SU1650337A1 - Method and device for automatic control of continuous metal casting process - Google Patents

Abstract

The invention relates to the automatic control of the flow rates of the chiller over the sections of the secondary cooling zone (VZO) of continuous casting machines. The aim of the invention is to improve the quality of the ingot and increase the yield of the metal due to the invariance of the thermal regime of each element of the ingot surface relative to the speed of its extrusion from the crystallizer. As a given control action, an optimal coolant flow rate distribution (ORRO) for the ingot surface element is used. According to ORRO, the costs of the cooler are determined for the areas of VZO based on the processing of information on the behavior of the ingot draw rate for the previous period of time. The effect of technological deviations. The parameters of the casting process, based on the quality of the ingot cooling, are eliminated by adjusting the optimal distribution of the flow rate of the chiller. Variants of the device for carrying out the method are proposed, in which the ORRO is corrected for deviations from the predetermined temperature of the metal supplied to the mold and the surface temperature of the ingot at the exit of the GVO. 2 sec. and 5 hp f-ly, 9 ill. AND

Description

The invention relates to metallurgy, more specifically to the continuous casting of metals, and can be used in automated process control systems for continuous casting machines (CCMs).

The aim of the invention is to improve the quality of the ingot and increase the yield of the metal due to the invariance of the thermal regime of each element of the ingot surface relative to the speed of its drawing out of the crystallizer.

1 shows a block diagram of an apparatus for carrying out the proposed

way; 2 shows graphs explaining the control method; FIG. 3 is a block diagram of a buffer memory block; figure 4 - block diagram of the meter deviations of the specific heat in the mold; Fig. 5 shows timing diagrams for the operation of the device; figure 6 is a block diagram of the device, the first option; in Fig.7 is a block diagram of the device, the second version of the Fig.8 is a block diagram of the computing unit; figure 9 - time diagrams of work: the known and proposed objects.

The block diagram of the device (Fig. 1) denotes a mold 1, an ingot 2, sections 3 of the secondary cooling zone (VZO), regulOs

ate about

CO 00

H

cooler flow meter 4 by sections, ingot pull rate meter 5, cooler flow rate setting master 6, algebraic adder 7, interpolation unit 8, first buffer memory unit 9, first computing unit 10, specific heat removal deviation meter 11 the crystallizer from a given, the second block 12 of the buffer memory, the second computing unit 13.

The device (figure 1) operates as follows.

Prior to casting, on the basis of technological requirements and the laws of heat engineering and thermodynamics, the optimal distribution of the flow rate of the chiller is calculated and entered into the setpoint controller. It contains information on the costs of the cooler in the areas of VZO with the nominal technological parameters, which include the rate of ingot drawdown. This information before casting goes through the algebraic adder 7 and interpolation unit 8 to the inputs of the controller 4, which determine the costs of the cooler in sections 3, which ensure the optimal distribution of the flow rate of the cooler d (g). After the start of casting, the costs of the chiller in the areas of VZO are generally redistributed as a function of deviations of the technological parameters of the casting from the specified ones.

Using the speed meter 5 and blocks 8-10, the coolant flow rates are redistributed to areas as a function of the draw rate, while the self-optical distribution of the coolant flow rate remains unchanged. With the help of speed meter 5, the algebraic adder 7, the first block 9 the memory, the meter 11 of the deviations of the specific heat, the second block 12 of the buffer memory and the second computation block 13 adjust the optimal distribution of the coolant flow rate by the amount deviation sensible heat in the secondary cooling from the ingot at a predetermined input.

The units included in the device (Fig. 1) perform the following functions.

Unit 6 contains the optimal distribution of the flow rate of the cooler in the form of a table for a finite number of values of the parameter m:

g (Tj),, j 1,2J.

The algebraic adder 7 performs the adjustment of the table values of the optimal distribution of the coolant flow rate without changing the values of the argument c. The algorithm of the algebraic adder 7:

g (rf) g (rf) + Ag (rf),

where Ag (rf) is the correction to the second input of the algebraic adder 7 from the output of the second computing unit 13;

g (rf) - values of the corrected optimal distribution of the flow rate of the cooler 5, coming from the output of the algebraic adder 7 to the first input of the interpolation unit 8.

