JPS6230806A - Method for controlling desiliconization of molten iron - Google Patents

Abstract

PURPOSE:To match the content of Si in a molten iron after a desiliconization treatment with a target grade with good accuracy in the stage of subjecting the molten iron tapped from a blast furnace to the desiliconization treatment in a molten iron spout by measuring the content of Si and the flow rate of the molten iron before and after the desiliconization, determining the necessary amt. of a desiliconizing agent from the final target content of Si and adding the desiliconizing agent at the determined amt. to the molten iron. CONSTITUTION:The desiliconizing agent consisting of a main agent 9 and auxiliary agent 9a is added to the molten iron 3 tapped from the blast furnace to desiliconize the molten iron in the stage when the molten iron flows down in the molten iron spout 6. The molten irons before the after the addition of the desiliconizing agent are respectively sampled by sampling devices 22, 23 and the contents of Si in the respective molten irons are measured; at the same time, the flow rate of the molten iron is actually measured with a measuring instrument 24. The measured values and the amt. of the added desiliconizing agent are inputted to an electronic computer 14 which calculates the necessary amt. of the desiliconizing agent from the target content of Si set in a setter 20 and controls the rotating speeds of rotary valves 11, 11a of desiliconizing agent hoppers 10, 10a via control devices 17, 18 so that the necessary amt. of the desiliconizing agent is added to the molten iron.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention aims to reduce the amount of silicon in hot metal after desiliconization in a continuous desiliconization process of hot metal that continuously desiliconizes hot metal tapped from a blast furnace. This relates to a control method for matching the values.

[Conventional technology]

In the preliminary desiliconization of hot metal, in addition to reducing the amount of silicon in the hot metal after desiliconization, in order to achieve the following benefits 1) to 3) in the downstream steelmaking stage,
At the same time, it is required to reach a desired target value halfway and to reduce variations.

1) Improving the accuracy of converter blowing control 2) Reducing the amount of processing agents such as quicklime used in converter blowing and reducing the width of generated slag 3) Increasing the amount of silicon in hot metal charged into the converter - Mixed pig iron for planting Omission of hot metal handling operations such as car assembly and pig iron receiving Figure 2 shows an embodiment of preliminary desiliconization of hot metal.

1 is the blast furnace, 2 is the tap hole, and 3 is the hot metal. Numeral 4 is a large gutter, where hot metal and slag are separated, and the slag is discharged from a slag discharge port 5, while one of the hot metals is guided to a mixer wheel 8 via a hot metal gutter 6 and a hot metal tilting gutter 7. 9 and 9a are desiliconizing agents, 10 and 10a are storage hoppers thereof, 11 and lla are rotary valves for cutting, 12 is a charging conveyor, and 13 is a desiliconizing agent addition port. Hot metal 3 is discharged from the tap hole 2 of the blast furnace, is separated from slag in a large gutter 4, and flows into a hot metal gutter 6. A desiliconizing agent is added to the hot metal 3 at an appropriate point in the hot metal sluice 6 or the hot metal tilting sluice 7, and the silicon contained in the hot metal is removed by forcible mixing at the time of addition and by mixing during the subsequent flow. It is intended to remove.

As a desiliconizing agent, the main agent 9 such as mill scale, sintered ore powder, and iron-containing dust, which is a solid oxygen source for oxidizing and removing silicon in hot metal, is used to preferentially adjust the properties of desiliconizing slag and perform desiliconizing reactions. An adjuvant 9a, such as quicklime, is used to make the mixture stand out. The base material 9 is added based on the degree of silicon C (hereinafter referred to as desiliconization) to be removed so that the target value of silicon concentration in the hot metal after desiliconization (hereinafter referred to as the target value of silicon content) and the tapping rate are determined.
11, and the amount of the auxiliary agent 9a to be added is generally determined in proportion to that of the main agent 9, and the two are mixed first and added continuously from the desiliconizing agent addition port 13.

