JPS6130168B2 - - Google Patents

Description

[Detailed description of the invention]

The present invention relates to a method for controlling combustion in a heating furnace, and more particularly to a method for controlling combustion in a multi-zone heating furnace by controlling oxygen concentration. Generally, the purpose of controlling the oxygen concentration in the combustion exhaust gas in a heating furnace is to (1) avoid incomplete combustion of the fuel (not generate black smoke), (2) provide the minimum amount of air necessary for combustion, and These include reducing fuel consumption and suppressing _NOx emissions. In other words, the goal is to optimize combustion conditions. Generally, when controlling the oxygen concentration to achieve the above objectives (1) and (2), a specific target value of the oxygen concentration is an oxygen concentration of 0.5 to 1.5%. Oxygen concentration is 0.5
In a combustion state exceeding ~1.5%, the generation of _NOx increases, which is a pollution problem, and thermal efficiency (fuel consumption rate) also deteriorates. On the other hand, in a combustion state where the oxygen concentration is 0.5 to 1.5% or less, black smoke is generated, which poses a pollution problem and also deteriorates thermal efficiency (fuel consumption rate). Therefore, in controlling the oxygen concentration of a multi-zone heating furnace, it is necessary to determine the excess air coefficient in the combustion control system for each zone so that the oxygen concentration in the exhaust gas in each zone reaches the above-mentioned target value. If this is not possible, the above control purpose
(1) and (2) cannot be achieved. The conventional method for controlling oxygen concentration in multi-zone heating furnaces is to sample the combustion exhaust gas at the bottom of the multi-zone heating furnace, analyze the oxygen concentration, and adjust the oxygen concentration from an oxygen concentration controller to the desired oxygen concentration. The air-fuel ratio is given to the combustion zone air-fuel ratio setting device in a cascade through the setting device that determines a predetermined distribution ratio for each combustion zone, and the air flow rate is controlled by correcting the air flow rate setting signal for each combustion zone. It was hot. However, such conventional oxygen concentration control methods control the oxygen concentration in the mixed combustion exhaust gas, which is the result of mixing combustion exhaust gases from each zone of a multi-zone heating furnace, to a target value. Since the oxygen concentration in the combustion exhaust gas produced by the control system does not always reach the target value, there is a problem in that the purpose of oxygen concentration control cannot be completely achieved. Therefore,
Install an oxygen concentration meter to measure the oxygen concentration in the flue gas in each combustion zone of a multi-zone heating furnace, and determine the excess air coefficient of the combustion control system so that the oxygen concentration reaches the target value. However, the exhaust gas measured by the oxygen concentration meter installed in each combustion zone includes not only the combustion exhaust gas generated by the combustion control system in that zone, but also the exhaust gas measured by the oxygen concentration meter installed in each combustion zone.
It is a mixture of exhaust gas from the combustion zone located upstream of the flue gas flow in the furnace, and the measured value of the oxygen concentration meter installed in each combustion zone is only the flue gas produced by the combustion control system of each combustion zone. It's not the oxygen concentration. Even if oxygen concentration control is performed independently in each combustion zone based on oxygen concentration measurement values that are affected by other combustion zones, the control objective cannot be achieved. Therefore, in the present invention, in order to optimize the combustion state in each combustion zone of a multi-zone heating furnace, an oxygen concentration meter is installed in each combustion zone. The excess air coefficient of the combustion control system for each combustion zone is determined by understanding how the exhaust gases in the combustion zone interact with each other and propagate to the oxygen concentration meter. That is, the gist of the control method of the present invention is to control the combustion of a multi-zone heating furnace in which the fuel flow rate and air flow rate supplied to each combustion zone of a heating furnace having a plurality of combustion zones is controlled according to a desired furnace temperature. In the method: The following state description and target oxygen concentration for each combustion zone are determined in advance, in which the amount of change in the air excess coefficient of each combustion zone is input and the amount of change in oxygen concentration in the exhaust gas of each combustion zone is output;
