JPH09303741A - Method and device for control of fluidized bed incineration furnace - Google Patents

Abstract

PROBLEM TO BE SOLVED: To correspond with a sudden variation of the operation state of a furnace, and perform a stable control so as to meet a target value by a method wherein a deviation amount which is added to an immediately previous manipulated variable, is determined based on a target locus, a specific equation model, an error being obtained in an error calculation process, and a control amount being obtained in an manipulated variable fixed time controlled variable calculation process. SOLUTION: At an equation model identification unit 3, an equation model to represented the relationship between a manipulated variable and a controlled variable from the present point to a point after a specified period of time, is identified in order for each specified control cycle, using an output from a simulator unit 6. At a manipulated variable determining unit 5, a deviation amount of the manipulated variable at the present point is determined by an estimation control method based on a target locus, an error being obtained by an error calculation unit 2, a controlled variable being obtained at a manipulated variable fixed time controlled variable calculation unit 4, and the equation model which has been identified the equation model identification unit 3. By adding this deviation amount to an immediately previous manipulated variable, the manipulated variable at the present point, is determined. By this method, the controlled variableis more stably controlled so as to meet the target value while taking the non-linear characteristics of the furnace under consideration.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and apparatus for controlling a fluidized bed incinerator, and more particularly, to a combustible material such as dust and fuel in a sand layer fluidized by blowing primary air. The present invention relates to a method and an apparatus for controlling a fluidized bed incinerator in which unburned gas is burned by blowing secondary air and steam generated by waste heat is circulated in the sand layer for heating.

[0002]

2. Description of the Related Art With the increasing demand for effective use of energy in recent years, power generation systems using fluidized bed furnaces such as refuse power generation systems are increasing. In order to recover more energy from the plants of these power generation systems, stable and efficient operation is required, and a control system to achieve this is required. However,
In particular, in an internal circulation type energy recovery furnace such as a fluidized bed incinerator, the system from charging and burning dust and fuel to obtaining superheated steam is complicated, and the existence of dead time and the influence of disturbance are large. Things are a problem. Furthermore, stable control is required for changes in the control target value when the furnace is started up or shut down, or when dust or fuel is input. For this reason, a control method has been proposed for controlling the amount of steam generated by using a predictive control method for an internal circulation type fluidized bed waste incinerator (for example, Japanese Patent Application No. 7- 233746). In this method, as shown in FIG. 6, the measured value of the blown-in amount of primary air up to the immediately preceding time point is substituted into a model M that represents the relationship between the blown-in amount of primary air and the steam generation amount. Based on this, the steam generation amount y _M (k) at the present time is calculated, and the error Δy of the calculated value with respect to the actually measured value y (k) of the steam generation amount at the present time and the target steam generation trajectory y _d (
τ + k + L) and the past blown amount of primary air are used to determine the blown amount of primary air at the present time by a predictive control method.

Here, the model M is

(Equation 7) And the predictive control method is

[0004]

(Equation 8) Is represented by

[0005]

However, the fluidized bed type incinerator has a feature that the operating state of the furnace is likely to change because the combustion state is more intense and the furnace is smaller than the stoker furnace. For example, when a large amount of dust or fuel suddenly drops into the furnace, rapid combustion occurs, and the operating state of the furnace greatly differs from the steady state. In such a case, the conventional predictive control method as described above using one fixed model cannot sufficiently cope with a sudden change in the operating state of the furnace. Further, the above-mentioned conventional method has a problem that it cannot sufficiently deal with the nonlinear characteristic of the furnace. The present invention has been made in view of the above circumstances, and an object thereof is to be able to sufficiently cope with a sudden change in the operating state of the furnace at any given time, and to take the nonlinear characteristics of the furnace into consideration. It is an object of the present invention to provide a control method and apparatus for a fluidized bed incinerator that can control the controlled variable to a target value more stably.

