JPH0380566B2 - - Google Patents

Description

[Detailed description of the invention] [Industrial application field]

The present invention relates to a rolling control method for a mandrel mill, and in particular to a method for controlling the rolling of a pipe by controlling a plurality of stands having slotted rolls and a mandrel bar, which is suitable for use in a hot rolling line for seamless steel pipes. This invention relates to an improvement in a rolling control method for a mandrel mill that gradually reduces thickness.

[Conventional technology]

Mandrel mills, which are widely used as continuous rolling mills for hot rolling lines for seamless steel pipes, are medium-thick-walled blank pipes that are formed by perforating round steel billets (billets) with viasers after heating them in a rotary heating furnace. (Rolled material)
As shown in FIG. 8, a mandrel bar 12 is inserted into the inside of the material 10, and the material is rolled between 14 pairs of rolling rolls (caliber rolls) to achieve the target uniform thickness and outer diameter over the entire length. The purpose is to form a finished tube with a diameter of
Furthermore, it is sent to a stretch reducer or the like and formed into a finished product. In such a mandrel mill, it is necessary to control the average wall thickness of the tube after rolling to a target value in order to make not only the average wall thickness uniform but also the wall thickness in the circumferential direction. This is because the roll hole shape of the final stands for finishing odd and even thicknesses is usually set so that when the average thickness matches the target value, the thickness in the circumferential direction becomes the most uniform. be. In addition, it is also necessary to control the wall thickness of the tubes at the bottom of the roll grooves of the final stands for odd-numbered and even-numbered finishing stands to be equal in order to make the wall thickness uniform in the circumferential direction. By the way, the main factors that influence the average wall thickness of the tube after rolling are considered to be wear of the mandrel bar and rolls, and variations in the dimensions of the raw tube and the rolling load. Therefore, in the past, for the wear of the mandrel bar, the diameter of the worn mandrel bar was calculated based on the average wall thickness of the pipe after rolling and the roll gap during the biting of the pipe. , a method of controlling the rolling position has been proposed. According to this control method, a steel pipe with the same wall thickness can be obtained every time, despite repeated use of the plug or mandrel bar. In addition, for example, Japanese Patent Laid-Open No.
No. 54-121266 discloses a method of changing the roll circumferential speed ratio, in other words, the tension (including compression force) based on the length of the raw pipe. In addition, in order to eliminate variations in the rolled length or average wall thickness of the tube, in Japanese Patent Publication No. 59-16847, the circumferential speed of the roll was determined based on the coefficient of friction between the outer surface of the tube and the roll and the coefficient of friction between the inner surface of the tube and the mandrel bar. A method of changing the ratio has been proposed.

[Problems to be solved by the invention]

However, in the method of determining the diameter of a worn mandrel bar and controlling the rolling position proposed in the above-mentioned Japanese Patent Application Laid-open No. 54-102270, it is important to accurately determine the roll gap during tube engagement. Since the zero adjustment of the rolling position of the mandrel mill was carried out without applying any rolling force, as shown in Fig. 9, variations in mill rigidity due to backlash in the mechanical system (housing) and nonlinear parts of mill rigidity (low There is a problem that sufficient accuracy cannot be obtained due to the influence of the load section). In addition, in the control method of changing the roll circumferential speed ratio based on the length of the raw tube, which was proposed in the above-mentioned Japanese Patent Application Laid-Open No. 54-121266, when rolling tension is applied, a predetermined pressure is applied to the tip and rear ends of the tube. Since no tension is applied and the tension varies each time the tube is inserted into or pulled out of each stand, there is a problem in that the longitudinal wall thickness and outer diameter of the tube after rolling vary. Furthermore, in the control method of changing the roll circumferential speed ratio based on the friction coefficient between the outer surface of the tube and the roll and the friction coefficient between the inner surface of the tube and the mandrel bar, proposed in the above-mentioned Japanese Patent Publication No. 59-16847, -121266, it has the problem that the longitudinal wall thickness and outer diameter of the rolled tube fluctuate. Furthermore, the rolling load varies depending on the friction characteristics of the inner and outer surfaces of the tube, as well as the deformation resistance and thinning rate of the tube, but so far, no control method has been proposed that properly takes these into consideration. , an appropriate control method has not been proposed to equalize the wall thickness of the tube at the bottom of the roll groove of each even-numbered finishing stand, and therefore, it is difficult to sufficiently control the wall thickness and outer diameter of the tube after rolling. The problem was that it could not be suppressed.

[Purpose of the invention]

The present invention was made in order to solve the above-mentioned conventional problems, and the rolling position and roll rotation are adjusted so that the average wall thickness of the pipe after rolling reaches the target value and the wall thickness in the circumferential direction is uniform. An object of the present invention is to provide a rolling control method for a mandrel mill that can control the number of rolling mills.

