JPH05209234A - In-furnace temperature control device of heating furnace - Google Patents

Abstract

PURPOSE:To improve the temperature control precision of an object to be heated by executing estimation operation of the specified fuel flow rate at the respective timepoints in future which is required to heat the object to be heated in the specified temperature rise pattern on the basis of a formula of the heat balance model, thereby setting this fuel flow rate for the heating furnace. CONSTITUTION:An in-furnace temperature control device is provided with a detecting means to detect the values of the heating value-influencing parameters (Fj, Arj, Twj, etc.), of the respective operating formulas I-V to operate the combustion heating value (Qhj), the preheated air heating value (Qaj), the furnace heat radiation value (Qwj), and the combustion gas carrying-out heating value (Qgj) of the formula of the heat balance model Q1=Qgj-1+Qhj+Qaj-Qsj-Qwj-Qgj=0 at the respective timepoints (i=1-n). The detected values at the respective timepoints (i=1-2) are introduced in the respective formulas I-V by this means, the coefficients of the respective formulas IV are calculated, and the formula of the heat balance model is updated to that employing the calculated coefficient values. This method allows the temperature estimation precision of the in-furnace material to be improved and the temperature control precision to be also improved.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to the temperature control of a heating furnace, and more particularly, it is required at each future time required to bake a material to be heated to a target temperature with a required temperature raising pattern based on a thermal equilibrium model equation. The present invention relates to a furnace temperature control for predicting and calculating a fuel flow rate and setting the fuel flow rate in a heating furnace at each future time.

[0002]

2. Description of the Related Art The temperature of a heating furnace is determined by the heat balance, that is, the amount of heat input to and the amount of heat released from it. Japanese Patent Laid-Open No. 2-1
In Japanese Patent No. 56017, the furnace temperature and the temperature of the material (material to be heated) currently in the furnace are estimated and calculated by using a thermal equilibrium model expression representing the relationship between the heat input amount, the temperature, and the heat radiation amount of the heating furnace. In addition, the material temperature at each future time is estimated and calculated while changing the operating conditions at each time, and the operating conditions that change in time series until baking to the target temperature are converted into evaluation values using a predetermined evaluation function. The operating condition at each time is changed so that the value becomes smaller, and the material temperature transition that minimizes the evaluation value, that is, the operating condition in time series is set. The thermal equilibrium model formula includes a parameter that is a linear heat loss coefficient, and the actual fuel flow rate and the actual in-reactor temperature are substituted into the thermal equilibrium model formula to calculate the value of the parameter, and the parameter of the thermal equilibrium model formula is calculated. -T is set to the calculated value, and the thermal equilibrium model equation updated in this way is used for calculating the furnace temperature. The thermal equilibrium model equation updated in this way follows a certain degree of change in the furnace state, and therefore has a certain effect.

FIG. 1A shows a schematic plan view of one heating furnace, and FIG. 1B shows a schematic vertical cross section. In the combustion control system of such a heating furnace, a thermal equilibrium model formula is used to estimate the temperature inside the heating furnace, especially the temperature inside the furnace. To improve the accuracy of the model, A learning method that automatically corrects model parameters following changes in conditions is effective. If the heating furnace is divided into several control regions and thermal equilibrium is taken into consideration in each region, and the temporal unsteady term is small and ignored, the following relational expression holds for the combustion gas. In the following, j indicates a belt, for example, j = 1 means furnace bottom, j = 2 means pre-tropical zone, ..., j = 5 means soaking zone.

[0004] _{Q j = Q gj-1 +} Q hj + Q aj -Q sj -Q wj -Q gj -Q Lj = 0 ··· (1) However, the unit is (Kcal / h).

