CN1990131A - Roughed plate bloom temperature control method in hot-rolled process - Google Patents

Abstract

The invention discloses a rough roll bar plate temperature control method in operation of heat rolling, the invention mainly uses the temperature when bar plate is pulled out of heating-furnace as start, according to the movement order and o running time of bar plate uses the recursive analytic temperature distribution model to calculating the temperature distribution of each points on the bar plate rolling operation path; and compares the average temperature of the measuring point on the bar plate after rough roll and the rough roll goal temperature, if the compared result is bigger then the permissible value of set temperature contrast of said both parties, then makes the last entrance temperature of rough roll as reference, and controls the temperature of bar plate through swing steel, in addition, makes self-adaptive amendment of the model radiation coefficient and water cooling coefficient in different areas, while according the rough roll goal temperature determines the swing time of bar plate on the middle roll table to controlling the temperature of bar plate accurately and improving the whole control level of rough roll procedure.

Description

A kind of in course of hot rolling roughed plate bloom temperature control method

Technical field

The present invention relates to the processing and forming of hot-rolled steel, more specifically refer to a kind of in course of hot rolling roughed plate bloom temperature control method.

Background technology

During hot rolling is produced, temperature is a very important technical parameters, the variations in temperature of each link that forecasts with unerring accuracy is to realize the computer-controlled important prerequisite of continuous hot-rolling mill, whether the forecast of rolling temperature is accurate, its whole setting is calculated and dimensional accuracy, plate shape and the mechanical property of improving rolled piece all has very important significance, and is the core place of improving rolling stability.

The model that adopts is usually calculated in board briquette control at present and forecast two classes, and a class is to simplify theoretical model, and another kind of is difference model.Before model I be to be distributed as the simplification theoretical model that obtains under the stable state hypothesis such as parabola in band steel inside, it has the fast advantage of computational speed, but computational accuracy is relatively poor, especially poorer when environmental catastrophe, this model can't provide the upper and lower surface temperature of slab, does not satisfy the board briquette control of hot-rolled process high request.Then a class difference model has the difference model of demonstration and recluse's difference model again, shows that difference model is not absolute stable, though recluse's difference model is absolute stable, amount of calculation is big.In order to realize the high-precision board briquette control of hot rolling roughing operation, just need resolve the computational stability of districution temperature control model and the contradiction of computational speed, make that districution temperature is controlled model computational accuracy height, computational speed is fast, computational stability good.

To entering the finish rolling inlet slab of strict temperature requirement is arranged, require slab to reach certain target temperature in the roughing outlet.This requirement realizes at delay table pendulum steel by slab, go out the temperature that slab arrives roughing outlet temperature measurement point by the board briquette model prediction, target temperature with technological requirement compares then, calculates the pendulum needed time of steel according to the temperature difference, and issues electric execution.This requires accurately to calculate the pendulum needed time of steel, and it depends on the computational accuracy of board briquette model, and is not only last computational accuracy, more requires the regional model calculating beyond radiation areas and radiation to have higher precision.Traditional rough rolling plate blank temperature computation model is paid close attention to the final result of calculation that reaches the roughing measurement point, and the actual measured value that does not make full use of temperature measuring instrument is carried out subregional correction to the model coefficient of middle process.Provide board briquette accurately for the calculating that solves the rough rolling plate blank temperature and control problem and to the calculating of technology amounts such as roll-force, just need to solve adaptive learning (be called correction again, the down with) problem of temperature distribution model.

As seen, in the hot rolling roughing operation slab rolling process,, need accurately forecast and control board briquette in order to improve the precision that whole setting calculates and to make the rolled piece temperature satisfy of the requirement of finish rolling inlet to middle board briquette.

Summary of the invention

The object of the present invention is to provide a kind of in course of hot rolling roughed plate bloom temperature control method, to realize good stability, the board briquette control that speed is fast and precision is high.

To achieve these goals, the present invention adopts following technical scheme,

A kind of in course of hot rolling roughed plate bloom temperature control method,

This temperature-controlled process may further comprise the steps:

A at first, is a starting point with the Temperature Distribution of slab when heating furnace is extracted out, presses the mobile order and the running time of slab, adopts solution by recursion formula eutectoid temperature distributed model, obtains the Temperature Distribution of each point on the slab rolling operating path;

B, then, slab measurement point after roughing is compared by mean temperature and the roughing target temperature that step a is tried to achieve, if comparative result is greater than the permissible value of the difference of both temperature that set, be reference then, put steel, slab is carried out radiation cooling with last passage inlet temperature of roughing, if comparative result less than setting the temperature permissible value, then carries out normally rolling to slab;

C, again the slab upper surface temperature measured value of slab at measurement point compared with the slab upper surface temperature that adopts solution by recursion formula eutectoid temperature distributed model to try to achieve at this place, when both deviation occurs, model radiation coefficient and water-cooled coefficient subregion are carried out the self adaptation correction, reach minimum and correction result is used on the follow-up slab of same size until deviation.

In the described step a),

Described hot rolling rough rolling plate blank solution by recursion formula eutectoid temperature distributed model mainly is meant according to boundary condition determines upper surface temperature, underlaying surface temperature, the mean temperature of slab in rolling process of rough rolling.