Measure the oxygen concentration in the exhaust gas, the fuel flow rate, and the air flow rate in each combustion zone at a predetermined period; Based on these measured values, identify the transfer matrix in the state description equation by applying a Kalman filter; Determine the amount of change in the air excess coefficient for each combustion zone from the target oxygen concentration of the zone and the state description equation; Determine the air-fuel ratio of the fuel flow rate and air flow rate to be supplied to each combustion zone from the determined amount of change in the air excess coefficient; Combustion control method using oxygen concentration in a multi-zone heating furnace, characterized by ): transfer matrix (oxygen concentration mutual interference coefficient matrix in exhaust gas in each combustion zone), u(k): amount of change in air excess coefficient, ω(k): white Gaussian noise, and k: time. The method for controlling combustion by controlling oxygen concentration in a multi-zone heating furnace according to the present invention will be explained below by taking a three-zone heating furnace as an example. FIG. 1 shows the state of mutual interference of exhaust gases in a three-zone heating furnace. In Figure 1, 1 is the soaking zone, 2
3 indicates an upper heating zone, 3 indicates a lower heating zone, and 4 indicates a flue provided on the charging side of the slab 5 in the heating furnace. 6,
7 and 8 indicate the positions of burners in the soaking, upper heating, and lower heating zones 1, 2, and 3, respectively (hereinafter referred to as burners). Reference numerals 9, 10, and 11 indicate oxygen concentration meters provided at the exhaust gas exits of the combustion zones 1, 2, and 3, respectively. The slab 5 moves in the furnace from the heating zones 2 and 3 toward the soaking zone 1 in the direction of arrow AR. The exhaust gas in each combustion zone flows in a direction opposite to the direction of movement of the slab 5 and reaches the flue 4. In burner 6 in soaking zone 1, excess air coefficient μ
The combustion exhaust gas resulting from the combustion of fuel and air in _{step 1} flows to the flue 4 through the soaking zone 1 and the upper and lower heating zones 2 and 3, so the oxygen concentration of the combustion exhaust gas in the combustion zone 1 is It affects not only the measured value O ₁ of the oxygen concentration meter 9 of No. 1 but also the measured values O ₂ and O 3 of the oxygen concentration meters 10 and 11 of the upper and lower heating zones 2 _{and 3} . Furthermore, in the burner 7 of the upper heating zone 2, the oxygen concentration in the combustion exhaust gas produced as a result of burning fuel and air with an air excess coefficient μ ₂ is not only the value O ₂ measured by the concentration meter 10 of the zone 2. , concentration meter 1 in lower heating zone 3
It also affects the measured value O ₃ of 1. Furthermore, in the burner 8 of the lower heating zone 3, the air excess coefficient μ ₃
The oxygen concentration in the flue gas resulting from the combustion of fuel and air in the heating zone affects the measured values O ₃ and O 2 of the densitometers 11 and 10 in the zone 3 and the upper heating zone ₂ . Now, burner 6 in soaking zone 1 has an excess air coefficient μ
₁ , fuel and air are supplied to burner 7 in upper heating zone 2 with excess air coefficient μ ₂ , and burner 8 in lower heating zone 3 with excess air coefficient μ ₃ for combustion. Tropical zone 1, heated zone 2 and heated zone 3
Let the combustion exhaust gas amounts of the above combustion zones 1, 2, and 3 be respectively F ₁ , F ₂ , and F ₃ , and the oxygen concentrations measured by the concentration meters 9, 10, and 11 in each combustion zone 1, 2, and 3 above as O ₁ , O ₂ , and O ₃ , respectively. , and (excess air ratio C = excess air coefficient -
The influence coefficient (transfer coefficient) that 1.0) has on the measured value of the oxygen concentration meter in other combustion zone i is assumed to be g _ij . 1: Soaking zone However, i, j = 2: Upper heating zone. 3: Lower heating zone In this case, input the state of mutual interference of exhaust gas in the furnace, that is, the excess air coefficient of each combustion zone,
The state description formula that outputs the oxygen concentration of the oxygen concentration meter in each combustion zone can be written as follows. However, μi: Air excess coefficient of the burner in combustion zone i Fi: Amount of exhaust gas produced when air and fuel are combusted with air excess coefficient μ _i Oi: Oxygen concentration measured with an oxygen concentration meter g _ij : Influence coefficient (transfer coefficient) of excess air ratio (μi-1.0) on the measurement value of oxygen concentration meter in combustion zone j