[0006]

In order to achieve the above-mentioned object, the method of the present invention comprises at least burning combustible substances and fuel in a sand layer fluidized by blowing primary air,
In a control method of a fluidized bed incinerator in which steam generated by the waste heat is circulated and heated in the sand layer, a target trajectory generation step of generating a target trajectory of the control amount based on a target value of the control amount and an actual measurement value. , The manipulated variable invariant time controlled variable calculation process that calculates the controlled variable when a constant manipulated variable is continuously added using the output from the simulator that calculates the behavior of the controlled variable with respect to the manipulated variable, and the above-mentioned simulator Using the output, a mathematical model identification step of sequentially identifying a mathematical model representing the relationship between the manipulated variable and the controlled variable from the present time for each predetermined control cycle, the measured value of the controlled variable at the present time, and the above An error calculation step for obtaining an error from the calculated value of the controlled variable at the present time using the output from the simulator, the target trajectory, the mathematical model identified in the mathematical model identification step, and the error meter Based on the error obtained in the process and the control amount obtained in the operation amount invariant time control amount calculation process, the deviation amount of the operation amount at the present time is determined by the predictive control method, and the deviation amount is set to the immediately preceding operation amount. The present invention is configured as a control method for a fluidized bed incinerator, characterized by comprising a manipulated variable determining step for determining a manipulated variable at the present time by adding.

The above mathematical model is

[Equation 9] The step input value of the manipulated variable is determined based on the deviation between the target value of the controlled variable and the actual measured value at the present time, and the coefficient matrix Ai of the mathematical model is determined by the step response to the step input value. be able to.

Further, the deviation amount of the manipulated variable at the present time determined by the predictive control method is

(Equation 10) Obtained by deriving a solution that minimizes the evaluation function given by

[Equation 11] Is represented by

In addition, the simulator includes a mass balance model, a heat balance model at a plurality of parts in the incinerator,
And the reaction model. Further, the device of the present invention is a fluidized bed type in which at least a combustible material and a fuel are combusted in a sand layer fluidized by blowing of primary air, and steam generated by waste heat thereof is circulated in the sand layer to be heated. In an incinerator control device, a simulator means for online calculating the behavior of the controlled variable with respect to the manipulated variable, a target trajectory generation means for generating a controlled trajectory target trajectory based on the controlled variable target value and the measured value, and the above simulator Using the output from the means, the manipulated variable invariant controlled variable calculation means for calculating the controlled variable when the constant manipulated variable is continuously added, and the output from the simulator means, A mathematical model identifying means for sequentially identifying a mathematical model representing the relationship between the manipulated variable and the controlled variable for each predetermined control cycle, an actual measured value of the controlled variable at the present time, and an output from the simulator means. Error calculation means for obtaining an error from the calculated value of the control amount at the present time, the target trajectory, the mathematical model identified by the mathematical model identification means, the error calculated by the error calculation means, Based on the control amount obtained by the control amount constant control amount calculation means, the deviation amount of the operation amount at the present time is determined by the predictive control method, and the operation amount at the present time is obtained by adding the deviation amount to the immediately preceding operation amount. It is configured as a control device for a fluidized bed type incinerator, which is characterized in that it comprises a manipulated variable determining means for determining. Further, all the control methods described above can be realized on this device.

[0010]

In the controller of the fluidized bed incinerator according to the present invention,
First, the target trajectory generation means generates the target trajectory of the controlled variable based on the target value of the controlled variable and the actually measured value. Then, the manipulated variable invariant time controlled variable calculation means calculates the controlled variable in the case where the manipulated variable at the previous time is continuously added using the output from the simulator means for online calculating the behavior of the controlled variable with respect to the manipulated variable. Next, in the mathematical model identifying means, the mathematical model representing the relationship between the manipulated variable and the controlled variable from the present time to a certain time after the current time is identified by using the output from the simulator means, for each predetermined control cycle, and the error is further identified. In the calculation means, the error from the calculated value of the controlled variable at the present time is obtained by using the output from the simulator means. In the manipulated variable determining means, the target trajectory, the mathematical model identified by the mathematical model identifying means, the error obtained by the error calculating means, and the control amount obtained by the manipulated variable invariant controlled variable calculating means are obtained. Based on this, the deviation amount of the operation amount at the present time is determined by the predictive control method, and the operation amount at the present time is determined by adding the deviation amount to the immediately preceding operation amount. Then, the actual process is controlled by repeating the above processing for each control cycle.

