JPH049260A - Device for controlling mold level - Google Patents

Abstract

PURPOSE:To execute mold level control, which hunting is not developed, by using fuzzy inference, obtaining correcting value of drawing velocity from deviation of the mold level to the preset level and deviation between nozzle opening degree on calculation and the actual nozzle opening degree and correcting the drawing velocity. CONSTITUTION:At the time of inputting information of the mold level detected with a level detector 8 and information of the nozzle 2 opening degree detected with a nozzle opening degree detector 15 into a correcting arithmetic unit 16 applying the fuzzy inference, the correcting arithmetic unit 16 inputs first - third membership functions corresponding to each fuzzy inference rule from the knowledge base 16-1, and suitably to the deviation of mold level is obtd. based on the information of mold level from the first membership function and also suitably to the deviation of nozzle opening degree is obtd. based on the information of nozzle opening degree from the second membership function and the optimum velocity corrected value V is obtd. from the third membership based on these suitability to correct the reference velocity given to the drawing velocity control means 14.

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention (Field of Industrial Application) The present invention relates to a mold control device in continuous casting equipment.

(Prior Art) As continuous casting equipment, there is one in which molten steel contained in a tundish is injected into a mold, and a slab that is cooled and solidified is continuously pulled out by a drawing device.

By the way, in such continuous casting equipment, in order to continuously draw slabs, the level of molten steel in the drawing mold (hereinafter referred to as mold level) is always constant, so that the molten steel is injected into the mold from the tundish. It is necessary to control the injection amount of molten metal and the drawing speed of the slab to be drawn from the mold depending on the mold level.

FIG. 4 shows an example of the configuration of such a conventional mold level control device. In Fig. 4, 1 is a tundish containing molten steel, and a sliding nozzle (hereinafter referred to as SN nozzle) 2 is installed at the bottom opening of this tundish 1.
is provided, and this SN nozzle 2 is connected to an SN nozzle drive device 3.
Its opening degree is controlled by. Further, 4 is a mold for cooling the molten steel injected from the tundish 1 through the SN nozzle 2, and the slab 5 solidified here can be pulled out from the bottom opening of the mold. Furthermore, 6 is the slab 5
The slab 5 is continuously pulled out from the bottom opening of the mold while being guided by a plurality of guide roller roughs.

On the other hand, 8 is a level detector that detects the mold level, 9
is a level controller to which the level detection signal detected by the level detector 8 is input, and this level controller 9 gives a control command to the SN nozzle drive device 3 according to level fluctuations, and adjusts the opening degree of the SN nozzle 2. It controls the Also,
Reference numeral 10 is a speedometer that is attached to the rotating shaft of the motor that serves as the drive source of the drawing device 6 and detects the drawing speed of the slab 5. Reference numeral 11 is a level detection signal detected by the level detector 8, which is inputted to determine the mold level. This is a correction device that calculates the change in drawing speed as a correction value ΔV from the broken line characteristic. Reference numeral 12 denotes an adder that adds the reference speed VSV set by the speed setter 13 and the correction value ΔV obtained by the correction device 11. Reference numeral 14 denotes a speed controller to which the speed detection signal V of the speedometer 10 and the addition output signal of the adder 12 are input. This is used to control the extraction device 6.

In the mold level control device having such a configuration, when the mold level fluctuates with respect to the set level, the level controller 9 gives a control command to the SN nozzle drive device 3 based on the detection signal of the level detector 8 to control the SN nozzle drive device 3. The opening degree of the nozzle 2 is controlled. Therefore, the amount of molten steel injected into the mold 4 from the tundish 1 is controlled so that the mold level is approximately at the set level. Further, the speed controller 14 adjusts the reference speed VSV using the correction value ΔV determined by the correction device 11 according to the mold level at that time.
is corrected, and the drawing speed of the slab 5 is controlled based on the deviation from the drawing speed detected by the speed meter 10.

(Problem to be Solved by the Invention) However, in the conventional mold level control device, the correction device 11 calculates the correction according to the mold level, regardless of the opening degree of the SN nozzle 2 controlled by the level controller 9. Since the drawing speed of the slab 5 is controlled by the deviation between the reference speed corrected by the value and the drawing speed detected by the speedometer 10, the SN
Even when the opening degree of the nozzle 2 is on the positive side with respect to the mold level fluctuation, the drawing speed of the slab 5 is controlled to be high in the same way as the control on the negative side. For this reason, hunting occurs in which the mold level alternately fluctuates to the + side and to the one side with respect to the set level, making it impossible to perform stable control.

