JPH115110A - Rolling method and equipment of h-shaped steel - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide the method and the equipment for the H-shaped steel which makes the operator easily understand the causal relation between the manipulated variable and the controlled variable and, also easily adjust the control system. SOLUTION: The nine manipulated variables are the level roll gap, the vertical roll average gap of the universal mill, the average gap of the edgier rolling mill, the positions at driving and free sides of an interlocking adjusting device, the axial direction position of the level roll, the leveling of the edgier rolling mill, the difference of the vertical roll gap between the free side and the drive side and the pass line. The nine quantities of state of the differences, each of which is controlled by the corresponding manipulated variables respectively, are the web thickness of the H-shaped steel the average thickness of the flange at four points, the average width of the flange at the free side and the drive side, the central polarization of the flange at the free side, imperfect alignment, the difference of the width of the flange between at the free side and the drive side, the difference of the thickness of the flange between the free side mean value and the drive side mean value and the difference of the thickness of the flange between the upper side average value and underside average value. As the whole control system are divided into the independent vectors with the nine state variables and the nine operation variables, the quantity of the state can be correspondent to the manipulated variable (the actuator).

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and an apparatus for rolling an H-section steel, and more particularly to an improvement in dimensional accuracy of an H-section steel.

[0002]

2. Description of the Related Art The present applicant has previously filed Japanese Patent Application No. 7-2603.
Patent Application No. 84, a bite adjusting device capable of adjusting the bite posture or position of the rolled material, a universal rolling mill in which a horizontal roll is supported movably in the axial direction, a drive side and a free side A rolling method including the following processing steps has been proposed as a rolling method using a group of universal rolling mills for rolling H-beams, each including an edger rolling mill in which a flange width reduction roll gap can be individually set. (A) At a position close to the upstream side of the universal rolling mill, the web thickness, four flange thicknesses, and four flange leg lengths of the H-section steel after forming and rolling were measured, and based on the measurement results, each target dimension and Calculating the deviation of (B) From the biting posture or position of the rolled material, the reduction rate at four flange thicknesses, the reduction rate of the web thickness, the reduction rate of the left and right flange widths, the deformation characteristics of the material to be rolled, and the characteristics of the rolling mill, Adjustment of biting posture or position of rolled material to obtain H-section steel having desired dimensions in subsequent rolling passes without disturbing shape, roll gap composed of upper and lower horizontal rolls and vertical rolls, edger rolling Calculating the correction amounts of the left and right roll gaps of the machine. (C) a step of changing the position of each roll based on the result of this calculation, and thereafter rolling in one or more passes.

[0003]

However, in the rolling method proposed above, if the dimensional correction is performed in one pass, 27 relational expressions (27 × 27 matrix) and the dimensional correction must be performed. 45 for 2 passes
ヶ (45 × 45 matrix) relational expression is required, ・ The control structure is complicated, ・ The causal relationship between the operation amount and the control amount is difficult for the operator to understand, ・ Where to adjust the matrix It was difficult to adjust the control system without knowing what to do.

The present invention has been made to solve such a problem, and has a simple control structure, makes it easy for an operator to understand the causal relationship between the operation amount and the control amount, and adjusts the control system. An object of the present invention is to provide a simple H-beam rolling method and apparatus.

[0005]

According to the present invention, the whole control is separated into independent vectors, and the control system is constituted by nine state variables and nine operation variables. The feature is that it is possible to cope with this. Specifically, it has the following configuration. (1) The method for rolling an H-section steel according to one aspect of the present invention includes:
A universal rolling mill in which a bite adjustment device capable of adjusting the bite posture or position of the rolled material is arranged on the rolling entry side, and a horizontal roll is supported so that the horizontal roll can move in the axial direction, and a flange width reduction on the drive side and the free side. In a rolling method using a group of universal rolling mills for rolling H-beams each provided with an edger rolling mill in which a roll gap can be individually set, a horizontal roll gap and a vertical roll average gap of the universal rolling mill, an average of the edger rolling mill Gap, the position of the drive side and the free side of the biting adjustment device, the position of the horizontal roll in the axial direction, the leveling of the edger rolling mill, the difference between the free side and the drive side of the vertical roll gap, and the pass line. The web thickness of the H-section steel, the average thickness of the four points of the flange, Average of free side and drive side of flange width, center deviation of drive side of flange, center deviation of free side of flange, axis deviation, difference between free side and drive side of flange width, average value of free side of flange thickness and drive The nine state quantities of the difference from the side average value and the difference between the upper average and the lower average of the flange thickness are controlled. The above-mentioned pass line is formed by upper and lower horizontal rolls of a universal rolling mill, and the position of the pass line is changed by moving the upper and lower horizontal rolls. The above-mentioned state quantities and operation quantities are described in the following embodiments using the following reference numerals.

