to the above [13]]The system for predicting the splash in a converter further comprises a machine learning computer for determining C, which is a threshold value for determination in expressions (1) to (3), by machine learning using 1 or more of a transition of emission intensity in oxygen blowing, an exhaust gas flow rate, an exhaust gas component, an oxygen supply rate from a top-blowing lance, and a lance height of the top-blowing lance₀、C₁、C₂The machine learning model of (1).

Effects of the invention

In the method and system for predicting the occurrence of slopping in a converter according to the present invention, the occurrence of slopping is predicted by measuring the emission spectrum of the combustion flame at the furnace mouth. Therefore, it is not necessary to provide a sensor for detecting the occurrence of the splash in the converter or in the vicinity of the converter, and the occurrence of the splash can be predicted without a time delay. In addition, according to the method for operating a converter of the present invention, since the splash prevention measures are taken at the time when the occurrence of the splash is predicted, the occurrence of the splash can be stably suppressed.

Drawings

Fig. 1 is a schematic diagram schematically showing the structure of a converter facility suitable for carrying out the present invention.

Fig. 2 is a graph showing the change in the time-series of oxygen blowing of the emission intensity index of the sputtered charge.

Fig. 3 is a graph showing the change in the time-series in oxygen blowing of the luminous intensity index of the charge in which no splashing occurred.

Fig. 4 is a schematic view schematically showing another configuration of a converter facility suitable for carrying out the present invention.

Detailed Description

The present inventors have conducted extensive studies with a view to predicting the occurrence of splash in oxygen blowing in real time without a time delay in decarburization refining in a converter that manufactures molten steel from molten iron by oxidizing and refining the molten iron. Specifically, in the decarburization refining in the converter, the furnace state of the converter at the time of occurrence of the splash is monitored in real time. It is known that: the splashing occurs in a state where slag in the converter is foamed (frothed).

As a result of the study, the inventors of the present invention paid attention to the mouth combustion flame of the converter as a factor for accurately grasping the in-furnace condition of the converter in real time, and thought that the emission spectrum of the mouth combustion flame was measured at predetermined time intervals in the decarburization refining. Here, the "mouth combustion flame" refers to a flame in the converter blown out from a flue duct extending upward from the mouth of the converter.

The emission spectrum of the converter mouth combustion flame includes information on CO gas generated by a decarburization reaction (C + O → CO) in the converter, and CO generated by natural ignition due to mixing of a part of the CO gas with air drawn in the converter mouth part₂Information relating to the gas. The emission spectrum also includes information on FeO (intermediate product) derived from iron atoms evaporated from a combustion point in the converter (a position where the oxidizing gas from the top-blowing lance collides with the molten iron surface). The inventors of the present invention have found that:if the emission intensity for each wavelength in the range of 580 to 620nm in the emission spectrum can be measured in real time, the in-furnace state of the converter can be easily estimated in real time.

The wavelength in the range of 580 to 620nm in the emission spectrum corresponds to "FeO orange system band" caused by the generation and disappearance of FeO (intermediate product), and is different from the wavelength region of the intermediate product of the hydrocarbon gas. Furthermore, the inventors of the present invention confirmed that: when FeO (intermediate product) is produced, an absorption peak is observed in the wavelength region, and when FeO (intermediate product) is disappeared, an emission peak is observed in the same wavelength region. Furthermore, it was confirmed that: the emission intensity is linked to the disappearance rate of FeO (intermediate product). Hereinafter, "FeO (intermediate product)" will be referred to simply as "FeO".

Then, in the decarburization refining in the converter, the emission spectrum of the mouth combustion flame of the converter was measured in a time series. The measurement of the emission spectrum of the mouth combustion flame of the converter is performed by mounting a spectroscopic camera 6 on the front surface of the converter 2 and imaging the mouth combustion flame 16 visible from the gap between the mouth 9 and the movable hood 10 as shown in fig. 1 (detailed description of fig. 1 will be described later). The captured image captured by the spectroscopic camera 6 is transmitted to the image analysis device 7. Then, an image is recorded in the image analysis device 7, and line analysis is performed on an arbitrary scanning line of the input image data, and the emission intensity for each wavelength of the emission wavelength is analyzed. The measurement of the emission spectrum and the analysis of the emission intensity were performed with the measurement time interval Δ t, which is the interval between the measurement points, set to 1 second, which was constant.

From the measurement results of the obtained emission spectra, the wavelength of 610nm, which had the largest variation width in the decarburization refining, was set as a specific wavelength (a wavelength used in analysis), and the emission intensity at the wavelength of 610nm at each timing measured in the decarburization refining was calculated, and the time-series variation of the emission intensity was obtained. When the time-series change of the emission intensity was determined, the emission intensity normalized to 1 in the image data obtained by imaging the furnace opening with the spectroscopic camera 6 before the start of oxygen blowing was defined as "emission intensity index", and the time-series change was determined using this emission intensity index. Of course, the time-series change can be obtained as it is without normalizing the emission intensity.

In this study, a converter (capacity 300 ton scale) capable of blowing an oxidizing gas from a top-blowing lance 3 and blowing an agitating gas from a bottom-blowing tuyere 4 at the bottom of the converter was used. Oxygen (industrial pure oxygen) was used as the oxidizing gas from the top-blowing lance, and argon was used as the stirring gas from the bottom-blowing tuyere. Further, as the top-blowing lance, a top-blowing lance having a laval (laval) type nozzle in which the number of oxygen nozzles provided at the tip is 5 and the injection angle is 15 ° was used. Here, the injection angle of the nozzle is a relative angle between the oxygen injection direction of the nozzle and the axial direction of the top-blowing lance.