The present invention provides stable control without notching by controlling the drawing speed of the slab in consideration of the SN opening degree and the mold level in response to variations in the mold level. The purpose of this invention is to provide a mold level control device that can perform the following steps.

[Structure of the Invention (Means for Solving the Problems)] In order to achieve the above object, the present invention injects molten steel contained in a tundish into a mold through a nozzle whose opening degree can be controlled at the bottom of the tundish, In continuous casting equipment in which the cooled and solidified slab is continuously pulled out by a drawing device, there is provided a level detector for detecting the level of molten steel in the mold (mold level), and an opening degree of the nozzle. a mold level control means for controlling the opening degree of the nozzle according to the deviation between the mold level detected by the level detector and a set level; and a slab pulled out by the pulling device. a speed detector for detecting the drawing speed of the drawing device; a drawing speed control means for controlling the drawing speed of the slab by the drawing device according to the deviation between the drawing speed detected by the speed detector and the reference speed; The mold level information detected by the level detector and the nozzle opening information detected by the opening detector are input, and based on these information, a correction value for the slab drawing speed is determined by fuzzy reasoning. A correction calculating means for correcting the reference speed, the correction calculating means calculates the first membership function expressed by the relationship between the deviation of the mold level with respect to the set level and the degree of conformity, and the drawing speed of the slab. The second membership function is expressed by the relationship between the deviation of the actual nozzle opening from the above nozzle opening and the degree of conformity, and the third membership function is expressed by the relationship between the speed correction value and the degree of conformity. When the molten steel level information and the nozzle opening degree information are inputted, a knowledge base prepared corresponding to each of a plurality of preset fuzzy inference rules, and a first to a first to corresponding to each fuzzy inference rule from the knowledge base are input. A third membership function is taken in to obtain a degree of fitness for the mold level deviation based on the mold level information from the first membership function, and a nozzle opening deviation based on the opening information is obtained from the second membership function. The system is configured with an inference section that calculates the degrees of conformity for the vehicle, and calculates the optimum speed correction value using a third membership function based on these degrees of conformity.

(Function) Therefore, in the mold level control device having such a configuration, the mold level information detected by the level detector and the nozzle opening degree information detected by the nozzle opening degree detector are processed using fuzzy reasoning. When inputted to the correction calculation means, the correction calculation means takes in the first to third membership functions corresponding to each fuzzy inference rule from the knowledge base and calculates the first membership function based on the mold level information from the first membership function. In addition to determining the degree of conformity to the mold level deviation, the degree of conformity to the nozzle opening deviation based on the nozzle opening information is determined from the second membership function, and based on these degrees of conformity, the third membership function is used to determine the optimal By determining the speed correction value and correcting the reference speed given to the drawing speed control means, stable mold level control without hunting is possible.

(Example) An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 shows an example of the configuration of a mold level control device according to the present invention. The same parts as those in FIG. 4 are given the same symbols, and the explanation thereof will be omitted, and only the different points will be described here.

In this embodiment, as shown in FIG. The signal and the opening detection signal of the SN nozzle 2 detected by the SN opening meter 15 are respectively input, and based on these detection signals, a correction value Δ■ for the drawing speed of the slab 5 is calculated using fuzzy control theory. A calculation device 16 is provided in place of a conventional correction device.

FIG. 2 shows the internal configuration of the correction calculation device 16. This correction calculation device 16 has a knowledge base 16 in which fuzzy inference rules and membership functions are stored, and information can be input and output through interaction with an operator. -1, the mold level detection signal, and the SN nozzle 2 opening detection signal are respectively input, the fuzzy inference rules and membership functions stored in the knowledge base 16-1 are referenced based on these signals. The inference section 16-2 is provided for calculating the optimum correction value for the drawing speed of the slab 5 at that time by inference.

FIG. 3 shows fuzzy inference rules and membership functions applied to fuzzy inference stored in the knowledge base 16-1. In FIG. 3, A11. A 1
2. A21. A22. A31. A32. A41
゜A42. B+, B2, Bi, and B4 are membership functions, and B+, R2, Ri, and R4 are fuzzy inference rules.

In the fuzzy inference of this embodiment, an inference method that also uses the min operation method is applied. Here, the input (premises) for inference
is the SN opening degree and the mold level, and is the one that connects the input (premise) and output (conclusion). Fuzzy inference rule R+
, R2, R3, and R4. That is, (Premise) Δ1-Δ]l, and Δ2-ΔZ1.