Q _gj-1 : heat quantity of combustion gas brought in Q _gj-1 = μ _gj-1 · C _gj-1 · V _j-1 · Θ _gj-1 (2) C _gj-1 : gas heat capacity ( Kcal / ° C ・ Nm ³ ) V _j-1 : Gas flow rate (Nm ³ / h) flowing from the region ( _j -1) Θ _gj-1 : Gas temperature (° C) in the region ( _j -1) μ _{gj- 1} : Coefficient Q _hj : Combustion calorific value Q _hj = μ _hj · h _j · F _j (3) h _j : Heat generation rate (Kcal / Nm ³ ) F _j : Fuel flow rate (Nm ³ / h) μ _hj : Coefficient Q _aj : Preheated air carry-in heat quantity Q _aj = μ _aj · C _aj · F _aj · Θ _aj・ ・ ・ (4) C _aj : Air heat capacity (Kcal / ° C · Nm ³ ) F _aj = Ar _j · F _j・ ・ ・ (5) Ar _j : Air-fuel ratio Θ _aj : Air temperature (℃) Θ _aj = μ _rj・ Θ _gj・ V _j・ ・ ・ (6) μ _aj : Coefficient μ _rj : Coefficient Q _sj : To material Transfer heat Q _sj = μ _sj · S _j · {Φ _gsuj · (T _gj ⁴ −T _suj ⁴ ) + Φ _gslj · (T _gj ⁴ −T _slj ⁴ )} (7) S _j : surface area of material , Φ: Overall heat absorption rate, T Temperature (° K) μ _sj: coefficients respectively lowercase, g is the gas, s materials, u is an upper surface, l
Means the lower surface.

[0006] Q _wj: furnace wall brought out the amount of heat _{Q wj = μ wj · S wj} · Φ gwj · (T gj 4 -T wj 4) ··· (8) S w: the area of the furnace wall (m ²⁾ μ _wj : Coefficient, lower case w means furnace wall.

Q _gj : heat quantity taken out of combustion gas Q _gj = μ _gj · C _gj · V _j · Θ _gj (9) V _j = G _j · ΣF _kj (10) G _j : generation of combustion gas Rate μ _gj : Coefficient Q _Lj : Heat loss and correction term Q _Lj = α ₁ · V _j + α ₀ (11) α ₁ : Coefficient, α ₀ : Constant Now, in the conventional method as described above, each addition heat balance model i.e. the on tropical formula _{(1), f a (T, F)} + α is expressed as _{1 · F + α 0 = 0} ··· (1a). T is the furnace temperature, F is the fuel flow rate (flow velocity),
α ₁ and α ₀ are model parameters, that is, learning parameters.
It is When the fuel flow rate V _ei (target value) is determined based on the above equation (1) for each time point i = 1 to n and the furnace temperature is estimated to be T _ei and the furnace temperature is controlled, that is, heating is performed at each time point. _Assuming that the furnace temperature is T _i (measured value) and the fuel flow rate is F _i (measured value) when fuel is supplied to the furnace with F _ei as a target value, the evaluation function P is Here, the first term on the right-hand side is an estimation operation expression, the second term on the right-hand side is an actual operation operation expression, and the estimation operation expression {f _a (T _ei ,
Since F _ei ) + α ₁ · F _ei + α ₀ } is zero from the equation (1), And α ₁ and α ₀ that minimize the evaluation function P.
Ask for.

[0008]

It is difficult to directly measure the material temperature in the heating furnace due to the influence of various disturbances. Generally, instead of actually measuring the material temperature, the above-mentioned estimation calculation is performed. The furnace temperature is calculated, and the material temperature is estimated and calculated based on this. In the past, only the input fuel flow rate (actual fuel flow rate value) and each zone temperature in the reactor were measured, and the others were calculated using model formulas, so the number of parameters that should be corrected should be limited and the actual process Cannot be reproduced sufficiently, the error of the thermal equilibrium model formula becomes large. Furthermore,
When updating the model coefficient α (α ₁ , α ₀ ) based on the actual value every predetermined time as in JP-A-2-254123, the physical meaning of the model coefficient is unknown and the operating conditions are compared. Since it is difficult to perform adjustment work when there is a large change in efficiency, versatility is low.

For example, in this case, the determination of α ₁ and α ₀ that minimizes the evaluation function P in the equation (12) is a linear regression calculation. However, since the in-furnace temperature T is a function of α ₁ and α ₀ and has non-linearity, α ₁ and α
The estimation error of ₀ becomes large, so that the estimation accuracy of the furnace temperature does not improve. Further, since the thermal equilibrium model formula itself has an error, the estimated temperature itself does not always match the actual temperature.

It is an object of the present invention to improve the accuracy of temperature estimation in the furnace and improve the accuracy of temperature control of the material to be heated.

