JPH04205604A - Method and system for predicting/controlling process - Google Patents

Abstract

PURPOSE:To expand the model scale of process and to improve the accuracy of a predicting value by predicting controlled variable in the near future through the use of the model of sub-process linking as the shapes of an upstream and a downstream in the predicting control of process such as a thermal power plant, etc. CONSTITUTION:A system consists of a target value predicting system 20 which obtains the predicting value of the set point of steam temperature in the thermal power plant 10, a steam temperature predicting system 30 which obtains the predicting value of steam temperature and a manipulated variable decision system 40 which decides manipulated variable based on the predicting value of the set point and the predicting value of steam temperature. The model of a heat exchanger whose exit steam temperature is controlled variable and that of, at least, one heat exchange at the upstream side of the heat exchanger are incorporated in the thermal power plant 10 and steam temperature in the near future is predicted by using the model. Thus, the fluctuation of the entrance steam state of the heat exchanger whose exit steam temperature is controlled variable is considered and the predicting accuracy of steam temperature is improved.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to a process predictive control method and a predictive control system, and in particular, even when the process is a multi-input multi-output system, it is possible to control the control amount in the near future). The present invention relates to a predictive control method and a predictive control system for a process that can accurately predict processes.

[Conventional technology]

In plant control, predictive control is effective when there is a large delay in the time response of a change in a controlled variable to a change in a manipulated variable. One way to do this is to use “Steam T
emperature r'redictjon
Con1: roll for Thermal Pow
er Plant” (IE3/PR51984Win
terMeeting, l1allas, Texas
s, U, S, A, , Ianuary 29-Fe
bruary3.1984, 841IIM236-6)
There are predictive control methods described above. An overview of this conventional predictive control method will be explained below.

FIG. 10 shows the configuration of a conventional predictive control method.

This method is a predictive control method that targets the main steam temperature of thermal power plants, and includes a built-in model of the final stage superheater (in this case, the secondary superheater), and uses this model to calculate the main steam temperature. is predicted, and the fuel flow rate is determined based on this prediction result.

For prediction, the current fuel flow rate is held constant,
Assuming that the artificial steam state of the final stage superheater is maintained, the main steam temperature in the near future (for example, 10 minutes in the future) is predicted. In addition, in determining the fuel flow rate based on the prediction result, the fuel flow rate is determined by applying proportional and integral calculations to the deviation between the near future "1" vote value and the predicted value of the main steam temperature in the near future. ing.

[Problem to be solved by the invention]

In the above conventional technology, only a model of the final stage superheater is built-in, and this model is used under the condition that the current fuel flow rate is kept constant and the artificial steam state of the final stage superheater is maintained. Since the main steam temperature is predicted in the near future, it is not possible to make predictions that take into account fluctuations in the state of steam in the final stage superheater, and there is a limit to the prediction accuracy. In other words, in a thermal power plant, upstream of the final stage superheater,
Even if there are multiple heat exchangers such as the primary superheater and the current fuel flow rate is maintained, the state of artificial steam in the final stage superheater will fluctuate, so conventional technology predictions that do not take this fluctuation into account There was a limit W to the improvement of prediction accuracy.

On the other hand, in the control of thermal power plants, a method has been proposed in which predictive control is applied not only to control of main steam temperature but also to control of reheat steam temperature.

FIG. 11 is an example thereof. That is, assuming a main steam temperature system 1a and a reheating steam temperature system 1b in the thermal power plant, the predicted values are obtained using the secondary superheater model 3a and the secondary reheater model 3b, respectively. Thereby, the operation amount is determined by the main steam temperature control section 4a and the reheat temperature control section 41).

In this case, the controlled variable is the reheat steam temperature, the manipulated variable is the recirculation gas flow, and it is treated as another phase input single output system. The model also uses a model of only the secondary reheater portion, and predicts the outlet steam quality of the secondary reheater, that is, the reheat steam temperature using this model. Therefore, there are two control systems: the main steam temperature system 11111 control system and the reheat steam temperature prediction control system.
There are two models, each predicting a different controlled variable and manipulating a different manipulated variable using a different model. In other words, 41 manpower
It can be considered that two separate IL output systems exist.