Described upper surface temperature, underlaying surface temperature, mean temperature expression formula are respectively,

$\mathrm{\θ}(\frac{H}{2},t)=\underset{i=0}{\overset{4}{\mathrm{\Σ}}}{(-1)}^i{A}_i\left(t\right)+\underset{j=1}{\overset{4}{\mathrm{\Σ}}}{(-1)}^{j-1}{B}_j\left(t\right),$

$\mathrm{\θ}(-\frac{H}{2},t)=\underset{i=0}{\overset{4}{\mathrm{\Σ}}}{(-1)}^i{A}_i\left(t\right)-\underset{j=1}{\overset{4}{\mathrm{\Σ}}}{(-1)}^{j-1}{B}_j\left(t\right),$

θ _m(t)=A ₀(t), and

Above various model coefficient obtain by the following formula recursion,

$\left[\begin{array}{c}{A}_0\left(t\right)\\ A_1\left(t\right)\\ A_2\left(t\right)\\ A_3\left(t\right)\\ A_4\left(t\right)\\ B_1\left(t\right)\\ B_2\left(t\right)\\ B_3\left(t\right)\\ B_4\left(t\right)\end{array}\right]=\left[\begin{array}{ccccccccc}1& 0& 0& 0& 0& 0& 0& 0& 0\\ 0& m_1\left(t\right)& 0& 0& 0& 0& 0& 0& 0\\ 0& 0& m_2\left(t\right)& 0& 0& 0& 0& 0& 0\\ 0& 0& 0& m_3\left(t\right)& 0& 0& 0& 0& 0\\ 0& 0& 0& 0& m_4\left(t\right)& 0& 0& 0& 0\\ 0& 0& 0& 0& 0& m_1^{*}\left(t\right)& 0& 0& 0\\ 0& 0& 0& 0& 0& 0& m_2^{*}\left(t\right)& 0& 0\\ 0& 0& 0& 0& 0& 0& 0& m_3^{*}\left(t\right)& 0\\ 0& 0& 0& 0& 0& 0& 0& 0& m_4^{*}\left(t\right)\end{array}\right]\left[\begin{array}{c}{A}_0\left(0\right)\\ A_1\left(0\right)\\ A_2\left(0\right)\\ A_3\left(0\right)\\ A_4\left(0\right)\\ B_1\left(0\right)\\ B_2\left(0\right)\\ B_3\left(0\right)\\ B_4\left(0\right)\end{array}\right]$

$+\left[\begin{array}{c}{a}^2\·({\mathrm{\φ}}_S+{\mathrm{\φ}}_I)\·t/(\mathrm{\λ}\·H)\\ p_1\·K\·(1-m_1\left(t\right))\\ p_2\·K\·(1-m_2\left(t\right))\\ p_3\·K\·(1-m_3\left(t\right))\\ p_4\·K\·(1-m_4\left(t\right))\\ q_1\·R\·(1-m_1^{*}\left(t\right))\\ q_2\·R\·(1-m_2^{*}\left(t\right))\\ q_3\·R\·(1-m_3^{*}\left(t\right))\\ q_4\·R\·(1-m_4^{*}\left(t\right))\end{array}\right]$

In the formula,

λ is the slab pyroconductivity, and ρ is a slab proportion, c _pBe slab specific heat, the three is the physical parameter that is heated slab, is given in the reality, A _i(0), B _i(0) be the initial parameter that model calculates, initial value is provided by the heating furnace model, and follow-up parameter recursion obtains,

$a^2=\frac{\mathrm{\λ}}{\mathrm{\ρ}\·c_p},$ Be the slab thermal diffusion coefficient,

$m_i\left(t\right)=\exp \{-\frac{4\·i^2\·a^2\·{\mathrm{\π}}^2\·t}{H^2}\}$ (i＝1～4)，

$m_j^{*}\left(t\right)=\exp \{-\frac{{(2j-1)}^2\·a^2\·{\mathrm{\π}}^2\·t}{H^2}\}$ (j＝1～4)，

T is the calculating step-length,

H is a slab thickness,

$K=\frac{({\mathrm{\φ}}_S+{\mathrm{\φ}}_I)\·H}{2\·\mathrm{\λ}},R=\frac{({\mathrm{\φ}}_S-{\mathrm{\φ}}_I)\·H}{4\·\mathrm{\λ}}$

φ _S, φ _IBe the hot-fluid of slab upper and lower surface, determine by boundary condition.

Among the described step b, when putting steel,

Pendulum steel bar spare is Δ θ＞T _Const, T wherein _ConstBe the constant of process stipulation,

Δ θ=θ _m-θ _{T arg et}, θ _mBe the slab mean temperature that model calculates, θ _{T arg et}It is the given target temperature of technology;

The pendulum steel time is Δ t=(Δ θ-T _Const/ 2) * η _t,

${\mathrm{\η}}_t=\frac{\mathrm{\Δ}{t}_a}{\mathrm{\Δ}{\mathrm{\θ}}_{\mathrm{out}}}$

Δ t _aBe slab radiated time perturbation amount before last passage,

Δ θ _OutBe slab in roughing outlet temperature measuring position, the slab mean temperature is with respect to Δ t _aVariable quantity.

Among the described step c, to the correction of solution by recursion formula eutectoid temperature distributed model parameter the time, mainly be to utilize slab to resolve the slab upper surface temperature of distributed model with the board briquette recursion respectively to compare, model radiation coefficient and water-cooled coefficient subregion are carried out the self adaptation correction respectively at this place at the slab upper surface measured temperature of measurement point between pony roughing mill and after the roughing.