The interpolation unit 8 determines the costs of the cooler for the local emergency zones by calculating the values of the corrected optimal distribution of the coolant flow rate at the current values of the parameter m. The latter are calculated by the computing unit 10 and fed to the second input of the interpolation unit 8. The algorithm of the latter is as follows.

Based on the ratio m - m m, where m is the value of the argument r calculated by block 10 for the J-ro of the SZR, the numbers J and () -1) of the values of the corrected optimal distribution of copper consumption are determined. Then, the consumption rate of the i-ro chiller is calculated using the formula

Y (r.) G (rr-i) + (g (rf). IJ Tj - T - 1

The parameter is determined by processing the output information of the ingot speed meter 5 using the buffer memory block 9 and the computing block 10, which operate in discrete time, 1.2 ...

From the output of the meter 5 to the input of the block 9, the current values of the ingot extrusion speed V (n) are supplied,

In block 9 of the buffer memory, the values of V (n) are stored for the preceding period of time, equal to the sum of the optimal cooling time for the ingot of the TPC and TQ and the maximum time for the ingot element to pass through the mold. The input of the computing unit 10 is supplied with the values of the velocity V (n) only for the preceding period of time. Earlier values of V (n) are used to calculate the transit time of the ingot elements through the mold in the computing unit 13 or in other computing units that can be entered into the device of FIG.

The first computing unit 10 operates according to the following algorithm, which implements the invariance condition

h / v (r) d g (1)

t - r

in discrete time.

The accumulation rate of the ingot drawdown rate stored in block 9 from the current time in a negative time direction is summed up until the condition

I

2 v (r-k-AT)

k 0azr

(2)

where li is the distance from the onset of VZO to the middle of its 1st section;

Лг - discreteness of computer time;

m is the number of summed speed values.

After that, using the obtained value of the number m, the parameter m is calculated

T Agt.

Blocks 8-10 can be executed both on a set of hardware and software of microprocessor technology, for example, KTS LIUS-2, and in hardware on hardware of discrete computing technology.

The automatic control method is illustrated by the graphs in FIG. a is a graph of the optimal time variation of the temperature of an element of the ingot surface (T0 is the time the ingot element is in the irrigation zone); b is a time graph of the flow rate of the cooler providing the temperature graph of FIG. 2; in - approximation of the graph of the intensity of the flow rate of the cooler, generated by the execution of ZVO in the form of sections; r is an example of the implementation of a given graph of the flow rate of the cooler in the VZO when the ingot drawdown rate decreases.

The essence of the method of automatic control of the process of continuous casting of metal is as follows.

Prior to the start of casting, according to a predetermined process schedule, the optimum temperature variation of an element of the ingot surface during its time at 3 VO (Fig. 2, a) by calculation, find the optimal distribution of the coolant flow rate on the element of the ingot surface (Fig. 2, b). The specified distribution is obtained by taking all the technological parameters of the casting process, including the ingot extrusion rate, nominal. Thus, the optimal distribution of the flow rate of the cooler is a numerical display of all process parameters.

casting, including design parameters of ZVO and thermal parameters of the city of ingot at the entrance of ZVO.

Due to the fact that at present

ZVO structurally performed in the form of serially connected sections, the intensity of the flow rate of the cooler throughout the entire length of each of which is constant, the optimal distribution of consumption

0, the cooler is approximated by a step function (Fig. 2c). Therefore, information on the optimal distribution of the flow rate of the cooler can be presented in the form of the flow rate of the cooler in each section of the ZVO at the nominal ingot draw rate. In addition, information on the optimal distribution of the flow rate of a cooler may include distances from the beginning of VZO to the middle of the sections | or the values of the time parameter corresponding to the time of passage of the ingot element of these distances at the nominal drawing speed

,

0 where V0 is the nominal ingot pulling rate.

In the process of casting, in the general case, the values of the optimal distribution of the flow rate by cooling (q) are used at arbitrary values of the argument m, which can be accomplished by interpolating graph b in figure 2, given by the final

the number of values of the argument t g

5 By the considered method of automatic control, two tasks are solved autonomously;

ensuring the invariance of the temperature regime of each element of the ingot surface relative to the speed of its drawing;

elimination of the ingot quality influence on deviations from the specified parameters of the thermal regime by means of adjusting

5 costs of the chiller for all existing sections of VZO.