[Table] 3: Lower heating zone Equation (1) expresses the state of mutual interference of combustion exhaust gas in the furnace by the excess air ratio (μi-1) of combustion zone i.
is the product of the exhaust gas amount Fi in combustion zone i and the transfer coefficient g _ij divided by the total amount of exhaust gas passing through the cross section of the furnace where the oxygen concentration meter in combustion zone j is installed. This is a state description equation that assumes that the excess air ratio of i affects the measured value of the oxygen concentration meter in combustion zone i. The amount of exhaust gas Fi in equation (1) can be determined by the following equation. μi=Fai/CaFfi...(2) Fi=GowFfi+(Fai-CaFfi)...(3) However, Gow: Theoretical amount of exhaust gas per unit fuel (including water vapor) Ca: Theoretical amount of air per unit fuel Ffi: Fuel flow rate Fai: Air amount when excess air coefficient μi Fai=μiCaFfi μiCa=Fai/Ffi: Air-fuel ratio Therefore, the transfer matrix of equation (1)

If [formula] can be determined, the optimal excess air coefficient μ to achieve the target oxygen concentration
i can be determined using the transfer matrix G. That is, it can be obtained using the following equation, which is a modification of equation (1). However, μi: excess air coefficient to achieve the target oxygen concentration i: target oxygen concentration value for each zone

[Table] 〓3: Lower heating zone
However, when the present inventor investigated the transfer matrix of equation (1) in an actual heating furnace, it was found that it contained white Gaussian noise with a certain dispersion. It was also found that the measured values of the oxygen concentration meter contained white Gaussian noise with a certain dispersion. Therefore, it is impossible to directly obtain the transfer matrix from equation (1). It is known that the Kalman filter is effective in such cases. Kalman filter is a method for estimating the internal state of a plant by knowing the input values and observed values, given the characteristics of the observation system and noise. Therefore, in the present invention, the input value to the plant is the air excess coefficient, the observed value is the oxygen concentration in the exhaust gas, and the internal state is regarded as the transfer matrix, so that the transfer matrix is identified by applying a Kalman filter. Below, a method for identifying the transfer matrix in the equation (1) describing the state inside the heating furnace using a Kalman filter will be described. Equation (1) is a state description equation that holds true when the measured value of the oximeter does not include noise, and the actual measured value of the oximeter contains white Gaussian noise with a certain average value. Since it is known that the above noise is included on the right side of equation (1), and the average value of the noise is canceled out, when the exhaust gas amount ratio is approximately constant, the mutual interference of the exhaust gas in the furnace is calculated. Suppose that is written in the differential format as shown below. However, ΔOi: Amount of change in exhaust gas oxygen concentration within a certain time Δμi: Amount of change in excess air coefficient within a certain time ωi: Observation noise Fi: Amount of combustion exhaust gas in each zone

[Formula]: Transfer matrix

[Table] 〓3: Lower heating zone
Since the observation noise ωi is white Gaussian noise and the magnitude of variance is known, the transfer matrix can be identified using a Kalman filter. Specifically, equation (5) is written as a discrete value system, and equation (5) is written as
Transform into equation (6). y(k)=B(k)u(k)+ω(k)...(6) However, the amount of change in exhaust gas oxygen concentration: (observed value) Transfer matrix: Excess air coefficient change amount: (Input value) White Gaussian noise: (Observation noise) However, ′ indicates the transposed matrix. Here, in order to identify B(k) in equation (6) by applying a Kalman filter, it is transformed as follows. h′(k)=[b ₁₁ (k)OOb ₂₁ (k)b ₂₂ (k)b ₂₃
(k) b ₃₁ (k) b ₃₂ (k) b ₃₃ (k)] ...(6e) Then, equation (6) becomes y(k)=M(k)h(k)+ω(k)...(7). Here, it is known that h, that is, G, contains noise, so the relationship between h(k) and h(k+1) is as follows, where v(k) is white Gaussian additive noise, h(k+1)=Ψ( k)h(k)+v(k)
...(8) becomes. If the sampling period (time between time k and time k+1) is set sufficiently long, equation (8) needs to express a steady state. In order to express the steady state, it is sufficient to make the transition matrix Ψ(k) a unit matrix, so h(k+1)=h(k)+v(k)...(9). B(k) in equation (6) can be identified using the basic equations of the Kalman filter according to equations (7) and (9). In order to identify the transfer matrix B(k) in equation (6), the observed value y(k) and the input value u(k) must be known. Observed value (change in exhaust gas oxygen concentration)
y(k) can be calculated from the measured value of the oxygen concentration meter, and the input value (excess air coefficient change amount) u(k) can be calculated using the measured values of air and fuel flow rates. Furthermore, in order to determine the transfer matrix G(k) from the transfer matrix B(k), the exhaust gas amount Fi(k) must be known. The exhaust gas amount Fi(k) can be calculated using the measured values of the air flow rate and the fuel flow rate, the theoretical exhaust gas amount, the theoretical air amount, and the excess air coefficient. in particular,
The exhaust gas oxygen concentration is measured at intervals of Δt seconds for t ₁ seconds (t ₁ /
Δt) times, and the average value is adopted. The excess air coefficient is obtained by taking in the measured values of air flow rate and fuel flow rate at intervals of Δt seconds, calculating each time according to equation (2) using the theoretical air amount, and calculating (t ₁ /Δ
t) times and use the average value. The exhaust gas amount is calculated by taking in the measured values of the air flow rate and fuel flow rate at intervals of Δt seconds and using the theoretical exhaust gas amount and theoretical air amount each time according to equations (2) and (3) _. /Δt) times and use the average value. The procedure for determining the transfer matrix G(k) in the oxygen concentration control of the present invention explained above is summarized as follows. (1) Measure the oxygen concentration, fuel flow rate, and air flow rate in each zone at intervals of Δt seconds for t ₁ seconds (t ₁ /Δt) times,
For each measurement, calculate the excess air coefficient and exhaust gas amount according to equations (2) and (3) based on the measured values of fuel and air, and calculate the average of the oxygen concentration, excess air coefficient, and exhaust gas amount. The values are Oi (k), μi (k)
and Fi(k). (2) The previous value Oi(k-
1) and μi(k-1), and calculate the difference between
Create an expression. (3) Create equations (7) and (9) and identify the transfer matrix B(k) using a Kalman filter. (4) Determine the transfer matrix G(k) using the transfer matrix B(k) and the exhaust gas amount Fi(k). That is, the oxygen concentration, fuel flow rate, and air flow rate in each zone are measured as described above, and based on these measured values, a Kalman filter is applied to the transfer matrix G in the state description equation (1). to identify and determine. In the present invention, the excess air coefficient for each zone is determined using the transfer matrix G(k) so that the oxygen concentration in each zone reaches the target value, and this procedure is as follows. (1) Determine the deviation between the measured value of oxygen concentration in each zone and the target oxygen concentration. (2) Using the transfer matrix G(k), the exhaust gas amount Fi(k), and the deviation of the target oxygen concentration according to equation (1), calculate the deviation from the excess air coefficient μi(k). demand. (3) From the above air excess coefficient μi(k) and the above deviation, determine the air excess coefficient that sets the oxygen concentration in each zone to the target value. Further, the present invention operates the excess air coefficient of the combustion control system of the furnace based on the excess air coefficient determined as described above. By the way, the equation (9) used to determine the transfer matrix B(k) in the present invention is an equation representing a steady state, so after manipulating the excess air coefficient of the furnace, t ₂ Wait a second and sample the oxygen concentration, air flow rate, and fuel flow rate again. Accordingly, the present invention operates the furnace air excess coefficient at intervals of (t ₁ +t ₂ ) seconds. Note that t ₂ seconds is a sufficient time for the furnace to stabilize with the new excess air coefficient and for the oxygen concentration meter to detect exhaust gas based on the new excess air coefficient. FIG. 2 shows a control device that implements the control method of the present invention. In this, 1 is a soaking zone, 2 and 3 are upper and lower heating zones, 6, 7, and 8 are burners for each zone 1, 2, and 3, and 9, 10, and 11 are each zone 1, 2, and 3. The oxygen concentration meter 12, 13, and 14 are thermocouples for each zone 1, 2, and 3. 15, 16, 17