The molten iron having a carbon concentration of 3.5 mass% was decarburized and refined in the converter described above. The supply of oxygen from the top-blowing lance was continued from the time point when the carbon content of the molten iron was 3.5 mass%, to the time point when the carbon content of the molten iron in the furnace became 0.04 mass%.

The flow rate of oxygen from the top-blowing lance is set to 800-1000 Nm³Min, the height of the top-blowing lance is set to 2.5-3.0 m, and the flow rate of the stirring gas from the bottom-blowing tuyere is set to 5-30 Nm³And/min. Here, the "lance height of the top-blowing lance" is a distance from the tip of the top-blowing lance to the molten iron liquid surface in a stationary state in the converter.

Fig. 2 shows the change in the time sequence in oxygen blowing of the emission intensity index calculated by the above method for the sputtered charge, and fig. 3 shows the change in the time sequence in oxygen blowing of the emission intensity index calculated by the above method for the non-sputtered charge. The decarburization refining time of the splashed charge shown in FIG. 2 was 19.5 minutes, and that of the non-splashed charge shown in FIG. 3 was 18.0 minutes. The oxygen blowing progressivity shown by the horizontal axis in fig. 2 and 3 is defined by the following formula (4).

Oxygen blowing in-spread (Q)_O2C/Q_O2)×100……(4)

Wherein Q is_O2CIs the cumulative amount of oxygen (Nm) from the start of oxygen blowing to an arbitrary point in time³)，Q_O2Is the cumulative oxygen amount (Nm) at the end of oxygen blowing³)。

As is clear from fig. 2 and 3, regardless of the occurrence of the splash, in the first half of the oxygen blowing (the range up to the oxygen blowing extension of 60 to 70%), the luminous intensity index increases as the oxygen blowing extension increases, while in the second half of the oxygen blowing, the luminous intensity index decreases as the oxygen blowing extension increases.

However, as shown in fig. 2, in the case of the charge in which the splash occurred, even in the first half of the oxygen blowing, the emission intensity index that started to increase with an increase in the oxygen blowing progress degree was temporarily decreased, and after that, the emission intensity index was again increased, the splash occurred.

In this regard, it can be considered that: when the slopping occurs, the slag in the furnace foams, so the apparent thickness of the slag increases, and the reduction reaction of FeO, that is, the decarburization reaction, is stagnated by the cutting effect achieved by the increase in the apparent thickness of the slag, and thus the luminous intensity index temporarily decreases. The reason why the luminous intensity index turns to increase again after that is considered to be: the reduction reaction of FeO stagnates, the amount of FeO in the slag becomes excessive, and the decarburization reaction starts again at the interface between the slag and the molten iron (FeO + C → Fe + CO) by the excessive amount of FeO, and the emission intensity index increases again.

From this result, the inventors of the present invention found that the time-series change of the luminous intensity index can be utilized in the prediction of sputtering.

With respect to the charge in which no splash occurs, as shown in fig. 3, the luminous intensity index increases as the blowing progresses, and exhibits a maximum value in the middle of the blowing. Then, until the end of the blowing, the reduction reaction rate of the iron oxide decreases, and thus the luminous intensity index decreases.

The inventors of the present invention compared the change in the time-series of the emission intensity index between the sputtered charge and the non-sputtered charge from the viewpoint that the emission intensity index forms a pattern that increases and decreases. As a result, it was found that: when sputtering occurs, the following characteristics are present in the change in the time-series of the emission intensity index. Namely, found that: when the current value of the emission intensity index is increased by 20% or more from the current emission intensity index at the measurement point 10 seconds ago and the emission intensity index at the measurement point 10 seconds ago is equal to or smaller than the emission intensity index at the measurement point 80 seconds ago from the current time, the splash occurs. In this regard, the measured value of the emission intensity as it is can be expressed as described above without being normalized.

That is, it is suggested that when splashing occurs: the emission intensity and the emission intensity index continue to decrease or exhibit behavior in which the stagnant slope turns into a large increase after a certain period of time (about 70 seconds) as described above (this behavior is defined as "inflection point" in the present specification). In other words, it suggests: by detecting the occurrence of an inflection point in the time-series change of the luminous intensity and the luminous intensity index, the occurrence of the splash can be predicted.

In the case of the material in which the splash occurred as shown in fig. 2, there were two periods (dips in the emission intensity index) during which the emission intensity index once decreased and then increased. In the first valley (timing at which the oxygen blowing progressivity is about 30%), the rate of increase in the emission intensity index with respect to the emission intensity index at the measurement point 10 seconds before each measurement point was small compared to the second valley (timing at which the oxygen blowing progressivity is about 45%). That is, it can be considered that: the first valley (timing of oxygen blow in spread of about 30%) suggests the occurrence of foaming of slag that does not cause 0 splashing. On the other hand, in the second valley (timing at which the oxygen blow-in spread is about 45%), the rate of increase in the luminous intensity index from the luminous intensity index at the measurement point before 10 seconds exceeded 20%, and the splash occurred after passing through the second valley. Namely, it can be considered that: by detecting an inflection point at which a second valley appears in the time-series change of the emission intensity and the emission intensity index, the occurrence of the splash can be predicted more accurately.