What is important in the continuous desiliconization process described above is to set the post-desiliconization target value (1) low and to reduce the variation by reaching this target value midway through. requires consideration.

As a conventional method for controlling the amount of desiliconizing agent added, the method disclosed in Japanese Patent Application Laid-Open No. 56-217 is generally known and consists of the following components.

(1) Silicon in the hot metal tapped from the blast furnace: The amount of silicon in the hot metal tapped from the blast furnace is determined by means such as prediction and estimation, and compared with a predetermined target value for the amount of silicon after desiliconization, the amount of silicon is determined at appropriate intervals. Calculate the desiliconization stiffness and determine the basic unit of the desiliconization agent to be added.

■ In order to achieve the desired desiliconizing agent consumption rate, measure the tapping rate [t/min], calculate the desiliconizing agent addition rate [kg/ml/nl], and set the desiliconizing agent cutting rotary. Be able to give the rotation speed instruction to the valve.

FIG. 3 is a block diagram of an automatic control system for adding a silica additive to realize the conventional method described in L. The electronic computer 14 includes a model A for estimating the amount of silicon in hot metal before desiliconization, and an arithmetic logic B for calculating the tapping speed.
, an arithmetic logic C for calculating the amount of silicon removed, and an arithmetic logic D for calculating the addition amount. 15 is a load cell for measuring the weight of hot metal in the car, and 16 is a hot metal weight detector.

The instrumentation control device a17 incorporates a pulse setting device E for signal conversion of the addition amount setting instruction value, and a main agent and adjuvant addition amount weigher F, and the electric control device 1B incorporates the addition amount setting instruction and the motor rotation speed instruction. An arithmetic unit G is built in to convert the data into .

Reference numerals 19 and 19a are load cells for detecting the actual addition amounts of the main agent and the auxiliary agent, respectively. 20 is a display and setting device for each control constant, and 21 is a typewriter for logging.

When operating using the various detection terminals and control system shown in Figure 3,
First, the amount of silicon in hot metal before desiliconization is estimated using estimation model A. On the other hand, constants such as a target value for the amount of silicon in the hot metal after desiliconization are given to the computer in advance from the display settings 20, and the required amount of desiliconization is sequentially calculated from both of them using logic C. On the other hand, using the signal from the hot metal weight detector 16, the tapping speed is calculated by logic relay B.

Logic D incorporates the relational expression between desiliconization and desiliconization agent addition unit in advance.The required desiliconization agent [median value is determined according to the required amount of desiliconization from logic C, and the amount obtained in logic B is calculated in advance. The addition rate of the desiliconizing agent is calculated based on the tapping rate.

After converting this into a signal in the instrumentation control device 17, the electrical control device W
At step 118, it is converted into a rotation speed command and transmitted to the rotary valve, and the rotary valve is driven accordingly, and the desiliconizing agent is cut out.

The characteristics of the conventional method described above are that the amount of silicon before desiliconization and the tapping speed are input to the system, and their fluctuations are measured or estimated, and the target value of silicon amount after desiliconization is achieved by controlling the oven looseness. It's where you are.

[Problem that the invention seeks to solve]

The conventional method has the following problems.

b) Since the conventional method does not measure the actual value [5i) f of silicon t: after desiliconization, it is not guaranteed that it matches the target value (Si).

The chemical reaction that progresses during the desiliconization treatment of hot metal, the so-called desiliconization reaction, involves a mixture of slag formation from the solid desiliconizing agent and a slag-metal reaction between the slag formed by the reaction product and the hot metal. Therefore, the reaction rate of the entire desiliconization reaction is controlled by various conditions, and the estimated amount of silicon before desiliconization (Si)ip
It is not possible to guarantee that the actual Dim (S i ) r matches the target value (Si) with only the information on the and tapping speed.