[0011]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments and examples of the present invention will be described below with reference to the accompanying drawings to facilitate understanding of the present invention. The following embodiments and examples are mere examples embodying the present invention, and do not limit the technical scope of the present invention. 1 is a block diagram showing a schematic configuration of a control apparatus for a fluidized bed incinerator according to an embodiment of the present invention, and FIG. 2 shows a schematic configuration of a fluidized bed incinerator to which the above control apparatus can be applied. FIG. 3 is a schematic diagram, FIG. 3 is a flow chart showing an operation procedure of the control device of the fluidized bed incinerator, FIG. 4 is a diagram showing an example of a method of setting the target trajectory y _d of the control value, and FIG. FIG. 6 is a diagram showing an example of the relationship between the quantity prediction value y ₀ , the deviation model Δy _m , and the step response model y _m, and FIG. 6 is a block diagram showing a schematic configuration of a conventional fluidized bed refuse incinerator control method. .

As shown in FIG. 1, the flow according to the present embodiment
The control device for the floor incinerator uses the primary air blowing
Burn combustible materials such as dust and fuel in the activated sand layer.
The secondary air is blown in to burn unburned gas,
The steam generated by the waste heat of is circulated in the sand layer for heating.
It is a controller for the actual process A, which is a fluidized bed incinerator.
You. In the target trajectory generation unit 1, the target value y of the controlled variable_rAnd actual measurement
Target trajectory y of the controlled variable based on the value y (k)_dGenerate
You. In the error calculation unit 2, the behavior of the controlled variable with respect to the manipulated variable
At present, using the simulator section 6 that calculates online
Calculated value y of the controlled variable at _sim(K) is calculated, and at the present time
Calculated value y of controlled variable_sim(K) and actual measurement of the controlled variable at the present time
An error Δy (k) from the value y (k) is obtained. When the manipulated variable is unchanged
In the control amount calculation unit 4, the output from the simulator unit 6 is output.
The control amount when the operation amount at the previous time is continuously added using
y₀Is calculated. In the mathematical model identification unit 3, the simulation
After a certain time from the current time, using the output from the
Mathematical model that represents the relationship between the manipulated variable and the controlled variable up to
(Described later) is sequentially identified for each predetermined control cycle. Manipulated variable
In the determination unit 5, the target trajectory y_dAnd the above error meter
The error Δy (k) calculated by the computing unit 2 and the manipulated variable invariant tense
Control amount y calculated by control unit 4₀And the above mathematical model
Predictive control based on the mathematical model identified by the constant part 3
Method (details will be described later)
The deviation amount Δu (k) is determined, and the immediately preceding manipulated variable u (k-1)
By adding the deviation amount Δu (k) to
The manipulated variable u (k) is determined.

Next, a specific structure of a fluidized bed type incinerator to which the above control device can be applied will be described with reference to FIG. In FIG. 2, dust and fuel are thrown into the hopper 11 and fall into the sand layer 12. By operating valves 13 and 14, primary air 15 and 16 are sent to the sand layer 12 separately in a plurality of parts, whereby the sand flows and the injected dust and fuel are primarily burned in the sand layer 12. Then, the generated gas flows to the freeboard 17.
The secondary air 19 is sent to the freeboard 17 by operating the valve 18, completely burns unburned gas such as CO, the exhaust gas passes through the exhaust gas outlet 20, and the waste heat boiler 2
It is sent to 1. In the waste heat boiler 21, saturated steam is generated by the thermal energy (heat amount) of the exhaust gas, and the saturated steam is further heated by passing through the heat transfer pipe 22 circulating in the sand layer 12 to become higher temperature superheated steam. Be recovered.
The main manipulated variables in this system are the amount of dust and fuel input, the amount of primary air and the amount of secondary air, and the major controlled variables are the temperature of the sand layer, the temperature at the top of the furnace, the temperature of the exhaust gas, the amount of superheated steam, There is superheated steam temperature.