When the emission intensity index is used for predicting the splash, the splash can be predicted even by comparing the emission intensity indexes of the instantaneous values (actual values) which are not moving averaged, as shown in fig. 2 and 3. However, it was found that: by using a moving average of the luminous intensity index over a certain period, the occurrence of splash can be predicted more accurately. Here, the moving average is a value obtained by dividing the sum of a certain range of changing data by the number of data, and is a method of smoothing time series data.

By moving average of the emission intensity (actual value) and the emission intensity index, the variation becomes small. By appropriately selecting the variable number of the moving average or the like, for example, the moving average of the luminous intensity index monotonously increases until the luminous intensity index assumes a maximum value in a charge in which no splash occurs, and monotonously decreases after the luminous intensity index assumes a maximum value. The emission intensity (actual value) also exhibits the same behavior as the emission intensity index.

Moreover, it was found that: by obtaining the time-series change of the emission intensity and the emission intensity index using the equation for determination based on the moving average, the occurrence of the splash can be predicted more accurately.

For example, the following expressions (1) to (3) can be used as the expressions for determination based on the moving average of the emission intensity index at the measurement point n. By using expressions (1) to (3), the inflection point can be easily detected. Here, the measurement point n is a measurement point at an arbitrary time point in the decarburization refining, and corresponds to the current measurement point.

[ mathematical formula 3]

I(n，m₀)≥C₀…(1)

Wherein, I (n, m)₀) Is from the measuring point n-m₀Luminescence up to the measurement point nMoving average of intensity index (a.u.), I (n-L.)₁,m₁) Is from the measurement point n-L₁-m₁To the measurement point n-L₁Moving average of the luminescence intensity indices so far (a.u.), I (n-2L.)₁,m₁) Is from the measurement point n-2L₁-m₁To the measurement point n-2L₁Moving average of the luminescence intensity indices so far (a.u.), I (n, m.)₂) Is from the measuring point n-m₂Moving average of emission intensity indexes up to measurement point n (a.u.), I (n-L)₂,m₂) Is from the measurement point n-L₂-m₂To the measurement point n-L₂Moving average of the previous luminous intensity indexes (a.u.), C₀、C₁、C₂Is a threshold value for determination, C₀>0、C₂>0、C₁<C₂，L₁、L₂Is a constant and is an integer of 1 or more, m₀、m₁、m₂Is constant and is an integer of 0 or more.

Here, the expression (1) means that only C will be possessed₀The data of the above values are used as data for determination. By adding this condition, it is possible to remove background noise and data of a period in which the field of view of the spectral camera is blocked and blackened, and to perform determination. (2) Formula (II) shows the measurement from the measurement point n-2L₁To the measurement point n-L₁The period until then, i.e., the amount of change in the luminous intensity index slightly before the present time. Further, the formula (3) shows the result from the measurement point n-L₂The period until the measurement point n, that is, the current immediately preceding expression of the change amount of the luminous intensity index.

The formulae (2) and (3) are represented by the formula I (n-2L)₁,m₁) And I (n-L)₂,m₂) In a normalized form, they are intended to remove the influence of the variation in the absolute value of the luminous intensity for each charge. In the present invention, the inflection point of the time-series change in which the luminous intensity index temporarily decreases in a period slightly before the current time and then the luminous intensity index is increased immediately before the current time is detected, so C is₂>0，C₁<C₂。

In addition, L₁、L₂Give an answerThe number of measurement points traced back from the current time. When the measurement time interval is set to Δ t (seconds), L₁×Δt、L₂The time period (second) x Δ t is the period to be traced back from the current time. m is₀、m₁、m₂The number of measurement points that become the rear moving average range is given. When the measurement time interval is set to Δ t (seconds), m is₁×Δt、m₂×Δt、m₃The time range (sec) of the moving average after the start of the operation is represented by x Δ t.

Using the above expressions (1), (2) and (3), it is assumed that when all expressions (1) to (3) are satisfied, splash occurs and the threshold value for determination is changed₀、C₁、C₂And constant L₁、L₂、m₀、m₁、m₂A test for predicting the occurrence of splash in decarburization refining was performed.

The test results are shown in table 1. In this test, the wavelength of 610nm was set to a specific wavelength, and even when it was determined that splash occurred, the operating conditions were not changed, and the occurrence of splash was not prevented.

[ Table 1]

Number of charges carried out 100 charges for each test

As shown in table 1, it was found that: by appropriately selecting the number of moving averages of the luminous intensity index and the threshold value for determination, it is possible to stably predict the splash. Here, the "determination success rate" in table 1 is a ratio of the charge in which the splash is successfully predicted before 60 seconds or more from the time point at which the splash actually occurs. The "normal detection rate" is a ratio of the charge in which the occurrence of the splash is not predicted among the charges in which the splash is not occurring, that is, a ratio of the charges in which the detection is not made erroneously.

The method and system for predicting the splash of a converter and the method for operating a converter according to the present invention have been completed based on the above findings and further additional studies. Hereinafter, a specific method for carrying out the method for predicting the splashing of the converter, the system for predicting the splashing, and the method for operating the converter according to the present invention will be described with reference to the drawings. Fig. 1 schematically shows a schematic view of a structure of a converter facility suitable for carrying out the present invention.

The converter installation 1 suitable for implementing the invention comprises: a converter 2; a top-blowing lance 3; a spectral camera 6 which is disposed around the converter 2 and can photograph the combustion flame 16 at the furnace mouth; an image analysis device 7 that records the captured image captured by the spectroscopic camera 6 in a removable manner and analyzes the captured image; and a control computer 8 for transmitting a control signal based on the data analyzed by the image analysis device 7.