(State quantity) 1a. H-section web thickness: tw 2a. The average thickness of 4 points of the _{flange: (tf 1 + tf 2 +} tf 3 + tf 4) / 4 3a. Free side and drive side of the mean value of the flange width: (Bd + Bf) / 2 or _{(bf 1 + bf 2 + bf} 3 + bf 4) / 2) 4a. Center deviation of drive side of flange: (bf ₁ -bf ₂ ) / 25a. Center deviation on the free side of the flange: (bf ₃ −bf ₄ ) / 26a. Axial _{deviation: (tf 3 + tf 2 -tf} 1 -tf 4) / 2 7a. Difference between free side and drive side of flange width: Bf-Bd Or _{bf 3 + bf 4 -bf 1 -bf} 2 8a. The difference between the free side average and the drive-side average value of the flange _{thickness: (tf 3 + tf 4 -tf} 1 -tf 4) / 2 9. The difference between the upper average and lower average flange _{thickness: (tf 1 + tf 3 -tf} 2 -tf 4) / 2 Note that it is _{Bd = bf 1 + bf 2 Bf} = bf 3 + bf 4.

(Amount of Operation) 1b. Horizontal roll gap: Sh 2b. Vertical roll average gap: (Svd + Svf) / 23b. Average gap of edger mill: (Sed + Sef) / 24b. Drive side position of biting adjustment device: Gd 5b. Free side position of bite adjustment device: Gf 6b. Horizontal roll shift: SF 7b. Edger mill leveling: -Sed + Sef 8b. Difference between free side and drive side of vertical roll gap: (Svf-Svd) / 29b. Pass line position: PL

(2) In a method for rolling an H-section steel according to another aspect of the present invention, a biting adjustment device capable of adjusting a biting posture or a position of a rolled material is arranged on a rolling-in side, and a horizontal roll has a shaft. Equipped with a universal rolling mill supported so as to be movable in two directions, and an edger rolling mill capable of individually setting roll width reduction roll gaps on the drive side and the free side.
In a rolling method using a group of universal rolling mills for section steel rolling, a horizontal roll gap and a vertical roll average gap of the universal rolling mill, an average gap of the edger rolling mill, positions of a drive side and a free side of the biting adjustment device, The position of the horizontal roll in the axial direction, the leveling of the edger rolling mill, the difference between the free side and the drive side of the vertical roll gap, and extracting a specific number of operation amounts from nine operation amounts of the pass line, The web thickness of the H-section steel, the average thickness of the four flanges, the average value of the flange width on the free side and the drive side, the deviation of the center of the flange on the drive side, and the center of the flange on the free side corresponding to the nine manipulated variables. Deviation, axis deviation, difference in flange width between free side and drive side, difference between flange side free side average value and drive side average value, Of the nine states of the difference between the upper average and lower average di- thickness, to control the state quantity corresponding to the extracted operation amount of the. (3) The method for rolling an H-section steel according to another aspect of the present invention, in the rolling method according to the above (1) or (2), further comprising: The steel web thickness, the flange thickness at four locations, and the flange leg length at four locations were measured, and based on the measurement results, H
Performs dimensional control of section steel.

(4) In a rolling device for H-section steel according to another aspect of the present invention, a biting adjustment device capable of adjusting a biting posture or a position of a rolled material is arranged on a rolling-in side, and a horizontal roll has a shaft. Equipped with a universal rolling mill supported so as to be movable in two directions, and an edger rolling mill capable of individually setting roll width reduction roll gaps on the drive side and the free side.
Universal rolling mill group for section steel rolling, horizontal roll gap and vertical roll average gap of the universal rolling mill, average gap of the edger rolling mill, drive side and free side positions of the biting adjustment device, the horizontal roll Axial position, leveling of the edger mill,
The difference between the free side and the drive side of the vertical roll gap,
And 9 pass lines as operation amounts, the web thickness of the H-section steel corresponding to the operation amount, the average thickness of four points of the flange, the average value of the free side and the drive side of the flange width,
Center deviation of the drive side of the flange, center deviation of the free side of the flange, axis deviation, difference between the free side of the flange width and the drive side, difference between the average value of the free side of the flange thickness and the average value of the drive side, and the upper average of the flange thickness 9 of the difference between
And an arithmetic and control unit for controlling the number of state quantities. (5) In the rolling device for H-section steel according to another aspect of the present invention, the biting adjustment device capable of adjusting the biting posture or position of the rolled material is arranged on the rolling entry side, and the horizontal roll moves in the axial direction. A group of universal rolling mills for rolling H-beams, each comprising a universal rolling mill supported so as to be able to move, and an edger rolling mill in which a roll width of a flange width reduction on a drive side and a free side can be individually set. Horizontal roll gap and vertical roll average gap, average gap of the edger rolling mill, drive side and free side position of the biting adjustment device, axial position of the horizontal roll, leveling of the edger rolling mill, the vertical A specific number of operation amounts are extracted from the difference between the free side and the drive side of the roll gap and the nine operation amounts of the pass line, and the nine operation amounts are extracted. Corresponding web thickness of H-section steel, average thickness of four points of flange, average value of flange width on free side and drive side, center deviation of drive side of flange, center deviation of free side of flange, axial misalignment, flange width Of the nine state quantities of the difference between the free side and the drive side, the difference between the free side average value of the flange thickness and the drive side average value, and the difference between the upper side average and the lower side average of the flange thickness, the above-mentioned extracted state quantities are extracted. And an arithmetic and control unit for controlling a state quantity corresponding to the manipulated variable. (6) The rolling device for H-section steel according to another aspect of the present invention is the rolling device according to (4) or (5) above, wherein the H-shape after forming and rolling is located at a position close to the upstream side of the universal rolling mill. The apparatus has a hot dimension meter for measuring the steel web thickness, the flange thickness at four locations, and the flange leg length at four locations, and the arithmetic and control unit controls the dimensions of the H-section steel based on the measurement results.