is the combustion control system for bands 1, 2, and 3, and 18 is an electronic computer. 19a, 19b, 19c are furnace temperature controllers, 20
a, 20b, 20c are furnace temperature controllers 19a, 19
fuel flow rate controllers 21a, 21b, 21c for adjusting the fuel flow rate according to commands from b, 19c;
is a flow control valve, 22a, 22b, 22c are orifices, 23a, 23b, 23c are piping, 24a,
24b, 24c are oscillators, 25a, 25b, 25
Reference numeral c denotes an air-fuel ratio setter, and 26a, 26b, and 26c represent air flow rate controllers for adjusting the air flow rate in accordance with commands from the air-fuel ratio setters 25a, 25b, and 25c. The target temperature in each zone 1, 2, and 3 is now displayed on the computer 18.
In addition, when the target oxygen concentration, theoretical air amount, and theoretical exhaust gas amount in each zone 1, 2, and 3 are set, the furnace temperature controllers 19a, 19b, and 19c are set to the thermocouples 12 and 1.
3 and 14, the fuel flow rate controllers 20a and 20 are adjusted so that the input deviation becomes zero.
Give the fuel flow setting value to b, 20c, and adjust the controller 20.
a, 20b, 20c are orifices 22, respectively.
The control valves 21a, 21b, 21c are adjusted so that the measured fuel flow rate obtained through the transmitters 24a, 24b, 24c that convert the differential pressure signals of a, 22b, 22c into electrical signals matches the above set value. do. On the other hand, air flow rate controllers 26a, 26b, 26c
controls the opening and closing of the control valves 22a, 22b, and 22c so that the measured air flow rate matches the air flow rate set value from the air-fuel ratio setters 25a, 25b, and 25c. The air flow rate settings from the air-fuel ratio setters 25a, 25b, 25c are determined by the fuel flow rate controllers 19a, 19.
b, fuel flow rate setting value from 19c and calculator 18
The air-fuel ratio is calculated and set at predetermined intervals. The computer 18 calculates the oxygen concentration from the concentration meters 9, 10, 11 of each combustion zone 1, 2, 3, and the combustion control system 15, 1 of each zone 1, 2, 3.
6 and 17 to capture fuel flow and air flow measurements. Moreover, the calculator 18 calculates (t ₁ +
_t2 ) Start capturing the measured values of oxygen concentration, fuel flow rate, and air flow rate for each zone 1, 2, and 3 at second intervals,
Capture is performed for t ₁ seconds (t ₁ /Δt) times at intervals of Δt seconds.
For the above oxygen concentration measurements, the average value is
On the other hand, for the measured values of fuel flow rate and air flow rate, the excess air coefficient and exhaust gas amount are calculated and stored using the theoretical air amount and theoretical exhaust gas amount each time they are taken in, and the average value is calculated and stored as μi. (k),
Store as Fi(k). Once the oxygen concentration Oi (k), excess air coefficient μi (k) and exhaust gas amount Fi (k) are determined, Calculator 1
8 is the oxygen concentration Oi (k) in each zone 1, 2, and 3 above.
and for the excess air coefficient μi(k), (t ₁ +
The _previous value Oi(k
-1) and μi(k-1) to create the above equation (6). Next, the calculator 18 creates the above equations (7) and (9),
The optimum value of the transfer matrix B(k) is identified using a Kalman filter. Next, a transfer matrix G(k) is determined from the transfer matrix B(k) and the exhaust gas amount Fi(k) according to equation (6b). Then each belt 1,
The oxygen concentration deviation of each zone 1, 2, and 3 is calculated from the oxygen concentration target value of 2 and 3 and the above oxygen concentration Oi (k) of each zone 1, 2, and 3, and the state description formula of equation (1) is calculated. According to the above transfer matrix G(k) and the exhaust gas amount Fi in each zone
(k) and the above oxygen concentration deviation, the deviation from the excess air coefficient μi(k) of each zone is calculated to make the oxygen concentration of each zone a target value. And the above air excess coefficient μ
By adding i(k) and the above deviation, the optimum value of the excess air coefficient to achieve the target oxygen concentration is obtained. The air-fuel ratio of each zone is calculated from the excess air coefficient and the theoretical air amount. The air-fuel ratio is determined by the combustion control systems 15, 16, 17 of each zone.
air-fuel ratio setters 25a, 25b, and 25c. After ₂ seconds have passed after setting the air-fuel ratio, the calculator 18
begins capturing measurements of oxygen concentration, fuel flow rate, and air flow rate for each zone. Therefore, the calculator 18 determines the optimum air excess coefficient for each zone to achieve the target oxygen concentration at intervals of (t ₁ +t ₂ ) seconds;
The air-fuel ratio of each zone is set by the air-fuel ratio setters 25a, 25b, 2.
Set to 5c. The present invention as described above keeps the oxygen concentration in each combustion zone constant by grasping from time to time how the combustion exhaust gases in each combustion zone interact with each other and propagate to the oxygen concentration meter in each combustion zone. Since the excess air coefficient of each combustion zone is determined so that the target value is reached, the combustion state of each combustion zone can be maintained optimally.