The top-blowing lance 3 includes a lance height control device 11 configured to be operated by a control signal transmitted from the control computer 8, and to adjust a lance height of the top-blowing lance 3, and an oxidizing gas flow rate control device 12 configured to adjust a flow rate of the oxidizing gas injected from the top-blowing lance 3. Further, the apparatus includes a bottom-blowing gas flow rate control device 13 for adjusting the flow rate of the stirring gas blown from the bottom-blowing tuyere 4, a sub-raw material input control device 14 for controlling the type and input amount of the sub-raw material stored in a hopper (not shown) on the furnace, and a movable hood height position control device 15 for controlling the height position of the movable hood 10.

The control computer 8 receives, for feedback control, the lance height measured by the lance height control device 11, the oxidizing gas supply rate measured by the oxidizing gas flow rate control device 12, the bottom-blowing gas flow rate measured by the bottom-blowing gas flow rate control device 13, the sub-raw material input amount measured by the sub-raw material input control device 14, and the movable hood height position measured by the movable hood height position control device 15. The control computer 8 is inputted with an exhaust gas flow rate measured by an exhaust gas flow rate measuring instrument (not shown) provided in the flue for measuring the flow rate of the exhaust gas discharged from the converter and components (CO, CO) of the exhaust gas discharged from the converter and provided in the flue for measuring the components₂、O₂) The exhaust gas component measured by the exhaust gas component measuring instrument (not shown).

The converter 2 used in the present invention can inject an oxidizing gas jet 17 from the top-blowing lance 3 toward the molten iron 5 in the converter and also can inject an agitating gas from the bottom-blowing tuyere 4 at the bottom of the converter. A spectral camera 6 capable of measuring the emission spectrum of the mouth combustion flame 16 of the converter is attached to the periphery of the converter 2. The furnace mouth combustion flame 16 visible from the gap between the furnace mouth 9 of the converter and the movable hood 10 is imaged by the spectroscopic camera 6 attached thereto.

The installation position of the spectroscopic camera 6 is only required to be a position where the load of heat, dust, and the like on the spectroscopic camera 6 is small, the durability is high, and the burner combustion flame 16 visible from the gap between the burner 9 and the movable hood 10 of the converter can be imaged. For example, when the spectroscopic camera 6 is attached to the front surface of the converter 2, it is possible to photograph the port combustion flame 16 which is visible through a small window (gap) for flame confirmation provided in the charging door. Further, an imaging window capable of imaging the burner combustion flame 16 may be provided in the furnace space (the side opposite to the operation chamber) and on the furnace side (the trunnion side) surrounding the wall of the converter body, and the spectroscopic camera 6 may be attached to the outside of the window to perform imaging. Alternatively, even inside the wall surrounding the converter body, the spectroscopic camera 6 may be used if there is a place where it can endure.

The captured images (image data) captured by the spectral camera 6 are sequentially transmitted to the image analysis device 7. The image analysis device 7 records the transmitted captured image (image data), and performs line analysis on an arbitrary scanning line of the image data to analyze the emission wavelength and the emission intensity for each wavelength. The image analysis device 7 predicts the occurrence of splash based on the analysis result of the transferred captured image (image data).

The image data of the burner combustion flame 16 analyzed by the image analyzer 7 is transmitted to the control computer 8 every time the occurrence of the splash is predicted. Similarly, the control computer 8 transmits operation data such as the oxygen supply rate from the top-blowing lance, the lance height of the top-blowing lance, the exhaust gas flow rate, and the exhaust gas component to the image analyzer 7 at each time.

When the occurrence of splash is predicted, the control computer 8 transmits control signals for operating the lance height control device 11, the oxidizing gas flow rate control device 12, the bottom-blowing gas flow rate control device 13, the sub-raw material charge control device 14, and the movable hood height position control device 15 separately or simultaneously, when receiving the predicted occurrence of splash from the image analysis device 7. In fig. 1, reference numeral 18 denotes an oxidizing gas supply pipe to the top-blowing lance, reference numeral 19 denotes a cooling water supply pipe to the top-blowing lance, and reference numeral 20 denotes a cooling water discharge pipe from the top-blowing lance.

In the present invention, the converter facility 1 is used to produce molten steel from molten iron 5 by blowing an oxidizing gas from a top-blowing lance 3 into molten iron 5 contained in a converter 2, or further blowing an oxidizing gas or an inert gas from a bottom-blowing tuyere 4 into molten iron 5 contained in the converter 2 to perform oxidation refining of the molten iron 5, that is, decarburization refining of the molten iron 5.

In the decarburization refining, the burner combustion flame 16 is photographed by the spectroscopic camera 6, and the obtained emission spectrum is analyzed to estimate a change in the furnace state in the decarburization refining in the converter 2 in real time. The splash is predicted based on the estimated change in the furnace condition. The spectroscopic camera 6 is preferably used to take an image of the burner combustion flame 16 and analyze the emission spectrum, from the viewpoint of improving productivity and improving iron yield, with the measurement time interval Δ t being set to 1 to 10 seconds.

The captured emission spectrum is recorded in the image analysis device 7 so as to be removable. The image analysis device 7 identifies the emission wavelength for the wavelength in the range of 580 to 620nm in the obtained emission spectrum of the burner combustion flame 16, and identifies and calculates the emission intensity for each emission wavelength.