However, there is only one thermal power branch 1~, and the reheat steam temperature systems such as the main steam temperature system interfere with each other.

If there are two separate +1 input single output systems, the predictability of the control amount will not be much improved.

(The following is a blank space) An object of the present invention is to provide a process predictive control method and a system for the same, which improves the accuracy of predicting the control amount of the process.

[Means to solve the problem]

In order to achieve the above [1], according to one aspect of the present invention, a model of the process is included, and using this model,
In a predictive control method for a process that predicts a controlled variable in the near future and determines a manipulated variable based on the predicted result, a subprocess whose state is a controlled variable and at least one on the first stream side of this subprocess A predictive control method for a process is provided, which includes a built-in model of two sub-processes and uses this model to predict control b() in the near future.

Further, according to another aspect of the present invention, predictive control of a process includes a built-in process model, uses this model to predict a control amount in the near future, and determines an operation amount based on the prediction result. A predictive control method for a process is provided, which is characterized in that a manipulated variable is determined such that a deviation from a target value in the near future, such as a predicted value of a controlled variable, is a predetermined value.

Further, according to the present invention, there is provided a control claw prediction system that predicts a control claw in the near future of a process based on an operation amount and a control amount inscribed in a process to be controlled, and a "single standard" that predicts a target value in the near future of a process. a value prediction system; and a manipulated variable determining system that determines the manipulated variable of the process based on the deviation between the predicted 1" target value and the control claw, and the controlled variable predicting system is configured to The control method is characterized by comprising means for forming a model of a sub-process that is idle and at least one sub-process upstream of this sub-process, and using this model to predict a control amount in the near future. A predictive control system for a process is provided.

Furthermore, according to the present invention, such a predictive control system l
A thermal power plant system is provided which includes a thermal power generation plan 1- and a thermal power generation plan 1- to be controlled.

In addition, there is provided a steam temperature prediction system for predicting steam temperature of a thermal power plant based on manipulated variables and control for thermal power generation plans 1 to 1, wherein at least one of the state variables is a steam temperature. and a steam temperature prediction system comprising a built-in model of at least one sub-process upstream of this sub-process, and means for predicting steam temperature in the near future using this model. Ru.

The model for the sub-process may be one that simulates, for example, a heat exchanger in Plan I-.

Specifically, for example, in a thermal power plant, a model of a heat exchanger whose outlet steam temperature is a controlled variable and at least one heat exchanger on the first stream side of this heat exchanger is built-in, and this model is It can be used to predict steam temperatures in the near future.

[For production]

For example, when the present invention is applied to thermal power generation plan 1, a model of a heat exchanger whose outlet steam temperature is a controlled variable in a thermal power plant, and at least one heat exchanger located upstream of this heat exchanger (1111) is used. This model is used to predict steam temperature in the near future, so fluctuations in the steam state can be taken into account and the accuracy of steam temperature prediction can be improved. You can improve.

It also has a built-in predictive model for steam temperature, which is a controlled variable in thermal power plants, and uses this model to predict steam temperature in the near future, and determine the deviation between this predicted value and a target value in the near future. If the manipulated variable is determined so that the value becomes zero, for example, tuning of the control parameters becomes unnecessary.

In this case, the predictive control system controls multiple control variables and manipulated variables.
In the case of a multi-input multi-output control system with
Since the manipulated variable can be determined in consideration of the variation in l, non-interference control that suppresses interference between controlled variables can be easily realized.

〔Example〕

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 shows an outline of the configuration of an embodiment of the present invention.

This embodiment includes a target value prediction system 2 for calculating a predicted value of a target value of steam temperature in a thermal power generation plan 1-1, a steam temperature prediction system 3 for calculating a predicted value of steam temperature, and a system for calculating a predicted value of a target value and steam temperature. It is configured to include a manipulated variable determination system 4 that determines the manipulated variable based on a predicted value of temperature. The details of this embodiment will be explained below.

First, an outline of an embodiment of thermal power generation plan 1- to which the present invention is applied will be explained.

FIG. 2 is an explanatory diagram showing an example of coal-fired thermal power generation plan 1.

Air from the forced draft fan 101 is preheated through an air preheater 102 and accelerated by a primary air fan 1.03 to blow into the coal mill]07. On the other hand, the coal in one coal bunker 104 is conveyed into the coal mill 107 by a coal feeder 106 driven by a coal feeder driving motor 105 . The coal that has been pulverized by the coal mill 107 is sent into the boiler 122 along with the air flow, where it is combusted.

Furnace water-cooled wall 110. Water drains through the primary superheat @ii1, becomes steam, passes through the spray 121, is further superheated in the secondary superheater 108, and passes through the main steam piping and the main steam control valve]]6 to the high-pressure turbine. Enter 117.

The steam leaving the high pressure turbine 117 is transferred to the secondary reheaters 1 and 09
It is resuperheated and sent to the medium/low pressure turbine 118. A generator 119 is immediately operated by a high-pressure turbine 17 and a medium/low-pressure turbine 118.

The steam exiting the medium/low pressure turbine 118 is condensed by a condenser 120, and this condensate is fed by a water supply pump 115.
It is sent again to the boiler's furnace water cooling wall 1.10. The condensate is also sent to the spray 121 via the spray control valve]]3. Furthermore, the boiler 122 is provided with a gas recirculation fan 112 that circulates the gas inside.

Among the control variables of thermal power generation plan 1-1, the control variables that are difficult to control are the steam temperature, which includes the main steam temperature Tus, the primary superheater 111 outlet steam temperature T, , and the reheat steam temperature TRs.
There are three. Furthermore, there are three manipulated variables for controlling these steam temperatures: fuel flow rate Fp, spray flow rate Fsp, and recirculation gas flow rate FOR.

First, the functions of each system shown in FIG. 1 will be explained along with their operating principles.

The target value prediction system 2 predicts the near future value of the target value of steam temperature using the following equation.

Here, P□(k, n): At the current moment k, the main steam 1
Predicted value r2 (lc, n) of 11 sampling destinations of the target value r1 of the talk rate go ljs: 'At time 1 (at the time 1), the target value r2 of the steam temperature T x s o
Predicted value r, (k, n) of sampling light: Q time 1 (at reheat steam temperature T
■1 of the target value r of R5 Predicted value rl(k) of the sampling destination: Current k of the target value r of steam temperature
(j-1 to 3) aI: Rate of change of target value r of steam temperature at current time k (i-1 to 3) 8T: Sampling period Steam temperature prediction system 3 uses a steam temperature system model to predict the near future value of steam temperature.

The model of thermal power generation plan 1- can be expressed by a differential equation derived from the Ymu law of conservation of mass and the law of conservation of energy. However, pressure systems have faster response and better controllability than temperature systems, and pressure fluctuations are smaller.

Therefore, the pressure remains at the set value or B1 during the predI period.
It can be treated as being held in the measured value. Here, only differential equations derived from the law of conservation of energy are used, without using differential equations derived from the law of conservation of mass.

FIG. 3 is a diagram showing an example of a modeling range.

By the way, heat exchangers for boilers are roughly divided into two types depending on the form of heat transfer. 1 is a heat exchanger mainly based on convection heat transfer, which includes a primary superheater 111, a secondary superheater 108, a primary reheater 132, and a secondary reheater 1.09 shown in FIG.
corresponds to this.

The other type is a heat exchanger mainly using radiation heat transfer, and the furnace ice wall 110 shown in FIG. 3 corresponds to this.

These two types of heat exchangers consist of water/steam, metal, and gas sections, as shown in FIG. next,
The model formula for each part will be explained.

Above all, the entire model can be obtained by combining the heat exchangers (subprocesses) shown in FIG.

(1) Water/steam part The model formula for the water and steam parts is as follows! I can get it.

-18=Q,,=A,,,ct,,(o,,-o,)
(4) Here, v6: Volume of water/steam γ・: Specific weight of water/steam I (,: Enthalpy of water/steam H, ,: Enthalpy of water/steam at inlet F,: Flow rate of water/steam Q9.: Heat transfer from metal to water/steam A: Heat transfer area from metal to water/steam α4: Convective heat transfer coefficient from metal to water/steam 06: Temperature of metal O: Water/steam Steam temperature Note that the enthalpy 1 and specific weight γ of water/steam are expressed as functions of temperature θ and pressure P by the following equation.