[Means to solve the problem]

The present invention relates to a rolling control method for a mandrel mill that gradually reduces the outer diameter and wall thickness of a tube using a plurality of stands having grooved rolls and a mandrel bar. In the rolling control method of a mandrel mill that gradually reduces the outer diameter and wall thickness of a tube using multiple stands and mandrel bars, each actual measured value of hollow material weight, hollow length, and hollow average outer diameter Wb, Lh , Dh, the scale loss rate μr in the heating furnace, and the hollow density ρh, the formula th=(Dh/2)×[1−√1−{(1−)}{
(2) ² }], find the actual wall thickness th of the hollow, the material weight of the hollow, the length of the shell, the actual outer diameter Ds obtained from the actual measured outer diameter of the shell, and the scale loss rate μr of the heating furnace, Based on the shell density ρs, the formula ts=(Ds/2)×[1−√1−{(1−)}{
(2) ² }], find the actual wall thickness ts of the shell, and calculate the roll gap Go(i), rolling load P(i), and mill spring constant K of each i stand before hollow biting.
Based on the formula G(i)=Go(i)+P(i)/K, calculate the actual roll gap G(i) of each stand when the hollow is engaged, and calculate the actual roll gap G(i) of each stand when the hollow is engaged. Roll gap G(i) and mandrel bar diameter Db
Based on the formula t(i)=(G(i)-Db)/2, calculate the actual thickness t(i) of each i-stand, and calculate the thickness of each odd and even number among the calculated actual thicknesses. Actual wall thickness t (odd) of thick finishing final stand
and t(even) and the actual thickness ts of the shell to find the gauge meter deviation ΔG corresponding to the wear of the roll and mandrel bar, and calculate the actual thickness t(i) and the gauge meter deviation ΔG.
Based on the formula tc(i)=t(i)+ΔG/2, calculate the corrected wall thickness tc(i) of each i-stand, and calculate the actual wall thickness th of the hollow and the corrected wall thickness tc
Based on (i), the formula Ro(i)=(tc(i-2)-tc(i))/(th-tc(odd)) Re(i)=(tc(i-2)-tc(i) )) /(th-tc(even)) Find the actual thinning distribution ratios Ro(i) and Re(i) for each odd and even i stand, and correct for the final stand with wall thickness finishing for each odd and even number. Based on the wall thicknesses tc (odd) and tc (even) and the actual wall thickness th of the hollow, so that the corrected wall thickness of the final stand for finishing each odd and even number of hollows to be rolled next time is equal. Formula Bo={th-(tc(odd)+tc (even))/2}/(th-tc(odd)) Be={th-(tc(odd)+tc (even))/2}/(th- tc (even)), find the odd-even thinning distribution ratios Bo and Be to be set for each odd and even i stand next time, and calculate the actual wall thickness th of the hollow to be rolled next time and the wall thickness tss set for the shell. Based on the actual thickness reduction distribution ratios Ro(i) and Re(i) of the odd and even stands, and the set odd-even thickness reduction distribution ratios Bo and Be, the formula Δta(i)=(th-tss)×Ro From (i)×Bo Δta(i)=(th-tss)×Re(i)×Be, set the set thickness reduction amount Δta(i) for each i-stand so that the wall thickness of the shell becomes the set thickness. Based on the set thickness reduction amount Δta(i), the set wall thickness ta(i) of each i-stand is determined from the formula ta(i)=ta(i-2)−Δta(i), and the hollow and the temperature θh and θb of the mandrel bar, the set thinning rate {1-(ta(i)/ta(i-2))
}
Taking into consideration, the predicted rolling load P(i) is calculated, and the calculated predicted rolling load P(i), the gauge meter deviation ΔG, the set wall thickness ta(i) of each i-stand,
Mandrel bar diameter Db, roll hole depth DK (i)
Based on the formula Ss(i)=2ta(i)+Db−ΔG −(P^(i)/K)−2DK(i), the set reduction position of each i-stand for each hollow
Find Ss(i), and based on the set wall thickness ta(i) of each i-stand,
Assuming the reference wall thickness as t0(i) and the coefficient as Ka(i), calculate the cross-sectional area correction ratio Ca(i) from the formula Ca(i)=Ka(i)×(ta(i)/t0(i)). Based on the obtained cross-sectional area correction ratio Ca(i), the set rotation speed at the standard wall thickness is set as N0(i), and from the formula Ns(i)=N0(i)/Ca(i), each i Find the set rotational speed Ns (i) of the stand, and set the determined lower position Ss of each i stand.
(i) and the set rotational speed Ns(i) are output to the control device of the mandrel mill to control the rolling position and rotational speed of each i-stand, thereby achieving the above object. Further, in the embodiment of the present invention, when calculating the actual wall thickness t(i) of each i-stand, rolling is performed while the upper and lower rolls of each i-stand are in contact and a preload load Ppr is applied. After performing zero adjustment of the position, find the roll gap at the time of hollow biting from the formula G(i)=Go(i)+(P(i)-Ppr)/K, and calculate the obtained roll gap G(i)
The actual wall thickness t(i) is determined based on the following.