As described above, the wavelength in the range of 580 to 620nm corresponds to FeO orange system band (iron oxide orange system band) caused by the generation and disappearance of FeO. The inventors of the present invention confirmed that: when FeO is produced, an absorption peak is observed in the wavelength region, and when FeO disappears, a luminescence peak is observed in the same wavelength region, and the luminescence intensity therein is linked with the speed of disappearance of FeO. That is, the wavelength in the range of 580 to 620nm reflects the reaction in the converter, and is a subject of measurement because it is a clue that the in-furnace state of the converter can be easily estimated. The emission intensity indicates the amount of emission energy when FeO changes from an excited state (FeO ×) to a ground state.

The image analysis device 7 calculates the obtained emission intensity and emission intensity index for each wavelength. Preferably, a moving average of the luminous intensity index is calculated. Then, the image analysis device 7 estimates a change in the furnace state based on the calculated time-series change in the emission intensity, the emission intensity index, and the moving average of the emission intensity index, and predicts the occurrence of the splash during the operation of the converter. In this case, it is preferable to detect the inflection point and predict the occurrence of the splash by detecting the inflection point.

Here, it is preferable to use expressions (1) to (3) as expressions for determination based on moving average used for prediction of splash, and it is determined that splash occurs when all of expressions (1) to (3) are satisfied. (1) C as the threshold value for the determination in the expressions (1) to (3)₀、C₁、C₂The emission intensity is measured and a preliminary test is performed so that the ratio of normal detection to splash by expressions (1) to (3) is maximized, depending on the imaging environment and operating conditions of each converter. In view of operational advantages, C may be determined so that the normal detection rate is maximized within a range in which the false detection is minimized₀～C₂。

With respect to other L₁、L₂、m₀、m₁、m₂When the respective constants of (a) are set to large values, the tendency becomes moderate, and excessive detection is unlikely to occur (it is determined that the state of not being splashed is splashed). However, if these values are set to be too large, sensitivity becomes sluggish, and it becomes difficult to detect the splash before the occurrence of the splash.

In addition, Δ t × L₁、Δt×L₂、Δt×m₀、Δt×m₁、Δt×m₂Taking into account the time (on the order of tens of seconds to hundreds of seconds) for carrying out the countermeasure after detecting the splash, the preliminary test is carried out in advance so that the proportion of normal detection of the splash by expressions (1) to (3) becomes maximum. At Δ t × L₁、Δt×L₂、Δt×m₀、Δt×m₁、Δt×m₂The blowing time is about 1 to 5% in length, and a relatively good detection rate can be obtained. In view of operational advantages, the normal detection rate can be determined to be the maximum within a range in which false detection is the lowest.

In the image analysis device 7, the operator can set C to₀、C₁、C₂Each threshold value and L₁、L₂、m₀、m₁、m₂Each constant of (2) is set to an arbitrary value. The image analysis device 7 is provided with a threshold value C determined by machine learning using 1 or 2 or more of the transition of emission intensity in oxygen converting, the exhaust gas flow rate, the exhaust gas component, the oxygen supply rate from the top-blowing lance, and the lance height of the top-blowing lance₀、C₁、C₂The machine learning model function of (1). That is, the image analysis device 7 is provided with the function of automatically setting C by machine learning₀、C₁、C₂The function of (c).

As shown in fig. 4, a machine learning computer 21 having a machine learning model function can be provided independently of the image analysis device 7. In this case, each constant can be set as follows. First, the operation data is transmitted from the control computer 8 or the like in which the operation data is recorded, and the data of the emission intensity is transmitted from the image analysis device 7 or the like in which the data of the emission intensity of the burner combustion flame is recorded, to the machine learning computer 21 in an off-line manner. The machine learning computer 21 performs machine learning based on the received data, determines each of the above constants, and transmits the determined values to the image analysis device 7. The image analysis device 7 receives the new constant, and performs determination using the new constant in the next and subsequent operations.

Fig. 4 is a schematic view schematically showing another configuration of a converter facility suitable for carrying out the present invention. The converter facility 1A shown in fig. 4 is configured by further arranging a machine learning computer 21 in the converter facility 1 shown in fig. 1. The other structures are the same as those of the converter facility 1 shown in fig. 1, and the same portions are denoted by the same reference numerals, and the description thereof is omitted.

The specific wavelength used for calculating the emission intensity index is determined by measuring in advance the wavelength having the largest amount of change in emission intensity in decarburization refining at a wavelength in the range of 580 to 620nm, or by monitoring a plurality of wavelengths within the wavelength range in the decarburization refining and determining the wavelength having the largest amount of change in emission intensity at each time.

In the method for operating a converter according to the present invention, at the time point when it is determined that splash is generated based on the calculated time-series change in emission intensity in decarburization refining, in order to prevent the occurrence of splash and damage to facilities, 1 or 2 or more of the flow rate adjustment of the oxidizing gas blown from the top-blowing lance, the adjustment of the lance height, the height position adjustment of the movable hood, the flow rate adjustment of the oxidizing gas or inert gas blown from the bottom-blowing tuyere, and the furnace charge of the calming material are performed. In this case, it is preferable that the time-series change of the emission intensity index is obtained using the formula for determination of the formulas (1) to (3), and it is determined that the splash occurs when all of the formulas (1) to (3) are satisfied.