[0010]

DESCRIPTION OF THE PREFERRED EMBODIMENTS An influence coefficient method used in an embodiment of the present invention will be described with reference to FIGS.
Assuming that the rolling mill is one process, as shown in FIG. 10, a material in an input state x (dimension) is input and an operation amount u
When (gap) is given, an output state y is obtained. The relationship between x, u, and y is non-linear, but can be expressed by the following equation if attention is paid to minute changes. Note that Δy, Δx, Δ
u is a variable matrix, and A and B are constant matrices.

[0011]

(Equation 1)

In the case of a section steel mill, multiple-pass rolling is performed. In the case of two-pass rolling, as shown in FIG. 11, the output of the first pass is the input of the second pass. Is represented by

[0013]

(Equation 2)

By the way, in the case of performing the dimensional control with this effect coefficient, the input state and the output state is given by the sensor signal or the target value (deviation 0), the operation amount Delta] u ₁ Delta] u ₂ meet it ... However, since the number of manipulated variables (the manipulated variable per pass × n passes) is too large for dividing the number of variables to be controlled (input or output state variables), some restriction is imposed on the manipulated variable to set the number of variables. Need to match. In the case of an H-section steel, for example, if the operation amount is selected so as to be the same as the number of state variables, and further the constraint that the operation amounts of all the paths are the same, the following equation is obtained (see FIG. 12). ).

[0015]

(Equation 3)

In the case of feedback control, by setting Δx _n to the sensor output deviation and setting the input state Δx ₀ to 0, the operation change Δu can be obtained by the following equation. However, in consideration of the sensor noise and the model error, the weight coefficient w
Multiply by ₁ to control the next material. The operation amount of the next material = the operation amount of the material + w ₁ Δu This w ₁ can be theoretically variously derived, but since sensor noise and model error cannot be accurately evaluated, empirically, w ₁ = 0. It is sufficient to adjust it experimentally as 5 to 1.0.

[0017]

(Equation 4)

In the case of feedforward control, ΔX ₀ is used as the sensor output deviation for the first sensor input of the target material, and the output state ΔX _n after n passes is set to 0, so that the operation change amount Δu is calculated by the following equation. Can be requested.

[0019]

(Equation 5)

For the second and subsequent sensor inputs of the target material, the disturbance on the model from the previous sensor input to the current sensor input can be estimated, so that the operation change amount including the disturbance compensation is obtained. Better control. It is better to model the most basic offset of the equipment (displacement from the zero point) as the disturbance. If the offset amount is Δd, the disturbance Δd can be estimated by processing as shown in FIG.

In this measurement, it is assumed that the state before rolling in the i-th pass is measured, and the deviation of the measured value is ΔX _i-1 . It is assumed that the previous measurement was performed before rolling in the ( _im ) pass, and the deviation of the measured value is ΔX _im−1 . It is assumed that an operation amount change amount Δu _im of feedforward is added between the ( _im ) path and the (i-1) path. ΔX _i-1
Is determined by the following equation.

[0022]

(Equation 6)

Therefore, in consideration of sensor noise and model error, Δd can be obtained by the following equation.

[0024]

(Equation 7)

It is to be noted that w ₂ is determined by a weighting factor and adjusted by experiments. Assuming that this disturbance is input with the same value after the current pass, the manipulated variable change amount Δu that makes the deviation n-pass ahead zero is similarly obtained by the following equation.

[0026]

(Equation 8)

In the case where the feedforward control and the feedback control are used together, when the control result of the feedforward of the target material is reflected in the feedback control of the next material, the amount of change in the operation amount of the feedforward affects the final dimension. The amount ΔX _FF is calculated in the same manner as in the above equation (a), and the measured value ΔX used for feedback calculation is calculated.
_{The FB} subtracts the effect of the feedforward control from the actual measurement value ΔX of the target material by ΔX _FF .