[Brief explanation of the drawing]

The drawings are all explanatory diagrams of one embodiment of the present invention, and FIG. 1 shows a state description formula in which the air excess coefficient of each combustion zone of a three-zone heating furnace is input and the oxygen concentration in the exhaust gas of each combustion zone is output. FIG. 2 is a side cross-sectional view for explaining the transfer matrix identification method in this description, and FIG. 2 is a block diagram showing an example of a control device for implementing the control method of the present invention in a three-zone heating furnace. 1: Soaking zone, 2: Upper heating zone, 3: Lower heating zone, 4: Flue, 5: Slab, 6, 7, 8: Burner, 9, 10, 11: Oxygen concentration meter, 12, 13,
14: Thermocouple, 15, 16, 17: Combustion control system,
18: Electronic computer, 19a, 19b, 19c: Furnace temperature controller, 20a, 20b, 20c: Fuel flow rate controller, 21a, 21b, 21c: Flow rate adjustment valve, 2
2a, 22b, 22c: Orifice, 23a, 2
3b, 23c: Piping, 24a, 24b, 24c:
Transmitter, 25a, 25b, 25c: Air-fuel ratio setter, 26a, 26b, 26c: Air flow rate controller.

Claims (1)

[Claims] 1. In a combustion control method for a multi-zone heating furnace, which controls the fuel flow rate and air flow rate supplied to each combustion zone of a heating furnace having a plurality of combustion zones according to a desired furnace temperature: The following state description formula that inputs the amount of change in the excess air coefficient of each combustion zone and outputs the amount of change in oxygen concentration in the exhaust gas of each combustion zone and the target oxygen concentration for each combustion zone are determined in advance; Measure the oxygen concentration in the exhaust gas, fuel flow rate, and air flow rate in each combustion zone;
Based on these measured values, identify the transfer matrix in the state description equation by applying a Kalman filter;
Determine the amount of change in the excess air coefficient for each combustion zone from the target oxygen concentration of each combustion zone and the state description equation; From the amount of change in the excess air coefficient determined, calculate the air-fuel ratio of the fuel flow rate and air flow rate supplied to each combustion zone. A combustion control method using oxygen concentration in a multi-zone heating furnace, characterized by: y(k)=B(k)u(k)+ω(k), y(k): amount of change in oxygen concentration, B (k): transfer matrix (oxygen concentration mutual interference coefficient matrix in exhaust gas in each combustion zone), u(k): amount of change in air excess coefficient, ω(k): white Gaussian noise, and k: time.