[Effect]

First, in order to accurately determine the roll gap during tube jamming, it is necessary to remove the effects of mechanical system play and nonlinear parts of mill rigidity (low load parts). Therefore, as shown in Figures 2 to 4, when assembling the rolls, the upper and lower rolls are kept in contact and a predetermined rolling force (hereinafter referred to as preload load) is applied, and the rolling position is zero-adjusted (hereinafter referred to as preload load). , preload zero adjustment). That is, the balance adjustment is performed in the state shown in FIG. 2, the preload zero adjustment is performed in the state shown in FIG. 3, and the reference gap is set in the state shown in FIG. 4. Therefore, the roll gap G when the tube is caught is:
It can be determined with high accuracy using the following formula. G=Go+(P- _Ppr )/K...(1) Here, Go is the roll gap before the tube is caught,
P is the rolling load, P _pr is the preload load, and K is the mill spring constant. Next, a method of controlling the rolling position and roll rotation speed of each roll stand for each raw pipe will be explained with reference to FIG. 6. In the description of the present invention, the raw tube will be referred to as a hollow tube, and the rolled tube will be referred to as a shell. First, based on the actual measurements of the hollow material weight, hollow length, hollow average outer diameter Wb, Lh, and Dh, the scale loss rate μr in the heating furnace, and the hollow density ρh,
From equation (3) below, calculate the actual wall thickness th of the hollow, and calculate the actual outer diameter D _s obtained from the material weight of the hollow, the length of the shell, the actual outer diameter of the shell, and the scale loss rate of the heating furnace. μr, based on the shell density ρs, described later (5)
Calculate the actual wall thickness ts of the shell from the formula. Also, calculate the actual roll gap G of each stand (at the time of tube jamming) based on the above formula (1), and use that value and the mandrel bar diameter (a value measured in advance or a set value) D _b From Equation (6), the actual wall thickness t(i)(i
is the stand number). In addition, by comparing the actual wall thicknesses t(odd) and t(even) of the final stands finishing odd and even wall thicknesses with the actual shell wall thickness _ts , the amount ΔG( (hereinafter referred to as gauge meter deviation) is determined, and this value is used the next time this mandrel bar is used. Also, gauge meter deviation ΔG and actual wall thickness t(i)
Corrected wall thickness from equation (10) given below for each stand based on
Find t _c(i) . Next, use this value and the actual hollow wall thickness t _h
Based on the equations (11) and (12) given below for odd and even numbers, the actual thinning distribution ratio of each stand is calculated.
Find R _p(i) and R _e(i) . Furthermore, based on the corrected wall thickness t _c(i) of the final stands for finishing the odd and even numbers, the corrected wall thicknesses for the final stands for finishing the odd and even numbers of the tube to be rolled next are made equal. , the set odd-even thinning distribution ratios B _p and B _e are determined from equations (13) and (14) below. Next, based on the actual shell wall thickness t _s and the actual wall thickness t _h of the hollow to be rolled next, the shell wall thickness of the next rolled pipe is adjusted to the target value (19), (20 )
Find the set amount of thinning Δt _a for each stand total from the formula. Based on this set thickness reduction amount Δt _a(i) , see (21) below.
The set wall thickness t _a(i) of each (i) stand is calculated from the formula. Actual thinning distribution ratio for each stand determined from the above
Based on R _p(i) , R _e(i) , the set odd-even thickness thinning distribution ratio B _p , B _e , and the set thinning amount Δt _a , set wall thickness t _a(i) for each stand.
and determine the set thinning rate 1-(t _a(i) /t _a(i-2) ). Next, the actual measured temperatures θ _b and θ _h of the mandrel bar and hollow, the set thinning rate 1-(t _a(i) /t _a(i-2) ), the actual thinning rate and the actual rolling load P ₍ Based on _i) , calculate the predicted rolling load P _(i) for each stand from equation (22) below. Furthermore, based on the predicted rolling load P(i) and the gauge meter deviation ΔG of the mandrel bar to be used, the set rolling position S _s(i) is determined from equation (23) given below, and the rolling position is adjusted to this value. . On the other hand, the roll rotation speed is controlled by the cross-sectional area correction ratio C _{a (i) obtained based on the set wall thickness t a(i)} of each stand, based on the constant mass flow law from equation (25) below. This is done by finding the set rotation speed N _s(i) that will result in tensionless rolling and adjusting it to this value. Therefore, it is possible to control the rolling of the mandrel mill in consideration of the deformation resistance and wall thinning rate of the tube, and to make the wall thickness of the tube at the bottom of the roll groove of the final stands of odd and even thicknesses equal. be able to control it.