The invariance of the temperature regime of all elements of the ingot surface in the SZR is achieved by redistributing the flow rate of the cooler based on the processing of information on the history of the ingot draw rate over the previous period of time T0. Necessary and sufficient condition

5 invariance assurance is the determination of the intensity of the flow rate of a chiller according to sections according to the optimal distribution of the flow rate of a chiller q (t) with the value of the argument t m. If t makes the equation

li / v (r) dr,

T - Tj

where I is the section number of ZVO;

li is the distance from the beginning of ZVO to the middle of its 1st part;

g- current time;

v (t) is the rate of ingot extrusion.

Therefore, the essence of the automatic control method applied to solving the first task is reduced, in addition to setting the optimal distribution of the flow rate of the cooler q (i), to calculating the current values of the argument t of the local emergency zones by solving equation (1) and determining the flow rates of the cooler by substitution the parameter g cooler.

The optimal distribution of the flow rate of the cooler is defined in the range of the argument 0 r t0 and is zero outside it. Therefore, if the calculated parameter TI is greater than G0, the flow rate of the coolant for the 1st section is set to zero, which corresponds to its shutdown.

The invariance provided by the fulfillment of condition (1) is complete, i.e. have a place in both established and transient modes. The resulting distribution of the flow rate of the cooler (Fig. 2, c) differs from the initial one only by the error of the stepped approximation generated by the constructive execution of the PZO in the form of sections.

The second task is to improve the quality of the ingot by compensating for deviations of the parameters of the thermal regime from those specified by controlling the costs of the cooler by sections. This compensation does not entail any change in the duration of the ingot cooling in the VZO, and therefore does not interfere with the solution of the first problem.

The optimal distribution of the flow rate of the cooler is specified in the form of a finite number of numbers equal, for example, to the number of sections of the ZVO pole, the initial and final value, i.e. d (o) and d (r0)

g (Tj) ,, j 1,2J,

where J is the number of numerical values.

Due to the fact that all technological parameters of the casting process, with the exception of the ingot pulling speed, vary within a limited range, the description of the adjustment for the optimal distribution of the coolant flow rate can be linearized. In this case, for a corrected optimal distribution of the flow rate of the cooler, the expression

g (rf) - g to) + k I d Q (k) tf) - O)

where N is the number of perturbing factors by which the optimal distribution is corrected;

A 9 W) amendments for the disturbing influence.

The calculation of corrections is generally made by the formula

15

A g (rf.r) ai A f (t - if - A tj), (4)

where t is the current time;

ag coefficients determined by the parameters of VZO and technical process; . ,,. s

Dg (r -TJ - Dt)) - disturbing effect;

D d) is the time from the moment the disturbing action is applied to the ingot element until the last occurrence at the SZO input;

g is the argument of optimal distribution of the coolant flow rate.

Other formulas for calculating corrections are possible depending on the type of disturbing influences and how they are evaluated.

The adjustment of the optimal distribution of the flow rate of the cooler according to formula (2) entails a variable in time change in the cost of the cooler in the IZO sections in accordance with the location of the element of the ingot subjected to the disturbing effect. To ensure such an adjustment, it is necessary to store in the memory the values of the disturbing action for the preceding period of time (t + ATJ).

As for the static coefficients u, they are functions of the casting process parameters and are determined by calculation before the casting as follows. Initially, the initial timeline b of figure 2 is constructed, which ensures obtaining the optimal temperature schedule of figure 2 of the surface element of the ingot with the nominal technological parameters. Then, a continuously acting perturbation in the form of a deviation Af is introduced into the process parameter under study and the calculation of the original time plot b in FIG. 2 is repeated to ensure that the same optimal temperature graph is obtained in FIG. 2.

The increments Ag (rf) of the original time plot b of figure 2 are calculated, affected by the disturbing effect. Static coefficients aj are calculated using the formula

aj-AgfaVAf.

. they are partial derivatives of the given values of the original time schedule b of figure 2, with the optimum temperature schedule a of figure 2 and zero perturbing influences.