FI, = f (L, P,) (5) γ,
=f Co,,p,) (6) Also, the temperature O
5 is given by the following equation as a function of enthalpy ■ and pressure P.

o,=f (II,, p,) (7) (ii) Metal part The model formula for the metal part is expressed by the following equation.

Q,,=A,,・α. (0 & -0□) (9) (In case of convective heat transfer) Qgh=4n+・αmm, (kxOg+ -(], + to -) 0y-L,) where 1M1: Metal mold plate C□: Metal Specific heat Q,: Amount of heat transfer from gas to metal A g, ,: Heat transfer area α from gas to metal, 9: Convection heat from gas to metal 1 Delivery rate 01: Temperature of gas 1 (□, to 2 : Constant β1.1: Radiant heat transfer from gas to metal (iii)
Gas Part The model equation for the gas part is given by the following equation.

Here, ■f: Volume of gas γ,: Specific weight of gas ■11: Enthalpy of gas Hl [Enthalpy of gas in two population Ff: Flow rate of gas, enthalpy of gas 11. and specific weight γ9 at temperature 0
1 and pressure P5 as a function of the following equation.

H,=f(13yo,P,) (12)γ,
, -f (0,,P,) (13) Also,
Temperature 011011 is enthalbin T-T,, TI,,
and pressure P1 as a function of the following equation.

0,=f (IT,, P,) (14
)01,==f(HuIt P,) (] 5
) Furthermore, in the furnace, the enthalpy of the inlet gas ■1
, , can be determined by the following equation.

=21-■,・F,=H,・F, 10H,・F,+TI,R・
FR (1, 6) where, II: Calorific value of fuel Fr: Fuel flow rate H,: Enthalpy of air F,: Air flow rate rr, R: Enthalpy of recirculating gas FtR: Recirculating gas flow rate Also, furnace The outlet gas flow rate 171, that is, the boiler gas 'tAt plate F-p, is determined by the following equation.

F-=Fr +FI0F-R(], 7) In the above, the model was created using a differential equation derived from the law of conservation of energy, but it is also derived from the law of conservation of mass. A model may be created by adding differential equations.

Further, although one example of the modeling range is shown in FIG. 3, the modeling range may be expanded as shown in FIG. 5. That is, in the example shown in FIG. 5, the feed water superheater 133 and the energy saver 134 are added to the modeling range shown in FIG. 4.

Furthermore, the main steam control valve 116 and the high-pressure turbine 117 may be added to the modeling range.

The manipulated variable determining system 4 determines the manipulated variable based on a predicted value of the target value in the near future and a predicted value of the steam temperature in the near future. Next, this will be explained.

First, at the current time k, the previous sampling time (l
For the case where the manipulated variable is held at the value of c-1), the deviation between the predicted value of the steam temperature in the near future and the predicted value of the target value in the near future is determined by the following formula 2).

Here, eta (k, n): Deviation (j = 1 to 3) , current time 1 (at the previous sampling time (k
-1) When each manipulated variable changes from the value, the deviation between the predicted value of each steam temperature and the predicted value of the target value is determined by the following equation.

Here, Xlj (k+n): Predicted value of the steam temperature when only the manipulated variable U, δuj, is varied from the value at the previous sampling point (k-1) at the current time l(1-1 to 3゜j= 1 to 3) eIj (k, n): Prediction of the steam temperature when only the manipulated variable U, δnj, is changed from the value at the previous sampling point (k-1) at the current moment k. Eye 1 vote (direct predicted value r・2 (
k, n) (1-1 to 3 + J = 1 to 3) Furthermore, at the current point k, the deviation from the previous sampling point (k
-1) When each manipulated variable changes from the value, the influence coefficient for the deviation between the predicted value of each steam temperature and the predicted value of the target value is determined by the following equation.

The influence coefficient (i = 1 to 3. Using the determined influence coefficient, the estimated value of the predicted deviation of the steam temperature when the manipulated variable 11J is changed from the value at the previous sampling point (1(-1)) at the current time k is expressed by the following equation.