Resolve the slab upper surface temperature value of distributed model and compare when revising and further comprise the steps: carrying out between pony roughing mill measured temperature and board briquette recursion at this place

C11 carries out deviation and decomposes;

C12 obtains the correction of radiation coefficient and water-cooled coefficient respectively;

C13, limiting value and smoothing processing;

C14, the board briquette that utilizes the roughing actual value to obtain first area terminal point measuring instrument position after the roughing distributes.

Resolve the slab upper surface temperature value of distributed model and compare when revising and further comprise the steps: carrying out after the roughing measured temperature and board briquette recursion at this place

C21 carries out deviation and decomposes;

C22 obtains the correction of radiation coefficient and water-cooled coefficient respectively;

C23, limiting value and smoothing processing.

In the technical solution adopted in the present invention, this method is a starting point with the Temperature Distribution of slab when heating furnace is extracted out mainly, press the mobile order and the running time of slab, adopt solution by recursion formula eutectoid temperature distributed model, obtain the Temperature Distribution of each point on the slab rolling operating path; And the mean temperature and the roughing target temperature of slab measurement point after roughing compared, if comparative result is greater than the permissible value of the difference of both temperature that set, be reference then with last passage inlet temperature of roughing, by the pendulum steel slab is carried out temperature control, in addition, also to model radiation coefficient and water-cooled coefficient subregion are carried out the self adaptation correction.Therefore, method of the present invention can be determined the Temperature Distribution of hot rolling roughing operation slab in each position rapidly and accurately, determines the pendulum steel time of slab at delay table according to the roughing target temperature simultaneously, accurately controls the temperature of slab.This method has solved traditional rough rolling plate blank temperature model and has had contradiction between computational speed, computational accuracy and the computational stability, have and calculate simply, speed is fast, the advantage of good stability, also solved the technological deficiency that traditional temperature distribution model is difficult to model parameter is carried out adaptive learning, the computational accuracy and the rough rolling plate blank control accuracy of temperature of rough rolling plate blank temperature model have been improved, and, improved the integral body control level of roughing operation for the calculating of technology amounts such as roughing roll-force provides reliable board briquette.

Description of drawings

Fig. 1 is for adopting control method schematic flow sheet of the present invention.

Fig. 2 is for adopting a flow process schematic diagram of control method of the present invention.

Fig. 3 is for adopting the another schematic flow sheet of control method of the present invention.

Fig. 4 is a hot rolling roughing unit area arrangements schematic diagram.

The specific embodiment

In order more to be expressly understood content of the present invention, describe in detail further below in conjunction with accompanying drawing and specific embodiment.

Embodiment one

See also shown in Figure 1ly, this temperature-controlled process may further comprise the steps:

A at first, is a starting point with the Temperature Distribution of slab when heating furnace is extracted out, presses the mobile order and the running time of slab, adopts solution by recursion formula eutectoid temperature distributed model, obtains the Temperature Distribution of each point on the slab rolling operating path;

B, then, slab measurement point after roughing is compared by mean temperature and the roughing target temperature that step a is tried to achieve, if comparative result is greater than the permissible value of the difference of both temperature that set, be reference then, put steel, slab is carried out radiation cooling with last passage inlet temperature of roughing, if comparative result less than setting the temperature permissible value, then carries out normally rolling to slab;

C, again the slab upper surface temperature measured value of slab at measurement point compared with the slab upper surface temperature that adopts solution by recursion formula eutectoid temperature distributed model to try to achieve at this place, when both deviation occurs, model radiation coefficient and water-cooled coefficient subregion are carried out the self adaptation correction, reach minimum and correction result is used on the follow-up slab of same size until deviation.

Specifically,

Utilize recursion to resolve the districution temperature model, determine that according to boundary condition the board briquette of rough rolling plate blank in this hot rolling rolling process of rough rolling distributes.

Described hot rolling rough rolling plate blank recursion is resolved districution temperature model tormulation formula:

Slab upper surface temperature:

$\mathrm{\θ}(\frac{H}{2},t)=\underset{i=0}{\overset{4}{\mathrm{\Σ}}}{(-1)}^i{A}_i\left(t\right)+\underset{j=1}{\overset{4}{\mathrm{\Σ}}}{(-1)}^{j-1}{B}_j\left(t\right)$

The slab underlaying surface temperature:

$\mathrm{\θ}(-\frac{H}{2},t)=\underset{i=0}{\overset{4}{\mathrm{\Σ}}}{(-1)}^i{A}_i\left(t\right)-\underset{j=1}{\overset{4}{\mathrm{\Σ}}}{(-1)}^{j-1}{B}_j\left(t\right)$

The slab mean temperature:

θ _m(t)＝A ₀(t)

Above various model coefficient obtain by the following formula recursion:

$\left[\begin{array}{c}{A}_0\left(t\right)\\ A_1\left(t\right)\\ A_2\left(t\right)\\ A_3\left(t\right)\\ A_4\left(t\right)\\ B_1\left(t\right)\\ B_2\left(t\right)\\ B_3\left(t\right)\\ B_4\left(t\right)\end{array}\right]=\left[\begin{array}{ccccccccc}1& 0& 0& 0& 0& 0& 0& 0& 0\\ 0& m_1\left(t\right)& 0& 0& 0& 0& 0& 0& 0\\ 0& 0& m_2\left(t\right)& 0& 0& 0& 0& 0& 0\\ 0& 0& 0& m_3\left(t\right)& 0& 0& 0& 0& 0\\ 0& 0& 0& 0& m_4\left(t\right)& 0& 0& 0& 0\\ 0& 0& 0& 0& 0& m_1^{*}\left(t\right)& 0& 0& 0\\ 0& 0& 0& 0& 0& 0& m_2^{*}\left(t\right)& 0& 0\\ 0& 0& 0& 0& 0& 0& 0& m_3^{*}\left(t\right)& 0\\ 0& 0& 0& 0& 0& 0& 0& 0& m_4^{*}\left(t\right)\end{array}\right]\left[\begin{array}{c}{A}_0\left(0\right)\\ A_1\left(0\right)\\ A_2\left(0\right)\\ A_3\left(0\right)\\ A_4\left(0\right)\\ B_1\left(0\right)\\ B_2\left(0\right)\\ B_3\left(0\right)\\ B_4\left(0\right)\end{array}\right]$