The block diagram (Fig. 8) of the buffer memory block 9 contains registers 14, key blocks 15, time delay elements 16, discrete time input channel 17, which is the input of block 9. Output signals arrive at the information input of the buffer memory block 9 meter 5 of the ingot pull rate, and it functions as a delay line, the elements of which are registers. From the outputs of the latter, all N values of the pulling rate can be simultaneously removed during the previous period of time. The operation of block 9 is as follows. After each discrete time pulse arrives at the input 17, the information of each register 14 is transferred to the next one through blocks of 15 keys. The time delay elements 16 are required in order for the information to be entered into this register only after the information contained therein is transmitted to the subsequent register.

By analogy with block 9, block 12 of the buffer memory can also be executed.

Now let us consider the operation of the meter 11 deviations of the specific heat withdrawal in the mold and blocks 12 and 13 for correcting the optimal distribution of the flow rate of the cooler in block 7.

The meter 11 determines the deviation of the specific heat in the mold from the set value, which is equivalent to the deviation of the amount of specific heat in the ingot at the entrance to the SIZ from the set one. The measurement results for a period of time equal to the sum of the optimal time of the ingot passage through the SCR g and the time from the center of the metal column in the mold to the lower cut of the latter are stored in the buffer memory block 12. The computing unit 13 determines the corrections for the values of the optimal distribution of the flow rate of the cooler, taking into account the history of the movement of the ingot through the mold.

The block diagram (Fig. 4) of the meter 11 contains the meters of the water inlet and outlet of the crystallizer 18 and 19, the first unit of comparison 20, the meter 21

0

five

0

five

0

five

0

five

0

five

water for cooling the mold, block 22 multiplying, block 23 division, unit 24 specific heat in the mold, block 25 comparison.

The meter 11 operates as follows.

Measurements 1b, 19 and 21, blocks 20 and 22 measure the intensity of heat extraction in the mold. From the output of dividing unit 23, the signal on specific heat removal in the mold is removed, and the setting unit 24 determines the required heat removal at the nominal parameters of the casting process. Therefore, the output signal of comparator unit 25 corresponds to the deviation of the amount of specific heat from the one specified in the ingot drawn from the mold.

The second computing unit 13 determines the corrections for the optimal distribution of the flow rate of the cooler using the formula

(n) l m Dt Dg} ™

where n is the current discrete time;

aj are the static coefficients determined by the parameters of VZO and process technology;

j is the number of the given value of the optimal distribution of the flow rate of the cooler;

TO - deviation of the specific amount of heat in the ingot from the set;

Aij (p) - the time delay corresponding to the passage of the element of the ingot from the center of the mold to its lower cut.

The time delay Atj (n) is determined by solving the equation

n-Jrf + AnfnH / Ar

Ag (Arf, n) ajAQ / n

Ik tkt

2 v (k)

(6)

k n - G | / At

where IK is the height of the metal column in the mold.

Equation 6 is solved in the same way as equation 4 in computational unit 10. After solving equation (6), the function argument AQ {n - Tf / Arj - Art (n) / At} is determined, the function value is selected from block 12 the buffer memory and the correction is calculated by the formula (5).

The correction of the optimal distribution of the cooling flow rate of the algebraic adder 7, the meter 11 and the blocks 12 and 13 is illustrated in the graphs shown in Fig. 5: g is the graph of the optimal distribution of the flow rate of the cooler g (tf); I is a graph of the deviation of the specific amount of heat in the ingot at the inlet of the VZO from the given one; W, W, and K are graphs of corrections for the values of the optimal distribution of the flow rate of the cooler at the output of block 13 as a function of time; l is the graph of the formed correction for the entire optimal distribution of the coolant flow rate as a function of the parameter g; m - corrected optimal distribution of the flow rate of the cooler at the output of the algebraic adder 7.

After the signal at the output of the meter 11 (Fig. 5) appears, the specified values of the optimal distribution of the flow rate of the cooler are corrected sequentially in time in accordance with the passage of the disturbing influence through the mold and the zones of the PZO. If you do not take into account the time delay required for the ingot element to pass through the mold, the time for correcting the optimal distribution of the coolant flow rate coincides with the optimal cooling time for the ingot in the VZO and does not depend on the rate of extrusion.

The described operation of the device ensures the invariance of the thermal mode of each element of the ingot surface relative to the drawing speed in both established and transient modes. At the same time, the quality of the nonstationarity of the heat removal parameter in the crystallizer is eliminated.