As a specific coping method, it is preferable to reduce the flow rate of the oxidizing gas blown from the top-blowing lance, to lower the lance height of the top-blowing lance, to raise the height position of the movable hood to prevent damage of the movable hood due to the slag, to increase the flow rate of the oxidizing gas or the inert gas blown from the bottom-blowing tuyere, or to inject the sedative. More preferably, 2 or more of the above operations are combined. By such adjustment, foaming of the slag or a rapid decarburization reaction can be suppressed before occurrence of the splash or at an extremely initial stage of occurrence of the splash, and ejection of the slag and the molten iron to the outside of the furnace can be avoided, whereby the iron yield can be improved.

Here, the reduction of the flow rate of the oxidizing gas blown from the top-blowing lance, the reduction of the lance height of the top-blowing lance, the increase of the flow rate of the gas blown from the bottom-blowing tuyere, and the input of the calming material are changes in the operating conditions for preventing the occurrence of splashing, and the raising of the height position of the movable cover is a change in the operating conditions for preventing damage to the equipment. Therefore, it is preferable to perform at least the change of the operation conditions for preventing the occurrence of the splash. In view of reducing the ejection of slag, it is also effective to reduce the height position of the movable hood to physically prevent the ejection of slag, but in this case, it is necessary to consider that the number of times the movable hood is used is reduced.

Here, the calming material is a steel-making sub-material for forming a degassing flow path in the slag in the converter by being charged into the converter, improving the degassing of the foamed (foamed) slag, and suppressing the foaming of the slag. As the calming material, a calming material obtained by granulating a carbonaceous material, a mill scale, a slag, or the like with water or grease is generally used, but other substances may be used.

The amount of reduction in the flow rate of the oxidizing gas blown from the top-blowing lance, the amount of decrease in the lance height of the top-blowing lance, the amount of change in the height position of the movable hood, the amount of increase in the flow rate of the oxidizing gas or inert gas blown from the bottom-blowing tuyere, the amount of input of the calming material, and the like are preferably determined in advance based on the ratio of the stirring force of the molten iron to the flow rate of the oxidizing gas.

In addition, in the converter facility 1 suitable for carrying out the present invention, it is preferable that: at the time point when all the formulas for determination of the above formulas (1) to (3) are satisfied, every time a control signal is sent from the control computer 8 to the lance height control device 11 to lower the lance height, or a control signal is sent to the oxidizing gas flow rate control device 12 to reduce the flow rate of the oxidizing gas injected from the top-blowing lance, or a control signal is sent to the movable hood height position control device 15 to raise the height position of the movable hood, or a control signal is sent to the bottom-blowing gas flow rate control device 13 to increase the flow rate of the oxidizing gas or the inert gas to be blown in, or a control signal is sent to the subsidiary-raw-material-charge control device 14 to charge a predetermined amount of the sedative material, or all these control signals are sent at the same time.

The oxidizing gas to be blown from the top-blowing lance 3 is generally oxygen (industrial pure oxygen), but a mixed gas of oxygen and a rare gas such as argon or helium, or nitrogen, air, oxygen-enriched air, or the like can be used. The "oxidizing gas" referred to herein is an oxygen-containing gas having an oxygen concentration equal to or higher than that of air. The gas blown in from the bottom-blowing tuyere 4 is an inert gas or an oxidizing gas, and when the oxidizing gas is blown in, it functions as an oxidizing gas for oxidation refining and also functions as a stirring gas.

In the determination of the furnace internal conditions by the spectral analysis of the burner flame 16, there is a case where erroneous detection occurs due to a change in conditions such as shielding of the visual field caused by the passage of a crane, deposition of the raw material metal on the burner, and the like. Therefore, it is preferable to use C, which is the threshold value for the determination in expressions (1) to (3)₀、C₁、C₂Varying for each converter operation for each charge.

Specifically, it is preferable to determine the threshold value, i.e., C, using 1 or 2 or more of the transition of the emission intensity in oxygen converting, the exhaust gas flow rate, the exhaust gas composition, the oxygen supply rate from the top-blowing lance, and the lance height of the top-blowing lance₀、C₁、C₂。

More preferably, C, which is a threshold determined by machine learning, is used as 1 or 2 or more of the transition of emission intensity in oxygen converting, the exhaust gas flow rate, the exhaust gas composition, the oxygen supply rate from the top-blowing lance, and the lance height of the top-blowing lance₀、C₁、C₂。

In addition, although the above description uses the emission intensity index to calculate expressions (1) to (3), expressions (1) to (3) may be calculated using the emission intensity itself at each time point.

As described above, according to the present invention, in the converter 2 for decarburization refining of molten iron 5, since the occurrence of slopping is predicted by measuring the emission spectrum of the burner combustion flame, it is not necessary to provide a sensor for detecting slopping in the converter or in the vicinity of the converter, and the occurrence of slopping can be predicted without a time delay. Further, since the spatter prevention measure is performed at a time point when the occurrence of spatter is predicted, the occurrence of spatter can be stably suppressed.

Example 1

Decarburization refining of molten iron 5 was performed using a top-and-bottom blowing converter (oxygen top-blown converter, argon bottom-blown converter) having a capacity of 300 tons of the same type as that of the converter 2 shown in fig. 1. The top-blowing lance 3 used was a lance in which 5 laval (laval) nozzle type nozzles were disposed at the tip end at an injection angle of 15 ° and equally spaced on the same circumference with respect to the axis of the top-blowing lance. The throat diameter d of the nozzle_tIs 73.6mm, the outlet diameter d_eIs 78.0 mm.

