JP2581832B2 - Temperature control method for continuous heating furnace - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a temperature control method for a continuous heating furnace.

(Prior art)

In the temperature control of a continuous heating furnace having a plurality of continuous control zones, the following techniques are conventionally known.

The required furnace temperature is determined for each slab in each zone, and the highest furnace temperature is controlled for each zone from the required furnace temperature group as a representative furnace temperature (JP-B-51-30526). .

In the heating furnace, the difference between the predicted temperature at the time of slab extraction and the target extraction temperature is included in the evaluation function, and the amount of change in the set furnace temperature that minimizes the evaluation function is determined by a linear programming method.
-25771).

In the heating furnace, a set furnace temperature is calculated according to a deviation between a predicted temperature of the slab and a predetermined heating pattern (JP-B-49-29403).

[Problems to be solved by the invention]

However, Japanese Patent Publication No. 51-30526 discloses a method in which only slabs that are difficult to be heated are heated, and other slabs are over-burned and fuel consumption is increased. Japanese Patent Publication No. 61-25771 discloses that in order to satisfy the manufacturing conditions of each slab, such as a target extraction temperature and a target extraction lower limit temperature, and to perform an energy-saving operation, a plurality of slabs are considered in an evaluation function. , It is necessary to have a plurality of conditions such as a target extraction temperature, a target extraction lower limit temperature, and an evaluation value of fuel consumption. In order to calculate the evaluation function by a linear programming method in a control cycle (two minutes), In addition to the enormous amount of calculation, there is a case where an optimal solution cannot be obtained due to a calculation error or the like, which is a factor for inhibiting productivity. Further, in the linear programming, there is no flexibility in setting conditions for obtaining an optimal solution, and it is not possible to cope with a change in an operation policy or the like. Tokiko 42-2940
In the case of Japanese Patent Publication No. 3, since the response of the slab temperature to the furnace temperature is slow, it is too late to control the temperature after detecting the deviation.

An object of the present invention is to improve the conventional problems as described above.

[Means for solving the problem]

The present invention has been made in view of such a point, and in a temperature control method for a continuous heating furnace having a plurality of continuous control zones, it is necessary to extract all slabs which are sequentially charged and extracted. When finding the optimum set furnace temperature of each control zone for satisfying the target extraction temperature and the target extraction lower limit temperature, and optimizing the combustion consumption to save energy, all of the charges that are sequentially charged and extracted The difference between the target extraction temperature and the predicted extraction temperature at the time of extraction, the difference between the target extraction lower limit temperature and the predicted extraction lower limit temperature, and the fuel consumption evaluation value for the slab at the time of extraction were calculated using the evaluation function expressed in the furnace temperature change amount. And performing partial differentiation on the evaluation function with the in-furnace temperature change amount to obtain an in-furnace temperature change amount as zero, and obtaining an optimum set in-furnace temperature change amount that minimizes the evaluation function.

[Action]

According to the present invention, when calculating the optimal setting furnace temperature change amount that minimizes the evaluation function, it is not necessary to perform a payout calculation for converging the result, and a unique result can be obtained. And the load on the computer is reduced. In addition, the weight function of the evaluation function can be adjusted and determined for each slab condition and the amount of furnace temperature change for each control zone. By adjusting and determining in advance, it can be dynamically changed online. can do. Thereby, for example, the optimum operation of the heating furnace can be performed in response to operation changes such as high quality and energy saving.

In a preferred aspect of the present invention, the optimum set furnace temperature is predicted for each control cycle (for example, every two minutes) from the current time to the extraction time for all slabs that are sequentially charged and extracted. This makes it possible to cope with operations under greatly different conditions, such as a hot piece and a cold piece, and a high-temperature extractor and a low-temperature extractor.

In a preferred embodiment, the slab heating influence coefficient of the evaluation function is obtained by integrating the coefficients of the ARMA model formula.
As a result, the slab temperature rise influence coefficient can be easily obtained, and the invention can be implemented very effectively when processing is performed by a computer.

Further, according to the present invention, since all slabs to be charged and extracted sequentially and a plurality of evaluation values can be set, the heating furnace can be optimally operated under various operating conditions and slab conditions.
In addition, by arbitrarily selecting the weighting function of the evaluation function for each slab and each control band, for example, a heating furnace optimal operation corresponding to an operation change such as quality-oriented and energy-saving oriented can be performed. Further, since the optimally set furnace temperature change amount can be uniquely obtained without repeatedly performing convergence calculation, the control cycle can be subdivided and the computer load can be reduced.

〔Example〕

Hereinafter, embodiments of the present invention will be described in detail. Here, the evaluation function J for obtaining the optimum amount of change in the furnace temperature is a secondary form in which the extraction temperature, the extraction lower limit temperature, and the fuel flow rate consumption are evaluated as in the following equation.

Evaluation function J = Σr ・ (Evaluation value of extraction temperature / extraction lower limit temperature) ² + Σq ・ (Evaluation value of fuel flow rate consumption amount) ² = (A ・ Δu-B) ^TR (A ・ Δu-B) -P ^T (A・ Δu
−B) + Δu ^T QΔu + W ^T Δu (1) where, A: Slab heating influence coefficient (constraint condition matrix) representing the amount of change in extraction temperature due to the change of the predicted heating history process (m × n) Δu: In the furnace A.Δu: Slab extraction temperature change (m × l) B: Slab extraction temperature required change (restriction condition vector) (m
× l) (considering both target extraction temperature-prediction extraction temperature and target lower limit temperature-prediction lower limit temperature) R: weight function (m × m) Q: weight function (n × n) P: weight function (non-negative vector) (M × l) W: weight function (non-negative vector) (n × l) m: 2 × (number of slabs) n: prediction time (minute) / control cycle (minute) × number of control bands

In this embodiment, a point at which the evaluation function J becomes a minimum value is obtained in order to obtain the optimum furnace temperature change amount. That is,
The point at which the evaluation function J is partially differentiated by Δu and the result becomes 0 is the minimum point of the evaluation function J, and the calculation formula of the optimum furnace temperature change amount is as shown in the following equation (2).

^{Δu * = (A T RA +} Q) -1 (A T RB-W / 2 + A T P / 2) ......
(2) According to the equation (2), the amount of change in the optimally set furnace temperature (Δu
^* ) Can be uniquely obtained, and the weighting function of the evaluation function can be arbitrarily selected for each slab and each control band, so that, for example, operations such as quality-oriented and energy-saving oriented as described later. Optimal operation of heating furnaces that respond to changes.

Here, the constraints forming the evaluation function will be described. First, the concept of the slab temperature rise influence coefficient (A) will be described. In a control cycle (Δt) of a certain control zone, a change amount (ΔT) of the slab temperature on the control zone exit side when the furnace temperature is changed by Δu can be expressed by the following equation (3).

ΔT = ∂T / ∂u ...... (3 ) Therefore, the furnace temperature change amount of the control zone, given the above equation extends from the current time t ₀ in increments Δt to t _n, the slab extraction temperature The amount of change is as shown in the following equation (4).

Here, ΔTk (t _out , x): x-point extraction temperature change in the slab thickness direction Δu ₁ , Δu ₂ , Δu ₃ : furnace temperature change in each control zone (ex pre-tropical zone, heating zone, uniform tropical zone) t ₁ , t ₂ , t ₃ : Remaining occupancy time of each control zone (ex pre-tropical zone, heating zone, isotropical zone) k: Slab number

Here, the extraction temperature / extraction lower limit temperature is defined as follows.

Extraction temperature = T (t _out , 0) Extraction lower limit temperature = T (t _out , 1) T (t _out , 0): Average temperature of the entire cross section T (t _out , 1): Average temperature of the on-skid section According to this definition When the slab temperature rise influence coefficient (A) is obtained based on the equation (4), the result is as shown in FIG. here,
The partial differential term can be easily obtained from the coefficient (a, b) of the ARMA model equation as shown in the following equation (5), as shown in equation (6). It is valid.

_{Tn = Σa j · T nj +} Σb j · u nj ...... (5) However, n: the current time j: Time T: slab temperature u: furnace temperature here, k: Slab number i: Burning tropics (= 1), heating zone (= 2), uniform tropics (= 3) It is.

Next, the concept of the required change amount (B) of the slab extraction temperature will be described below.

Here, Ti (t _out , 0): Average predicted temperature of all cross sections at the time of extraction of Si slab (A
Ti (t _out , 1): Average predicted temperature of on-skid section cross section at the time of extraction of Si slab (predicted value by ARMA model).

As described above, the equation (2) can be solved by obtaining the constraint conditions (A, B). An example of how the weight coefficient, which is one of the features of the present invention, corresponds to the operation change is shown below. .

For example, when the slab condition is the extraction lower limit temperature regulating material or the like, by changing the above-mentioned weight, the optimum operation in which the slab is given priority can be performed.