Here, e+ (k+ r+): The estimated value of the difference in the prediction section of the steam temperature when changing the manipulated variable uj from the value of the previous sampling point (1 (-1) at the current time 1 (j = 1 ~ 3, j = 1 to 3) Using equation (26), the estimated value of the predicted deviation of steam temperature e
The amount of change in the manipulated variable Δu, (10) at which t(1(,n) becomes a determined value, for example, zero, is obtained by the following equation.

Here, [AI-1, multi helix [AI inverse 71
~Using the change in the manipulated variable Δu, (k) obtained by the Ricks equation (27), the manipulated variable ut (k
) can be determined.

Each of the above systems is realized, for example, by installing an arithmetic program in a hardware system as shown in FIG. 8 and executing this program.

The system shown in FIG. 8 includes a central processing unit 1 to (CPU) 801 that executes various calculations, a read-only memory (ROM) 802 that stores programs executed by the CPU 801 and fixed data, and a read/write capable of reading/writing the following. memory(
RAM) 803 and analog data to analog/de, f
Analog people who change and incorporate digital changes 1~ (AI
) 804, digital human power units l to (DI) 805 that take in digital input data, analog output units h (AO) 806 that output analog signals, and digital output units I and (T') that output digital signals.
O) 807, a communication interface 808, and an internal bus 809 connecting these. Note that the communication interface 808 is connected to the communication bus 310.

Next, the procedure for determining the manipulated variable by each of the above systems will be explained with reference to FIG.

First, at present point 1, the predicted value of the target value of steam temperature in the near future is determined using equation (1) (Step 1)
.

At current time 1 (, the previous sampling time (k −
1,), the deviation between the predicted value of the steam temperature in the near future and the predicted value of the target value in the near future is determined using equation (18) (step 2).

At current time 1 (, the previous sampling time (k-1
), the deviation between the predicted value of each steam temperature and the predicted value of the target value when each manipulated variable varies from the value of (19), (20)
, (21) (step 3).

At the current moment k, the previous sampling time (k-1,
), the influence coefficient on the deviation between the predicted value of each steam temperature and the predicted value of the target value when each manipulated variable changes from the value of (
22), (23), and (24,) (Step 4).

Using the determined influence coefficient, at the current point k, the estimated value of the predicted deviation of the steam temperature when each manipulated variable is changed from the value at the previous sampling point (k -1,) is determined by equation (25) (step 5).

The estimated value of the predicted deviation of steam temperature is a determined value, e.g.
The amount of change in the manipulated variable that becomes zero is determined using equation (26) (step 6).

Using the obtained change in the manipulated variable, the manipulated variable at the current moment k is determined by equation (27) (step 7).

Next, an embodiment of a control system for thermal power generation plan 1- to which the present invention is applied will be described.

FIG. 9 shows an overview of the configuration of the thermal power plant control system of this embodiment.

The thermal power plant control system of this embodiment includes a master controller 1 to roller], OO, and this mask controller].
, OO, and includes subloop controllers 410 to 4.70 that control lower control elements, and a predictive control controller 200 that provides an operation amount to the mask controller 100.

-, 11- The predictive control controller 200 includes the target value prediction system 10, the steam temperature prediction system 1130, and the manipulated variable determination system b40 described above.

The manipulated variable determination system 40 outputs fuel flow rate demand correction values (changes) ΔF, D, spray flow rate demand correction values (changes) ΔFSPD, and recirculation gas flow rate demand correction values (changes) 8 as manipulated variables. F abn is output,
master controller] 00.

Note that the contents of each of these systems have already been described, so the description will not be repeated.

The sub-loop controller 1-0-ra 4-1.0 has a turbine steam flow control section 411 and controls the main steam control valve 116. The sub-loop controller 1-roller 420 includes a water supply flow rate control section 42 and controls the water supply pump 15. The sub-loop controller I roller 430 includes a fuel flow rate control section 431 and controls the coal feeder 106. The subloop controller 1 to the roller 440 include an air flow rate control section 441 and control the forced fan 101. Sub loop controller 1-roller 4
50 includes an exhaust gas flow rate control section 451 and controls the induction fan 114. The subloop controller 460 includes a spray flow rate control section 461 and controls the spray valve 113. The subloop controller 1 roller 470 is the recirculating gas flow rate controller 4
71 and controls the gas recirculation fan 112.