[0011]

According to the present invention, in order to improve the above-mentioned problems, the thermal equilibrium model equation Q _j = Q _gj-1 + Q _hj + Q _aj -Q _sj -Q _wj -Q _gj = 0.・ Combustion heat generation amount (Q _hi ), preheat air heat amount (Q _aj ), furnace heat radiation amount of (13 _j )
(Q _wj ) and each calculation formula for calculating the amount of heat taken out of combustion gas (Q _gj ), Q _gj-1 = μ _gj-1 , C _gj-1 , G _j-1 , ΣF _kj-1 , Θ _gj-1・ ・ ・ (9 _j-1 ) Q _hj = μ _hj・ h _j・ F _j・ ・ ・ (3 _j ) Q _aj = μ _aj・ C _aj・ Ar _j・ F _j・ Θ _aj・ ・ ・ (4 _j ) Θ _aj = μ _rj・ Θ _gj・ V _j・ ・ ・ (6 _j ) Q _sj = μ _sj・ S _j・ {Φ _gsuj・ (T _gj ⁴ −T _suj ⁴ ) + Φ _gslj・ (T _gj ⁴ −T _slj ⁴ )} ・ ・ ・ (7 _j ) Q _wj = μ _wj・ S _wj・ Φ _gwj・ (T _gj ⁴ −T _wj ⁴ ) ・ ・ ・ (8 _j ) Q _gj = μ _gj・ C _gj・ G _j · ΣF _kj · Θ _gj heat effect of ··· (9 _j) parameters - data _{(F j, Ar j, Θ} aj, Θ gj,
T _wj , Θ _gj ) is provided with detection means (21 to 24, 31 to 35, etc.) for detecting the value of each time (i = 1 to n), and each calculation of the thermal equilibrium model formula (13 _j ) is performed. In the formula (3 _j , 4 _j , 6 _j , 8 _j , 9 _j ), each time (i = 1
~ N) is introduced, K _5j-1 = C _gj-1 · G _j-1 · ΣF _kj-1 · Θ _gj-1 , K _1j = h _j · F _j , K _2j = C _{aj · Ar j · F j · Θ} gj · V j, K 3j = S j · {Φ gsuj · (T gj 4 -T suj 4) + Φ gslj · (T gj 4 -T slj 4)}, K 4j = S _wj · Φ _gwj · (T _gj ⁴ −T _wj ⁴ ), K _5j = C _gj · G _j · ΣF _kj · Θ _gj are calculated, and these are substituted into the thermal equilibrium model formula (13 _j ). Time coefficient matrix μ _gj-1・ K _5j-1 ＋ μ _hj・ K _1j ＋ μ _aj・ μ _rj・ K _2j −μ _sj・ K _3j −μ _wj・ K _4j −μ _gj・ K _5j = 0 ( 14 _j ) (n pieces) based on the coefficient values (μ _hj , μ _aj・ μ)
_rj , μ _wj , μ _gj ) are calculated, and the thermal equilibrium model formula (13 _j ) is updated to one using the calculated coefficient value. That is, the coefficient values (μ _hj , μ _aj ) of the respective calculation formulas for calculating the combustion calorific value (Q _hi ), the preheating air heat value (Q _aj ), the furnace heat _release value (Q _wj ), and the combustion gas carry-out heat value (Q _gj ).・_Update μ _rj , μ _wj , μ _gj ) by learning.

The amount of combustion gas carried in μ _gj-1 · K _5j-1 =
The actual calculated value of Q _gj-1 is forward (upstream in the material feed direction)
This is the actual calculation value of the combustion gas carry-out heat quantity of the side zone, and this is the actual calculation value of the combustion gas carry-out heat quantity Q _gj-1 of the front zone. The actual calculated value of the amount of heat transferred to the material μ _sj · K _3j = Q _sj can be obtained based on equation (7) by measuring the surface temperature and back surface temperature of the heating target material in the j band. Therefore, the actual calculated value is used in the zone where the front surface temperature and the back surface temperature of the heating target material can be easily measured. In this case, μ _{sj will} also be updated by learning. When such temperature measurement is difficult, the value Q _sj calculated in the intermediate process of the temperature estimation calculation of the heating target material at each time (i = 1 to n) is used. In this case μ _sj is not updated.

[0013]

According to the present invention, at least the combustion heating value of the thermal equilibrium model equation _{(13 j) (Q hi)} , preheated air heat (Q _aj), the furnace heat radiation amount (Q
_wj ) and the amount of heat taken out of combustion gas (Q _gj ), the coefficient values (μ _hj , μ _aj · μ _rj , μ _wj ,
Since μ _gj ) is updated by learning, the actual number of parameters for correcting the thermal equilibrium model equation is larger than in the past, and the error in the thermal equilibrium model equation is smaller, and the reproducibility of the actual process is high.
Therefore, there is little adjustment work when the operating conditions change relatively large, and versatility is high. As a result, the accuracy of estimating the temperature inside the furnace is increased, and the accuracy of temperature control of the material to be heated is improved.