[0028]

(Equation 9)

FIG. 1 is an explanatory diagram showing the relationship between the equipment of the H-section steel size control device according to one embodiment of the present invention and the influence coefficient parameter. In FIG. 1, reference numeral 10 denotes a universal mill of the previous pass, 11 denotes an edger rolling mill, and 12 denotes a universal mill of the corresponding pass. Here, a specific method of calculating the influence coefficient matrix will be described. Although FIG. 1 does not show a bite adjusting device capable of adjusting the bite posture or position of the rolled material, it is incorporated as a bite control model (δ1, δ2) of the universal mill for the corresponding path. . (1) Description of Symbol Here, a state at a specific position in one pass is represented. In deriving the influence coefficient matrix of the entrance state and the exit state of one pass of rolling, it is easier to arrange the entrance state and the exit state at a common position for each pass. As a representative position. However, it can be easily guided between any states. It should be noted that the dashes added to the reference numerals are the data on the exit side of the edger rolling mill. (A) Universal mill exit side state of previous pass: twin, tf1i
n, tf2in, ft3in, tf4in, bf1in, bf2in, bf3in
, Bf4in (b) U-mill exit side state of the corresponding path: twout, tf1out, tf
2out, tf3out, tf4out, bf1out, bf2out, bf3out, bf4o
ut (c) Flange width after edger rolling: bf1 ', bf2', bf
3 ', bf4' (Subscripts 1 to 4 are based on common agreement) (d) Flange thickness after edger rolling tf1 ', tf2', tf3 ', tf4' (Subscripts 1 to 4 are common agreement (E) Operation amount: Edger roll opening Se1, Se2 Horizontal roll opening of universal mill Sh Horizontal roll pass line PL Vertical roll opening Sv1, Sv2 Horizontal roll shift SF Biting adjustment amount δ1, δ2

FIG. 2 is a diagram for explaining the dimensions of each part of the H-section steel. tw is the web thickness, tf1 to tf4 is the flange thickness,
bf1 to bf4 indicate the flange width from the center, respectively. In the above, the subscript in indicates the exit side of the previous path, that is, the entry side of the relevant path, and out indicates the exit side of the relevant path.

Note that the left and right vertical roll opening and the upper and lower horizontal roll shift have only three degrees of freedom in controlling the dimensions, even though there are four variables. The assumption is made that the upper and lower rolls move the same amount in opposite directions. In the case of performing reciprocating edger rolling, two-pass edger rolling is performed, but it is sufficient to calculate that one-pass edger rolling is performed with a minimum edger roll opening of two passes.

(2) Model of Universal Mill (2.1) Gauge Meter Type FIG. 3 is an explanatory view of the gauge meter type. In this model, COSθ = 1 can be set, so that it is expressed by the following equation. twout = Ph / Mh + offset 1 ... (1) tf1out = (Pv1 + Pv2) / Mv + (Pv1-Pv3) / Mhs + Sv1 + SF + (Sh / 2 + PL) * tan (.theta.) + (Pv1-Pv2) / MVD + offset 2 ... ( 2) tf2out = (Pv2 + Pv1) / Mv + (Pv2−Pv4) / Mhs + Sv1−SF + (Sh / 2−PL) * tan (θ) + (Pv2−Pv1) / MVD + offset 3 ... (3) tf3out = (Pv3 + Pv4) / Mv + (Pv3-Pv1) / Mhs + Sv2-SF + (Sh / 2 + PL) * tan (θ) + (Pv3-Pv4) / MVD + Offset 4 ... (4) tf4out = (Pv4 + Pv3) / Mv + (Pv4-Pv2) / Mhs + Sv2 + SF + (Sh / 2 + PL) * tan (θ) + (Pv4−Pv3) / MVD + offset 5… (5) where θ: Roll taper angle Mh: Mill constant for horizontal rolling Mv: Mill constant for vertical rolling Mhs: Horizontal Mill constant in roll thrust direction MVD: Mill constant for vertical load difference in vertical rolling

(2.2) Load Formula In the model shown in FIG. 3, the load is represented by the following formula. Ph = fh (twin, tf1 ', tf2', tf3 ', tf4', bf1 ', bf2', bf3 ', bf4', twout, tf1out, tf2out, tf3out, tf4out, LH) (6) LH is FIG. Are the dimensions shown in FIG. Pv1 = fv (twin, tf1 ', tf2', tf3 ', tf4in, bf1', bf2 ', bf3', bf4 ', twout, tf1out, tf2out, tf3out, tf4out) (7) For Pv2, Pv3, Pv4 , (7), the subscript of the variable of the argument of the fv function may be changed. For example, when calculating Pv2, tf1 → tf2, tf2 → tf1, tf3 → tf4, tf4 → tf
3 and calculated by the following formula. Pv2 = fv (twin, tf2 ', tf1', tf4 ', tf3', bf2in, bf1in, bf4in, bf3in, twout, tf2out, tf1out, tf4out, tf3out) (8)

(2.3) Width Model The output side width of the universal mill is expressed by a widening model function fb.
And the term _C.delta . bf1out = bf1in + fb (twin, tf1in, tf2in, tf3in, bf4in, bf1in, bf2in, bf3in, bf4in, bf1 ', bf2', bf3 ', bf4', twout, tf1out, tf2out, tf3out, tf4out) + C δ1 ... (11 For bf2out, bf3out, and bf4out, only the index of the argument of fb needs to be changed in the same manner as when Pv is obtained. The term C _δ due to the bite adjustment is obtained experimentally, but is sufficient as the following equation. _{C δ1 = (K δ / 2} ) (bf1in -bf2in) + K δ δ 1 ... (15) C δ2 = - (K δ / 2) (bf1in -bf2in) -K δ δ 1 ... (16) C δ3 = ( _{K δ / 2) (bf3in -bf4in} ) + K δ δ 2 ... (17) C δ4 = - (K δ / 2) (bf3in -bf4in) -K δ δ 2 ... (18) where K _[delta] is experimentally It is a coefficient to be obtained and takes the following value. K _δ = 0.1-0.3 (19)