Embodiments Hereinafter, embodiments of a rolling control device for a mandrel mill to which the present invention is applied will be described in detail with reference to the drawings. As shown in FIG. 6, this embodiment includes a plurality of sets of rolls 14 arranged in series for sequentially rolling a tubular hollow 10 whose inner diameter is regulated by a mandrel bar 12 inserted therein. The present invention is applied to a mandrel mill, and the rolling load P^ _(i)
Various sensors including a load cell 20 for detecting the rotation speed of the roll 14 _i of the i-th stand, a pulse generator, etc. for detecting the rotation speed of the roll 14 i of the i-th stand, and based on the sensor output, the actual hollow wall thickness t _h and the actual shell thickness From the thickness t _s , determine the set thickness reduction amount Δt _a for the total of each stand so that the average wall thickness of the pipe after rolling becomes the target value, and then calculate the thickness finishing final stand for each of odd and even numbers (in this example, 5th and 6th stands)
Based on the actual wall thicknesses t ₅ and t ₆ of the exit pipe, the odd-even thickness reduction distribution ratios B p and B e are calculated so that the wall thickness of the pipe after rolling is uniform in the circumferential direction, and the set reduction ratios B _p and B _e are determined. Set the wall thickness of the outlet pipe of each stand based on the wall thickness Δt _a , the set odd-even wall thinning distribution ratio B _p , B _e , and the actual wall thinning distribution ratio R _p(i) , R _e(i) of each stand. The thickness t _a(i) is determined, and the temperatures θ _h and θ _b of the hollow and the mandrel bar, the set thinning rate 1-(t _a(i) /t _a(i-2) ) of each stand, and each Calculate the predicted rolling load P (i) for each stand based on the stand's actual thinning rate 1 - (t _{c (i)} / t _{c (i-2)} ) and the actual rolling load P _{( i)} ,
The rolling position and roll rotation speed are set for each hollow based on the set wall thickness t _a (i) of the pipe on the exit side of each stand and the predicted rolling load P^ (i) of each stand, and the roll rotation speed is set for each hollow. and an arithmetic device 24 that outputs the control amount of the reduction position to a control device (not shown). Hereinafter, the operation of the arithmetic unit 24 will be explained with reference to FIG. 5 mentioned above. First, roll gap G(i) during tube jamming.
Perform preload zero adjustment in order to obtain accurate values.
In this embodiment, this preload zero adjustment is performed in the roll shop as shown in FIGS. 2 to 4. That is, as shown in FIG. 2, a solid cylindrical plug gauge 16 is inserted into the roll bit and a downward pressure is applied to adjust the horizontal and vertical balance. Next, as shown in FIG.
40ton), perform zero adjustment of the reduction position. Thereafter, as shown in FIG. 4, it is set at a predetermined lowered position. In addition, the reference numeral 1 in FIGS. 3 to 5
7 is a coupling, 18 is a motor, and 19 is a cercin. Therefore, the relationship between the roll gap and the rolling force at zero adjustment in this embodiment is as shown in FIG. Next, the calculation device 24 first calculates the actual measured values of billet weight, hollow length, and hollow average outer diameter.
The actual hollow wall thickness (average value) _th is calculated from W _b , L _h , D _h , the scale loss rate μ _r in the rotary heating furnace, and the hollow density ρ _h using the relationship of the following formula. ρ _h L _h πt _h (D _h −t _h )=W _b (1−μ _r ) ……(2) t _h = (D _h /2)×[1−√1−{ _b (1− _r ) }
{ _{h h} ( _h 2) ² }]...(3) Next, calculate the actual shell outer diameter _Ds . Since the outer diameter of the shell is usually elliptical, it is calculated from the following approximation formula for the circumference of an ellipse based on the measured values D _sl and D _ss of the outer diameter of the shell corresponding to the long and short sides. D _s = (2/π) × {0.9827D _sl + 0.311D _ss + 0.2867 (D _ss ) ² /D _sl } ...(4) Next, calculate the actual shell thickness t _s . The actual shell wall thickness t _s is the same as the actual hollow wall thickness t _h described above (2).
It is calculated using the relationship of the following formula, which is a modified version of the formula. t _s = (D _s /2)×[1−√1−{ _b 1− _r )}/{
ρ _s L _s π(D _s 2) ² }]...(5) where ρ _s is the density of the shell. Next, calculate the actual wall thickness t _(i) of each stand. The actual wall thickness t _(i) of each stand is calculated from the gap G _(i) of the roll during tube engagement (at the bottom of the groove) and the diameter D _b (actual value or set value) of the mandrel lever, using the relationship of the following formula. calculate. t _(i) = (G _(i) − D _b )/2 ...(6) The roll gap G _(i) during the tube jamming is:
Calculated using the following relationship. G _(i) = S _(i) +2DK _(i) + (P _(i) - P _pr ) / K ... (7) Here, S _(i) is the rolled down position (the roll flange part before the pipe is caught) gap), DK is the depth of the roll hole, P _(i) is the rolling load, P _pr is the preload load, K
is the Mill spring constant. Next, calculate the gauge meter deviation ΔG. This gauge meter deviation ΔG is determined by the actual wall thicknesses t ₍ 5) and t ₍₆₎ of the odd and even final stands (fifth and sixth stands in the example) and the actual shell wall thickness t _s. Based on this, it is calculated from the relationship of the following formula. ΔG＝α _g ×ΔG ^l-1 + (1−α _g )×2 × {t _s − (t ₍₅₎ + t ₍₆₎ )/2} ...(8) Here, α _g is the smoothing coefficient The coefficient ΔG ^l-1 indicates the gage meter deviation before one cycle of the mandrel bar, that is, during the previous rolling. In addition, gauge meter deviation ΔG