Solid-liquid reactions are controlled by conditions such as (Si)i and other hot metal components, hot metal temperature, solid-liquid mixing state, solid-liquid reaction interface area, and desiliconization agent composition. The main conditions are those that can be defined under different names. Some examples are as follows.

a) Hot metal temperature: Since the same tap iron 11 is normally used every 1 tap, the temperature of the hot metal trough drops when it is outside the tap, and the hot metal temperature is affected by temperature fluctuations in the furnace. In addition, fluctuations due to temperature rise in the hot metal sluice are added.

b) Particle size and components of the desiliconizing agent: Generally, the desiliconizing agent is not limited to one type, but a mixture of multiple types is used, and each type has its particle size composition, chemical composition, and other properties. are different.

When these powders and granules are used in the same powder and powder, the particle size and composition of the desiliconizing agent cut out will vary considerably over time due to particle size segregation within the powder and powder. Therefore, among the above conditions, it is a mixture! S, the reaction interfacial area also changes over time.

C) Tapping speed and wear condition of the hot metal runner: In the conventional method, the tapping speed is used to determine the command value for the addition amount of desiliconizing agent, but this is only expected to play a role in satisfying the material balance. Fluctuations in the tapping speed change the flow state of hot metal and affect the solid-liquid mixing state. Gutter wear also affects hot metal flow conditions.

[Means for solving problems]

The present invention was developed to solve the above problems, and its technical means are as follows: (1) Measurement and prediction of silicon content (Si)i before desiliconization (when predicted, it is expressed as (Si)ip) and silicon 1 after desiliconization (
Measurement of Si)f is carried out by sampling the corresponding hot metal before and after the desiliconization treatment at regular time intervals, and analyzing the amount of silicon in each hot metal.

(Si)i is measured on the upper casing side of the desiliconizing agent inlet, and (Si)f is measured at a point 1 downstream of the desiliconizing agent inlet, for example, at the end of a hot metal trough, a hot metal tilting trough, or a mixing wheel.

(2) Measure the tapping speed. In order to measure the hot metal before and after desiliconization treatment, and for use in fine calculations of desiliconization agent addition, the tapping speed is measured using online measurements such as the correlation method and spatial filter method upstream of the desiliconization agent inlet. I do.

(3) Find the maximum likelihood estimate of the reaction rate of the desiliconization reaction and use it in one calculation for adding the desiliconization agent. (Si)i like the conventional method,
(Si) Eliminating the method of calculating the microcalculation value of the desiliconization agent addition by material balance calculation from the iron tap rate and the iron tapping rate, the maximum likelihood estimate k regarding the reaction rate constant of the desiliconization reaction is calculated as (S i) i *
(S i) ip.

(Si)f, determined by a Kalman filter using the actual values of the tapping rate and the amount of desiliconizing agent added・k・C3i)i, [S
i) Calculate the microcalculation value of desiliconizing agent addition from the iron tapping speed.

[Effect]

The configuration diagram of the control device used to implement the method of the present invention is divided into three parts according to the zero function shown in FIG.

(D detection unit Sampling device 2-22゜23 for hot metal before and after desiliconization treatment, device 24 for measuring the tapping rate, and device 2 for analyzing the amount of silicon in hot metal
5 and load cells 19 and 19a for fine calculation of desiliconizing agent addition.
The respective output signals are transmitted to the Tf child computer 14.

(2) Ml is included in the calculation section electronic computer 14, and maximum likelihood estimation is made for the desiliconization reaction rate constant from the signal from the detection section.
2, and further calculates the command value for adding the desiliconizing agent.

(2) Control section This is the section that controls the amount of desiliconizing agent added based on the calculation results.

A method for determining the maximum likelihood estimate of the desiliconization reaction rate constant will be explained. The Kalman filter is a method for statistically analyzing operating results to efficiently obtain the maximum likelihood estimate of the next desiliconization reaction rate constant.