Next, referring to FIG. 3, an outline of the control procedure (steps) of the fluidized bed incinerator and its operation will be described step by step. (Step S1) The target value y _r of the controlled variable is set. When there are a plurality of controlled variables, the target value y _r is a vector. (Step S2) The target trajectory y _d is set in the target trajectory generator 1 according to the target value y _r . The method of setting the target trajectory y _d is, for example, the method shown in FIG. This is a measured value y of the controlled variable at the present time (k), by using the target value y _r (k + i), the target trajectory _{y d (k + i) =} (I-C i) y r (k + i) + C i y
(K) where I: identity matrix (I = 1 when the control amount is 1) C: diagonal matrix in which real numbers 0 <c _j <1 are arranged in diagonal elements (0 when the control amount is 1 <C <1), which is to approach y _r (k + i) from y (k) at a constant rate.

(Step S3) The manipulated variable invariant time controlled variable calculation unit 4 uses the simulator unit 6 to simulate a case where the manipulated variable u (k-1) at the previous time point is kept constant and is added, The value of the control amount at this time is y ₀ . That is, Δu (k) = Δu (k + 1) =. . . = Δu (k + p-
1) = 0, the simulation value of the control amount is y ₀ (k +
p). Here, the configuration of the simulator section 6 which is operated online will be described. The simulator section 6 is composed of three models of a mass balance model, a heat balance model, and a reaction model for each of the sand layer section 12 and the freeboard 17 divided into a plurality of sections. The mass balance model expresses the inflow and outflow of a substance at the elemental level, and is modeled as the difference between the inflow and outflow accumulated or released. For example, in the sand layer portion 12, dust, fuel, primary air, input sand, etc. are the input side substances, and exhaust gas, dust, discharge sand, etc. from the sand layer portion 12 to the freeboard 17 are output side substances, and the difference is accumulated. Or released. Specifically, for example, the mass balance model of carbon content and oxygen content in the sand layer portion 12 is determined as follows.

[0016]

(Equation 12) The heat balance model expresses the heat generated by the heat input and output and the reaction by enthalpy, and is modeled as the accumulation or release of the difference between the input and output. For example, the sand layer 12
In sensible heat of waste and fuel, sensible heat of primary air, sensible heat of input sand, etc., the sensible heat of exhaust gas and dust from the sand layer part 12 to the freeboard 17, sensible heat of discharged sand, sand layer The amount of heat given to the steam passing through the heat transfer tube 22 circulating in the part 12 and the heat loss are heat output, and the reaction heat due to the primary combustion is the generated heat, and the difference between them is the temperature of the sand layer part 12. Affects raising and lowering. Specifically, for example, the heat balance model of the sand layer portion 12 is defined as follows.

[0017]

(Equation 13) The reaction model is a model that defines the reaction rate and reaction rate of a substance, and changes depending on the temperature, oxygen concentration, flow state, etc. of each part in the furnace at that time. Specifically, for example, considering the residual carbon in the sand layer portion 12,

[0018]

[Equation 14] Set the reaction type as follows. Then, the respective reaction ratios w: x: y: z are set as a function of the sand layer temperature, the residual carbon amount, and the primary air amount. The reaction ratio is set as follows, for example. Percent of carbon remaining intact in the sand layer: ratio of w = 1-x-y- z carbon leaving the left sand layer portion of the _{carbon: x = C x + α x} t + β x c / a + γ x a sand layer becomes carbon monoxide Proportion of carbon coming out of the part: y = C _y + α _y t + β _y c / a + γ _y a Proportion of carbon going out of the sand layer part as carbon dioxide: z = C _z + α _z t + β _z c / a + γ _z a , T: temperature of sand layer c: amount of residual carbon in sand layer a: amount of primary air C _x , C _y , C _z , α _x , α _y , α _z , β _x , β _y , β
_z , γ _x , γ _y , γ _z : Coefficients identified by experiment