Other objects and features of the present invention will become apparent from the following description of embodiments with reference to the drawings.

[0015]

DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 shows the outline of one embodiment of the present invention. The heating furnace 1 has a furnace bottom 1 as shown in FIG.
1, a pre-tropical zone 12, a second heating zone 13, a third heating zone 14 and a soaking zone 15. As shown in FIG. 1 (b) showing a longitudinal section, in each zone, in-furnace temperature (Θg) detectors 21 to 24 are provided.
In addition to the furnace wall temperature (Tw) detectors 31 to 35 and the flow rate detectors that detect the actual fuel flow rate values F of each zone, V (combustion gas flow rate), Ar (air-fuel ratio), Θa (air) of each zone temperature)
Equipped with various detectors for measuring etc. The fuel flow rate control device 4 is provided in each of the pretropical zone 12 to the soaking zone 15
0 supplies fuel for each belt. The schedule calculator 51 of the optimal control device 50 gives the fuel flow rate control device 40 a fuel flow rate target value for each band, and the fuel flow rate control device 40
The fuel flow rate of each band is controlled so that the detected value matches the target value by referring to the detected value of the fuel flow rate detector of each band. Measured values (F, V, Θg, Ar, Θa, Tw) of each zone measured by the furnace temperature detectors 21-24, the furnace wall temperature detectors 31-35 and other detectors are It is given to the schedule calculator 51.

The optimum controller 50 has an input / output device 52 for exchanging data with the fuel flow rate controller 40 and for reading the detected value, and an input data for storing the read data for a predetermined period. The data memory 53, the output data memory 54 for storing the output data for a predetermined period, the calculated value data memory 55 for storing the calculated data, and the material charging and extraction of the heating furnace 1. An input / output device 56 for exchanging data with the control computer 60 to be managed is provided.

FIG. 2 shows the outline of the control operation of the schedule computer 51 of the optimum control device 50, FIG. 3 shows the contents of “calculation of Fei” (2) shown in FIG. 2, and FIG. The contents of "Update of thermal equilibrium model formula" (10) shown in Fig. 2 are shown.

First, the outline of the control operation of the optimum control device 50 will be described with reference to FIG. At the beginning of starting the furnace temperature control of the heating furnace 1, the schedule computer 51 uses the "initial value setting" (1) to display the current heating furnace state information shown in Tables 1 and 2 from the control computer 60. In response to this, the data is written in the input data memory 53, and various measured values such as temperature measured values are read and written in the data memory 53.

[0019]

[Table 1]

[0020]

[Table 2]

The schedule computer 51 controls the temperature of each of the pretropical zone 12 to the soaking zone 15. However, only the temperature control of the soaking zone 15 (j = 5) will be described below. The temperature control of the other zones (j = 1 to 4) is the same as the temperature control of the soaking zone 15.

The schedule computer 51 next calculates the current time (i = 1), the time after dt1 (= 12 × 10 sec) (i = 2), and the time after dt1 (i = 2).
3), ... Calculate the fuel flow rate Fei that should be set in the soaking zone 15 at each of the end times (i = n) (step 2: hereinafter, the word step is omitted in parentheses),
These n pieces of data are given to the fuel flow rate control device 40 as the fuel flow rate target values at each time, and these data are output to the output data memory 54, and the calculation data required for the calculation process. Is written in the calculated value data memory 55 (3). The fuel flow rate control device 40 reads the target value data at each given time, sets the reference value at the first time (i = 1), and then changes the reference value to the elapsed time in conjunction with the passage of time. Corresponding ((i
= 2,3,4 ...) and controls the fuel flow rate of the soaking zone 15 so that the fuel flow rate (actual value) detected by the flow rate detector (not shown) matches the reference value. ..