The initial value of is calculated using the following formula. ΔG=a ₁ + b ₁・N _r +c ₁・N _b ...(9) Here, N _r is the number of roll rolling, N _b is the number of mandrel bar rolling, and a ₁ , b ₁ , and c ₁ are calculated in advance. This is the calculated coefficient. Next, the corrected wall thickness t _c(i) of each stand is calculated.
This corrected wall thickness t _{c (i)} is calculated from the following relationship based on the actual wall thickness t _(i) and the gauge meter deviation ΔG. t _c(i) = t _(i) + ΔG/2 ...(10) Next, the actual thickness reduction distribution ratio of each stand R _p(i) , R _e(i)
Calculate. Here, R _{p (i)} indicates the actual thickness reduction distribution ratio on the odd-numbered stand side, and R _{e (i)} indicates the actual thickness reduction distribution ratio on the even-numbered stand side. The actual thickness reduction distribution ratios R _{p (i)} and R _{e (i)} of each of these stands are calculated based on the actual hollow wall thickness t _h and the corrected wall thickness t _{c (i)} according to the relationship of the following equation. R _p(i) = (t _c(i-2) −t _c(i) ) / (t _h −t _c(5) ) ……(11) R _e(i) = (t _{c(i-2) )} −t _c(i) ) /(t _h −t _c(6) ) ……(12) In the above formula, t _c(-1) and t _c(0) are the actual hollow wall thickness t _h
Set it as equal to . Next, the set odd-even thinning distribution ratios B _p and B _e are calculated.
These set odd-even thickness reduction distribution ratios B _p and B _e are the corrected wall thicknesses t _{c (5)} of the fifth and sixth stands of the pipe to be rolled next time,
This is to ensure that t _c(6) is equal.
Note that B _p indicates the set thinning distribution ratio for the odd-numbered stands, and B _e indicates the set thinning distribution ratio for the even-numbered stands. The set odd-even thickness reduction distribution ratios B _p and B _e are calculated from the following relationship based on the corrected wall thickness t _{c (i)} of each stand and the actual hollow wall thickness _th . B _p = {t _h −(t _c(5) +t _c(6) ) /2}/(t _h −t _c(5) ) ...(13) B _e = {t _h −(t _{c(5) )} +t _c(6) ) /2}/(t _h −t _c(6) ) ...(14) Next, calculate the rolling load learning coefficient C _p(i) for each stand. The rolling load learning coefficient C _p(i) for each stand
is a coefficient used when calculating the predicted rolling load P _(i) of each stand, which will be described later. The rolling load learning coefficient C _p(i) is calculated from the relationship of the following equation based on the smoothing coefficient α _p , the mandrel bar temperature influence coefficient C _b , the hollow temperature influence coefficient C _h , and the like. C _p(i) = α _p・C _p(i) ^n-1 + (1−α _p ) × [P _(i) / [a _2(i) + b _2(i) × {1−(t _{c( i)} /t _c(i-2) )} ×C _b ×C _h ……(15) Here, a _2(i) and b _2(i) are the coefficients obtained by analyzing each steel type in advance, C _p(i) ^n-1 indicates the rolling load learning coefficient between the tube and the main body. Note that the initial value is C _p(i) =1. Next, calculate the shell thickness deviation Δt _f . This shell thickness deviation Δt _f can be calculated from the following relationship. Δt _f = α _t × Δt _f ^n-1 + (1−α _t ) × (t _ps − t _s ……(16) where α _t is the coefficient for smoothing, and t _ps is the target shell thickness. The initial value is Δt _f = 0. Next, the set shell thickness t _ss is calculated. The set shell thickness t _ss is calculated from the relationship of the following equation: t _ss = t _ss ^{n -1} + Δt _f ... (17) The initial value is t _ss = t _ps , that is, the target shell thickness. Using the actual values calculated as above, next, various settings for the hollow to be rolled are set. The amount is calculated as follows. First, the set thickness reduction amount Δt _a is obtained by subtracting the set shell wall thickness t _ss from the actual hollow wall thickness t _h . In other words, the total of each stand is determined by the following equation. Calculate the set thickness reduction amount Δt _a . Δt _a = t _h −t _ss ... (18) Next, calculate the set thickness reduction amount Δt _a(i) for each stand. Set thickness reduction amount for odd-numbered stands Δt _a(i) is calculated using the following relationship: Δt _a(i) = (t _h −t _ss ) × R _p(i) × B _p ……(19) On the other hand, the set thickness reduction amount for even-numbered stands Δt _a(i) is calculated using the following relationship: Δt _a(i) = (t _h −t _ss ) × R _e(i) × B _e ……(20) Here, the actual thickness reduction distribution ratio R The initial values of _p(i) and R _e(i) are predetermined values, and the initial values of the set odd-even thinning distribution ratio B _p and B _e are 1. Next, the set wall thickness t _a( Calculate _i) .
This set wall thickness t _a(i) is the set wall thickness reduction amount of each stand from the set wall thickness t _a(i-2) of the stand one step before the desired stand between each odd numbered stand and between each even numbered stand. It can be found by subtracting Δt _a(i) . That is, the set wall thickness t _a(i) of each stand is calculated from the relationship of the following equation. t _a(i) = t _a(i-2) −Δt _a(i) ...(21) Note that t _a(-1) = t _a(0) = t _h . Next, the predicted rolling load P^ _(i) for each stand is calculated using the relationship of the following equation. P^ _(i) = [a _2(i) +b _2(i) × {1−(t _a(i) /t _a(i-2) )}] ×C _p(i) ×C _b ×C _h ...(22) However, C _b =k _b・(θ _0b /θ _b ), C _h =k _h・(θ _0h /