The above three models are used to simulate the behavior of the controlled variable with respect to the manipulated variable in the furnace. The outline procedure is shown below. (1) For the sand layer part, the input values are the amount of dust input, the amount of primary air, the composition, mass, and temperature of the input sand. (2) The composition and mass of the substance leaving the sand layer or the substance remaining in the sand layer can be obtained from the input substance and the sand layer residual substance by the reaction model and the mass balance model. (3) Since the composition and mass of the substances that enter the sand layer, the remaining substances, and the substances that leave the sand layer have been determined, the heat value of each substance can be determined by enthalpy calculation. Also, the heat of reaction and the heat of vaporization of water are obtained based on the reaction model. (4) Based on the temperature of the sand layer, the steam temperature on the inlet side of the steam superheater and the steam flow rate, the amount of heat absorbed by the steam superheater and the superheated steam temperature on the outlet side of the superheater can be determined. (5) The sand layer temperature at the next time can be obtained from the heat balance model. (6) For the freeboard section, the temperature of the freeboard section and the composition / mass / temperature of the exhaust gas can be obtained in the same way. By repeating the above calculation, the behavior of the controlled variable with respect to the manipulated variable in the furnace can be simulated.

(Steps S4 to S6) In the mathematical model identification unit 3, using the output of the simulator unit 6,
A mathematical model (step response model) representing the relationship between the manipulated variable and the controlled variable from the present time to a fixed time later is sequentially identified for each predetermined control cycle. First, the step response model creation unit 3
In b, by using the output of the simulator unit 6 to input the deviation of the step-like operation amount with respect to the operation amount u (k-1) at the previous time point, the following deviation model Δy _m of the control amount is obtained. Identify (S5). That is, the coefficient matrix A _{i of the following} equation is determined.

[0021]

(Equation 15) Prior to that, the value of the deviation input of the manipulated variable when identifying the deviation model Δy _m of the control amount is calculated based on the deviation between the target value of the control amount and the actual measurement value. It is created in 3a (S4). This is to focus on the nonlinear characteristic of the furnace, in which the rate of change in the controlled variable is different on the side of increasing the manipulated variable and on the side of decreasing the manipulated variable, and it is intended to compensate for this nonlinear characteristic when creating the deviation model Δy _m. is there.
For example, when the superheated steam temperature is controlled by changing the flow state around the heat transfer tube 22 with the primary air 15, if the measured value of the superheated steam temperature is lower than the target value, the primary air 1
The deviation model is identified by inputting the deviation input in the direction of increasing 5 and conversely, when the measured value of the superheated steam temperature is higher than the target value, by inputting the deviation input in the direction of decreasing the primary air 15. Further, the magnitude (absolute value) of the deviation input may be changed depending on the magnitude (absolute value) of the deviation between the actually measured value of the superheated steam temperature and the target value. By the method as described above, it is possible to create the deviation model in consideration of the rate of change of the superheated steam temperature which differs between the rising side and the falling side of the amount of the primary air 15. In addition, for a multi-input multi-output system that controls multiple controlled variables with multiple manipulated variables, the sign and absolute value of the deviation input for each manipulated variable can be determined from the magnitude relationship between each controlled variable and the corresponding target value. The same problem can be solved by setting the reference table in advance.

Further, the step response model creating section 3b
And the deviation model Δy identified above _mAnd the above steps
Predicted value y of manipulated variable invariant control variable obtained in S3₀When,
Considering the dead time τ, the following step response model y_m
Is determined (S6).

(Equation 16) The manipulated variable invariant time controlled variable predicted value y ₀ , deviation model Δ
y _m, and an example of the relationship between the step response model y _m shown in FIG. (Step S7) Next, the error calculation unit 2 obtains an error _Δy (k) between the actual measured value y (k) of the control amount at the present time and the calculated value y _sim (k) of the simulator unit 6. Δy (k) = y (k) −y _sim (k)

(Steps S8-S11) Δy
Considering that (k) is due to disturbance, and assuming that this disturbance will continue in the future, the predicted value y (τ + k + p) of the controlled variable

[Equation 17] (S8). Since it is sufficient to perform control so that the predicted value y of the control amount matches the target trajectory y _d , the input deviation vector {Δu (k) of the operation amount that minimizes the evaluation function given by , Δu (k + 1) ,. . . , △ u
(K + p-1)}) is determined.

[0024]

(Equation 18) This problem is solved as a least squares problem, and the deviation vector Δu (k) of the current manipulated variable is derived as follows (S9).