$+\left[\begin{array}{c}{a}^2\·({\mathrm{\φ}}_S+{\mathrm{\φ}}_I)\·t/(\mathrm{\λ}\·H)\\ p_1\·K\·(1-m_1\left(t\right))\\ p_2\·K\·(1-m_2\left(t\right))\\ p_3\·K\·(1-m_3\left(t\right))\\ p_4\·K\·(1-m_4\left(t\right))\\ q_1\·R\·(1-m_1^{*}\left(t\right))\\ q_2\·R\·(1-m_2^{*}\left(t\right))\\ q_3\·R\·(1-m_3^{*}\left(t\right))\\ q_4\·R\·(1-m_4^{*}\left(t\right))\end{array}\right]$

In the following formula:

λ is the slab pyroconductivity, and ρ is a slab proportion, c _pBe slab specific heat, the three is the physical parameter that is heated slab, is given in the reality, A _i(0), B _i(0) be the initial parameter that model calculates, initial value is provided by the heating furnace model, and follow-up parameter recursion obtains.

Wherein:

$a^2=\frac{\mathrm{\λ}}{\mathrm{\ρ}\·c_p},$ Be the slab thermal diffusion coefficient.

$m_i\left(t\right)=\exp \{-\frac{4\·i^2\·a^2\·{\mathrm{\π}}^2\·t}{H^2}\}$ (i＝1～4)

$m_j^{*}\left(t\right)=\exp \{-\frac{{(2j-1)}^2\·a^2\·{\mathrm{\π}}^2\·t}{H^2}\}$ (j＝1～4)

T: calculate step-length

H: slab thickness

$K=\frac{({\mathrm{\φ}}_S+{\mathrm{\φ}}_I)\·H}{2\·\mathrm{\λ}},R=\frac{({\mathrm{\φ}}_S-{\mathrm{\φ}}_I)\·H}{4\·\mathrm{\λ}}$

φ _S, φ _IBe the hot-fluid of slab upper and lower surface, determine by the different boundary condition of back.

p ₁＝-0.106694，p ₂＝0.03125，p ₃＝-0.0183，p ₄＝0.007812

q ₁＝0.821068，q ₂＝-0.101245，q ₃＝0.0452020，q ₄＝-0.03248

When definite hot rolling rough rolling plate blank temperature, the heat flow when requiring out slab to be positioned at air cooling zone, water-cooled district and rolling deformation district.

Concrete definite method is as follows:

The calculating of air cooling heat flow:

φ _s＝ε·σ·[(T _air+237) ⁴-(θ _s+273) ⁴]

Wherein σ is Si Difen-graceful constant of bohr thatch.

T _AirIt is the air cooling environment temperature.

θ _sBe the belt steel surface temperature, distributed model calculates.

ε is a radiation coefficient, ε ∈ [0.75,0.88], and occurrence can be determined by experiment.

The calculating of water-cooled heat flow:

φ _s＝a _w·(T _w-θ _s)

T wherein _WBe the temperature of cooling water, a _wBe the water-cooled heat exchange coefficient, determine by experiment.

The calculating of the heat flow that belt steel rolling is out of shape and the band steel contacts with roll and produce:

φ _s＝(T _FOR-T _LTG)·C _P·ρ·δ/t _c

T wherein _FORBe rolling deformation heat, T _LTGBe the heat loss that the band steel contacts with roll, C _P: band steel specific heat, ρ: band steel proportion, δ: half belt steel thickness, t _c: band steel and roll time of contact.

The hot T of rolling deformation _FORDetermine as follows:

$T_{\mathrm{FOR}}=\frac{C\·\mathrm{HART}\left(I\right)\·\mathrm{FEPS}\left(I\right)\·\mathrm{SWM}}{\mathrm{BRH}\left(I\right)\·\sqrt{\mathrm{AWR}\left(G\right)\·H_{\mathrm{EIN}}\left(I\right)}}$

Wherein: C is the work of deformation invariant.

FEPS (I) is the influence coefficient of depressing relatively distortion, calculates according to relative reduction EPS (I), promptly $\mathrm{FEPS}\left(I\right)={\left(\frac{\mathrm{EPS}{\left(I\right)}^3}{1-\mathrm{EPS}\left(I\right)}\right)}^{1/2}.$

HEAT (I) is a material hardness.

SWM is the roll torque coefficient of heredity of passage special use.

BRH (I) is a band steel throat width.

AWR (G) is that working roll flattens radius.

H _EIN(I) be inlet thickness.