(Fuzzy inference rule) R3: If Δ1 = A + 1, ΔZ-A, and 2, ΔvmB.

R2: If Δ1-A 21 Te, ΔZ-A22 Naraba,
ΔvmB2.

R1: If Δ1-A, □ and ΔZ-A, 2, then Δv
-83.

R4: If Δl −A 4t, Δz−A42,
Δ■1wB4.

(Conclusion) ΔV−Δv1.

However, Δ1. Δ11: Mold level deviation, ΔZ,
ΔZ,: SN nozzle opening degree, ΔV, Δvl: speed correction value.

Next, the correction calculation device 16 uses the mold level and the opening degree of the SN nozzle as input information to correct the speed standard using a correction value Δ.
The fuzzy inference rules and membership functions for finding V will be explained.

■Fuzzy inference rule R Membership function AI indicates the case where the mold level is higher than the set value, the horizontal axis is the mold level deviation,
The vertical axis is the goodness of fit ω.

The membership function A1□ is the speedometer 10 in Figure 1.
The difference between the SN opening calculated from the withdrawal speed detected by the (calculated SN opening) and the SN opening detected by the SN opening meter 15 (actual SN opening difference) is large on the + side ( The calculated SNN opening difference (actual SN opening difference) is shown, the horizontal axis is the SN opening difference, and the vertical axis is the degree of conformity ω.

Here, the fitness degree ω1 for the mold level deviation Δ11 with the membership function A1H is compared with the fitness degree ω2 for the SN opening difference Δz1 with the membership function A1□, and when the fitness degree is smaller, the membership function B1 cut. The ΔV coordinate of the center of gravity of the figure of this cut membership function 81 is the fuzzy inference rule R
The speed correction value is inferred by 1.

■Fuzzy inference rule R2 Membership function A21 indicates the case where the mold level is higher than the set value, the horizontal axis is the mold level deviation,
The vertical axis is the goodness of fit ω.

The number of memberships A22 is the calculated SN opening degree and the actual SN
The degree to which the opening difference is large on one side (calculated SN opening difference minus actual SN opening difference) is shown, the horizontal axis is the SN opening difference, and the vertical axis is the degree of conformity ω.

Here, a mold level deviation Δ! of membership function A21 is given. 1 and the fitness degree ω4 of the membership function A2□ for a certain SN opening degree difference Δzl, and the membership function 82 is cut at the smaller fitness degree. The ΔV coordinate of the center of gravity of the figure of this cut membership function B2 is the fuzzy inference rule R
The speed correction value is inferred by 2.

■Fuzzy inference rule R3 Membership function A31 indicates the case where the mold level is lower than the set value, the horizontal axis is the mold level deviation,
The vertical axis is the goodness of fit ω.

Membership function A3□ is calculated SN opening and actual SN
The degree to which the difference from the opening is large on the positive side (calculated SN opening difference ≧ actual SN opening difference) is shown, the horizontal axis is the SN opening difference, and the vertical axis is the degree of conformity ω.

Here, the goodness of fit for a certain mold level deviation Δ11 of the membership function A31 and the membership function A3
Compare the goodness of fit for the SN opening difference ΔZl with ゜,
The membership function 83 is cut at the smaller fitness level. This cut menpantsubu function B3
The ΔV coordinate of the center of gravity of the figure becomes the speed correction value inferred by the fuzzy inference rule R3.

■Fuzzy inference rule R4 Membership function A41 indicates the case where the mold level is lower than the set value, the horizontal axis is the mold level deviation,
The vertical axis is the goodness of fit ω.

Membership function A4□ is calculated SN opening and actual SN
The degree of the difference from the opening degree is shown to one side (the calculated SN opening difference minus the actual SN opening difference), the horizontal axis is the SN opening difference, and the vertical axis is the degree of conformity ω.

Here, the goodness of fit for a certain mold level deviation Δ1 of the membership function A41 and the membership function A
4□ Compare the degree of fitness for a certain SN opening degree difference ΔZ1, and use the membership function B4 at the smaller degree of fitness.
cut. This cut membership function B
The ΔV coordinate of the center of gravity of the figure No. 4 becomes the speed correction value inferred by the fuzzy inference rule R4.