[0025]

[Equation 19] The above-mentioned Δu (k) is added to the previous operation amount u (k−1) and the operation amount u (k) obtained by (S10) is input to the actual process A. Further, the operation amount u (k) becomes an input operation amount to the error calculation unit 2 in the next control cycle (S1).
1). The operation amount determination unit 5 performs the above-described steps S8 to S11. The processing from step S1 to step S11 is repeated for each control cycle.

The features of the control method and apparatus described above are as follows. (1) Since the future behavior of the real process A is predicted online using the simulator 6 and the step response model (numerical model) is recreated for each control cycle, the characteristics of the fluidized bed incinerator are in a steady state. Even if it greatly deviates from
The desired control characteristics reflected in the weighting coefficient matrix of the evaluation function can be stably realized. (2) Since the deviation input of the manipulated variable at the time of creating the above step response model is determined based on the deviation between the target value of the controlled variable and the actual measured value at the present time, it is determined whether the manipulated variable is increased or decreased. It is possible to realize control that takes into account the nonlinear characteristics of the furnace in which the change characteristics of the controlled variable are different. (3) For the parts that could not be modeled by the simulator 6 and the presence of disturbances due to the instantaneous combustion peculiar to the fluidized bed incinerator, the difference between the measured value and the calculated value by the simulator 6 is considered as feedback. Therefore, it is robust against these modeling errors and disturbances. (4) From the input of dust and fuel, the primary combustion in the sand layer 12, the secondary combustion in the freeboard 17, the steam generation in the boiler 11, and the superheated vaporization in the sand layer 12 In the internal circulation type fluidized bed incinerator that is passed through, it is possible to control the finally required stable superheated steam amount and temperature while considering the dead time existing in each process. As described above, according to the controller of the fluidized bed incinerator of the present embodiment, it is possible to sufficiently cope with the momentary changes in the state of the furnace, and to consider the nonlinear characteristics of the furnace to make it more stable. In addition, the controlled variable can be controlled to the target value.

[0027]

Since the control method and apparatus for a fluidized bed incinerator according to the present invention are configured as described above, it is possible to sufficiently cope with a sudden change in the operating state of the furnace. Taking the nonlinear characteristics of the furnace into consideration, the controlled variable can be controlled more stably to the target value.

[Brief description of drawings]

FIG. 1 is a block diagram showing a schematic configuration of a control device for a fluidized bed incinerator according to an embodiment of the present invention.

FIG. 2 is a schematic diagram showing a schematic configuration of a fluidized bed incinerator to which the above control device can be applied.

FIG. 3 is a flowchart showing an operation procedure of the control device of the fluidized bed incinerator.

4 is a diagram showing an example of a method of setting the target trajectory y _d of the control value.

FIG. 5 is a diagram showing an example of a relationship between a manipulated variable invariant time controlled variable predicted value y ₀ , a deviation model Δy _m , and a step response model y _m .

FIG. 6 is a block diagram showing a schematic configuration of a conventional fluidized bed refuse incinerator control method.

[Explanation of symbols]

 1 ... Target trajectory generation unit 2 ... Error calculation unit 2a ... Current control amount calculation unit 3 ... Mathematical model identification unit 3a ... Model creating step response input creating unit 3b ... Step response model creating unit 4 ... Control when operation amount does not change Quantity predictive value calculation unit 5 ... Manipulation amount determination unit A ... Actual process 11 ... Hopper 12 ... Sand layer portion 13 ... Valve 14 ... Valve 15 ... Primary air 16 ... Primary air 17 ... Freeboard 18 ... Valve 19 ... Secondary air 20 ... Exhaust gas outlet 21 ... Waste heat boiler 22 ... Heat transfer tube

Continuation of the front page (51) Int.Cl. ⁶ Identification number Reference number within the agency FI Technical display location F23G 5/46 ZAB F23G 5/46 ZABB (72) Inventor Akira Kitamura 1-5-5 Takatsukadai, Nishi-ku, Kobe-shi, Hyogo No.5 Kobe Steel Works, Ltd. Kobe Research Institute

Claims (10)

