JPS6124083B2 - - Google Patents

Description

[Detailed description of the invention]

The present invention relates to a method for controlling the thickness of a section steel, and particularly to an automatic thickness control method suitable for application to H-section steel. H-section steel is usually rolled using a universal mill and an edge mill as shown in FIG. In FIGS. 1a and 1b, 1a and 1b are horizontal rolls of a universal mill, 2a and 2b are vertical rolls of a universal mill, and 2a is a work side 2b is a drive side vertical roll. 3
3a and 3b are rolls of an edge mill, 4 is an H-section steel, 4a is a work side flange, 4b is a drive side flange, and 4c is a web. In conventional H-shaped steel rolling, the roll rolling position of each roll is fixed at a position determined according to the rolling schedule. However, in general shape steel rolling, as in simple plate rolling, there are disturbances such as temperature and entrance plate thickness. accuracy is reduced. When rolling is performed at a predetermined rolling position as in the past, it has been difficult to improve these precisions. The main reason why automatic plate thickness control was not performed in H-section steel rolling is that the web and flange rolling interfere with each other, so it was necessary to avoid the negative effect of one plate thickness control on the other plate thickness control. There was a point where I couldn't do it. Conventionally, the next pass or next stand is determined by comparing the values obtained by actually measuring the dimensions of each part of the rolled material after the first pass, that is, the web thickness, flange thickness, and flange width, with the preset target exit dimensions. A device for correcting the reduction set value based on the setting standard pass schedule (Japanese Patent Application Laid-open No. 7873/1983),
A method for setting rolls in a universal mill (Japanese Patent Laid-Open No. 48-66557) and a roll gap setting device (Japanese Patent Laid-open No. 65158-1982) are known to some extent. In view of the drawbacks of the conventional method, the purpose of the present invention is to
An object of the present invention is to provide a control method for producing a section steel in which both the flange thickness and the web thickness are controlled with high precision by providing an automatic sheet thickness control device for the section steel that takes into consideration the special characteristics of section steel rolling. The present invention determines the web thickness, work side flange thickness, and drive side flange thickness by considering the relative positional relationship between the horizontal roll and vertical roll of the universal mill and the elastic distortion of the mill, and then adjusts the web thickness, work side flange thickness, and drive side flange thickness to each target value. The feature is that the horizontal roll, work side vertical mill and drive side vertical roll are driven and controlled separately. Before explaining the present invention, a method for detecting webs and flanges will be explained using H-section steel rolling as an example. Second
The figure shows the movement of a horizontal mill and a vertical mill during rolling. The broken line in the figure indicates the rolling position of the roll before the H-section steel is bitten by the mill. At this time, the roll opening degrees S _HO and S _VO are defined as shown in the figure.
When the H-shaped steel is bitten by a universal mill, rolling reaction forces P _H and P _V act on the horizontal roll and vertical roll. Since the rolling mill is an elastic body, each part of the mill is displaced by this reaction force. Assuming that the spring constant in the horizontal roll direction of the mill is K _H , the spring constant in the vertical direction is K _V , and the taper of the roll is α°, the deviation between the horizontal roll direction and the vertical roll direction is P _H /K _H , P _V / _KV . Furthermore, it is assumed that the rolling positions of the horizontal roll and the vertical roll are corrected by ΔS _H and ΔS _V /cosα by automatic plate thickness control, which will be described later. The positions of the horizontal roll and vertical roll at this time are represented by solid lines in FIG. Since the H-section steel is at a very high temperature, its elastic recovery after being rolled in this mill can be ignored. Therefore, it can be seen that the horizontal roll spacing shown by the solid line in FIG. 2 represents the web thickness h _W on the exit side of the mill, and the spacing between the horizontal roll side surface and the vertical roll represents the flange thickness h _F . i.e.

[Table] Here, the roll opening degree is S _H = S _HO +ΔS _H , S _V = S _V
If _O + ΔS _V is defined, (1) and (2) become (3) and (4) as follows.