In hot metal desiliconization treatment, a desiliconizing agent added to molten iron continuously reacts with silicon in the hot metal, but at this time, the rate is determined by the diffusion process of silicon in the hot metal, and the following equation (1) holds true.

un((Si)f/(Si)i) = -W... (1) Here, (Si)i - Silicon in hot metal before desiliconization, also star [%] (Si)
f: Silicon in hot metal after desiliconization j^ [%] W: Silica base φ position C
kg/t-pig) K: Desiliconization reaction rate constant As shown in L, the desiliconization reaction rate constant is an unknown parameter that is influenced by various factors and changes over time. Therefore, a Kalman filter is used to efficiently estimate the desiliconization reaction rate constant K.

The current time is represented by k and the current data is, for example, (S i)
The data sampled i (k), − times before is expressed as (S i) i (k −1), etc. Then, equation (1) becomes )i (k)) = -K (k)W (k) ... (2) Here, -K (k) is replaced with K(k), and y (k) - sentence n((Si) f (k)) - sentence n ((Si)i (k)) +e(k) ...(3), then
Equation (2) becomes the following equation (4).

y(k)=K (k)W (k)+e (k)...(
4) Here is the error of the intersection H([5i)f(k))−sn((Si)i(k))] caused by the analysis error of e(k)+(Si).

Further, a stochastic process in which the unknown parameter K varies depending on various factors is expressed as K (k+1)=K (k)+v (k)-(5).

(4) Is the Kalman filter for equation (5) 1? (k
)=R(k-1)+P(k)X(V(k)-W
(k) R(k-1) )...(6) P(k)=m(k)W(k)/(m(k)→・σe)...(7)m(k)
= (1-P (k-1) W (k-1) )Xm(
k-1)+σY...(8)

Here, σe: variance of e(k) σv: variance of v(k) P(k): filter weighting coefficient m(k): unknown parameter K(k) when Y(k-1) is known. ) is the root mean square value of the difference between the maximum likelihood estimated value E (K (k) /y (k −1)) and its actual value K (k), that is, the variance of the error.

The Kalman filter expressed by equations (6) to (8) can be realized as shown in FIG.

Next, a method for calculating the optimum amount of desiliconizing agent to be added using this Kalman filter will be explained with reference to FIG.

Measuring device i! The silicon amounts (Si)i(k) and (Si)f(k) before and after desiliconization measured at 122.23 are input to the electronic computer 14 and converted to y(k) using equation (3).

Tapping speed vp (t
) (t/min) and the mean product value of the desiliconizing agent additive w (t
) (kg/min.
Si) measurement time (..., -1°k, +1...
・) From the respective values Vp(k) and w(k), W(k) in equation (4) can be calculated as
9) Ask for more. The setter 20 is used to set m(0), σe and σV that are safe for Kalman filter calculations, and the calculation loop shown in FIG. 1. Find the maximum likelihood estimate 1'e(k) of the current K.

R(k), (S i)i (k), vp (t) and the desiliconized silicon target value (Si
) The optimum desiliconizing agent addition; w(t) is calculated from the formula (10).

w (t) = ((Vp (t))/R(k))XJ
ln ((Si)i (k)/ (Si) tree)...
(10) Here, (time corresponding to k)≦t<(time corresponding to (k+1)).

FIG. 5 illustrates the timing of the above-mentioned calculations in two consecutive times of tapping.

In one tapping, if there is some time between the start of tapping and the first sampling of hot metal, or if analysis requires time, until the measured value (Si)i is input into the computer 14, 10) In cases where it is not possible to calculate w (H), the estimated value (Si)ip can be used instead of the measured value (Si)i. It is determined by applying a furnace thermal model using statistical methods.

In addition, during the -th tapping, the value of K before the time when k at the beginning of the drum is calculated is, for convenience, the value at the end of the previous same tap or the previous same tap. It is sufficient to use the average value during tapping at .

