JPH0262326B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a coiling temperature control method for controlling the coiling temperature and cooling rate of a hot rolled steel sheet to desired values.

Conventionally, the coiling temperature control method for hot-rolled steel sheets detects the plate thickness, temperature, and coiling temperature at the exit side of the finishing rolling mill, and controls the cooling amount in the upper and lower cooling zones to the amount necessary to obtain the target coiling temperature. This was done by manipulating only the However, with conventional methods, it is difficult to control the cooling rate, which has recently become important in order to reduce variations in the strength of steel sheets.

The present invention has been made in consideration of the current situation, and uses only a speed detector, finishing rolling mill exit side plate thickness gauge, finishing rolling machine exit side thermometer, and winding thermometer as detectors. To provide a method for controlling the cooling rate by estimating the temperature of a hot rolled steel sheet at several points in the cooling zone using a Kalman filter, and at the same time accurately controlling the coiling temperature by using multivariable control theory. It is something.

The present invention will be explained in detail below.

Figure 1 shows the geometrical arrangement of a general winding temperature control system, where 1 is a plate thickness gauge on the exit side of a finishing rolling mill, and 2
3 is the finish rolling mill exit side thermometer, 3 is the winding thermometer, 4,
5 is an upper and lower cooling zone (these cooling bodies are divided into N zones), 6 is a final finishing mill,
7 is a speed detector, 8 is a coiler, and 9 is a hot-rolled steel plate flowing from the finishing mill to the coiler. The speed of the hot rolled steel plate 9 rolled by the finishing mill 6 is measured by the speed detector 7, and the temperature at the exit side of the finishing rolling mill is measured by the thermometer 2.
The plate thickness is measured with a thermometer 1, and the coiling temperature is measured with a thermometer 3. Using these measured values, target winding temperature CT
The physical model that is the basis for calculating the amount of cooling required to ensure this is shown below.

ΔT(i)=T(i)/CpρH(i) 1/No {Nu(i)・αu(i)＋Nl(
i)・αl (i)} Δt(i)+2ε(i) {T(i)+273} ⁴ /CpρH(i)Δt(i) (1) ΔT(i): Temperature drop in the i-th zone [℃ ] T(i): Temperature of the steel plate at the entrance of the 1st zone [℃] Nu(i), Nl(i): Number of open upper and lower headers in the i-th zone αu(i), αl(i): i-th Upper and lower heat transfer coefficients in the zone [Kcal/m ² hr℃] σ: Stefan-Poltzmann constant = 4.88×10 ^-8
[Kcal/m ² hr〓] H(i): Steel plate thickness in the i-th zone [m] Δt(i): Passage time in the i-th zone [hr] ρ: Density of the hot-rolled steel sheet [Kg/m ³ ] ε (i): Emissivity in the i-th zone [Kcal/m ³ hr
〓 ⁴ 〕 Cp: Specific heat of hot steel plate [Kcal/Kg℃] No: Total number of headers to the top and bottom of the i-th zone Equation (1) shows the amount of temperature drop in the i-th zone. Here, the number of open upper and lower headers Nu(i) and Nl(i) in the i-th zone becomes the control operation amount. In order to apply multivariable control theory to the winding temperature control system based on the physical model shown in equation (1), an equation of state is derived from equation (1).

If the subscript k is an arbitrary sampling time, then from equation (1), T _k+1 (i+1) = {1−α ^* k(i)Δt _k (i)/CpρH _k (i)}
T _k (i)−2ε(i)σ{T _k (i)+273} ⁴ /CpρH _k (i)Δt _k (i)(
2) However, α ^* _k (i) = 1/No {Nu _k (i) αu _k (i) + Nl _k (i) αl _k (i)} (3) Now, if we take the deviation of equation (2), we get , that is, considering the deviation from the operating point determined by equation (2), dT _k+1 (i+1)=∂T _k+1 (i+1)/∂H _k (i)dH _k (i)+
∂T _k+1 (i+1)/∂α ^* _k (i)dα ^* _k (i) +∂T _k+1 (i+1)/∂Δt _k (i)dΔt _k (i)+∂T _k+1 (
i+1)/∂T _k (i)dT _k (i)(4) The differential coefficient of each deviation in equation (4) is as follows according to equation (2).

∂T _k+1 (i+1)/∂T _k (i)={1−α ^* k(i)/CpρH _k (
i)Δt _k (i)}−8ε(i)σΔt _k (i)／CpρH _k (i){T _k (i)+27
3} ³ (5) (=a _i defined) ∂T _k+1 (i+1)/∂α ^* _k (i)=−Δt _k (i)T _k (i)/CpρH
_k (i)(6) (=b _i is defined) ∂T _k+1 (i+1)/∂Δt _k (i)=α ^* _k (i)/CpρH _k (i)T _k (
i)−2ε(i)σ{T _k (i)+273} ⁴ /CpρH _k (i)(7) (=c _i is defined) ∂T _k+1 (i+1)/∂H _k (i)= {α ^* _k (i)Δt _k (i)／CpρT
_k (i)＋2ε(i)σ{T _k (i)273} ₄ Δt _k (i)／Cpρ}1/H _k (i
) ₂ (8) (Define = e _i ) dT _k (i) (i≠1) is the state, dα ^* _k (i) is input, dT _k
By selecting (1) and dH _k (i) as disturbances and collecting equation (4) for all zones and displaying it in a matrix, the equation of state shown by equation (9) is obtained.