The heat loss T that the band steel contacts with roll _LTGContact temperature loss TLT (I) and rolling deformation heat with roll additional to contact warm damage TFLG (I) definite by not considering to be out of shape when hot the band steel:

TLTG(I)＝TLT(I)+TFLG(I)

TLT (I) is calculated as follows:

$\mathrm{TLT}\left(I\right)=\mathrm{GTL}\·(T_{\mathrm{EIN}}\left(I\right)-T_W)\·\frac{\mathrm{GLG}\left(I\right)\·\mathrm{WDUZ}\left(I\right)}{H_{\mathrm{AUS}}\left(I\right)\·V_{\mathrm{AUS}}\left(I\right)}$

Wherein: GLT is the temperature conduction weighted factor, is 0.4986 * 10 ^-3(mmm/sec)/and Z, Z is iron scale thickness (mm).

T _EIN(I) the band steel inlet mean temperature of calculating for model, T _WRoller temperature, GLG are rolling contact arc length, and WDUZ (I) is that heat exchange coefficient is obtained by experiment.

H _AUS(I) be band steel exports thickness, V _AUSBe band steel exports speed.

TFLG (I) is calculated as follows:

TFLG(I)＝GFLG·[1-WDUZ(I)]·T _FOR

Wherein: GFLG is the weighting constant of distortion heat to heat conduction influence.

Determine the pendulum steel of hot rolling rough rolling plate blank on delay table during the time, main mean temperature and the roughing target temperature that calculates according to the model of slab measurement point after roughing, and follow following formula and obtain the pendulum steel time:

Pendulum steel bar spare is Δ θ＞T _Const, T wherein _ConstBe the constant of process stipulation, Δ θ=θ _m-θ _{T arg et}, θ _mBe the slab mean temperature that model calculates, θ _{T arg et}It is the given target temperature of technology.

Pendulum steel time Δ t is:

Δt＝(Δθ-T _const/2)*η _t

η wherein _tBe calculated as follows:

${\mathrm{\η}}_t=\frac{\mathrm{\Δ}{t}_a}{\mathrm{\Δ}{\mathrm{\θ}}_{\mathrm{out}}}$

Δ t _aFor slab radiated time perturbation amount before last passage, got 20 minutes.

Δ θ _OutBe slab in roughing outlet temperature measuring position, the slab mean temperature is with respect to Δ t _aThe variable quantity that causes.

When the rough rolling plate blank recursion is resolved the model parameter correction, utilize slab at the slab upper surface temperature measured value of measurement point between pony roughing mill and after the roughing, the slab upper surface temperature of calculating at this place with board briquette recursion analytic modell analytical model compares, model radiation coefficient and water-cooled coefficient subregion are carried out adaptive learning respectively, and the result uses on the follow-up slab of same size.Correction thes contents are as follows:

● the condition of correction

$T_{\mathrm{gate}}^{\min }<\left|\mathrm{\&Delta;}{\mathrm{\&theta;}}_s\right|<T_{\mathrm{gate}}^{\max }$

T wherein _Gate ^Max, T _Gate ^MinBe the constant of process stipulation, Δ θ _s=θ _s-θ _Act, θ _sBe the slab upper surface temperature that model calculates, θ _ActIt is the steel slab surface temperature of thermometer actual measurement.

● deviation is distributed

As Δ θ _s=θ _s-θ _Act＞0 o'clock,

$\mathrm{\&Delta;}{\mathrm{\&theta;}}_w=\mathrm{\&beta;}*(\mathrm{\&Delta;}{\mathrm{\&theta;}}_s-T_{\mathrm{gate}}^{\min }/2)$

$\mathrm{\&Delta;}{\mathrm{\&theta;}}_r=(1-\mathrm{\&beta;})*(\mathrm{\&Delta;}{\mathrm{\&theta;}}_s-T_{\mathrm{gate}}^{\min }/2)$

As Δ θ _s=θ _s-θ _Act＜0 o'clock,

$\mathrm{\&Delta;}{\mathrm{\&theta;}}_w=\mathrm{\&beta;}*(\mathrm{\&Delta;}{\mathrm{\&theta;}}_s-T_{\mathrm{gate}}^{\min }/2)$

$\mathrm{\&Delta;}{\mathrm{\&theta;}}_r=(1-\mathrm{\&beta;})*(\mathrm{\&Delta;}{\mathrm{\&theta;}}_s-T_{\mathrm{gate}}^{\min }/2)$

β ∈ [0,1] wherein; Δ θ _w, Δ θ _rBe respectively the model calculation deviation that water-cooled convection coefficient and radiation coefficient error cause.

● the model coefficient correction

Radiation coefficient for the first area is modified to:

Δε ₁＝Δθ _r/M

$M=\mathrm{\&Delta;}{\mathrm{\&theta;}}_{\mathrm{out}}^1/\mathrm{\&Delta;}{\mathrm{\&epsiv;}}_v$

Wherein, Δ ε ₁For eliminating temperature deviation Δ θ _rThe needed change amount of first area radiation coefficient, M is the influence coefficient of radiation coefficient to board briquette,

Δ ε _rFor the perturbation amount of first area radiation coefficient, get 0.2.

Δ θ _Out ¹Be that slab is the first area between pony roughing mill, the temperature measuring set position, slab upper surface temperature is with respect to Δ ε _rVariable quantity.

