DD285935A5 - METHOD FOR PLANNING CONTROL IN ROLLERS ON ROLLED RUGS - Google Patents

Abstract

The invention relates to a method for flatness control in rolling mills on rolling mills with 4 or more rollers, in which the tensile stress distribution of the rolled strip is measured over the bandwidth and compared with a desired distribution and coefficients are formed from the analysis of the difference between actual and desired distribution serve by means of suitable transfer functions for adjusting the actuators of the rolling stand. Flatness control; Tape; Rolling mill; Target distribution; tensile stress; approximation; Flatness; Cold rolling}

Description

is representable, where y is the difference of actual and desired curve at the point χ, χ is the normalized coordinate of the bandwidth and Ti, T ₂ , and T ₃ Chebyshev polynomials of the form

T ₁ = X

T ₂ = 2x ² - 1

T ₃ = 8X ⁴ - 8x ² + 1

and whose coefficients a ,, a ₂ , a ₃ serve to adjust the actuators.

2. The method according to claim 1, characterized in that for producing the flatness of the strip to be rolled, the support or intermediate roll bending is changed such that, if I a ₃ 1> g, without changing the work roll bending a ratio of ka _3rd = a ₂ would be just after re-measurement and approximation, and the Arbnitswalzenbiegung is changed so that the after changing the support or Zwischenwalzenbiegung theoretically adjusting coefficients a ₂ and a _{3 are} minimized, and that if a ₃ 1 <g, the backup or intermediate roll bend is changed so that the a ₂ coefficient is minimized until I a ₂ 1 <g, where g is a boundary value characterizing the flatness criterion and k is a scaffold dependent factor.

3. The method of claim 1 and 2 for rolling stands with axially movable work or intermediate rolls, characterized in that when the adjustment range of the support or intermediate roll bend is exhausted, to adjust the necessary ratio of the coefficients a ₂ and a ₃ and to minimize the a ₂ coefficient, the working or intermediate roller displacement is used.

4. The method according to claim 1-3, characterized in that for determining the necessary actuator changes transfer functions of the form

kx.i = bo + b, f (Fw) + b ₂ h, + [b ₃₁ + b ₃₂ f (F _w )] X ₂ + [b ₄₁ + b ₄₂ f (F _w ) l X ₃

be used, where

k _x - modification of the coefficient a ₂ or a ₃ bj related to the actuator change - factors dependent on technical-technological process parameters h _f - thickness of the rolled strip

Xi - mean instantaneous size of the other actuators.

For this 2 pages drawings

Field of application of the invention The invention relates to the production of flat belts when rolling strip on rolling stands with 4 or more rolls. Characteristic of the known technical solutions

The rolling process of tapes is generally characterized by variations in the profile of the hoop over the width and length of the belt, rolling force variations and other instabilities. These lead to fluctuations in the flatness of the rolled strip.

In order to be able to compensate for these variations, rolling stands with 4 or more rolls are generally provided with roll bending devices and partly with axial roll shifting mechanisms and equipped with devices for measuring and controlling the flatness.

Known is a control arrangement on strip and plate rolling mills (Patent DE 3115461), in which the output signals of the Bandzugmeßeinrichtung are given to two evaluation elements, one of which generates a signal to adjust the work roll deflection and the other a signal to adjust the intermediate roll displacement. A method for regulating the flatness by targeted change of the work roll and intermediate roll bending is described in the patent DE 3331335. In this method, the difference between the measured by a Bandspannungsmeßeinrichtung tape tension distribution and the tape tension setpoint distribution in the edge and middle region of the tape is approximated by a respective Ausgleichsparabel whose coefficients a _K (edge region) and a ** (center region) as a deviation in each case the bending of the work rolls (a _K ) or the intermediate rollers (a _M ) influence. The problem here is the determination of the boundary between the boundary and center area, which is fluid depending on the technological conditions. To take into account the influence of the work roll and intermediate roll bending on the center or edge area of the bed, mutual correction of the two coefficients takes place according to the size of the other. The disadvantage of this method is that a separate consideration of the center and edge region takes place and the influence of the center or edge region can be taken into account by the work roll or intermediate roll bending only by a complex coefficient correction.