(3) Edger Rolling Down Model, Edger Gauge Meter Type FIG. 4 is an explanatory diagram of the operation of the edger rolling mill. Since the rolling load does not significantly affect the overall accuracy of the edger, it is sufficient to assume that the upper and lower rolling amounts are equal, and the following formula is obtained. Be1 = bf1 '+ bf2' ... (20) Be2 = Bf3 '+ bf4' ... (21) Pe1 = fe ((tf1in + tf2in) / 2, bf1in + bf2in, Be1) ... (22) Pe2 = fe ((tf3in + tf4in) / 2 , Bf3in + bf3in, Be2) ... (23) The relationship of the gauge meter thickness is shown in Fig. 5 of the screw-down device (screw).
Since the values are different in the left-right direction, the following equation can be derived from the geometric relationship. Be1 = (Se1 * c1 + Se2 * c2) + (Pe1 * d1 + Pe2 * d2) / Me (24) Be2 = (Se1 * c2 + Se2 * c1) + (Pe1 * d2 + Pe2 * d1) / Me (25) c1 = (1 + H) / Span) / 2 (26) c2 = (1-H / Span) / 2 (27) d1 = (1+ (H / Span) ² ) / 2 (28) d2 = (1- (H / Span) ) ² ) / 2… (29) Me: Edger Mill constant Note that Equations (26) and (27) represent the screw position (Sp
An) indicates the amount of leveling actually exerted on the H-section steel. Equations (28) and (29) show the displacement.

Based on the assumption that the upper and lower reduction amounts are the same, bf1 '= bf1in- (bf1in + bf2in-Be1) / 2 (30) bf2' = bf2in- (bf1in + bf2in-Be1) / 2 (31) bf3 ' = Bf3in- (bf3in + bf4in-Be2) / 2 (32) bf4 '= bf4in- (bf3in + bf4in-Be2) / 2 (33) The change in flange thickness after edger rolling is described in the literature ("H-shape by influence coefficient method"). Rolling Dimension Control Method for Steel ", Iron and Steel Vol.79
(1993)), the model is further simplified, and the average flange thickness is given by the following equation. tf1 '= tf1 * bf1in / bf1' * k (34) tf2 '= tf2 * bf2in / bf2' * k (35) tf3 '= tf3 * bf3in / bf3' * k (36) tf4 '= tf4 * bf4in / bf4 '* k (37) k = 0.8 to 0.9 (38)

(4) Linearization According to the above equations (2) and (3), the solution of the exit state is uniquely determined with respect to the entry state and the operation amount. If an attempt is made to directly solve this equation, it is nonlinear and it is not possible to obtain an explicit solution from the incoming state to the outgoing state. However, what is obtained here is a linearized solution for a minute change, and a linear relational expression between the input state and the output state can be derived by linearizing each expression as follows.

(4.1) Explanation of Variables In order to deal with minute changes, variables are defined as follows. (A) Small changes in the exit state of the universal mill in the previous pass: Δtwin, Δtf1in, Δtf2in, Δtf3in, Δtf4in
, Bbf1in, bbf2in, bbf3in, bbf4in (b) Small changes in the universal mill exit side state of the corresponding path: twtwout, ttf1out, ttf2out, ttf3out, ttf4o
ut, Δbf1out, Δbf2out, Δbf3out, Δbf4out (c) Small change in flange width after edger rolling: ΔBe1
, ΔBe2 (d) Small change in flange width after edger rolling: Δbf
1 ', Δbf2', Δbf3 ', Δbf4' (e) Small change in flange thickness after edger rolling: Δtf
1 ', ttf2', ttf3 ', ttf4' (f) Operation amount: Slight change of edge roll opening SeSe1, SeSe2 Slight change of universal mill horizontal roll opening ShSh Slight change of vertical roll opening SSv1, SSv2 Horizontal roll shift Small change ΔSF Small change in bite adjustment amount Δδ1, Δδ2

Here, when the state is represented by a vector, the state is represented by the following equation.

[0040]

(Equation 10)

(4.2) Relational Expression of Input / Output Variables of One Path By linearizing the above expressions (1) to (18), the following expression can be derived.

[0042]

[Equation 11]

FIG. 6 is an explanatory diagram of a model in which edger rolling is not performed. In this case, there is no intermediate state of the edger, and only the treatment of the expression for widening is different. The above expression (46) becomes the following expression, and can be easily transformed into the above expression (50). .

[0044]

(Equation 12)

(5) Coefficient of influence of finishing mill The finishing mill has not been modeled because it is under light pressure, and the above model has a roll taper angle of 0, so the degree of freedom is reduced by one. Although the matrix cannot be solved as dimensional control, empirically, the gauge meter and load model are the same model as the coarse mill, but in analysis, it is practical to use the roll with a taper angle similar to that of the coarse mill. The level can be used for control.

However, the above solution still has the following problems. a. Since A and B described above are 9 × 9 matrices, there are 9 × 9 × 2 elements in total, and it is difficult to see how each element affects the control result. Cannot adjust the element. b. It is assumed that all 9 components can be measured and that all 9 operating devices are available. If any failure occurs in the equipment during operation, the effects of mutual interference cannot be predicted, and basically Not applicable and lacks flexibility.

Therefore, in the present embodiment, the above-mentioned problems are solved by separating the nine components into independent small groups, thereby facilitating adjustment and increasing flexibility against obstacles. Next, the method will be described.

(§1) The target is limited to a general H-shaped steel,
Separation into groups that do not interfere with each other by using symmetry of dimensions: In the above discussion, the general theory was developed without mentioning the target values of the four flange thicknesses and widths.
If the range is limited to the shape steel, the target values of the four flange thicknesses and widths are equal, the cross section is top and bottom, left and right, and point symmetric at the center. Therefore, the influence coefficient A of the input side state on the output side state during one pass is tw and tfi (i = 1, 2, 3, 4). (All are the same.) The degree of influence between the input and output sides of tw and bfi (i = 1, 2, 3, 4) is not related to i. The relationship between tfi and tfj is as follows due to point symmetry:
There is only one set of independent values, and the others can be obtained simply by changing the arrangement. The relationship (same position) with i = 1 and j = 1 is (i = 2,
j = 2) (i = 3, j = 3) (i = 4, j = 4). The relationship (up and down position) with i = 1 and j = 2 is (i = 2,
j = 1) (i = 3, j = 4) (i = 4, j = 3). The relationship (horizontal position) with i = 1 and j = 3 is (i = 2,
j = 4) (i = 3, j = 1) (i = 4, j = 2). The relationship (opposing position) with i = 1 j = 4 is (i = 2,
j = 3) (i = 3, j = 2) (i = 4, j = 1). The relationship between tfi and bfj and the relationship between bfi and bfj are the same as described above. That is, as shown in FIG. 7, the same relationship holds even when rotated by 90 degrees, so that 9 × 9 = 8
As shown in the following table, there are only 21 independent elements out of one element, and the redundancy is large and there is room for arrangement.

[0049]

(Equation 13)

Therefore, if the matrix H for converting the state variable X into X ^* is appropriately selected, the matrix A is a block matrix of 3 × 3, 2 × 2, 2 × 2, 2 × 2 only in the diagonal part, and the other matrix matrix is used. 0 can be converted to a matrix A ^* .

[0051]

[Equation 14]

[0052]

(Equation 15)

[0053]

(Equation 16)

The matrix B also has the following shape due to symmetry (b1 to b18 are constants determined by the rolling model).

[0055]

[Equation 17]

Similarly, if the matrix G for converting the matrix u into the matrix u ^* is appropriately selected, the matrix B becomes the matrix B ^* and the matrix A ^*
The matrix may be a non-zero block matrix only at the diagonal portion in the same position and configuration as in FIG.

[0057]

(Equation 18)

[0058]

[Equation 19]

[0059]

(Equation 20)

Note that the horizontal roll shift is treated here assuming that the upper and lower rolls move by the same amount in the opposite directions. However, even if only one of the rolls moves, it can be dealt with only by converting the operation variables as described below. .

A description will be given of a variable conversion method for using this analysis when the roll shift device is provided on only one of the upper and lower rolls. For equipment that moves only one side, a dash will be added to the variable name, and SF '
Is defined as the roll shift amount from the neutral position. If the vertical roll gaps are Svd 'and Svf', SF, Svd and Svf in this analysis may be converted using the following equation. SF = SF '/ 2 Svd = Svd'-SF' / 2 Svf = Svf + SF '/ 2

By dividing the state and the operation amount into four groups by the above processing, control can be performed without mutual interference. In each group, the operation principle of the operation amount with respect to the state is also shown in the following table.
The physical meaning is clear. That is, the state variables in Table 1 correspond to the manipulated variables in Table 2, respectively. For example, the web thickness in the first term (No. 1) in Table 1 and the horizontal value in the first term (No. 1) in Table 2 The gap of the roll corresponds.

[0063]

[Table 1]

[0064]

[Table 2]

Therefore, the present embodiment has the following advantages. -By simply changing variables, it is possible to derive an expression whose causal relationship is easy to understand without approximation. -In the case of evaluation by experiment, if you decide which part of the matrix to evaluate, the experimental method is clearly determined and adjustment is easy. -The use or non-use of control can be switched in units of four groups. However, for the third term (No. 3) in Tables 1 and 2, since the combination of variables has little merit in operating the actual equipment, the control method including the following approximation is better.