First, after charging scrap iron into a converter, 300 tons of molten iron having been subjected to desulfurization and dephosphorization in advance and having a temperature of 1310 to 1360 ℃ are charged into the converter. The chemical composition of the molten iron is shown in table 2.

[ Table 2]

Then, while argon gas as a stirring gas was blown into the molten iron from the bottom-blowing tuyere 4, oxygen gas as an oxidizing gas was blown toward the molten iron liquid surface from the top-blowing lance 3, and decarburization refining of the molten iron was started. The charging amount of scrap iron was adjusted so that the temperature of molten steel after completion of decarburization refining became 1650 ℃.

Then, in the decarburization refining, quicklime was charged as a CaO-based flux from an upper hopper (not shown) of the furnace, and the decarburization refining was performed until the carbon concentration in the molten iron became 0.05 mass%. The amount of quicklime charged was adjusted so that the basicity ((mass% CaO)/(mass% SiO)) of the slag formed in the furnace was adjusted₂) ) to 2.5.

In decarburization refining, the ratio of the average value of the measured time interval Δ t: the furnace mouth combustion flame 16 visible from the gap between the furnace mouth 9 of the converter 2 and the movable hood 10 was continuously imaged for 1 second by the spectroscopic camera 6 provided on the substantially front surface of the converter 2.

From the obtained captured image, the emission spectrum (image data) is measured by the image analyzer 7, and the emission wavelength at each time point is identified and the emission intensity index for each wavelength is calculated for the wavelength in the range of 580 to 620nm in the obtained emission spectrum. The wavelength (specific wavelength) used was 610 nm. The analysis is performed by performing line analysis on an arbitrary scanning line of the image data.

The above expressions (1) to (3) were calculated using the obtained emission intensity index at the specific wavelength at each time point. In this case, the threshold for the determination of expressions (1) to (3) is set to C₀＝15、C₁＝0.65、C₂0.7, constant is set to L₁＝L₂＝10、m₀＝m₁＝m₂＝20。

When all of the expressions (1) to (3) are satisfied, it is determined that the splash is generated, and at the time point when the splash is determined to be generated, any one of 1 or 2 or more of the adjustment of the oxygen flow rate from the top-blowing lance, the adjustment of the lance height of the top-blowing lance, the adjustment of the height position of the movable hood, the adjustment of the flow rate of the oxidizing gas or the inert gas blown from the bottom-blowing tuyere, and the in-furnace charge of the sedative material is performed.

In particular, the adjustment of the oxygen flow from the top-blowing lance is from 1000Nm³From/min to 833Nm³Min is reduced, the height of the spray gun is adjusted from 3.0m to 2.5m, and the flow of bottom blowing gas is adjusted from 15Nm³30 Nm/min³The/min increases. The height position of the movable hood is adjusted to a position 500mm higher than the height position of the movable hood at the time point when it is determined that splash is generated, and the amount of the fed sedation material is set to 500-1500 kg.

When it is determined that the splash occurs, the control computer 8 immediately sends control signals to the lance height control device 11, the oxidizing gas flow rate control device 12, the bottom-blowing gas flow rate control device 13, the sub-raw material charge control device 14, and the movable hood height position control device 15 to operate the above adjustments. The oxygen flow rate from the top-blowing lance, the lance height of the top-blowing lance, the height position of the movable hood, and the flow rate of the oxidizing gas or inert gas blown from the bottom-blowing tuyere were returned to the values before adjustment at any time point when the flow rates no longer satisfied any of expressions (1) to (3), and decarburization refining was continued.

By practicing the present invention, the incidence of splashing was reduced to about 1/3 compared to the incidence prior to practicing the present invention.

Example 2

Decarburization refining of molten iron 5 was carried out in the same manner as in example 1 using the same converter facility (top-and-bottom blowing converter) as in example 1.

In the decarburization refining, the measurement time interval Δ t was set to 1 second, and the mouth combustion flame 16 visible from the gap between the mouth of the converter 2 and the movable hood was continuously imaged by the spectroscopic camera 6 in the same manner as in example 1. From the obtained captured image, the emission spectrum (image data) is measured by the image analyzer 7, and the emission wavelength at each time point is identified and the emission intensity index for each wavelength is calculated for the wavelength in the range of 580 to 620nm in the obtained emission spectrum. The wavelength (specific wavelength) used was 610 nm. The analysis is performed by performing line analysis on an arbitrary scanning line of the image data.

The above-mentioned expressions (1) to (3) were calculated using the obtained emission intensity index at the specific wavelength at each time point. In this case, the constants L of the expressions (1) to (3)₁、L₂、m₀、m₁、m₂The same procedure as in example 1 was followed, except that C was the threshold value determined₀、C₁、C₂The same operation data of 200 charges as in the decarburization refining described in example 1 was divided into 4 zones based on the magnitude of the average value of the oxygen flow rate in the oxygen blowing, and the threshold value of each zone was determined. That is, the respective thresholds of expressions (1) to (3) are set to 4 types based on the magnitude of the average value of the oxygen flow rate.