Water-cooled convection coefficient for the first area is modified to:

$\mathrm{\&Delta;}{a}_w^1=\mathrm{\&Delta;}{\mathrm{\&theta;}}_w/N$

$N=\mathrm{\&Delta;}{\mathrm{\&theta;}}_{\mathrm{out}}^1/\mathrm{\&Delta;}{a}_w^v$

Wherein, Δ a _w ¹For eliminating temperature deviation Δ θ _wThe needed change amount of first area water-cooled convection coefficient, N is the influence coefficient of water-cooled convection coefficient to board briquette, Δ a _w ^vFor the perturbation amount of first area water-cooled convection coefficient, get 1000.Δ θ _Out ¹Be that slab is the first area between pony roughing mill, the temperature measuring set position, slab upper surface temperature is with respect to Δ a _w ^vVariable quantity.

Correction for second area radiation coefficient and water-cooled convection coefficient is identical with the model coefficient correction of first area, just the starting point of model calculating is the terminal point of first area, the recursive parameter of recursion analytic modell analytical model is to utilize the parameter that obtains behind the model of first area, utilizes the actual information recursion to the recursive parameter of the terminal point of the first area initial parameter as the recursion analytic modell analytical model.

● limiting value inspection and smoothing processing

The method that bound is checked is as follows:

Δε _i＝MIN(Δε _i，Δε _uper)

Δε _i＝MAX(Δε _i，Δε _lower)

Wherein, Δ ε _i, i=1,2 is the change amount of first area or second area radiation coefficient, Δ ε _UpperBe radiation coefficient change amount higher limit, Δ ε _LowerIt is radiation coefficient change amount lower limit.

$\mathrm{\&Delta;}{a}_w^i=\mathrm{MIN}(\mathrm{\&Delta;}{a}_w^i,\mathrm{\&Delta;}{a}_{\mathrm{uper}})$

$\mathrm{\&Delta;}{a}_w^i=\mathrm{MAN}(\mathrm{\&Delta;}{a}_w^i,\mathrm{\&Delta;}{a}_{\mathrm{lower}})$

Wherein, Δ a _w ⁱ, i=1,2 is the change amount of first area or second area water-cooled convection coefficient, Δ a _UpperBe the change amount higher limit of water-cooled convection coefficient, Δ a _LowerIt is the change amount lower limit of water-cooled convection coefficient.

The method of smoothing processing is as follows:

$\mathrm{\&Delta;}{\mathrm{\&epsiv;}}_i={\mathrm{\&beta;}}_b\&CenterDot;\mathrm{\&Delta;}{\mathrm{\&epsiv;}}_i^N+(1-{\mathrm{\&beta;}}_b)\&CenterDot;\mathrm{\&Delta;}{\mathrm{\&epsiv;}}_i^O$

$\mathrm{\&Delta;}{a}_i={\mathrm{\&beta;}}_a\&CenterDot;\mathrm{\&Delta;}{a}_i^N+(1-{\mathrm{\&beta;}}_a)\&CenterDot;\mathrm{\&Delta;}{a}_i^O$

β wherein _b, β _aFor revising smoothing factor, i=1,2 expression first area and second areas.

Δ ε _i ^NBe first area or the new change amount of second area radiation coefficient, Δ ε _i ^OFirst area or the old correction of second area radiation coefficient.

Δ a _i ^NBe first area or the new change amount of second area water-cooled convection coefficient, Δ a _i ^OFirst area or the old correction of second area water-cooled convection coefficient.

Embodiment two,

See also shown in Figure 2ly, the step of this embodiment specifically describes as follows:

At first, obtain the temperature distribution model parameter and the determined data of rolling mill practice of extracting slab from heating furnace out, as physical parameter, the slab speed of service and the water spraying mode etc. of slab thickness, slab.

Utilize the recursion analytic modell analytical model to obtain the Temperature Distribution of slab air cooling radiation before arriving out the squama case, obtain and relate to two parameters in the process, one is the running time of slab, one is radiation coefficient, they determine that because no matter be air cooling zone or water-cooled zone, its length is known, so slab is foreseeable in the running time of respective regions, radiation coefficient is that model parameter is known.

Obtain board briquette distribution and the pony roughing mill cooling water preceding board briquette distribution through air cooling of slab by the water-cooled convection current of De-scaling box.And judge whether and need before frame, spray water to slab, need cooling then obtain the board briquette distribution of the preceding water-cooled of process frame again and obtain the Temperature Distribution that slab contacts with roll through rolling deformation behind the water spray; Do not need spray cooling then directly to obtain the Temperature Distribution that slab contacts with roll through rolling deformation.

After this need water spray after judging frame again, and the board briquette of obtaining through water-cooled after the frame distributes, if last passage, also needing to obtain temperature measuring set position after the roughing distributes through the board briquette of air cooling radiation, if not last passage is then obtained the preceding board briquette through the air cooling radiation of pony roughing mill cooling water again and is distributed.

Need to judge whether the pendulum steel at last, if need the pendulum steel, before then entering water-cooled with last passage is starting point, obtain the pendulum steel time, this time is added to the preceding air cooling of last passage in the time, and recirculation after this turns back to obtains pony roughing mill cooling water preceding board briquette distribution and each later step through the air cooling radiation; If do not need to put steel, then finishing control process.

Embodiment three

The board briquette distributed model subregion that control method of the present invention is described in conjunction with Fig. 3 and Fig. 4 again process that radiation coefficient and water-cooled coefficient are revised below.