The block diagram (Fig. 6) of the first embodiment of the device in which the automatic adaptation of the optimal distribution of the flow rate of the cooler to the temperature of the metal supplied to the crystallizer includes a unit 26 and a temperature gauge 27 of the metal supplied to the crystallizer, the comparator unit 28, the third unit 29. buffer memory, the third computing unit 30.

The device (6) operates as follows.

The operation of all the blocks, with the exception of the newly introduced ones, does not differ from their operation in the device of FIG. The work of the newly introduced units is as follows. Block 28 compares the specified metal temperature with the actual and at each instant of discrete time and supplies the results of the comparison to the input of the third block 29 of the buffer memory. In the latter, all values of temperature deviations from the specified cooling time equal to the sum of the optimal cooling time are stored. ZVO T0 and the time the element of the ingot of the mold passes from the meniscus to the lower cut at the nominal drawing speed. Computing unit 30 calculates corrections for each value of the optimal distribution of the flow rate of the cooler using the formula

Agi (n);

C ATI p - -g1-4A g t ATI (n) i

Lt (

where j is the number of the given value of the optimal distribution of the flow rate of the cooler;

n - current discrete time;

DT-discrete time;

/ l type) -

H - deviation

LL ii ± l J

set temperature from actual;

Dt) (p) - the time delay caused by the passage of the element of the ingot through the mold and used in the n-th time;

bi - static coefficients.

The time delays Dt (p) in the computing unit 30 are determined in the same way as in the computing unit 13.

Formulas are used.

Ik kg

p -C / Dg - M

 P -

Dg | (p) M Dt.

The optical coefficients in the static coefficients used in the computing unit 13 are also determined during the preparation of the metal casting process.

If during the casting the temperature of the metal supplied to the mold coincides with the set one, the signals coming from the output of the computing unit 30 to the third input of the algebraic adder 7 are equal to zero. When a metal temperature deviates from a predetermined as the section of the ingot with a temperature defect passes through the mold and the ZVO section, correction signals appear at the output of the computing unit 30 for each value of the optimal distribution of the coolant flow. The timing diagrams in FIG. 5 are also valid for explaining the operation of the device in FIG. 6. With the disappearance of the temperature perturbing effect, the numerical values of the optimal distribution of the flow rate of the cooler are also restored sequentially in time.

The block diagram (Fig. 7) of the second variant of the device contains a gauge 31 of deviations of the temperature of the ingot surface at the output of the ZVO from the predetermined, accumulating adder 32 and fourth computing unit 33. The functional purpose of the inserted blocks is to adapt the optimal distribution of the coolant flow rate to the deviations of uncontrolled technological

parameters ZVO from the calculated. Correction of the optimal distribution of the flow rate of the cooler is carried out by extracting, accumulating and processing information about the deviations of the temperature of the ingot surface at the outlet from the set.

The meter 31 determines the deviation of the measured temperature from the specified during the passage of the sections of the ingot, the length of which, to ensure the convergence of the adaptation process, is equal to the distance from the lower section of the mold to the temperature meter (pyrometer) at the exit of the PZV.

The accumulating adder 32 operates according to the algorithm

D T (go) L T (t-1) + D T (t) § D T (g).

I 0 v

where D T (t) is the output of the accumulating adder;

. m is the measurement serial number;

DT (g) - the output signal of the meter 31;

j is the coefficient ensuring the optimal convergence of the adaptation process in the presence of random temperature measurement errors.

Computing unit 33, on the basis of the output information of the accumulating adder, at each instant of discrete time, calculates and feeds the input of the algebraic adder 7 corrections for all values of the optimal distribution of the coolant flow rate at the same time. The formula for calculating the corrections is Ddg (p) (p),

where Cj are the coefficients determined by the parameters of ZVO and ingot; „

D T (n) is the value of the signal D T (t) at the nth instant of discrete time.

The coefficients C, as well as the coefficients aj and bi, are calculated by the preparation of the casting process.

First, the values of g (rf) of the optimal distribution of the flow rate of the cooler are calculated at the nominal parameters of the technological process, which include the specified surface temperature of the ingot at the exit of the hot water supply. Then, the calculation is repeated when the specified ingot surface temperature deviates by the value of D T. The coefficients Cj are found by the formula

g (rf) -§ ()

h 5t.

where 3 (rf) is the optimum distribution of the flow rate of the cooler, obtained as a result of the second calculation.