In actual operation, the average of the oxygen flow rates is calculated one by one, and 1 threshold value out of the 4 kinds determined by the average of the oxygen flow rates is used. When all of the expressions (1) to (3) set as described above are satisfied, it is determined that the splash is generated, and at the time point when the splash is determined to be generated, any one of 1 kind or 2 or more kinds of the adjustment of the oxygen flow rate from the top-blowing lance, the adjustment of the lance height of the top-blowing lance, the adjustment of the height position of the movable hood, the adjustment of the flow rate of the oxidizing gas or the inert gas blown from the bottom-blowing tuyere, and the in-furnace charge of the sedative material is performed on the same basis as in example 1. The oxygen flow rate from the top-blowing lance, the lance height of the top-blowing lance, the height position of the movable hood, and the flow rate of the oxidizing gas or inert gas blown from the bottom-blowing tuyere are returned to the values before adjustment at the time point when any of the expressions (1) to (3) is no longer satisfied, and the decarburization refining is continued.

By determining the threshold value for the determinations in the expressions (1) to (3) in this way, the occurrence frequency of the splash in the decarburization refining becomes equal to or less than that in example 1, and it was confirmed that the decarburization refining is stable.

Example 3

In the determination of the in-furnace condition based on the spectral analysis of the burner flame 16, as described above, there may be a case where the detection is erroneously performed due to a change in the condition such as the shielding of the visual field caused by the passage of the crane, the deposition of the raw material metal on the burner, or the like. Therefore, it is preferable to use C, which is the threshold value for the determination in expressions (1) to (3)₀、C₁、C₂Varying for each converter operation for each charge.

Then, offline analysis data of the 2000 charge materials, which was subjected to the spectral analysis of the burner combustion flame 16, was used as training data, and neural network type machine learning was performed. The input data were 30 items such as molten iron mass, scrap mass, molten iron temperature before decarburization refining, amount of auxiliary raw material charged, oxygen flow rate per degree of forward progress of blowing (oxygen supply rate from top-blowing lance), bottom-blowing flow rate, lance height, exhaust flow rate, exhaust composition, movable hood height, and the like, and the hidden layer was set to 5 layers.

The threshold C for the determination of expressions (1) to (3) is set using the method for determining a threshold for machine learning as described above₀、C₁、C₂Decarburization refining of molten iron was performed in the same manner as in example 1 using the same converter facility (top-and-bottom blowing converter) as in example 1. (1) L is a constant of the formulae (I) to (3)₁、L₂、m₀、m₁、m₂The procedure was as in example 1.

In the same manner as in example 1, in the total blowing time of decarburization refining, the ratio of time at a predetermined time interval Δ t: the combustion flame 16 at the mouth of the converter 2 blown out from the mouth was continuously captured by the spectroscopic camera 6 for 1 second, and the emission spectrum (image data) was measured and recorded by the image analyzer 7 based on the captured image.

When all of the expressions (1) to (3) set as described above are satisfied, it is determined that the splash is generated, and at the time point when the splash is determined to be generated, any one of 1 kind or 2 or more kinds of the adjustment of the oxygen flow rate from the top-blowing lance, the adjustment of the lance height of the top-blowing lance, the adjustment of the height position of the movable hood, the adjustment of the flow rate of the oxidizing gas or the inert gas blown from the bottom-blowing tuyere, and the in-furnace charge of the sedative material are performed on the same basis as in example 1 (inventive example 3).

In order to compare examples 1 to 3, threshold value C for judging expressions (1) to (3) described in example 1 was also used₀、C₁、C₂The threshold value C for determining expressions (1) to (3) is set based on the magnitude of the average value of the oxygen flow rate in oxygen blowing as described in the decarburization refining (inventive example 1) and example 2 which are performed with a predetermined value set in advance₀、C₁、C₂And decarburization refining was carried out (inventive example 2).

Inventive example 1, inventive example 2 and inventive example 3 were charged with 100 charges, respectively. In any of the operations, at the time point when it was determined that the splash occurred, 1 or 2 or more kinds of the flow rate of oxygen from the top-blowing lance, the height of the top-blowing lance, the height position of the movable hood, the flow rate of the oxidizing gas or the inert gas blown from the bottom-blowing tuyere, and the furnace charge of the sedative material were performed by the same criteria as in example 1.

The oxygen flow rate from the top-blowing lance, the lance height of the top-blowing lance, the height position of the movable hood, and the flow rate of the oxidizing gas or inert gas blown from the bottom-blowing tuyere were returned to the values before adjustment at any time point when the flow rates no longer satisfied any of expressions (1) to (3), and decarburization refining was continued.

Table 3 shows the operation results in inventive example 1, inventive example 2, and inventive example 3. Table 3 also shows the operation results of the conventional decarburization refining (conventional example) performed without predicting the splashing. The spattering incidence shown in Table 3 is the percentage of the number of spattering generating charges with respect to the total number of charges (100 charges).

[ Table 3]

As is apparent from Table 3, in inventive example 3, the incidence of slopping was low, the blowing extension due to slopping was reduced, and the amount of the sedative material used was reduced.

Description of the reference symbols

1 converter installation

1A converter equipment

2 revolving furnace

3 top-blowing spray gun

4 bottom blowing tuyere

5 molten iron

6-beam splitting camera

7 image analysis device

8 computer for control

9 furnace mouth

10 movable cover

11 spray gun height control device

12 oxidizing gas flow rate control device

13 bottom blowing gas flow control device

14 auxiliary raw material feeding control device

15 height position control device for movable cover

16 burner flame

17 jet of oxidizing gas

Oxidizing gas supply pipe for 18-way top-blowing lance

19-direction top-blowing spray gun cooling water supply pipe

20 Cooling water discharge pipe from top-blowing lance

21 machine learning computer.