In this way, the correction calculation device 16 according to the present embodiment calculates each fuzzy inference rule R1 to R1 stored in the knowledge base 16-1 based on the input information of the mold level and the SN nozzle opening.
Membership function A corresponding to R4 11. A
H2~A 4H, A 42 to Δ1ΔZ, respectively, find the fitness, compare the fitness, cut the membership function B, -84 at the smaller fitness, and cut the membership function 81. The coordinates of ΔV of the center of gravity of the figures B4 to B4 are determined as the speed correction value. Therefore, by supplying this speed correction value ΔV to the adder 13 to correct the reference speed, stable control without causing hunting in the mold level can be performed.

It should be noted that the present invention is not limited to the above-mentioned embodiments, and can be implemented in the same manner as described above with the following configuration.

(1) In FIG. 1, it is the case where an SN nozzle is used, or even in the case of a stopper type, the same can be implemented by changing the SN nozzle characteristic to the stopper characteristic.

(2) A mold level control device equipped with a safety device that sets the withdrawal speed to 0 at the highest and lowest limits for the mold level can also be applied within the normal control range.

(3) Membership function All shown in FIG.

A) 2. A21・A22・A31・A32・A
41·A 42f3+ , B2 , B3 , and B4 are to be adjusted when this mold level control device is applied to a plant, and their shapes can be varied.

(4) In Figure 3, the membership function is defined by two elements, but the mold level can be determined by three or more membership functions including other variable elements such as mold width, type of steel material, and SN opening characteristic. The above-mentioned method can also be implemented by controlling .

[Effects of the Invention] As described above, according to the present invention, the calculated nozzle opening obtained from the mold level deviation with respect to the set level and the drawing speed and the actual Based on the deviation from the nozzle opening, the optimum drawing speed correction value at that time is determined, and the drawing speed of the slab is corrected using this correction value, making it possible to perform stable control without hunting. A mold level control device can be provided.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an embodiment of a mold level control device according to the present invention, FIG. 2 is a block diagram of a correction calculation device applying fuzzy inference in the same embodiment, and FIG. An explanatory diagram of fuzzy inference rules and membership functions applied to inference, and FIG. 4 is a block diagram showing a conventional mold level control device. 1... Tanditshu, 2... SN nozzle, 3...
・SN nozzle drive device, 4... Mold, 5... Slab, 6... Pulling device, 7... Guide roller, 8...
・Level detector, 9... Level controller, 10... Speed meter, 12... Reference speed setter, 13... Adder,
14... Speed controller, 15... SN opening degree detector, 1
6...Correction calculation device. Applicant's agent Patent attorney Takehiko Suzue 2 (3 accounts)

Claims (1)

[Claims] Continuous casting equipment in which the molten steel contained in the tundish is injected into the mold through a nozzle whose opening can be controlled at the bottom of the tundish, and the slab is cooled and solidified here and then continuously pulled out by a drawing device. , a level detector that detects the molten steel level in the mold (mold level), an opening detector that detects the opening of the nozzle, and a deviation between the mold level detected by the level detector and the set level. a mold level control means for controlling the opening degree of the nozzle according to the drawing speed; a speed detector for detecting the drawing speed of the slab to be drawn by the drawing device; and a drawing speed detected by the speed detector and a reference speed. a drawing speed control means for controlling the drawing speed of the slab by the drawing device according to the deviation of the drawing device; and mold level information detected by the level detector and nozzle opening information detected by the opening degree detector. speed correction calculation means for correcting the reference speed by determining a correction value for the drawing speed of the slab by fuzzy reasoning based on the input information, and this speed correction calculation means calculates the deviation of the mold level from the set level. The first membership function is expressed by the relationship between the degree of conformity and the degree of conformity, and the second membership function is expressed by the relationship between the degree of conformity and the deviation of the actual nozzle opening from the calculated nozzle opening obtained from the drawing speed of the slab. a knowledge base prepared by associating a third membership function expressed by a relationship between a membership function, a speed correction value, and a degree of fitness with a plurality of preset fuzzy inference rules, and the mold level information. When the nozzle opening degree information is input, the first to third membership functions corresponding to each fuzzy inference rule are imported from the knowledge base, and the mold level based on the mold level information is determined from the first membership function. In addition to determining the degree of conformity to the deviation, the degree of conformity to the nozzle opening deviation based on the opening degree information is determined from the second membership function, and based on these degrees of conformity, the third membership function performs optimal speed correction. A mold level control device comprising: an inference section that calculates a value.