(§2) Based on the calculation result of the rolling model,
Approximate separation of independent components When the above-mentioned matrix A ^* B ^* in the rolling model is calculated, for the other groups except the first group (since all six components have a bias relationship, hereinafter referred to as bias components) Is that the diagonal components of the matrix A ^* B ^* are sufficiently large compared to the other components (for example, in an A ^* B ^* matrix through five-pass rolling, the elements except for the diagonal part of the bias relationship are on the same line) 10% or less of the diagonal part of Therefore, it can be said that approximately all the bias components can be handled independently. This ensures that there is little mutual interference, so that the control can be flexibly switched according to the state of the equipment as follows.

When some of the measured values cannot be collected or when it is determined that the reliability is insufficient, the corresponding state quantity input data is set to 0. For example, when one flange thickness of the measuring instrument could not be measured during operation, use the control of No.1 to No.4, No.6, No.8 in Table 1 and Table No.5, No. .7 and No.9 are switched to a combination of non-use, and the information is utilized to the maximum extent possible. If the output side is multiplied by a gain for adjustment and the operation equipment does not move partly or does not want to move for convenience, the gain is set to 0, but a small value of 1 or less. In this case, it is necessary to multiply the gain for each independent component, and one gain is obtained for three average components and six gains are obtained for bias components. In order to make the causal relationship easier to understand, the center deviation-biting adjustment relationship is D (4, 4) and (6, 6) of the matrices A ^* and B ^* , which are almost equal.
The S side (drive side) and the FS side (free side) may be handled independently, and a state variable and an operation variable may be selected as shown in Table 3 below.

[0068]

[Table 3]

FIG. 9 is a block diagram showing the configuration of the control system of the present embodiment. In the drawing, reference numeral 30 denotes a hot dimension meter, and reference numeral 40 denotes an arithmetic and control unit. The hot dimension meter 30 is arranged at a position close to the upstream side of the universal rolling mill, and has a web thickness (tw), a flange thickness of four places (tf1 to tf4) and a flange leg length of four places (tf1 to tf4) after forming and rolling. bf1 ~
bf4) is measured. The arithmetic control device 40 performs the above-described arithmetic control. When introducing new equipment,
Initially, the reliability and accuracy of measuring instruments and operating equipment are often unstable, or the accuracy of model formulas is often insufficient. Therefore, the above (§1), which is a theoretically exact solution, is the core, and its surroundings (input, output, guidance) are the state variables and manipulated variable system of (§2). At the beginning of the equipment launch,
Online adjustment is performed with the adjustment gain of (§2), and the adjustment gain is set to “1” by gradually increasing the accuracy of the model equation, and the system is operated substantially by the solution of (§1). Further, as described in (§2), the input is set to 0, or the output is multiplied by a gain, and the control is flexibly responded by switching the control depending on the equipment status.

[0070]

As described above, according to the present invention, the control system is composed of nine state variables and nine operation variables so that the correspondence between the state quantity and the operation quantity (actuator) can be obtained. The following effects can be obtained. (1) Since the control structure is simple and the causal relationship between the operation amount and the state amount is easy to understand, adjustment (tuning) of the control system is easy. (2) Partial degenerate operation is possible (for example, tw, tf,
Only control of three dimensions of Bf, etc.).

[Brief description of the drawings]

FIG. 1 is an explanatory diagram (model) showing a relationship between equipment of an H-section steel size control device and an influence coefficient parameter according to an embodiment of the present invention.

FIG. 2 is an explanatory diagram of dimensions of each part of an H-section steel.

FIG. 3 is an explanatory diagram of a gauge meter type.

FIG. 4 is an operation explanatory view (part 1) of the edger rolling mill.

FIG. 5 is an operation explanatory view (part 2) of the edger rolling mill.

FIG. 6 is an explanatory diagram of a model in a case where edger rolling is not applied in the embodiment of FIG. 1;

FIG. 7 is a diagram for explaining that symmetry can be used in an H-section steel.

FIG. 8 is an explanatory diagram of a roll shift amount.

FIG. 9 is a block diagram illustrating a configuration of a control system according to the present embodiment.

FIG. 10 is an explanatory diagram (part 1) of an influence coefficient method used in the present embodiment.

FIG. 11 is an explanatory diagram (part 2) of the influence coefficient method used in the present embodiment.

FIG. 12 is an explanatory diagram (part 3) of the influence coefficient method used in the present embodiment.

FIG. 13 is an explanatory diagram of a method for estimating a disturbance offset amount in the present embodiment.

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Tatsuro Udagawa 1-2-1, Marunouchi, Chiyoda-ku, Tokyo, Japan Inside Nihon Kokan Co., Ltd. (72) Inventor Yukio Takashima 1-2-1, Marunouchi, Chiyoda-ku, Tokyo, Japan (72) Inventor Makoto Watanabe 1-1-2 Marunouchi, Chiyoda-ku, Tokyo Japan 1-2.Inventor Naoki Kataoka 1-1-2 Marunouchi, Chiyoda-ku, Tokyo Japan 1-2 Inside the corporation

Claims (6)