In this way, during tapping, at every control period of the desiliconizing agent addition acid, the electronic calculation fi14 is used to calculate the desiliconizing agent addition rate command value w (t
) is sent to the instrumentation control device 17, where it is converted into a signal, and then converted into a rotational speed command by the electric control device 18. By controlling the rotational speed of the rotary valve, the addition amount of the desiliconizing agent is determined. control. Further, the actual value of the desiliconizing agent addition needle is sent from the output of the hopper load cells 19, 19a to the electronic calculation fi14 via the instrumentation control device ff117.

In this way, during tapping, the maximum likelihood estimate of the desiliconization reaction rate constant is determined based on the actual data, and this is used to determine the optimal amount of desiliconization agent addition and control of desiliconization agent addition. be able to.

〔Effect of the invention〕

According to the present invention, the silicon e degree after the treatment in the cast bed desiliconization treatment can be made to closely match the target value.

The present invention is not limited to casting bed desiliconization of hot metal, but can be widely applied to 11-continuous hot metal pretreatment using a solid treatment agent, continuous steel manufacturing, etc., regardless of the type of reactor. In addition, this technology has a wide range of applications, not only in the steel industry, but also in cases where components are measured before and after adding a treatment agent, and the optimum amount of treatment agent to be added is determined to control the components after treatment.

[Brief explanation of drawings]

Fig. 1 is a block diagram of an embodiment of the present invention, Fig. 2 is a flow diagram showing an embodiment of hot metal desiliconization treatment, Fig. 3 is a block diagram of a conventional desiliconization agent addition amount control system, and Fig. 4 is a flowchart of the method for estimating the desiliconization reaction rate constant using the Kalman Fist filter,
FIG. 5 is a time chart showing the timing of various measurements and calculations during two rounds of tapping. l...Blast furnace 2...Tapping port 3...Hot metal 4...Gutter 5...Slag discharge port
6...Hot metal trough 7...Hot metal tilting trough 8...
- Mixed car wash 9... Main desiliconizing agent (main agent) 9a... Sub-silicifying agent (adjunct agent) 10, lOa... Storage hopper 11. lla... Rotary valve 12... Charging conveyor 13... Desiliconizing agent addition port 1
4...', [Sub computer 15...Hot metal car 1-1 measurement load cell 16...Hot metal type ↑4 detector 17...Instrumentation control device 18...Electric control device 20... Each control constant display setting device 22...Hot metal sampling device before desiliconization 23...Hot metal sampling device after desiliconization 24...Tapping rate measuring device 25...Silicon needle analyzer A in hot metal...Desiliconization Presiliconism silicon content estimation model B...
Arithmetic logic C for calculating the tapping speed...Arithmetic logic for calculating the amount of silicon removed. D...Arithmetic logic for calculating addition amount E...Pulse setter F...Main ingredient and adjuvant addition amount weigher G...Arithmetic unit Fig. 2

Claims (1)

[Scope of Claims] 1. A method for desiliconizing hot metal, which adds a desiliconizing agent to hot metal tapped from a blast furnace so that the amount of silicon in the hot metal after desiliconization is controlled to a predetermined target value. The amount of silicon in the hot metal is measured by sampling the corresponding hot metal before and after the silica treatment at regular intervals, and the tapping rate is continuously measured, and the measured value or estimated value of the amount of silicon in the hot metal before the desiliconization treatment; The maximum likelihood estimated value of the desiliconization reaction rate constant is calculated using the measured value of the amount of silicon in the hot metal after desiliconization treatment, the actual values of the tapping rate and the amount of desiliconization agent added, and the maximum likelihood estimated value, The amount of silicon removal agent added is determined from the amount of silicon in the hot metal before silicon treatment, the target amount of silicon after desiliconization treatment, and the tapping speed, and by adding the amount of desiliconization agent to the hot metal, silicon in the hot metal after desiliconization treatment is determined. A hot metal desiliconization control method characterized by controlling the amount to be kept constant.