θ _h ). Here, k _b and _kh are experimentally determined coefficients, θ _0b is the reference mandrel bar temperature, θ _b is the measured mandrel bar temperature, θ _0h is the reference hollow temperature, and θ _h is the measured hollow temperature. Although the thermal expansion of the mandrel bar diameter due to the mandrel bar temperature can be ignored, the state of lubricant application on the mandrel bar surface changes depending on the mandrel bar temperature, which changes the friction characteristics between the tube inner surface and the mandrel bar. Therefore, the rolling load fluctuates. Therefore, the above-mentioned mandrel bar temperature influence coefficient takes into account the influence of this mandrel bar temperature.
C _b and this mandrel bar temperature influence coefficient C _b
By using this, fluctuations in rolling load due to changes in the frictional characteristics between the inner surface of the tube and the mandrel bar are corrected. Also, even if the hollow temperature is different even for the same steel type, the deformation resistance will be different, so this is taken into consideration in the hollow temperature influence coefficient _{C h} mentioned above. This corrects fluctuations in deformation resistance. In addition, other factors that are difficult to estimate, such as the influence of the friction characteristics between the tube outer surface and the roll, are corrected using the learning coefficient C _p(i) described above. Next, the set rolling position S _s(i) of each stand is determined by the set wall thickness t _a(i) , the mandrel bar diameter D _b , the gauge meter deviation ΔG, the predicted rolling load P^ _(i) , and the depth of the roll hole shape. difference
Calculate based on DK _(i) using the following relationship. S _s(i) =2t _a(i) +D _b −ΔG −(P^ _(i) −P _pr )/K−2DK _(i) ……(23) Next, the cross-sectional area correction ratio C _a of each stand _(i) is calculated based on the set wall thickness t _a(i) according to the relationship of the following equation. C _a(i) = k _a(i) × (t _a(i) / t _p(i) ) ...(24) Here, k _a is the coefficient calculated in advance, and t _p is the reference wall thickness. shows. Next, the set rotational speed N _s(i) of each stand is calculated based on the cross-sectional area correction ratio C _a(i) using the following relationship. N _s(i) = (N _p(i) / C _a(i) ) ... (25) Here, N _p indicates the set rotation speed at the standard wall thickness. The set reduction position S _s(i) and set rotation speed N _s(i) obtained in this way are output to the control device of the mandrel mill to control the drive motor that drives the roll of each stand, and also to control the roll Activate the reduction device. Therefore, in this example, the predicted rolling load
By calculating P _(i) by taking into account the variation in the frictional characteristics between the inner surface of the tube and the mandrel bar, the variation in deformation resistance due to temperature changes in the hollow, and the influence of the frictional characteristics between the outer surface of the tube and the roll. , the rolling position can be controlled so that the wall thickness of the tube in the circumferential direction after rolling becomes uniform. Thereby, it is possible to control the average wall thickness of the tube after rolling to a target value. Also, the set wall thickness t _a(i) of each stand is changed to the corrected wall thickness t _c(i)
The set odd-even thickness reduction distribution ratio B _p calculated based on and the actual hollow wall thickness t _h , B _e and the actual thickness reduction distribution ratio R _p(i) ,
By calculating from R _e(i) and the set thickness reduction amount Δt _a , the set thickness reduction rate {1-(t a( _{i )} /t _{a(i) -2)} )} and calculate the predicted rolling load P^
_(i)
By calculating , it is possible to control so that the wall thickness of the tube at the bottom of the roll groove of the final stand with an odd number and an even number of wall thicknesses is equal. Therefore, the wall thickness of the rolled tube in the circumferential direction can be made uniform. Also, calculate the cross-sectional area correction ratio C _{a (i)} from the set wall thickness t a (i) and the actual hollow wall thickness t _h , and use this cross-sectional area correction ratio C _{a (i) to} calculate the set rotation speed N _{s (} By calculating _i) , it becomes possible to perform tensionless rolling, and it is possible to prevent variations in the wall thickness and outer diameter in the longitudinal direction of the tube after rolling. Particularly, in this embodiment, by performing preload zero adjustment, the roll gap during tube engagement can be determined with high precision, and therefore the rolling position can be controlled with high precision. In other words, in the past, the zero adjustment of the rolling position of a mandrel mill was carried out without applying any rolling force, and due to variations in mill rigidity due to backlash of the housing, etc.
As shown in FIG. 9, the gap during rolling had an error within a certain width. On the other hand, in this embodiment, as shown in FIG.
By adjusting the preload to zero, it is possible to eliminate the influence of variations in mill rigidity and to accurately determine the roll gap during rolling. This is because by applying the preload load P _pr , the origin of the roll gap can be moved to the right in FIG. 7, and the influence of the nonlinear part of the mill rigidity can be eliminated.

【Effect of the invention】

As explained above, according to the present invention, it is possible to control the wall thickness of the tube at the bottom of the roll groove of the final stands for odd and even thickness finishing to be equal, and therefore, the circle of the tube after rolling can be controlled to be equal. It has the excellent effect of making the wall thickness uniform in the circumferential direction and controlling the average wall thickness of the rolled tube to a target value.

[Brief explanation of drawings]

FIG. 1 is a flowchart showing the gist of the rolling control method for a mandrel mill according to the present invention, and FIG. 2 is a front view showing a state in which the balance of the rolling rolls is adjusted.
FIG. 3 is a front view showing a state in which the preload of the roll is adjusted to zero; FIG. 4 is a front view showing a state in which the roll is set to a reference gap;
FIG. 5 is a block diagram showing the configuration of a calculation device used in an embodiment of a rolling control device for a mandrel mill in which the present invention is adopted, and FIG. FIG. 7 is an enlarged vertical sectional view including a block diagram and a side view showing the configuration of the embodiment of the control device and the state of deformation of the pipe, and FIG. 8 is a side view schematically showing a conventional mandrel mill, and FIG. 9 is a diagram showing the relationship between roll gap and rolling load or rolling force according to the conventional method. 10... Hollow, 12... Mandrel bar, 1
4...Roll, 24...Arithmetic device.