X(k+1)=A(k)X(k)+B(k)U(k)+E(k)W(k)(9
) Here, A(k), B(k), and E(k) represent the influence coefficients (variation coefficients) of the state, input, and disturbance, respectively, shown in equation (10).

X(k)=[dT(2)……dT(N+1)] ^T (11) U(k)=[dα ^* (1)dα ^* (N)] ^T (12) W(k)=[dT( 1)dH(1)……dH(N)dΔt(1)……dΔt
(N)] ^T (13) In other words, qualitatively, X(k) is the variable to be controlled (state variable vector); U(k) is the control criterion (control variable vector), and W(k) is the observable Variables (disturbance variable vectors) A and B represent respective coefficient matrices.

In the present invention, the evaluation function J=Σ(X ^T (k)QX(k)+ ^UT (k) RU(k)) ...(14) However, Q and R are positive definite matrices, that is, the sum of squares of the deviation from the target temperature in each zone (the first term of equation (14)) and the equipment constraints on cooling capacity. The sum of squares of the deviation from the corresponding reference cooling amount ((14)
Basically, J=Σ{(q ₂・dT _k ² (2)+q ₃ dT _k ² (3)+…+q _N+1・dT _k (
N+1)) +(r ₁・dα _k ^*2 (1)+r ₂・dα _k ^*2 (2)+…+r _N・dα _k ^*2
(N)...(14') The evaluation function (the weighting coefficients consisting of qi and ri are determined from the balance between the strictness of the required temperature accuracy and the margin of cooling capacity in each zone) is Find the optimal control solution to minimize U ^* (k) U ^* (k) = −KX(k) − K′W(k) ……(15) and convert it into the required number of operation headers N _U and N _l By controlling the upper and lower headers,
In other words, it is possible to control the temperature even in the middle of the steel sheet cooling zone.
The winding temperature is controlled so that the cooling rate can be controlled in accordance with a desired temperature pattern.

In addition, the temperature of the steel plate in the middle of the cooling zone is estimated using a Kalman filter, and its characteristics will be described.

The temperature drop from the finish rolling mill outlet temperature FT to the coiling temperature CT is calculated by modifying equation (1), ΔT(i)=T(i)Nu(i)Δt(i)/CpρH(i)Noα _u ( i)＋T(i)N
lΔ(i)Δt(i)／CpρH(i)Noαl(i)+2σ(T(i)+273) ⁴
Δt(i)/CpρH(i)ε(i) (16) i=1, 2, ..., N (N: number of zones of cooling body)
Therefore, FT−CT=[f _i | g _i | h _i ] [αu(1)……αu(N) | αl(1
)...α _l (N) | ε(1)...ε(N)] ^T (17) f _i =T(i)Nu(i)Δt(i)／CpρH(i)N _p , g _i = T(i)Nl(i)Δ
t(i)/CpρH(i)N _p h _i =2σ(T(i)+273) ⁴ Δt(i)/CpρH(i) i=1, 2...N. Therefore, considering equation (17) as the observation equation Y(k)=M(k)X'(k) (18) k=1, 2, 3,..., the transition equation is k) (19) Assuming that k = 1, 2, 3, ..., estimate X'(k), that is, αu, αl, and ε for each zone, using a Kalman filter, and use equation (1) to estimate the Estimate the steel plate temperature.

Furthermore, the control logic according to the invention is represented in flowchart form in FIG. In FIG. 2, start to end is repeated at each sample time.

FIG. 3 shows an outline of a winding temperature control device that performs the control method according to the present invention.

In FIG. 3, 10 and 11 are A/D converters,
14 is a winding temperature control calculator. In particular, 12 is a calculation unit that outputs the optimal control input U ^* (k) to the upper and lower cooling zones 4 and 5 from the input described later in the procedure described below, and 13 is a Kalman filter that calculates the steel plate temperature in the middle of the cooling zone from the input described later. This is an arithmetic unit that estimates by

Control computer 14, especially optimal solution calculation unit 12
The above-mentioned equations (1) to (15) are programmed, and each physical property value in equation (1) is set as a constant.

The plate thickness detected by the plate thickness meter 1, the temperature detected by the thermometer 2, and the speed detected by the speed detector 7 are converted into digital signals by the A/D converter 10, and the optimal solution calculation calculation unit 12 and is input to the Kalman filter calculation unit 13. Further, the signal detected by the thermometer 3 is converted into a digital signal by the A/D converter 11 and input to the Kalman filter calculation section 13. In the Kalman filter calculation section, the above-mentioned equations (16) to (17) and the Kalman filter algorithm are programmed.
Using the above input and using a Kalman filter, the steel plate temperature T(i) in the middle of the cooling zone in equation (1) is estimated, and the deviation value from the operating point determined from equation (2) is calculated.
dT(i) (i≠1) is obtained and output to the optimum solution calculation calculation unit 12.