For example, the latter-stage load operation that saves energy is also
It is possible by changing the above weights. Here, the relationship between the weights R and P is expressed by the following equation, and P is determined by the coefficient K ₁ (1 × m).

P = K ₁ · R K ₁ and increasing the extraction temperature, it is safe size for extracting the lower limit temperature, the extraction temperature to reduce K _1, the optimum direction operation to the extracted minimum temperature.

The relationship between the weights Q and W is expressed by the following equation, and the coefficient K ₂ (1 ×
The weight W is determined by n).

Increasing W = K ₂ · Q K ₂ actively lowers the set furnace temperature and increases K ₂
If the value of is reduced, the operation is to keep the set furnace temperature constant.

FIGS. 1 and 2 show examples in which the weight of the evaluation function is changed using the control method described above.

In the example shown in FIG. 1, the weight is set to bake in the safe direction, and as is clear from the figure, all the target temperatures are satisfied. In the example shown in FIG. 2, the weight is set for baking in an energy-saving orientation, and as can be seen from the figure, it follows that the target temperature is favorably followed. The unit fuel consumption was much better than when the operator intervened quite frequently.

〔The invention's effect〕

As described above in detail, according to the present invention, a weighting function is set for each of a plurality of slabs by using a target extraction temperature and a target extraction lower limit temperature at the time of extraction, and an evaluation coefficient including an evaluation value of fuel consumption. Since the selection can be made arbitrarily for each slab and each control zone, the heating furnace optimal operation corresponding to the operation change such as quality-oriented and energy-saving-oriented can be performed, and the frequency of operator intervention and the fuel consumption rate can be reduced.

In addition, in the calculation that minimizes the evaluation function for obtaining the optimum set furnace temperature change amount, iterative calculation for convergence is not required, and the result can be obtained uniquely. It is significantly reduced as compared with the method, and the computer load can be reduced.

[Brief description of the drawings]

1 and 2 are graphs showing extraction temperature characteristics in one embodiment of the present invention, and FIG. 3 is a map showing a set of slab temperature rise influence coefficients (A) obtained from equation (4). is there.

Claims (3)

(57) [Claims] 1. A method for controlling the temperature of a continuous heating furnace having a plurality of continuous control zones, wherein a target extraction temperature and a target extraction lower limit temperature at the time of extracting all slabs to be charged and extracted sequentially are satisfied. hand,
In addition, when calculating the optimum set furnace temperature of each control zone for optimizing combustion consumption and saving energy,
The difference between the target extraction temperature and the predicted extraction temperature at the time of extraction, the difference between the target extraction lower limit temperature and the predicted extraction lower limit temperature, and the evaluation value of fuel consumption for all slabs that are charged and extracted sequentially are calculated as the furnace temperature. Using the evaluation function represented by the change amount, partial differentiation is performed on the evaluation function with the furnace temperature change amount to obtain a furnace temperature change amount as zero, and an optimum set furnace temperature change amount that minimizes the evaluation function is obtained. And a method for controlling the temperature of a continuous heating furnace. 2. The temperature of the continuous heating furnace according to claim 1, wherein, for all the slabs to be charged and extracted sequentially, the optimum set furnace temperature is predicted for each control cycle from the current time to the extraction time. Control method. 3. The evaluation function is defined as the following evaluation function J,
The slab heating effect coefficient (constraint condition matrix A)
The fuel flow rate consumption value is obtained by integrating the coefficients of the model formula, and the (fuel flow rate consumption evaluation value) ² is represented by the sum of the furnace temperature fluctuation evaluation value (Δu ^T QΔu) and the furnace temperature change amount evaluation value (W ^T Δu). Item 1
A method for controlling the temperature of a continuous heating furnace as described in the above. Evaluation function J = Σr ・ (Evaluation value of extraction temperature / extraction lower limit temperature) ² + Σq ・ (Evaluation value of fuel flow rate consumption amount) ² = (A ・ Δu-B) ^TR (A ・ Δu-B) -P ^T (A・ Δu
−B) + Δu ^T QΔu + W ^T Δu where A: slab heating influence coefficient (constraint condition matrix) (m × n) Δu: furnace temperature change (n × l) A · Δu: slab extraction temperature change (m × l) B: Slab extraction temperature required change amount (constraint condition vector) (m
× l) R: weight function (m × m) Q: weight function (n × n) P: weight function (non-negative vector) (m × l) W: weight function (non-negative vector) (n × l) m: 2 X (number of slabs) n: prediction time (minutes) / control cycle (minutes) x number of control bands.