There are no known methods that take into account the influence of the other actuators, the rolling force and the thickness of the rolled strip on the transmission behavior of this actuator in determining the need to correct the flatness image actuator change.

Object of the invention

The aim of the invention is to significantly improve the flatness during cold rolling of countries.

Explanation of the essence of the invention

The invention has for its object to determine by a suitable method, the automatic flatness control on rolling stands with 4 or more rolls control deviations and eliminate. The object is achieved according to the invention as follows:

1. The determined by a Bandspannungsmeßeinrichtung belt tension distribution of the rolled strip is compared with a desired distribution and determines their discrete difference. The difference in the belt tensions is translated into a difference in belt elongations across the width, which serve as a measure of the unevenness (waviness) of the rolled strip. According to the invention, this difference is determined by means of a suitable approximation method by a compensation curve of the shape

Υ = B ₀ + BiT + B ₂ T ₁ + 8jT4

approximated.

The advantage of using the Chebyshev polynomials (T ₁ ) for approximating the discrete difference curve is that the obtained approximation coefficients become geometrically interpretable, ie that z. For example, with an ideal quadratic curve of the bandgap differential curve and a maximum banding difference of 301-unit, the quadratic coefficient (a ₂ ) of the compensation curve would yield 30 and the other coefficients zero.

The a, coefficient characterizes a one-sided ripple that can be corrected by one-sided Walzenansteilung.

The a ₂ coefficient characterizes the degree of expression of the quadratic part of the waviness (edge ripple). The expression of the non-quadratic part of the waviness (off-center waviness, "quarter buckling" or simultaneous edge and center waviness) is described by the a ₄ coefficient.

2. The transfer behavior of the work roll bend is such that it affects the square and the non-quadratic portion of the waviness in the same direction and in a constant, dependent on the roll geometry ratio k. The bending of the support or intermediate rolls, however, only affects the square portion of the waviness. In order to bring the appearing ripples (unevenness), ie the deviation of the difference curve from the zero position, to zero, the non-square component of the ripple is first checked. If it is smaller than a predetermined limit g, which is considered as a criterion for flatness, then by means of the intermediate roll bending, which does not influence the non-squared part of the flatness, the a ₂ coefficient becomes | a ₂ | <g minimizes.

If the non-squared portion is greater than the predetermined limit g, it must be reduced by means of the work roll bending, but also the quadratic component (in the same direction) is changed, ie, in order to ensure the flatness of the band, must first computationally the ratio a _; / a ₃ = k can be set. This is done with the help of 'Jer support or intermediate roll bending.

After reaching the ratio S ₂ Za ₃ = k, the coefficients are proportionally reduced by using the work roll bend.

The setting of the ratio of the coefficients is an intermediate step of the control algorithm for determining the total actuator operation, which is not executed as a single step, both actuators are changed simultaneously.

3. The transfer behavior of the work rolls or intermediate roll displacement is similar to that of the backup roll or intermediate roll bend. If the adjustment range of the bend to establish the necessary ratio of a ₃ and aj Koefflzienten or to minimize the aj coefficient at | a _} | <g is exhausted, the axial wall displacement can be used here. The same principle can be used to restore the operationality of the bend when its operating range is exhausted.

4. In order to determine the necessary size of the actuator insert for the production of the desired curve of the flatness, transfer coefficients of the form

ky = - X

necessary, where a is the change of the coefficient a ₂ or a ₃ at an actuator change of X means. The transfer behavior of an actuator {roll bends, axial displacement of the rolls) is influenced by the rolling force, the thickness of the rolled strip, the level of the other actuators and stand-specific characteristics, ie the transfer coefficients are variable during the rolling process as the technological parameters change. To determine the size of the actuator insert for the actuator Xi is the relationship

k _x , = bo + b, f (F _w ) + b, h, + [b ₃₁ + b _M f | Fw)] X ₂ + Ib ₄ , + b ₄₂ (Fw) l X ₃

used.