[Claims] 1. A fluidized bed incinerator in which at least a combustible material and a fuel are combusted in a sand layer fluidized by blowing primary air, and steam generated by waste heat of the combustible material and fuel is circulated and heated in the sand layer. In the control method, the target trajectory generation process that generates the target trajectory of the controlled variable based on the target value and the measured value of the controlled variable, and the output from the simulator that calculates the behavior of the controlled variable with respect to the manipulated variable online Using the output from the above-mentioned simulator and the operation amount invariant time control amount calculation step for calculating the control amount when the operation amount is continued to be added, a mathematical expression representing the relationship between the operation amount and the control amount from the present time to a certain time later Mathematical model identification process that sequentially identifies the model for each predetermined control cycle, and obtains the error between the actual measured value of the controlled variable at the present time and the calculated value of the controlled variable at the present time using the output from the simulator. Error calculation process, the target trajectory, the mathematical model identified in the mathematical model identification process, the error calculated in the error calculation process, and the control amount calculated in the operation amount invariant time controlled variable calculation process. Based on, the deviation amount of the current manipulated variable is determined by the predictive control method,
A control method for a fluidized bed incinerator, comprising a step of determining the operation amount at the present time by adding the deviation amount to the immediately preceding operation amount. 2. The method for controlling a fluidized bed incinerator according to claim 1, wherein the simulator comprises a mass balance model, a heat balance model, and a reaction model at a plurality of parts in the incinerator. 3. The above mathematical model is expressed by the following equation: The method for controlling a fluidized bed incinerator according to claim 1 or 2, wherein 4. The step input value of the manipulated variable is determined based on the deviation between the target value of the controlled variable and the actual measured value at the present time, and the coefficient matrix Ai of the mathematical model is determined by the step response to the step input value. The method for controlling a fluidized bed incinerator according to claim 3, which is determined. 5. The deviation amount of the manipulated variable at the present time determined by the predictive control method is expressed by the following equation: Obtained by deriving a solution that minimizes the evaluation function given by, The method for controlling a fluidized bed incinerator according to any one of claims 1 to 4, represented by 6. A fluidized bed incinerator in which at least a combustible material and a fuel are combusted in a sand layer fluidized by blowing in primary air, and steam generated by waste heat of the combustible material and fuel is circulated and heated in the sand layer. In the control device, a simulator means for online calculating the behavior of the controlled variable with respect to the manipulated variable, a target trajectory generation means for generating a controlled trajectory of the controlled variable based on the target value and the measured value of the controlled variable, and the simulator means Using the output, the operation amount invariant time control amount calculation means for calculating the control amount when a constant operation amount is continuously added, and the output from the simulator means, The mathematical model identifying means for sequentially identifying the mathematical model expressing the relationship of the controlled variables at every predetermined control cycle, the measured value of the controlled variable at the present time, and the output from the simulator means were used. An error calculating means for obtaining an error from the calculated value of the control amount at the present time, the target trajectory, the mathematical model identified by the mathematical model identifying means, the error calculated by the error calculating means, and the manipulated variable invariant. Based on the control amount calculated by the time control amount calculating means, the deviation amount of the operation amount at the present time is determined by the predictive control method, and the operation amount at the present time is determined by adding the deviation amount to the immediately preceding operation amount. A control device for a fluidized bed incinerator, comprising: an operation amount determining means. 7. The simulator means comprises a mass balance model, a heat balance model at a plurality of sites in the incinerator,
6. The control device for a fluidized bed incinerator according to claim 5, wherein the control device comprises a reaction model and a reaction model. 8. The mathematical model described above is rewritten as The control device for a fluidized bed incinerator according to claim 6 or 7. 9. A step input value of a manipulated variable is determined based on a deviation between a target value of a controlled variable at present and an actual measured value, and a coefficient matrix Ai of the mathematical model is determined by a step response to the step input value. The control device for a fluidized bed incinerator according to claim 8, which is determined. 10. The deviation amount of the manipulated variable at the present time determined by the predictive control method is Which is obtained by deriving a solution that minimizes the evaluation function given by The control device for a fluidized bed incinerator according to any one of claims 6 to 9, which is represented by: