Summary of the invention
The object of the present invention is to provide dynamical roll gap compensation method during a kind of cold continuous rolling dynamic variable specification, low, that thickness the is overproof end to end technical problem of thickness control accuracy when solving dynamic variable specification; Improve the lumber recovery of cold rolled sheet.
The present invention includes two parts, a part is that the roll gap at position while welding calibrator place before inlet of rolling mill compensates; Another part is that position while welding compensates at the roll gap of each milling train outlet calibrator position.Because AGC function most of in dynamic variable specification process can not drop into, so roll gap regulation value will be determined by the model system calculated value of process computer to a great extent.Thus the thickness control accuracy when counting accuracy of model system directly affects dynamic variable specification.The actual entry thickness replacement raw thickness that calibrator place obtains by adopting actual calibrator before inlet of rolling mill, improve the primary data precision of setting model, revise the setting value of FGC roll gap according to the front and back coil of strip roll gap deviate calculated, thus improve the setting accuracy of roll gap.In each milling train outlet calibrator position, by dynamic self-adapting, obtain more accurate roll-force precision, improve the setting accuracy of roll gap, and then gap values between rollers during dynamic variable specification is compensated.Processing step is as follows:
1) calibrator place before milling train section entrance, according to measured data, carries out setup algorithm, improves model specification precision, carries out dynamic compensation to fixed value of roller slit;
2) each milling train outlet calibrator place, according to real-time measuring data, by dynamic self-adapting, improves calculation accuracy of rolling force, carries out dynamic compensation to fixed value of roller slit;
As a further improvement on the present invention, step 1) in comprise the following steps:
The first step, the band steel actual entry thickness H that milling train section porch calibrator is measured
aalternative materials thickness H, obtains actual reduction ratio according to actual entry thickness simultaneously, constant under the pressure of 2#, 3#, 4#, 5# frame;
Second step, according to the actual entry thickness obtained and reduction ratio, use rolling force model, roll torque model, rolling power model, coefficient of friction model, deformation resistance model, flattening Radius Model, forward slip model, roll gap computation model, plastic deformation Modulus Model, the rolling parameter of (stable state at a high speed, shear low speed and dynamic variable specification) under calculating different as-rolled condition, the roll gap difference of coil of strip before and after finally during acquisition dynamic variable specification as-rolled condition, is handed down to L1 level PLC as roll gap offset and controls executing agency.
As a further improvement on the present invention, step 2) in comprise the following steps:
The first step, each milling train outlet calibrator place, according to collecting the actual rolling parameter (actual belt steel thickness, mill speed, roll-force, advancing slip, roll gap, power) coming from Process Control System, calculate Dynamic Rolling Process force adaptive coefficient and advancing slip coefficient; Concrete steps are as follows:
A. each milling train outlet calibrator place, collects the actual rolling parameter coming from Process Control System, obtains actual roll-force p
a, simultaneously according to actual rolling parameter, obtain calculating roll-force p
ca, then calculate actual roll-force adaptation coefficient zp
a=p
ca/ p
a.
B. by actual roll-force adaptation coefficient zp
awith the roll-force adaptation coefficient zp in feedback form
fcarry out exponential smoothing, obtain new Dynamic Rolling Process force adaptive coefficient zp=zp
f(1-ε)+zp
a; Wherein ε is index balance factor.
C. according to collecting the actual rolling parameter coming from Process Control System, the advancing slip value of reality of each frame is calculated.
Second step, uses dynamic self-adapting coefficient in rolling Mathematical Modeling, recalculates rolling parameter values, comprise mill speed, roll-force, roll gap, thus carry out dynamic compensation to the gap values between rollers at rack outlet calibrator place; Concrete steps are as follows:
Calculate rolling parameter to be calculated by following steps and formula, first whether in the reasonable scope to check dynamic self-adapting coefficient; Then under different as-rolled condition, (comprise stable state high speed, low velocity shear, dynamic variable specification) according to rolling parameter model, calculate the rolling parameters such as coefficient of friction, roll-force, flattening radius, roll torque, rolling power, roll gap; Before and after finally during acquisition dynamic variable specification as-rolled condition, the roll gap difference of coil of strip, is handed down to basic automatization as roll gap offset.
The invention has the advantages that, carrying out dynamical roll gap compensation calculation by obtaining real data to inlet of rolling mill and outlet calibrator place, improve model intermesh determination precision, decreasing head thickness deviation during dynamic variable specification.
Detailed description of the invention
Below in conjunction with accompanying drawing, the specific embodiment of the present invention is described in detail.
This dynamical roll gap compensation method is to whether to be provided with calibrator before and after rolling mill production line milling train relevant, only has weld seam just can carry out dynamic compensation calculating when running into calibrator.This inventive method is applied to certain cold rolled silicon steel production line, and this factory uses five frame six rolling systems, has altogether installed 4 calibrators, is provided with calibrator 2 respectively before and after #1 milling train, installs calibrator 2 before and after #5 milling train.
So this dynamical roll gap compensation method calculates at #1 inlet of rolling mill place respectively, calculate in the outlet of #1 milling train, the outlet of #4 milling train and #5 milling train exit.The dynamic compensation at inlet of rolling mill calibrator place calculates the dynamic compensation exporting calibrator place with milling train and calculates not identical.
First, the dynamical roll gap compensation calculation at #1 inlet of rolling mill calibrator place is carried out.In the cold continuous rolling operation of rolling during dynamic variable specification, when spot welds is by #1 inlet of rolling mill calibrator, basic automatization sends real time data to Process Control System, and setup algorithm model uses actual (real) thickness to replace raw thickness, and calculates actual reduction ratio.
h[0]=h
a
r[0]=1-h[1]/h[0]
Wherein, h
afor the inlet thickness of #1 frame, the reduction ratio that r [i] is each frame, the thickness that h [i] is each frame.The reduction ratio r [0] of #1 frame is obtained by actual entry THICKNESS CALCULATION, and the reduction ratio of other frames is constant.
According to gateway thickness and the reduction ratio of each frame, roll data and other initial data, call setup algorithm Mathematical Modeling, the rolling parameter values such as mill speed, tension force, roll-force, flattening radius, roll gap of (stable rolling state, low velocity shear state, dynamic variable specification state) under calculating various as-rolled condition.Calculation flow chart as shown in Figure 1.
Be below computational process:
1) according to the comparative analysis of the maximum entrance velocity of #1 frame and the inlet thickness of supplied materials and maximum muzzle velocity and finished product thickness, obtain rolling calculate in maximum discharge per second;
2 according to steel grade, inlet thickness, width, the exit thickness equal-specification of rolled band steel, and inquiry constant form, sets the tension force of band steel, comprises tension force between frame and coiling tension etc.;
3) flatten radius, coefficient of friction and advancing slip according to calculated with mathematical model, then strip speed, the speed of rolls and motor speed are set;
4) according to rolling parameters such as calculated with mathematical model roll-force, roll torque, rolling powers.
The Explicit Form adopting Bland-Ford-Hill model and Hitchcock model connection to separate in theory is rolled power and calculates, and its basic theoretical model is as follows:
Wherein: p is roll-force, b is for rolling material width, k
pfor resistance of deformation, μ is coefficient of friction, D
pfor the calculated factor of hill, k is tension force factor of influence, and R ' is for flattening radius, and H is inlet of rolling mill thickness, and h is milling train exit thickness, and r is reduction ratio.
Wherein, the computing formula of tension force influence coefficient is as follows:
Wherein, k
sfor static deformation drag, t
bfor being with the backward unit tension of steel, t
ffor forward direction unit tension.
Wherein, the computing formula of dynamic deformation drag is as follows:
Wherein, k
sfor static deformation drag,
for sensitivity coefficient
5) according to calculated with mathematical model roller gap value, and the roll gap changing value caused due to the change (as broadened by narrow) of contact length;
K=b
1·log(x)-b
2
Wherein, K is milling train modulus, p
cfor roll-force constant, p is for calculating roll-force, and s is gap values between rollers, and s ' causes the modifying factor of roll gap variable quantity, s for modifying factor bending roller force
0for dead-center position during zeroing, s
zfor Roll-gap position during zeroing, C
pfor modifying factor; B1, b2 are constants, and x is the contact length of roll.
6) by corresponding formula of mathematical, calculate motor speed, then according to mathematical formulae calculating voltage and electric current, finally calculate power-balance and the mill speed of each frame;
7) whether by the computational mathematics formula of plastic deformation coefficient, distinguishing last frame is after mao roller, then calculates entrance, outlet plastic deformation coefficient successively
and tension force plastic coefficient
8) according to the gap values between rollers of coil of strip before and after during dynamic variable specification, when the roll gap difference before and after calculating during the dynamic variable specification of coil of strip and low velocity shear state and dynamic variable specification state time roll gap the difference (this value is generally 0) of variable quantity.
ds[0]=dsw[0]-dsb[0]
Wherein dsw [0] is the gap values between rollers of next coiled strip steel of #1 frame during dynamic variable specification, and dsb [0] is the gap values between rollers of #1 frame current coiled strip steel during dynamic variable specification.Ds [0] is the roll gap difference of coil of strip before and after #1 frame during dynamic variable specification.
Weld seam, after the dynamical roll gap at inlet of rolling mill place has compensated, when weld seam is through at the outlet of #1 milling train, the outlet of #4 milling train and #5 milling train outlet calibrator place, carries out dynamical roll gap compensation calculation again.
When spot welds is by milling train outlet calibrator, basic automatization sends real time data to Process Control System, now first carries out dynamic self-adapting calculating, calculates Dynamic Rolling Process force coefficient and advancing slip value, and the real data of collecting is checked, calculation process is as shown in Figure 2; And then according to the calculation procedure of weld seam at inlet of rolling mill place, carry out setup algorithm, obtain dynamical roll gap offset, be handed down to basic automatization.
Dynamic Rolling Process force adaptive coefficient calculating step is as follows:
A. each milling train outlet calibrator place, collects the actual rolling parameter coming from Process Control System, obtains actual roll-force p
a, simultaneously according to actual rolling parameter, obtain calculating roll-force p
ca, then calculate actual roll-force adaptation coefficient zp
a=p
ca/ p
a.
B. by actual roll-force adaptation coefficient zp
awith the roll-force adaptation coefficient zp in feedback form
fcarry out exponential smoothing, obtain new Dynamic Rolling Process force adaptive coefficient zp=zp
f(1-ε)+zp
a.
Wherein ε is index balance factor, p
afor actual roll-force, p
cafor calculating roll-force, zp
afor actual roll-force adaptation coefficient, zp
ffor feedback roll-force adaptation coefficient, zp is Dynamic Rolling Process force adaptive coefficient.
After using dynamic self-adapting to calculate dynamic self-adapting roll-force coefficient, zp is used in model specification calculating, obtains more accurate rolling force value.
p=p
c×zp
Wherein, p is roll-force, p
cfor not using the calculating roll-force of adaptation coefficient, zp is Dynamic Rolling Process force adaptive coefficient.
Leave zp result in dynamic setting table, dynamically set for #1 outlet, #4 outlet and #5 rack outlet place.The setting accuracy of roll gap variable quantity when dynamic self-adapting can improve dynamic variable specification, and there is reference record analytic function, for the optimization of parameter provides condition.
After dynamic self-adapting has calculated, then carry out dynamic setup algorithm in #1 outlet, #4 outlet and #5 exit successively, carry out dynamical roll gap compensation.First setup algorithm Mathematical Modeling is called, the rolling parameter values such as mill speed, tension force, roll-force, flattening radius, roll gap of (stable rolling state, low velocity shear state, dynamic variable specification state) under calculating various as-rolled condition, calculation flow chart as shown in Figure 3.
The rolling parameter values of calculating is stored in corresponding data file.
Finally, according to the gap values between rollers of coil of strip before and after during dynamic variable specification, when the roll gap difference before and after calculating during the dynamic variable specification of coil of strip and low velocity shear state and dynamic variable specification state time roll gap the difference (this value is generally 0) of variable quantity.
ds[0]=dsw[0]-dsb[0]
Wherein dsw [0] is the gap values between rollers of next coiled strip steel of #1 frame during dynamic variable specification, and dsb [0] is the gap values between rollers of #1 frame current coiled strip steel during dynamic variable specification.Ds [0] is the roll gap difference of coil of strip before and after #1 frame during dynamic variable specification.
In order to dynamical roll gap compensation method during dynamic variable specification at the scene practical effect evaluate, according to calibrator actual measurement data, the mean value of the head thickness deviation length of 52 coiled strip steels of dynamical roll gap compensation method and 57 coiled strip steels after using dynamical roll gap compensation method, afterbody thickness deviation length, maximum and minimum of a value is not used when calculating dynamic variable specification respectively, and the statistics to thickness deviation length end to end;
Take 3um as evaluation index, each coiled strip steel head before and after dropping into Optimal Parameters respectively gets 50m, afterbody gets 30m, find first overproof point within the scope of this after, then with the 10m after this point for boundary, if there is overproof point in 10m, then using in 10m, last is put as next overproof point, so repeatedly calculate, find last the overproof point in strip steel head 50m and afterbody 30m, as the overproof length of this coiled strip steel.
According to said method, calculate each selected coiled strip steel Out-gauge length, calculate the maximum of these overproof length, minimum of a value and mean value.
By the computational analysis to the overproof length of thickness end to end of each coil of strip before and after Optimal Parameters, there is following comparing result.
The overproof length vs of longitudinal tolerance:
The overproof length data contrast of steel longitudinal tolerance is with by table 1
(using 3um as standard)
Fall actual measurement data according to limit, calculate the lateral thickness difference data of 41 coiled strip steels before dynamical roll gap compensation method use and each sampling period of 48 coiled strip steels (200ms) correspondence after using respectively;
Calculate respectively dynamical roll gap compensation method use before and after each coiled strip steel in the maximum of each sampling period lateral thickness difference data, minimum of a value, mean value and mean square deviation;
Take 10um as evaluation index, calculation optimization parameter drops into the statistic of attribute result of the lateral thickness difference of front and back each coiled strip steel head, afterbody and entirety respectively, comprises maximum, minimum of a value, mean value and mean square deviation;
Above statistic of attribute result is analyzed;
The overproof length data contrast of the thick difference of the horizontal head of steel is with by table 2
(using 10um as standard)
Dropping into dynamical roll gap compensate function as can be seen from Table 1 and Table 2 during dynamic variable specification can make the overproof length mean value of the overproof length of strip steel head longitudinal thickness, lateral thickness difference reduce 1.15m, 4.46m respectively.It can also be seen that from the mean square deviation of strip steel head lateral deviation uses dynamical roll gap compensate function that the control of strip steel head lateral thickness difference can be made more stable.