With the help of the transfer function can now by the evaluation unit (process computer) of the rolling mill, the Transmission coefficients are calculated as a function of the technical-technological parameters and the necessary Actuator changes are determined in accordance with the deviations from the desired curve of the flatness. The determined Stelländerände ^r ments are then used to adjust the actuators. embodiment

The cold rolling process on a β-roll stand with work and intermediate roll bending and axial intermediate roll displacement is characterized by the following quantities:

- Thickness of the starting material 2.5mm

- incoming thickness 1.26mm

- leaking thickness 0.875mm

- Bandwidth 1250mm

- rolling force 11kN.

The flatness criterion is 2 l-unit. The basic setting of the actuators is zero. The actuators are adjustable in the following ranges:

- Stripper Bend: -430 ... + 43OkN

- Intermediate roll bending: -800 ... + 80OkN

- intermediate roller displacement: SOOmmjeWalze.

The coefficient k is 3.3 for the given framework. The reaction is described to three types of flatness errors:

a) A deviation from the target curve was determined, which after the regression values of a ₂ => 15 and a ₃ = 0 (Figure 1, curve Hergibt.

Since aj <2 l-unit, the response to the waviness is through the intermediate roll bend. By means of the transfer function an intermediate roll bending change of + 22OkN is determined. After setting this value, the regression after re-measurement gives width of 8; · = 0.4 and B ₁ = 0.1 (Fig.1, curve 2). Since both coefficients are smaller than the flatness criterion, the target curve is reached.

b) A deviation from the nominal curve was determined, which after the regression yields values of a ₂ = 20 and a ₃ = 10 (FIG. 2, curve 1).

Since 8)> 2l-unit, a ratio of 8 ₂ / a ₃ = 3.3 must first be established, ie, a _{2 should} assume a value of about 33. For this purpose, an intermediate roll bending change -20OkN is determined with the aid of the transfer function. The coefficients aj = 33.4 and a ₃ = 10.0 (FIG. 2, curve 2) would now result. In order to minimize both coefficients, a change of the work roll bending of + 19OkN is determined by means of the transfer function. After setting the new values for working and intermediate roll bending, the coefficients of a _t = 0.2 and a ₃ = 0 are obtained with renewed measurement and regression (FIG. 2, curve 3). This is the target curve is reached.

c) A deviation from the target curve was determined which, after the regression, gives values of 8} = 15 and S ₁ = 15 (Figure 3, curve H, since 8)> 2I, first oin ratio on a ₂ / a ₃ ° 3,3, ie, a ₂ would have to assume a value of -50, determined by means of the transmission "'' inMion wiiu ome intermediate roll deflection change of + 1050kW. Since this change the Stellbereicr. exceeds the intermediate roll deflection, the maximum intermediate roll deflection of + 80OkN is assumed. The coefficients are 8] = -38.2 and B ₁ = -14.9 (FIG. 3, curve 2). In order to realize the necessary coefficient ratio, the Zwischenwalzenverschinbung must be used. With the transfer function, the necessary intermediate roller displacement of 1210 mm is determined. The coefficients would now result in a ₂ = -54.9 and a ₃ = -16.9 (FIG. 3, curve 3). To minimize dioser coefficients, the change of the work roll deflection with -31OkN is now determined. After adjustment of all three actuators result in re-measurement and regression, the coefficients of the compensation curve to a ₂ = -0.2 and a ₃ = -0.5 (Fig. 3, curve 4). This is the target curve is reached.

Claims (1)

1. A planarity control method in rolling mills with 4 or more wheat, in which the tensile stress distribution of the rolled strip is measured and compared with a Sc '! Distribution and coefficients are formed from the analysis of the difference between actual and target distribution, which serve to adjust the actuators of the roll stand, characterized in that the discrete difference of the actual and desired distribution of the tensile stress is approximated by means of an approximation method by a compensation curve, the polynomial of the form y = a ₀ + a, T ₁ Ix) + a ₂ T ₂ (x) + a ₃ T ₃ (x)