The optimum solution calculation calculation unit 12 calculates the steel plate temperature deviation dT(i) in the middle of the cooling zone, the plate thickness deviation value (dH(1)...dH(N)) obtained from the above input group from the actual process, and the temperature deviation amount ( dT(1)), speed deviation amount (dΔt(1),...dΔt
(N)), according to the state equation of equation (9),
Optimal control variable vector (standard cooling amount deviation of upper and lower cooling zones: dα ^* (1)...
...dα ^* (N)) is calculated and added to the standard cooling amount of the upper and lower cooling zones required in advance, and the required operational cooling amount is output to the upper and lower cooling zones 4 and 5. Based on this required operational cooling amount, the upper and lower cooling amounts are determined by the required operational header numbers Nu and _Nl , and the steel plate 9 is cooled.

Example Shumireshin conditions ●Cooling zone length = 80m ●Average heat transfer coefficient = 2000Kcal/m ² hr℃ ●Reference temperature after finishing = 910℃ ●Plate thickness = 2.5mm ●Target winding temperature = 700℃ (However, top part U Pattern inclination = 3℃/m) ● Sketch mark: Frequency = 0.3Hz Plate thickness variation = ±50μm Temperature variation = ±60℃ ●Threading speed = 750mpm ●Top speed = 1000mpm ●Acceleration rate = 25mpm/SEC ●Zooming start = coil length 120m section ●Zooming end = Coil length 400m section Figure 4 shows the finished exit side temperature deviation dT(1) = 60℃, finished exit side plate thickness deviation dH(1) = 50μm in the example device, corresponding to the skid marks. The diagram shows the change in the winding temperature deviation when changing at a cycle of 0.3 Hz, where the solid line shows the case of the method of the present invention and the dashed line shows the case of the conventional method. As is clear from this, in the method of the present invention, not only is it possible to minimize the coiling temperature deviation with less influence from disturbances, but also the temperature of the hot rolled steel sheet at several points in the cooling zone is monitored as a state vector, so cooling Speed control is also possible.

As described above, according to the method of the present invention using multi-variable control theory, the target coiling temperature can be optimally controlled, and the steel plate temperature during cooling can also be monitored, so strict cooling rate control is also possible. This method greatly contributes to quality improvement, such as improving coiling temperature accuracy and reducing variations in the mechanical strength of steel sheets.

[Brief explanation of drawings]

Fig. 1 is a geometric layout diagram of a general winding temperature control system, Fig. 2 is a flowchart for explaining the present invention, and Fig. 3 is a schematic explanatory diagram of a winding temperature control device implementing the present invention. FIG. 4 is a drawing showing a comparison between the winding temperature control method according to the present invention and the conventional method. 1...Finishing mill exit plate thickness gauge, 2...Finishing rolling machine exit side thermometer, 3...Coiling thermometer, 4...Upper cooling zone, 5...Lower cooling zone, 6...Finish final rolling machine, 7... speed detector, 8... coiler, 9...
Hot rolled steel plate, 10...A/D converter, 11...A/
D converter, 12... Optimal solution calculation calculation unit, 13...
Kalman filter calculation section, 14... winding temperature control calculator.

Claims (1)

[Scope of Claims] 1. The cooling zone is composed of a plurality of zones, and each zone is equipped with cooling equipment consisting of a plurality of individually operable cooling headers, a finish rolling mill outlet plate thickness gauge, a finish rolling mill outlet thermometer, In a coiling temperature control method for a hot rolling mill that is equipped with a coiling thermometer to control the coiling temperature of a hot rolled steel sheet to a desired value, the temperature of the hot rolled steel sheet at several points in the middle of the cooling zone is measured. Deviation dT(2), dT(3), ...dT
(N) to the state variable vector X(k), and the reference cooling amount deviation dα ^* (1), dα ^* (2)...
dα ^* (N) is the control variable vector U(k), the speed deviation of the steel plate dΔt(1)...dΔt(N), the plate thickness deviation at the exit side of the final rolling mill dH(1)...dH(N) , and the finishing rolling mill outlet temperature deviation dT(1) as the disturbance (observable) variable vector W(k), the state equation of the winding temperature control system X(k
+1) = A(k)X(k)+B(k)U(k)+E(k)W(k) and the above state variable vector X(k), control variable vector U(k)
The evaluation function J= expressed as the time integral value of the quadratic form of
〓 Write ^k=i (X ^T (k)QX(k)+U ^T (k)RU(k)), estimate the state variable vector value X(k) using a Kalman filter, and minimize the above evaluation function J. Find the value of the control variable vector U ^* (k) = −KX ⁽ k) − K′W(k) to
A method for controlling the coiling temperature of a hot-rolled steel sheet, characterized in that the cooling amount of the upper and lower cooling zones is controlled by adding dα ^* (2)...dα ^* (N) to the standard cooling amount.