After obtaining the passage related data of just rolling steel, as obtain the actual water spray state of each passage frame, the conveyance time of each air cooling zone slab etc., adopt the described control step of second embodiment to obtain the board briquette distribution that last passage slab of frame R1 arrives the first area terminal point, as shown in Figure 4, i.e. the board briquette of temperature measuring set position distribution between frame R1 and frame R2.Utilize the roughing actual value, be retracted to the first area terminal point from heating furnace and obtain the board briquette distribution, the steel slab surface measured temperature that temperature measuring set between slab upper surface temperature and frame is obtained compares, and discontented podolite is practised condition and then entered second area, satisfies condition and then learns.Model parameter study divided for three steps, the one, deviation is decomposed, resolve into owing to radiation-induced deviation with because the deviation that water-cooled causes, the 2nd, respectively the radiation coefficient and the water-cooled coefficient of recursion analytic modell analytical model are revised, the 3rd, new model coefficient correction is carried out limiting value inspection and smoothing processing.The model parameter correction and the first area of second area shown in Figure 4 are identical, it is the starting point difference, the correction of second area model parameter is that the terminal point with the first area is a starting point, it is to adopt the actual value of just rolling steel that the board briquette of first area terminal point distributes, and adopt the just corrected model parameter in first area, utilize the recursion analytic modell analytical model to obtain the first area terminal point and obtain.

For instance, if the radiation coefficient of first area model gets 0.85, the radiation coefficient of second area gets 0.8, and obtaining step-length is slab handling time t/20, the coefficient of water-cooled up and down in high pressure descaling district is 5400 and 3600, and the coefficient of water-cooled up and down before and after the frame is 4650 and 3200.Model mean temperature and target temperature at roughing outlet temperature measuring instrument place are compared, obtain the pendulum steel time.After slab rolling is finished, utilize actual value, carry out the model parameter adaptive learning, model parameter to first area and second area, be that radiation coefficient and water-cooled coefficient are revised respectively, deviation decomposition coefficient β=N/ (M+N), wherein M and N are the radiation obtained and the influence coefficient of water-cooled, the perturbation coefficient of radiation coefficient is 0.2, and the perturbation coefficient of water-cooled coefficient is 1000.After this model parameter correction classification that obtains is preserved, and learning outcome is used on next piece slab of same size.

Claims (7)

1, a kind of in course of hot rolling roughed plate bloom temperature control method,

It is characterized in that,

This temperature-controlled process may further comprise the steps:

A at first, is a starting point with the Temperature Distribution of slab when heating furnace is extracted out, presses the mobile order and the running time of slab, adopts solution by recursion formula eutectoid temperature distributed model, obtains the Temperature Distribution of each point on the slab rolling operating path;

B, then, slab measurement point after roughing is compared by mean temperature and the roughing target temperature that step a is tried to achieve, if comparative result is greater than the permissible value of the difference of both temperature that set, be reference then, put steel, slab is carried out radiation cooling with last passage inlet temperature of roughing, if comparative result less than setting the temperature permissible value, then carries out normally rolling to slab;

C, again the slab upper surface temperature measured value of slab at measurement point compared with the slab upper surface temperature that adopts solution by recursion formula eutectoid temperature distributed model to try to achieve at this place, when both deviation occurs, model radiation coefficient and water-cooled coefficient subregion are carried out the self adaptation correction, reach minimum and correction result is used on the follow-up slab of same size until deviation.

2, as claimed in claim 1 in course of hot rolling roughed plate bloom temperature control method,

It is characterized in that,

In the described step a),

Described hot rolling rough rolling plate blank solution by recursion formula eutectoid temperature distributed model mainly is meant according to boundary condition determines upper surface temperature, underlaying surface temperature, the mean temperature of slab in rolling process of rough rolling.

3, as claimed in claim 2 in course of hot rolling roughed plate bloom temperature control method,

It is characterized in that,

Described upper surface temperature, underlaying surface temperature, mean temperature expression formula are respectively,

$\mathrm{\&theta;}(\frac{H}{2},t)=\underset{i=0}{\overset{4}{\mathrm{\&Sigma;}}}{(-1)}^i{A}_i\left(t\right)+\underset{j=1}{\overset{4}{\mathrm{\&Sigma;}}}{(-1)}^{j-1}{B}_j\left(t\right),$

$\mathrm{\&theta;}(-\frac{H}{2},t)=\underset{i=0}{\overset{4}{\mathrm{\&Sigma;}}}{(-1)}^i{A}_i\left(t\right)-\underset{j=1}{\overset{4}{\mathrm{\&Sigma;}}}{(-1)}^{j-1}{B}_j\left(t\right),$

θ _m(t)=A ₀(t), and

Above various model coefficient obtain by the following formula recursion,

$\left[\begin{array}{c}{A}_0\left(t\right)\\ A_1\left(t\right)\\ A_2\left(t\right)\\ A_3\left(t\right)\\ A_4\left(t\right)\\ B_1\left(t\right)\\ B_2\left(t\right)\\ B_3\left(t\right)\\ B_4\left(t\right)\end{array}\right]=\left[\begin{array}{ccccccccc}1& 0& 0& 0& 0& 0& 0& 0& 0\\ 0& m_1\left(t\right)& 0& 0& 0& 0& 0& 0& 0\\ 0& 0& m_2\left(t\right)& 0& 0& 0& 0& 0& 0\\ 0& 0& 0& m_3\left(t\right)& 0& 0& 0& 0& 0\\ 0& 0& 0& 0& m_4\left(t\right)& 0& 0& 0& 0\\ 0& 0& 0& 0& 0& m_1^{*}\left(t\right)& 0& 0& 0\\ 0& 0& 0& 0& 0& 0& m_2^{*}\left(t\right)& 0& 0\\ 0& 0& 0& 0& 0& 0& 0& m_3^{*}\left(t\right)& 0\\ 0& 0& 0& 0& 0& 0& 0& 0& m_4^{*}\left(t\right)\end{array}\right]\left[\begin{array}{c}{A}_0\left(0\right)\\ A_1\left(0\right)\\ A_2\left(0\right)\\ A_3\left(0\right)\\ A_4\left(0\right)\\ B_1\left(0\right)\\ B_2\left(0\right)\\ B_3\left(0\right)\\ B_4\left(0\right)\end{array}\right]$