It is expressed by the formula. h _W = S _H + P _H /K _H ... (3) h _F = S _V + P _V / K _V・cosα + 1/2 (S _H − S _HO + P _H /K _H ) sinα ... (4) However If the roll openings S _H and S _V and the rolling loads P _H and P _V are detected, the plate thicknesses h _W and h _F can be determined, so if this relationship is used for control, the target H-section steel can be obtained. is obtained. Hereinafter, a method for controlling the plate thickness of H-beam steel using the above relational expressions h _W and h _F will be described. FIG. 3 shows an embodiment of the present invention. 1a in the figure,
1b is a horizontal roll, 2a and 2b are vertical rolls, 4
5a to 5c are load cells for measuring rolling load, 6a to 6c are rolling screws, 7a to
7c is a reduction drive device that moves the reduction screw; 8
9a to 9e are multipliers that multiply by a preset gain; 10a to 10c are roll opening calculation units that calculate the roll opening degrees S _H and _SV from the screw down position. , 11a to 11c represent changeover switches. First, the relationship between the position of the reduction screw and the roll opening degree will be explained. Prior to rolling the H-shaped steel, the upper horizontal roll 1a is moved until it contacts the lower horizontal roll 1b, and the rolling position q _HO at the time of contact is memorized. The roll opening degree S _H of the horizontal roll at a given time can be calculated using the rolling position q _H at that time and the memorized value q _HO .
It is expressed by equation (5). S _H = q _H − q _HO ...(5) Now, the roll opening degree of the vertical roll is defined as the distance from the horizontal roll side (broken line in Figure 2) before rolling starts, and the roll opening of the vertical roll on the work side is Let S _VW be the opening, S _VD be that of the drive side, and let q _VWO , q _VDO be the vertical roll rolling positions when each vertical roll is in contact with a horizontal mill in vertical contact, then the rolling positions q _VW , q Roll opening degree S at _VD
VW and S _VD are expressed as in the following equations (6) and (7). S _VW = S _HO /2・sinα+(q _VW −q _VWO )・cos α…
…(6) S _VD = S _HO /2・sinα+(q _VD −q _VDO )・cos α…
…(7) Furthermore, the vertical roll opening setting value before the start of rolling is set to S.
If _VO , then the position of the vertical roll reduction screw q ^~ _VW , q ^~ _VD can be calculated from the above equations (6) and (7) using the following equation (8).
，
It is expressed by equation (9). q ^~ _VW = (S _VO − S _HO /2・sinα)/cosα+q _VWO
…… (8) q ^~ _VD = (S _VO −S _HO /2・sinα)/cosα+q _VDO
... (9) Furthermore, the positions q ^to _H of the screw down for horizontal roll reduction before the start of rolling are expressed by equation (10). q ^~ _H = S _HO + q _HO (10) In the above equations (8) to (10), q _VWO , q _VDO , and q _HO can be determined by roll contact as described above. Therefore, if the roll opening settings S _HO and S _VO are given in accordance with the rolling schedule, the positions of the rolling screws q ^~ _H , q ^~ _VW , and q ^~ _VD before the start of rolling are determined. The following description will proceed on the assumption that the rolling screws are set at q ^~ _H , q ^~ _VW , and q ^~ _VD prior to the start of rolling. The roll opening degrees 10a, 10b, and 10c are q _H , q _VW , and q _VD detected by Celsin 8 a, 8 b, and 8 c.
is input, and the roll opening degrees S _H , _SVD , and _SVD are calculated based on the above equations (5), (6), and (7). It is assumed that S _HO , q _VWO , and q _VDO necessary for calculation have been set and inputted in advance. As will be described later, when controlling the plate thickness, the switches 11a, 11b,
11c is connected to the contact a side, and the contact b side is open. The rolling load P _H acting on the horizontal rolls is measured by load cells 5 installed on the left and right horizontal roll shafts.
It can be obtained as the sum of the detected values of a and 5b. At this time, the reduction command (value at point A in FIG. 3) Z _HP to the reduction motor 7a is expressed by the following equation (11). Z _HP =h _W −(S _H +α _H /K _H・P _H ) ...(11) Here, h _W is the target value or lock-on value of the web thickness (described later), and S _H is the roll opening degree calculator 10a.
The roll opening degree of the horizontal roll calculated by α _H is the control gain, which is a constant close to 1. Lowering motor 7a
Correct the position of the lowering screw 6a until Z _HP becomes zero. At this time, S _H +α _H /K _H P _H is controlled to h _W from equation (11). It is clear from equation (3) that when α _H is 1, S _H +α _H /K _H P _H represents the current thickness h _W of the web, so by controlling the reduction screw described above, h _W
It can be seen that is controlled by h _W . When α _H is less than 1, h _W is controlled to a value slightly thinner than h _W . This deviation becomes larger as α _H becomes smaller. Flange thickness control is performed independently on the work side and drive side, but since the method is exactly the same, only the work side will be explained. Input to the work side lowering motor 7b (third
The value at point B in the figure) Z _VW is expressed by the following formula. Z _VW =h _FW −{S _VW +α _V /K _V・P _VW・cosα +1/2 (S _H −S _HO + _PH /K _H ) sin α}……(1
2) Here, h _FW is the target value or lock-on value of the flange thickness on the work side (described later), α _V is the gain of the reduction control (a constant close to 1), and S _VW is the vertical roll value calculated by the roll opening calculator 10b. roll opening degree, P
_VW represents the rolling load detected by the load cell 5b, S _H represents the roll opening of the horizontal roll determined by the roll opening detector 10a, and P _H represents the rolling load acting on the horizontal roll. The lowering motor 7b moves the position of the lowering screw 6a until _ZVW becomes zero. Z _VW
When becomes zero, the second term on the right side of equation (12) becomes equal to h _FW . If α _V of the second term on the right side of equation (12) is set to 1, it can be seen from equation (2) that this term is equal to the current thickness h _F of the flange. Therefore, if α _V is 1, the flange thickness h _F is the target value or lock-on value h _F
It can be seen that it is controlled by _W. When α _V is smaller than 1, h _F is controlled to a value slightly smaller than h _FW . From the above explanation, it will be clear that even if a plate thickness deviation occurs temporarily due to various disturbances, it is immediately controlled to h _W , h _FW , h _FD or a value close to it. Gains α _H and α _V are adjustment gains for stabilizing plate thickness control, and are adjusted to about 0.6 to 0.9 in normal rolling. Next, a method for locking on the thickness of the web and flange will be explained with reference to the flow chart shown in FIG. In step 41, it is determined whether or not rolling is in progress, and if rolling is in progress, step 42 is performed to determine whether lock-on has been completed.
Judge by. And if the lock is not completed,
It is determined whether a certain period of time has elapsed or not. If the certain period of time has not elapsed, the lock-on values W, FW, and FD are calculated at step 44. If there is a delay of a certain period of time, lock-on values _W , _FW , and _FD are calculated at step 45. The lock-on calculator uses the detected values of the horizontal roll rolling load P _H , the vertical roll rolling loads P _VW , P _{VW ,} and the roll opening computed by the roll opening computing section, S _H , S _VW , S _VD .
It has the function to input. First, before rolling the H-beam, the switches 11a, 11b, 11 shown in FIG.
Connect the b side of c and leave the a side open. Therefore, the roll opening degrees of the horizontal roll and the vertical roll are S _HO and S _VO . When the H-shaped steel is bitten by the mill, input the data using the input function described above, and calculate the thickness of the web and flange, h _W and h _FW , based on equations (3) and (4).
h Calculate _FD . Since the area near the tip of the rolled material often has an abnormal shape, this calculated value will also be abnormal.
The calculated values of the thickness of the web and flange at the time when the abnormal part has passed after a certain period of time has elapsed since the rolled material was bitten are calculated as the lock-on values _W , _FW , _F
Fixed as _D. When this lock-on calculation is completed, the lock-on calculator 15 outputs this calculated value and at the same time switches the interlock switch from b to a. By switching this switch, the automatic plate thickness control device (AGC) in the horizontal and vertical roll directions starts operating, and its target values _W , _FW , and _FD are also given. At the same time as the tail end of the H-shaped steel comes out, 11a, 11
b, 11c are again connected to the contact b side, and the a side is opened. According to the embodiment described above, since the rolling down of the horizontal rolls and the rolling down of the vertical rolls can be carried out in coordination, it is possible to control the plate thickness with extremely high precision. Here, the equations h _W and h _F expressed by equations (3) and (4) can be transformed into equations (3') and (4'). h _W =S _H +P _H /K _H =q _H -q _HO +P _H /K _H ...(3') h _F =S _V +P _V /K _V・cosα+1/2(S _H +S _HO +P
_H /K _H ) sin α = {S _HO /2・sin α+(q _V −q _VO )・cos α
} +P _V /K _V・cosα+1/2{q _H −q _HO −q ^~
_H −p _{H O} ) + P _H /K _H }・sinα = q _H −q _HO /2・sinα+(q _V −q _VO )・co
sα + P _V /K _V cos α + 1/2 (p _H −q ^~ _HO + P _H
/K _H ) sinα = (q _V −q _VO +P _V /K _V )cosα +1/2 (q _H −q _HO +P _H /K _H ) sin α……(
4') Equation (4') holds true for both the drive side and work side flanges, so no particular distinction will be made in the following explanation. In the above formula, the rolling position after a certain period of time after the tip of the H-shaped steel is engaged is q ^~ _H , q ^~ _V , and the rolling load is q ^~ _H ,
If q ^~ _V , then from equations (3') and (4'), the lock-on plate thickness
_W and _F are expressed by equations (13) and (14). _W = q ^~ _H -q _HO +PH / _{K H} ... (13) _F = (q ^~ _V -q _VO + _PV /K _V )・cosα +1/2 (q ^~ _H -q _HO + _PH /K _H ) sinα……
(14) In the case of formulas (3'), (4'), (13), and (14), it is only necessary to know the rolling position and rolling load, and there is no need to calculate the roll opening degree. An embodiment based on this idea is shown in FIG. Among the symbols in the figure, those with the same symbols as in Figure 3 indicate the same thing, and furthermore, 9f~
9k is a gain multiplier, 16a to 16c are storage devices, and 17a to 17c are changeover switches. In the following, α _H =1 and α _V =1 are assumed for simplicity. Prior to H-beam rolling, selector switch 17
Connect a to 17c to the contact b side. At this time, the deviation Za' between the input value q _H -q _HO to the storage device 16a and the set roll opening S _HO is output to the reduction drive device.
That is, the output Za' is Za' = (q _H - q _HO ) - S _HO (15) Since the rolling down drive device moves until Za' becomes zero, the rolling position is adjusted to q _HO + S _HO . Similarly, the following values Zb' and Zc' are output to the work side and drive side rolling drive devices. Zb′=q _VW −q _VWO −(S _VO −S _HO /2・sinα)…
… (16) Zc′=q _VO −q _VDO −(S _VO −S _HO /2・sinα)…
... (17) Since the reduction drive devices 7b and 7c are driven until Zb' and Zc' become zero, the roll opening degrees q _VW and q _VD are
It is adjusted to q ^~ _VW and q ^~ _VD expressed by equations (8) and (9).
When equations (15), (16), and (17) become zero, the outputs of memory locations 16a-16c are locked. The memory devices 16a to 16c release their output locks after a certain period of time after the H-shaped steel is bit into the universal mill. Since the input to the storage device at this time is expressed by equations (13) and (14) from the circuit configuration shown in FIG. 5, the outputs are also _W and _F in equations (13) and (14). At the same time as this output lock is released, the changeover switch 17
Connect a to 17c to the side of a. Therefore, in subsequent rolling, the outputs of the memory devices 16a to 16c are maintained at _W and _F in equations (13) and (14). As is clear from the circuit configuration, the value that is compared with the output of this storage device at the output side of this storage device is expressed by equation (3').
Corresponds to the right side of equation (4′). Therefore, if there is a deviation between h _W , h _F and _W , _F representing these values, the rolling drive devices 7a to 7c operate until the deviation is reduced to zero. In this way, h _W and h _F are controlled to _W and _F. Target value when control gains α _H and α _V are not 1
_W and _F are (1-α _H )・P _H /K _H , (1-α _V )・_PV /K _V , respectively.
It becomes smaller by cosα+1/2·(1−α _H )·P _H /K _H ·sin α, and the controlled h _W and h _F also become smaller. By the way, the third term on the right side of equation (4) represents the change in flange thickness due to the change in the roll gap of the horizontal rolls. From the perspective of the flange thickness control system, this term plays a role in compensating for fluctuations in the horizontal roll, as is clear from FIG. Therefore, it is desirable to move the vertical roll by this amount in synchronization with the fluctuation of the horizontal roll. However, there is generally a response delay in the rolling drive system, so if the horizontal roll gap S _H is grasped and added to the vertical roll rolling drive system as shown in the third term on the right side of equation (4), there will be a delay on the vertical roll side. Occur. FIG. 6 shows an embodiment that solves this drawback. As can be seen from the above embodiments, plate thickness control is performed by changing the rolling position depending on the deviation between the target value and the exit side plate thickness. Determining the thickness deviations Δh _W and Δh _F between the web and the flange from (3'), (4'), (13), and (14), we get Δh _W = h _W − _W = (q _H + P _H /K _H ) − (q ^~ _H + P _H /K _H )
...(18) Δh _F = h _F − _F [(q _V + P _V / K _V ) cos α + 1/2 (q _H + P _H /
_{K H} )・sin α] - [(q ^~ _V + PV / K _V ) cos α + 1/2・(q ^~ _H
+P _H /K _H ) sin α] ... (19) From the above equations (18) and (19), Δh _W and Δh _F have q _H
It can be seen that _O , _qVO are not related. Therefore, the apparent target plate thicknesses _<W> , _<F> , and the apparent exit side plate thicknesses _<hW> and _<hF> at arbitrary times are defined as follows. < _W > = q ^~ _H + P / K _H ... (20) < _F > = (q ^~ _V + P _V / K _V ) cosα + 1/2 (q ^~ _H + P _H / K _H ) sin α ... (21) <h _Wr > = q _H + P _H /K _H ...... (22) <h _F > = (q _V + P _V / K _V ) cosα + 1/2 (q _H + P _H /K _H ) sin α ... (23) At this time, from equations (18) to (23), the following holds true: Δh _W =〈h _W 〉−〈 _W 〉 (24) Δh _F =〈h _F 〉− _〈F〉 ……(25). Therefore, _W in the embodiments up to FIG. 5,
Use < _W >, < _F > instead of _F , h _W , h _F
The same effect can be obtained by using <h _W > and <h _F > instead. In fact, the control method and device become simpler because there is no need to measure or calculate the rolling down positions S _HO and S _VO when the rolls are in contact with each other. In the embodiments shown in FIGS. 3 and 5, the plate thickness h _W , h _F
W , _hFD is the plate thickness detected immediately after the start of rolling
Controlled by _W , _FW , and _FD . However, _W , _FW ,
Since _FB is not necessarily the finished target plate thickness, h _W , h _FW , and h _FD include errors. Furthermore, if there is an error in the values of the mill constants K _H and K _V or if there is nonlinearity, a certain error will occur in the plate thicknesses h _W , h _FW , and h _FD . In addition to this, the zero point position q _HO , q
If there is a detection error in _VWO , _qVDO , or if _qHO , _qVWO , _qVDO changes due to roll wear during rolling, DC-like errors may occur. FIG. 6 shows an embodiment for improving such DC errors. In Figure 6, h _W , h _FW , h _F
D is the lock-on plate thickness used in the embodiments of FIGS. 3 and 5. Further, 18a to 18c are integrators, 19a to 19c are switches, and 20 is a plate thickness detector provided on the outlet side of the universal mill, which are already commercially available. 20a is a gamma ray source, 20b~
Reference numeral 20d denotes a thickness detection unit that determines the plate thickness based on the amount of γ-rays transmitted through the web and flange. Plate thickness tester 20
The detected plate thickness is the finishing target plate thickness h _WP , h _FW
P ,h compared with _FDP . When the tip of the H-shaped steel reaches the plate thickness detector 20 and the signal from the detector 20 rises, the switches 19a to 19c are turned on.
After this, the deviation between the target finishing plate thickness and the signal from the detector 20 is integrated. When this integral value is added to the lock-on plate thickness, the rolling down device is activated and the plate thickness changes. This control loop controls the plate thickness so that the input to the integrators 18a to 18c becomes zero, so
Ultimately, the plate thickness detected by the detectors 20a to 20c coincides with the finishing target value. Since this plate thickness detector is placed at a position away from the universal, there is a dead time between when the plate is actually rolled and when it is detected. Therefore, the control loop by these detectors 20a to 20c is caused by plate thickness fluctuations with a period longer than this dead time.
This is particularly effective against the DC component of the plate thickness variation mentioned above. Higher frequency disturbances can be removed by the method shown in the embodiments of FIGS. 3 and 5.
By using the circuit shown in FIG. 6 which feeds back the measured value of the plate thickness in this way, it becomes possible to control plate thickness variations over a wide frequency range. _{Note that the} apparent thicknesses _{<W> , <FW> , <FD> ,}
When using <h _W >, <h _FW >, and <h _FD >, exactly the same effect can be obtained by adding the output of the integrator to < _W >, < _FW >, and < _FD >. Furthermore, in the above embodiments, for example, a method was explained in which the vertical rolls of a universal mill are controlled independently on the work side and the drive side. good. Furthermore, the vertical roll may be fixed and the web may be controlled only by the horizontal roll. Although highly accurate control cannot be expected in the above two modified examples, they are meaningful as simple systems. Although the present invention has been described above with reference to H-section steel, which is a typical type of section steel, it goes without saying that it can also be applied to general section steel. In the case of H-beam steel, a yield improvement of about 1% can be expected. Furthermore, since the web thickness, drive side flange thickness, and work side flange thickness are controlled independently, it can also be applied to non-symmetrical sections.

[Brief explanation of the drawing]

Fig. 1 is an explanatory diagram of a universal mill and an edge mill, Fig. 2 is an explanatory diagram of the operation of a horizontal roll and a vertical roll, Fig. 3 is an example of an H-beam steel according to the present invention, and Fig. 4 is a lock-on control method thereof. 5 shows another embodiment of the present invention, and FIG. 6 shows a partial diagram of an improved method of FIG. 5. 1a, b...Horizontal roll, 2a, b...Vertical roll, 3...Edge mill roll, 4...H section steel, 5a-c...Load cell, 6a-c...Reduction screw, 7a-c... Lowering drive device, 8a~
c...Selsin, 9a-c...gain multiplier, 1
0a~c...Roll opening calculation unit, 11a~c...
Changeover switch.

Claims (1)

[Claims] 1. In a universal mill consisting of horizontal rolls and vertical rolls, the rolling position and rolling load of each of the horizontal rolls and vertical rolls are detected, and the web is determined from the detected rolling position and rolling load of the horizontal rolls. Calculate the thickness, calculate the flange thickness from the detected rolling position of each roll, rolling load, and roll taper α, and generate a deviation signal between the calculated web thickness, flange thickness, and each thickness as a target value. 1. A method for automatically controlling the thickness of a sectioned steel, characterized in that: is used as a signal to control the rolling reduction of each roll. 2. According to claim 1, the calculated value of the thickness of the web and flange after a predetermined time has elapsed since the tip of the material to be rolled is bitten by the mill is set as the lock-on value. Automatic plate thickness control method for section steel. 3. The calculation of the flange thickness as set forth in claim 1 is calculated by multiplying the rolling position of the horizontal roll and the linear sum of the rolling load acting on the horizontal roll by the cosine of the roll taper, and the value obtained by multiplying the cosine of the roll taper. An automatic plate thickness control method for section steel, characterized in that the thickness is determined from the sum of the rolling position and the linear sum of rolling loads acting on vertical rolls multiplied by the sine of the roll taper.