While the control device 40 controls the fuel flow rate in this way, the schedule computer 51 has the furnace gas temperature Θg, the furnace wall inner surface temperature Twi, and the fuel flow rate actual values Fi, Vi (combustion) in a cycle of 10 seconds. Gas flow rate), Ari (air-fuel ratio), Θa (air temperature) are read and written in the input data memory 53 (5, 5).
6). When the measured value is read 12 times, the average value of each measured value is calculated and written in the i-time average value register (A9). By repeating this, each time (correctly 10 × 12 each) is stored in each of the i (= 1 to n) time average value registers.
When the average value of (sec period) is written, the coefficients (μ _hj , μ _aj · μ _rj , μ _wj , μ _gj ) of the thermal equilibrium model formula (13 _j ) are updated (1
0). Then, the latest in-furnace material information (Table 2) is received from the control computer 60 and the corresponding information in the input data memory is updated to this (11). And again, n × d from the current time
The fuel flow rate Fei to be set in the soaking zone 15 is calculated at each time i up to t1 (2), and these n pieces of data are set in the fuel flow rate control device 40 as the fuel flow rate target value at each time. Give (3). The same applies hereinafter.

Next, referring to FIG. 3, "calculation of Fei"
The contents of (2) will be described. Here, the schedule calculator 51 first calculates the fuel flow rate target value Fei (i = 1) at the current time (21 to 30).

In the calculation (21 to 30) of the fuel flow rate target value Fei (i = 1) at the current time, first Fei (i = 1) is calculated as the current fuel flow rate actual value (the latest value read in step 6) Feip Assuming a value Feie smaller than the set amount Fai (22), Q _gj-1 = μ _gj-1 · C _gj-1 · G _j-1 · ΣF _kj-1 · Θ _gj-1 ··· (9 _{j -1} ) Q _hj = μ _hj · h _j · F _j・ ・ ・ (3 _j ) Q _aj = μ _aj・ C _aj・ Ar _j・ F _j・ Θ _aj・ ・ ・ (4 _j ) θ _aj = μ _rj・ Θ _gj・ V _j・ ・ ・ (6 _j ) Q _wj = μ _wj・ S _wj・ Φ _gwj・ (T _gj ⁴ −T _wj ⁴ ) ・ ・ ・ (8 _j ) Q _gj = μ _gj・ C _gj・G _j · ΣF _kj · θ _gj (9 _j ) values Q _gj-1 , Q _hj , Q _aj , Q _wj , and Q _gj are calculated (23
A-23E). For Q _gj-1 , the value calculated for the third heating zone 14 (combustion gas carry-out heat amount of the third heating zone) is used. These values Q _gj-1 , Q _hj , Q _aj , Q _wj , and Q _gj are substituted into the thermal equilibrium model equation (13j) to calculate the amount of heat transferred to the material Q _sj (24).

The calculation of the temperature of the material and the temperature of the furnace wall is carried out by using the following approximate expression obtained by solving the one-dimensional unsteady heat conduction model using the weighted residual method.

T (t, x) = f ₀₀ + K _a · α · t + {(f ₀₁ −K _b ) · g ₁ (t) + K _b } · f ₁ (x) + {(f ₀₂ −K _{a / 2) · g 2 (t} ) + K a / 2} · f 2 (x) - (5/12) · (f 01 -K b) · g 1 (t) · f 3 (x) - (7 / _{8) · (f 02 -K a} / 2) · g 2 (t) · f 4 (x) ··· (15) _{However, g 1 (t) = exp} (- (5/2) · α · t ) (16) g ₂ (t) = exp (− (60/7) · α · t) (17) f ₁ (x) = x (18) f ₂ (x) ^{= x 2 -1/3 ··· (19)} f 3 (x) = x 3 - (3/5) · x ··· (20) f 4 (x) = x 4 - (6/7) · x ² +3/35 (21) where f ₀₀ , f ₀₁ , f ₀₂ , K _a , α and K _b are constants, the temperature T is the average temperature T = 1, and the position x is within the region. Becomes (-1 to +1) Sea urchin, has been standardized.
First, in the temperature calculation of the material, X is taken in the thickness direction of the material, the lower surface is X = −1, the center is X = 0, and the upper surface is X = + 1.

The boundary condition is

[0029]

[Equation 22]

[0030]

[Equation 23]

In this case, the upper and lower heat fluxes of the material in the furnace are q _suj = σ · Φ _gsuj · (T _gj ⁴ −T _suj ⁴ ) ... (24) q _slj = σ · Φ _gslj · (T _gj ⁴ −T _slj ⁴ ) ... (25). Where λ _{sj is} the thermal conductivity of the material (Kcal / hm ° C), Φ is the overall heat absorption coefficient, T is the temperature (° K), lowercase letters are g for gas, s for material, u for the upper surface, and l for the lower surface. Means

Next, in the temperature calculation of the furnace wall, X is taken in the thickness direction of the furnace wall, the inside of the furnace is X = -1, the center is X = 0, and the outer wall is X = +.