The adder 32 and block 33 should be implemented on the technical means of microprocessor technology supported by the software. Possible and hardware implementation of them. Fig. 8 shows a block diagram of the hardware execution of the computing unit 33. It comprises a generator 34 of coefficients Cj and J multiplication blocks 35, where J is the number

set values of the optimal flow rate of the cooler.

The presence of the meter 31, the adder 32 and the block 33 in the device (Fig.7) tells it the following advantages over the device of figure 1:

changes in the parameters of ZVO in time during the casting process do not entail deterioration of the quality of the ingot at the output;

an automatic adjustment is made to the optimal distribution of the flow rate of the chiller if errors were made in its calculation, caused, for example, by unreliable a priori information about the characteristics of the cast metal.

The advantages of the proposed facility over the known can be illustrated by examples.

Example 1. Let us have a working VZO with the following technological parameters: the nominal extraction rate of an ingot VQ is 1 m / min, the optimum cooling time of an ingot is 0 20 minutes, which corresponds to an irrigated area of 20 meters.

Let us compare the consequences for the quality of the ingot after a stepwise decrease in the draw rate from 1.0 m / min to 0.2 m / min.

When controlling in a known manner

the following will happen. At the moment of the jump, the extent of the irrigation zone abruptly decreases from 20 to 4 m with the inclusion of intermediate costs of the cooler. Then, at certain intervals in the area from 0 to 4 m, the costs corresponding to the new drawing speed are set.

The part of the ingot, which was located in ZVO at the time of the jump in the area from 4 to 20 m, already

will not receive cooling and will be rejected due to overheating.

When controlling the proposed method after a speed jump, the length of the irrigated area smoothly varies from 20

up to 4 mm linear for 20 min. In this case, each element of the ingot section located in the VZO at the time of a speed jump will receive cooling according to the optimal distribution of the coolant flow rate, i.e. marriage in this case will not be.

The example in question is illustrated in FIG. 9 by time diagrams, where n is a graph of the ingot extrusion rate; O, n — graphs of the irrigated plots, respectively, with known and proposed control methods; p is a graph of the variation of the parameter TU used to calculate the intensity of cooling of an element of the surface of an ingot located at a time of velocity jump at a distance of more than 4 mm from the beginning of VZO; c is a graph of the optimal distribution of the flow rate of the cooler d (g); t and y are graphs of the actual cooling intensity of the said element, respectively, with the known and proposed control methods.

A similar picture takes place also in the case of an abrupt transition from a low drawing rate to a nominal one. With the known control method, the ingot section that has already received the required cooling from 4 to 20 m in the ZVO is re-cooled, which leads to rejection due to overcooling. The proposed control method excludes the occurrence of a marriage in this case.

PRI mme R 2. Let there be ZVO with the same technological parameters as in Example 1, which constructively consists of sections. Let us see what happens to the ingot with the known and proposed control methods if the pulling rate changes abruptly from the nominal value to zero.

When controlling cooling in a known manner, the total length of all sections of the emergency water treatment should be reduced to zero and intermediate costs of the cooler should be established. All the ingot in the ZVO actually loses cooling. The unit costs of the cooler, corresponding to the altered drawing rate, can be set along the sections only through the time required for the ingot surface element to pass from the metal meniscus in the mold to the level corresponding to 0.1-0.3 of the 5th ZVO section. Flow switching cannot take place, as the waiting time due to the zero pulling rate becomes infinite. Thus, the known method turns out to be inoperable, which leads to a marriage of the ingot section in the ZVO due to overheating of the surface.

When controlling the cooling process by the proposed method, stepwise

a change in the exhaust rate to zero entails a smooth redistribution of the costs of the coolant in all sections of the VZO. All sites, starting with the last,

disconnect sequentially. After a period of time r0, which is the optimal cooling time for the element of the ingot surface, the first section also turns off. As a result, each surface element that appeared in the ZVO after the ingot is stopped, will receive cooling in accordance with the optimal distribution of the flow rate of the cooler, except for the approximation errors caused by the sectional design of the ZVO. Consequently, the proposed method provides optimal cooling of the ingot stuck in the SZO, thereby preventing rejection in this case.