$+\left[\begin{array}{c}{a}^2\&CenterDot;({\mathrm{\&phi;}}_S+{\mathrm{\&phi;}}_I)\&CenterDot;t/(\mathrm{\&lambda;}\&CenterDot;H)\\ p_1\&CenterDot;K\&CenterDot;(1-m_1\left(t\right))\\ p_2\&CenterDot;K\&CenterDot;(1-m_2\left(t\right))\\ p_3\&CenterDot;K\&CenterDot;(1-m_3\left(t\right))\\ p_4\&CenterDot;K\&CenterDot;(1-m_4\left(t\right))\\ q_1\&CenterDot;R\&CenterDot;(1-m_1^{*}\left(t\right))\\ q_2\&CenterDot;R\&CenterDot;(1-m_2^{*}\left(t\right))\\ q_3\&CenterDot;R\&CenterDot;(1-m_3^{*}\left(t\right))\\ q_4\&CenterDot;R\&CenterDot;(1-m_4^{*}\left(t\right))\end{array}\right]$

In the formula,

λ is the slab pyroconductivity, and ρ is a slab proportion, c _pBe slab specific heat, the three is the physical parameter that is heated slab, is given in the reality, A _i(0), B _i(0) be the initial parameter that model calculates, initial value is provided by the heating furnace model, and follow-up parameter recursion obtains,

$a^2=\frac{\mathrm{\&lambda;}}{\mathrm{\&rho;}\&CenterDot;c_p},$ Be the slab thermal diffusion coefficient,

$m_i\left(t\right)=\exp \{-\frac{4\&CenterDot;i^2\&CenterDot;a^2\&CenterDot;{\mathrm{\&pi;}}^2\&CenterDot;t}{H^2}\}$ (i＝1～4)，

$m_j^{*}\left(t\right)=\exp \{-\frac{{(2j-1)}^2\&CenterDot;a^2\&CenterDot;{\mathrm{\&pi;}}^2\&CenterDot;t}{H^2}\}$ (j＝1～4)，

T is the calculating step-length,

H is a slab thickness,

$K=\frac{({\mathrm{\&phi;}}_S+{\mathrm{\&phi;}}_I)\&CenterDot;H}{2\&CenterDot;\mathrm{\&lambda;}},R=\frac{({\mathrm{\&phi;}}_S-{\mathrm{\&phi;}}_I)\&CenterDot;H}{4\&CenterDot;\mathrm{\&lambda;}}$

φ _S, φ _IBe the hot-fluid of slab upper and lower surface, determine by boundary condition.

4, as claimed in claim 1 in course of hot rolling roughed plate bloom temperature control method,

It is characterized in that:

Among the described step b, when putting steel,

Pendulum steel bar spare is Δ θ＞T _Const, T wherein _ConstBe the constant of process stipulation,

Δ θ=θ _m-θ _Target, θ _mBe the slab mean temperature that model calculates, θ _TargetIt is the given target temperature of technology;

The pendulum steel time is Δ t=(Δ θ-T _Const/ 2) * η _t,

${\mathrm{\&eta;}}_t=\frac{\mathrm{\&Delta;}{t}_a}{{\mathrm{\&Delta;\&theta;}}_{\mathrm{out}}}$

Δ t _aBe slab radiated time perturbation amount before last passage,

Δ θ _OutBe slab in roughing outlet temperature measuring position, the slab mean temperature is with respect to Δ t _aVariable quantity.

5, as claimed in claim 1 in course of hot rolling roughed plate bloom temperature control method,

It is characterized in that:

Among the described step c, to the correction of solution by recursion formula eutectoid temperature distributed model parameter the time, mainly be to utilize slab to resolve the slab upper surface temperature of distributed model with the board briquette recursion respectively to compare, model radiation coefficient and water-cooled coefficient subregion are carried out the self adaptation correction respectively at this place at the slab upper surface measured temperature of measurement point between pony roughing mill and after the roughing.

6, as claimed in claim 5 in course of hot rolling roughed plate bloom temperature control method,

It is characterized in that:

Resolve the slab upper surface temperature value of distributed model and compare when revising and further comprise the steps: carrying out between pony roughing mill measured temperature and board briquette recursion at this place

C11 carries out deviation and decomposes;

C12 obtains the correction of radiation coefficient and water-cooled coefficient respectively;

C13, limiting value and smoothing processing;

C14, the board briquette that utilizes the roughing actual value to obtain first area terminal point measuring instrument position after the roughing distributes.

7, as claimed in claim 5 in course of hot rolling roughed plate bloom temperature control method,

It is characterized in that:

Resolve the slab upper surface temperature value of distributed model and compare when revising and further comprise the steps: carrying out after the roughing measured temperature and board briquette recursion at this place

C21 carries out deviation and decomposes;

C22 obtains the correction of radiation coefficient and water-cooled coefficient respectively;

C23, limiting value and smoothing processing.

Images