The boundary condition is

[0034]

[Equation 26]

[0035]

[Equation 27]

The heat flux in this case is q _wej = μ _wj · σ · Φ _gwj · (T _gj ⁴ −T _wej ⁴ ) ... (28) q _woj = μ _woj · β _oaj · (T _oaj − T _woj ) ... (29).

Where λ _{wi is} the thermal conductivity of the furnace wall (Kcal
/ Hm ° C), Φ: overall heat absorption coefficient of the inner wall of the furnace wall, β: heat transfer coefficient of the outer wall (Kcal / hm ² ° C), T is temperature (° K), lowercase letters are g for gas, w for furnace wall,
e means the inside of the furnace, and o means the outer wall.

From the equation (15), the internal temperature of the furnace wall is x =
When −1 and t = 0 are set, T _wej = B _1j · μ _woj + B _0j (30)

The difference between the temperature (estimated value) of each part of the material thus obtained and the target value, that is, the error (each part of each material) is calculated, and the "sum of errors" is calculated (26, 27 in FIG. 3). ).

Similarly to the above-described calculation of the "sum of errors" described above, the fuel flow rate assumption value Feie is similarly increased by one step (Δ) to Feip + Fai, and similarly for each step (each fuel flow rate assumption value). Execute (28 in FIG. 3,
29, 23A-23E-24-27). Then, the fuel flow rate hypothetical value for which the smallest one among the "sum of errors" of these steps is obtained is set as the fuel flow rate target value Fei (i = 1) of i = 1 (current time) (30). .. In this “calculation of F _e i” (2), the optimum value of the fuel flow rate may be calculated by using a known descent method in order to increase the calculation speed. For example, Feie = Feip-Fai, Feie = Feip-Fai
/ 2, Feie = Feip, Feie = Feip + Fai / 2 and Fe
The above "sum of errors" for the five points ie = Feip + Fai
Is the result of calculating
= Feip-Fai / 2, then Feie = Feip- (3 /
4) Calculate the "sum of errors" of Fai and Feie = Feip- (1/4) Fai, and the fuel flow rate that minimizes it is Feie
= Feip-Fai to Feie = Feip- (3/4) Fai region, F
eie = Feip- (3/4) Fai to Feie = Feip-Fai / 2 region, Feie = Feip-Fai / 2 to Feie = Feip- (1 /
4) Area of Fai and Feie = Feip- (1/4) Fai to Fei
It is determined which of the areas of e = Feip, the area in which the minimum value exists in this way is gradually narrowed, and finally the fuel flow rate at which the "total error" becomes the minimum value in the minimum unit of the fuel flow rate. Is determined to be the optimum value. When such a steep descent method is used, the calculation time for calculating the optimum fuel flow rate is shortened.

In the same manner as the determination of "i = 1 (current time) fuel flow rate target value Vei (i = 1)" described above, dt1 (= 1
After 0 × 12 sec) (i = 2), the fuel flow rate target value Vei (i = 2) to the fuel flow rate target value Vei (i = n) after n × dt1 are determined (31, 32, 22-). 30).