[Claims] 1. A universal rolling mill in which a bite adjusting device capable of adjusting the bite posture or position of a rolled material is arranged on a rolling-in side, and a horizontal roll is supported so as to be movable in an axial direction.
In a rolling method using a group of universal rolling mills for H-section steel rolling, each of which includes an edger rolling mill in which a flange width reduction roll gap on a drive side and a free side can be individually set, a horizontal roll gap and a vertical roll of the universal rolling mill Average gap, average gap of the edger rolling mill, position of the drive side and free side of the biting adjustment device, axial position of the horizontal roll, leveling of the edger rolling mill, free side and drive of the vertical roll gap The web thickness of the H-section steel corresponding to the operation amount with the difference from the side and the nine operation lines as the operation amount,
Average thickness of four points of flange, average value of flange width on free side and drive side, center deviation of drive side of flange, center deviation of free side of flange, axis deviation, difference between free side and drive side of flange width, flange A method for rolling an H-section steel, comprising controlling nine state quantities of a difference between a free-side average value and a drive-side average value of a thickness and a difference between an upper average and a lower average of a flange thickness. 2. A universal rolling mill in which a bite adjusting device capable of adjusting a bite posture or position of a rolled material is arranged on a rolling-in side, and a horizontal roll is supported so as to be movable in an axial direction.
In a rolling method using a group of universal rolling mills for H-section steel rolling, each of which includes an edger rolling mill in which a flange width reduction roll gap on a drive side and a free side can be individually set, a horizontal roll gap and a vertical roll of the universal rolling mill Average gap, average gap of the edger rolling mill, position of the drive side and free side of the biting adjustment device, axial position of the horizontal roll, leveling of the edger rolling mill, free side and drive of the vertical roll gap A specific number of manipulated variables were extracted from the manipulated variables of the nine differences of the path line and the pass line, and the web thickness of the H-section steel corresponding to the nine manipulated variables and the four points of the flange, respectively. Average thickness, average value of flange width on free side and drive side, center deviation of drive side of flange, center deviation of free side of flange, Axis deviation, the difference between the free side and the drive side of the flange width, the difference between the free side average value of the flange thickness and the drive side average value, and the difference between the upper side average and the lower side average of the flange thickness are nine state quantities. A method of rolling an H-section steel, wherein a state quantity corresponding to the extracted manipulated variable is controlled. 3. The web thickness, four flange thicknesses, and four flange leg lengths of the H-section steel after forming and rolling are measured at a position close to the upstream side of the universal rolling mill, and the dimensions are determined based on the measurement results. 3. The method for rolling an H-section steel according to claim 1, wherein the control is performed. 4. A universal rolling mill in which a bite adjusting device capable of adjusting the bite posture or position of a rolled material is disposed on a rolling entry side, and a horizontal roll is supported so as to be movable in an axial direction.
A group of universal rolling mills for rolling H-beams each including an edger rolling mill in which a flange width reduction roll gap on the drive side and the free side can be individually set, and a horizontal roll gap and a vertical roll average gap of the universal rolling mill, The average gap of the edger rolling mill, the position of the drive side and the free side of the biting adjustment device, the position of the horizontal roll in the axial direction, the leveling of the edger rolling mill, the free side and the drive side of the vertical roll gap. The web thickness of the H-section steel corresponding to the operation amount with the difference and nine of the pass lines as the operation amounts,
Average thickness of four points of flange, average value of flange width on free side and drive side, center deviation of drive side of flange, center deviation of free side of flange, axis deviation, difference between free side and drive side of flange width, flange An H-shaped steel having an arithmetic and control unit for controlling nine state quantities of a difference between a free side average value and a drive side average value of the thickness and a difference between an upper average and a lower average of the flange thickness. Rolling equipment. 5. A universal rolling mill in which a bite adjusting device capable of adjusting a bite posture or position of a rolled material is arranged on a rolling-in side, and a horizontal roll is supported so as to be movable in an axial direction.
A group of universal rolling mills for rolling H-beams each including an edger rolling mill in which a flange width reduction roll gap on the drive side and the free side can be individually set, and a horizontal roll gap and a vertical roll average gap of the universal rolling mill, The average gap of the edger rolling mill, the position of the drive side and the free side of the biting adjustment device, the position of the horizontal roll in the axial direction, the leveling of the edger rolling mill, the free side and the drive side of the vertical roll gap. By extracting a specific number of manipulated variables from the manipulated variables of the difference and nine of the pass lines, the web thickness of the H-section steel corresponding to each of the nine manipulated variables, the average thickness of four points of the flange, Average value of flange width on free side and drive side, center deviation of drive side of flange, center deviation of free side of flange, axis deviation, flange width Of the nine state quantities of the difference between the free side and the drive side, the difference between the free side average value of the flange thickness and the drive side average value, and the difference between the upper side average and the lower side average of the flange thickness, the above-mentioned extracted state quantities are extracted. And an arithmetic and control unit for controlling a state quantity corresponding to the manipulated variable. 6. A hot dimension meter for measuring the web thickness, four flange thicknesses, and four flange leg lengths of the H-beam after shaping and rolling at an adjacent position on the upstream side of the universal rolling mill. 6. The rolling device for H-section steel according to claim 4, wherein the arithmetic and control unit controls the dimension of the H-section steel based on the measurement result.