Claims (7)

1. Automatic control method the process of continuous casting of metals, including the supply of metal to the mold, the drawing of an ingot from a mold with a variable speed, cooling of the ingot in the secondary cooling zone with a cooler dissipated by grouped nozzles grouped in sections, measuring the ingot extrusion rate and controlling the intensity of cooling along the secondary cooling zone with simultaneous changes in the number of working sections of it, that, in order to improve the quality of the ingot and increase the yield of the metal by ensuring the invariance of the thermal regime of each element of the ingot surface relative to the speed of its extrusion from the crystalline s previously determined optimum flow distribution of d (r) for the cooler element ingot surface, provided where m is the residence time of the current element of the ingot in the secondary cooling zone, z0 - optimal cooling time, and during casting, deviations of the thermal mode parameter of the casting process from the specified technology are continuously measured, correct the optimal distribution of the coolant flow rate based on the measurement results while maintaining the limits of change C and determine the coolant flow rate gi over the secondary cooling zone sections by the expression gi gWn, where I is the number of the section of the secondary cooling zone; d (t) is the corrected optimal distribution of the flow rate of the cooler; tu - time values t ensuring fairness of equality li / v (r) dr, Mr-tj where h is the distance from the beginning of the secondary cooling zone to the middle of the t-ro of its portion; v (r) is the ingot extrusion rate. 2. Method of pop. 1, characterized in that the specific heat selection in the crystallizer is used as a parameter of the thermal mode of the casting process. 3. A method according to claim 1, characterized in that the temperature of the metal fed to the mold is used as a parameter of the thermal mode of the casting process. 4. A method according to claim 1, characterized in that the temperature of the casting process at the exit of the secondary cooling zone is used as a parameter of the thermal mode of the casting process. 5. A device for automatic control of the process of continuous casting of metals, containing cooler flow controllers for the sites and an ingot drawing rate meter, characterized in that, in order to improve the quality of the ingot and increase the yield of the suitable metal by ensuring the invariance of the heat regime of each element the ingot surface relative to the speed of its drawing out of the crystallizer, the unit for setting the optimum flow rate distribution of the cooler, the meter for deviation of the specific selection t in the mold from a given, algebraic adder, interpolation unit, two block buffer memory and two computational units, the output of the unit for the optimal flow rate distribution of the cooler connected to the first input of the algebraic adder, the output of which is connected to the first input of the interpolation unit, input the first block of the buffer memory is connected to the output of the drawing speed meter, and the output is connected to the input of the first computing unit; the second input of the interpolation unit is connected to the output of the first computing unit Its outputs are connected to the chiller flow regulator inputs by sections, the meter of specific heat extraction deviations in the mold is connected to the output of the drawing rate meter and the output to the input of the second buffer memory unit, the second computing unit is connected to the outputs of the first and the second blocks of the buffer memory, and the output to the second input of the algebraic adder. 6. The device according to claim 5, characterized in that, in order to improve the quality of the ingot by adapting the optimal distribution of the flow rate of the cooler to the temperature of the metal supplied to the mold, a sensor and a temperature meter of the metal supplied to the mold, a comparison unit, a third block are introduced into it the buffer memory and the third computing unit, the inputs of the comparator unit are connected to the outputs of the setter and the temperature meter of the metal supplied to the crystallizer, and the output through the third block of the buffer memory is connected to the The third input of the third computing unit, the second input of which is connected to the output of the first buffer memory, and the output to the third input of the algebraic adder. 7. The device according to claim 5, in which, in order to improve the quality of the ingot by adapting the optimal distribution of the coolant flow rate to changes in the technological parameters of the secondary cooling zone, a deviation meter is introduced into it the temperature of the ingot surface at the outlet of the secondary cooling zone from a given one, the accumulating adder and the fourth computing unit, which are connected in series and connected to the fourth input of the algebraic adder. to Shch-i) Vfrl block. five iLn . t, w v ixt 4Fye .H t, t at t, t v (n-m) t Ufa-) uln-H) 17 18 20 nineteen 21 Xv Ld, N) ldgk) about  w ugju g, ABOUT Adm fffr; rf i i; T-qp: i i i i (bus. 5 14 I123 chr-f  well 6 issue1 2 tilt) From flea 5 Fy x ,and T S - V y g , l xrr; To Yo -H s4 I "five   I / 1SW4XNSSS3 / y i ABOUT v о со -four f I f t), M / PON t, HUH 9