Claims (1)

[Claims] 1. A rolling control method for a mandrel mill in which the outer diameter and wall thickness of a tube are gradually reduced by a plurality of stands having grooved rolls and a mandrel bar, comprising: Based on the measured values Wb, Lh, and Dh of the average outer diameter, the scale loss rate μr in the heating furnace, and the hollow density ρh, the formula th=(Dh/2)×[1−√1−{(1−)} {
(2) ² }], find the actual wall thickness th of the hollow, the material weight of the hollow, the length of the shell, the actual outer diameter Ds obtained from the actual measured outer diameter of the shell, and the scale loss rate μr of the heating furnace, Based on the shell density ρs, the formula ts=(Ds/2)×[1−√1−{(1−)}{
(2) ² }], find the actual wall thickness ts of the shell, and calculate the roll gap Go(i), rolling load P(i), and mill spring constant K of each i stand before hollow biting.
Based on the formula G(i)=Go(i)+P(i)/K, calculate the actual roll gap G(i) of each stand when the hollow is engaged, and calculate the actual roll gap G(i) of each stand when the hollow is engaged. Roll gap G(i) and mandrel bar diameter Db
Based on the formula t(i)=(G(i)-Db)/2, calculate the actual thickness t(i) of each i-stand, and calculate the thickness of each odd and even number among the calculated actual thicknesses. Actual wall thickness t (odd) of thick finishing final stand
and t(even) and the actual thickness ts of the shell to find the gauge meter deviation ΔG corresponding to the wear of the roll and mandrel bar, and calculate the actual thickness t(i) and the gauge meter deviation ΔG.
Based on the formula tc(i)=t(i)+ΔG/2, calculate the corrected wall thickness tc(i) of each i-stand, and calculate the actual wall thickness th of the hollow and the corrected wall thickness tc
Based on (i), the formula Ro(i)=(tc(i-2)-tc(i))/(th-tc(odd)) Re(i)=(tc(i-2)-tc(i) )) /(th-tc(even)) Find the actual thinning distribution ratios Ro(i) and Re(i) for each odd and even i stand, and correct for the final stand with wall thickness finishing for each odd and even number. Based on the wall thicknesses tc (odd) and tc (even) and the actual wall thickness th of the hollow, so that the corrected wall thickness of the final stand for finishing each odd and even number of hollows to be rolled next time is equal. Formula Bo={th-(tc(odd)+tc (even))/2}/(th-tc(odd)) Be={th-(tc(odd)+tc (even))/2}/(th- tc (even)), find the odd-even thinning distribution ratios Bo and Be to be set for each odd and even i stand next time, and calculate the actual wall thickness th of the hollow to be rolled next time and the wall thickness tss set for the shell. Based on the actual thickness reduction distribution ratios Ro(i) and Re(i) of the odd and even stands, and the set odd-even thickness reduction distribution ratios Bo and Be, the formula Δta(i)=(th-tss)×Ro From (i)×Bo Δta(i)=(th-tss)×Re(i)×Be, set the set thickness reduction amount Δta(i) for each i-stand so that the wall thickness of the shell becomes the set thickness. Based on the set thickness reduction amount Δta(i), the set wall thickness ta(i) of each i-stand is determined from the formula ta(i)=ta(i-2)−Δta(i), and the hollow and the temperature θh and θb of the mandrel bar, the set thinning rate {1-(ta(i)/ta(i-2))
}
Taking into consideration, the predicted rolling load P(i) is calculated, and the calculated predicted rolling load P(i), the gauge meter deviation ΔG, the set wall thickness ta(i) of each i-stand,
Mandrel bar diameter Db, roll hole depth DK (i)
Based on the formula Ss(i)=2ta(i)+Db−ΔG −(P(i)/K)−2DK(i), the set reduction position of each stand i for each hollow
Find Ss(i), and based on the wall thickness ta(i) set for each i stand,
Assuming the reference wall thickness as t0(i) and the coefficient as Ka(i), calculate the cross-sectional area correction ratio Ca(i) from the formula Ca(i)=Ka(i)×(ta(i)/t0(i)). Based on the obtained cross-sectional area correction ratio Ca(i), the set rotation speed at the standard wall thickness is set as N0(i), and from the formula Ns(i)=N0(i)/Ca(i), each i Find the set rotational speed Ns(i) of the stand, and set the determined lowering position Ss of each i stand.
A rolling control method for a mandrel mill, comprising: outputting (i) and the set rotational speed Ns(i) to a control device of the mandrel mill to control the rolling position and rotational speed of each i-stand. 2. When calculating the actual wall thickness t(i) of each i-stand, the roll-down position is zero-adjusted with the upper and lower rolls of each i-stand in contact and the preload load Ppr applied. The roll gap at the time of hollow biting is determined from the formula G(i)=Go(i)+(P(i)−Ppr)/K, and the determined roll gap G(i)
The method for controlling rolling of a mandrel mill according to claim 1, wherein the actual wall thickness t(i) is determined based on the following.