Next, the contents of the "update of thermal equilibrium model formula" (10) will be described with reference to FIG. Note that the "update of thermal equilibrium model equation" (10) is performed in step 2 before n × dt1.
(FIG. 2: Details are shown in FIG. 3) The results, that is, the actual values (measured values in steps 4 to 8 in FIG. 2) that are represented by the fuel flow rate target values Vei at the respective time points (i = 1 to n) given in FIG. The actual values of K _ij , K _2j , K _4j and K _{5j in} the equation (14j) are calculated by substituting them into the equation (13j), and these and the combustion gas carry-out calorific value actual calculated value Q _{gj- of} the third heating zone 14 are calculated. ₁ = μ _gj-1 · K
_5j−1 , and the calculated heat transfer amount to the material Q _s = μ _sj · K _3j calculated when the fuel flow rate target value Vei was calculated, at each time i = 1 to n (13
The actual value introduction formula of j), that is, the formula (14 _j ) is calculated to obtain n coefficient matrices (101 to 104). Then, the coefficient μ of the calculation formula (3 _j ) of the combustion calorific value Q _hi is calculated by multiple regression calculation from this coefficient matrix.
_hj , coefficient μ _aj of calculation formula (4 _j ) of preheated air heat quantity Q _ai , coefficient μ _wj of calculation formula (8 _j ) of furnace heat _release quantity Q _wi and calculation formula (9 _j ) of combustion gas carry-out heat quantity Q _gi Calculate the coefficient μ _gj (10
5). Next, the coefficient used in the calculation of the fuel flow rate target value Vei and the weighted smoothing process are performed.
Thermal equilibrium model formula used to calculate the next fuel flow rate target value Vei
The coefficient value of (13j) is set (107). The coefficient μ _woj of the equation for calculating the inner surface temperature of the furnace wall T _wej = B _1j · μ _woj + B _0j (30) is similarly calculated and updated using this formula.

FIG. 5 shows changes in soaking temperature before and after the present invention is carried out. Before the implementation, the fluctuation of about 30 ° C remains, but after the implementation, the fluctuation is about 10 ° C.
It has become a degree. The operating condition at this time is the width of the material 76
0-1420mm, thickness 250mm, charging temperature 50-9
The temperature is 00 ° C and the extraction target temperature is 1100 to 1230 ° C.

[0044]

According to the present invention, the model error correction amount is not concentrated on a specific parameter of the thermal equilibrium model equation, and more accurate estimation can be performed. Even if there is a change in operation, many coefficient values in the thermal equilibrium model formula are automatically updated with high accuracy, so stable control can always be performed as compared with the conventional case. Therefore, it is possible to accurately estimate the optimum fuel for heating a wide variety of materials, and it is possible to secure the quality of baked materials and obtain a large energy saving effect.

[Brief description of drawings]

FIG. 1 is a block diagram showing one heating furnace, a heating furnace, and an embodiment of the present invention, in which (a) is a plan view of the heating furnace;
(B) is a longitudinal sectional view.

FIG. 2 is a schedule computer 51 shown in FIG. 1 (b).
3 is a flowchart showing the contents of the arithmetic processing of.

FIG. 3 is a schedule computer 51 shown in FIG. 1 (b).
2 is a flow chart showing the contents of the arithmetic processing of the above, and shows the contents of "calculation of _Vei " (2) shown in FIG.

FIG. 4 is a schedule computer 51 shown in FIG. 1 (b).
2 is a flow chart showing the contents of the calculation processing, and shows the contents of the "update of the heat balance model formula" (10) shown in FIG.

FIG. 5 is a graph showing a change in temperature in a soaking zone.

[Explanation of symbols]

1: Heating Furnace 11: Furnace Bottom 12: Pre-Tropical Zone 13: Second Heating Zone 14: Third Heating Zone 15: Soaking Zone 21-24: Furnace Temperature Detector (Detecting Means) 31-35: Furnace Wall Temperature Detector (Detection means) 40: Fuel flow rate control device (fuel flow rate control means) 50: Optimal control device 51: Schedule calculator (fuel flow rate calculation means, coefficient calculation means, update means)

Claims (1)

[Claims] 1. A thermal equilibrium model expression is used that represents the relationship between heat input, temperature, and heat release, including the combustion heat generation amount of fuel, preheat air heat amount, furnace heat release amount, combustion gas carry-out heat amount, and transfer heat amount to the material to be heated. Then, the amount of heat transferred to the material to be heated at each future time is estimated and calculated, and the temperature of the material to be heated is estimated and calculated based on this to calculate the fuel flow rate at each time required to bake the material to be heated to the target temperature. Fuel flow rate calculation means for
And a fuel flow rate control means for controlling the fuel flow rate of the heating furnace to the calculated fuel flow rate at each time, wherein the combustion heat generation amount, preheating air heat amount, furnace heat radiation amount and combustion Detecting means for detecting the value at each time of the calorific value influence parameter of each calculation formula for calculating each heat quantity taken out of gas; introducing the detected value at each time into each calculation formula of the heat balance model formula And coefficient updating means for calculating the coefficient value of each calculation formula based on the obtained coefficient matrix; and updating means for updating the thermal equilibrium model expression to one using the calculated coefficient value. A furnace temperature control device for a heating furnace.