The mask controller 100 includes a main steam pressure control section 151 that outputs a manipulated variable for the deviation between the actual main steam pressure and a set value, and a gas 02 control section that outputs a manipulated variable for the deviation between the actual gas 02 and the set value. 152 and a furnace pressure control section 153 that outputs a manipulated variable for the deviation between the actual furnace pressure and a set value, as a mass system control means with a relatively quick response time. Note that these may be provided separately from the master controller 10o.

In addition, the master controller 100 receives a load command E L D (Economic L
oad Dispatching) and A F C (A
utomatjcFrequency Control
))) and a mask unit 16 which limits the rate of change to a constant rate of change based on the sum of -], and a main steam pressure correction unit 1 which corrects the load command - based on a command from the main steam pressure control unit.
62 and the above fuel flow rate demand correction value (change amount) ΔFF
Fuel flow rate demand correction unit 1 that corrects the fuel flow rate by D
63, and the gas 0 is
2, a furnace pressure correction section 165 that corrects the furnace pressure based on the output of the furnace pressure control section 153, and a furnace pressure correction section 165 that corrects the furnace pressure based on the output of the furnace pressure control section 153; It includes a spray flow rate demand correction unit 166 that performs correction, and a recirculation gas flow rate demand correction unit 167 that performs correction of recirculation gas meteors using a recirculation gas flow rate demand correction value (variation) ΔFGRD.

The control system 11 for this type of thermal power plant is disclosed, for example, in the specification of Japanese Patent Application No. 57-44.709.

The mask controller 1 roller 100, the predictive control controller 200, the communication bus controller 300, and the subloop controllers 410 to 470 are connected by a communication bus 310 to transmit information.

Further, the various controllers can be configured using the hardware system shown in FIG. 8, similar to the predictive control controller 200 described above.

With such a configuration, in the operation of the thermal power generation plan 1-, the main steam pressure control section 151 and the gas 02 control section 16 control pressure, gas, etc. that respond quickly to the load command.
4 and the furnace pressure control unit 153, and on the other hand, things with a slow response such as steam temperature are controlled by the predictive control controller 200.

The predictive control controller 200 is configured as follows:
Since the sub-process model allows highly accurate predictions, errors in predicted values are reduced and a plant that is easy to operate can be realized.

In the embodiment described below, a model of the final stage sub-process and another sub-process on the upstream side is built in, and this model is used to predict the controlled variable in the near future. Fluctuations in the state quantities of sub-processes on the downstream side, which affect fluctuations, can be taken into account, improving the accuracy of predicting control variables.

Furthermore, in the embodiments described above, the manipulated variable is determined so that the deviation between the predicted value of the control amount in the near future and the target value in the near future becomes a predetermined value, for example, zero. Therefore, tuning of control parameters becomes unnecessary. Furthermore, if the predictive control system is a multi-input, multi-output control system that has multiple controlled variables and manipulated variables, the manipulated variable can be determined by taking into account fluctuations in other controlled variables. Interference can be suppressed and non-interference control can be easily realized.

It should be noted that if interference is not so much of a problem, other control methods may be applied instead of the above-mentioned control method. For example, a control method can be applied in which the manipulated variable is determined by performing proportional/integral calculations on the deviation between the predicted value of the controlled variable and a target value in the near future.

An example is shown in FIG.

The control system shown in the figure is equipped with a target value prediction system 1120 and a steam temperature prediction system 30 according to the same article as the above-mentioned embodiment, and for thermal power generation plan j-10, a target value prediction system 1120 and a steam temperature prediction system 30 are provided. Based on the deviation from the predicted value of steam temperature, proportional and integral calculations are performed to determine fuel flow rate F2 and spray flow rate Fs.
p and recirculation gas flow rate F. The PT controller 5], 52, and 53 each output K as a manipulated variable.

According to this embodiment, as in the above-described embodiments, prediction accuracy can be improved, and since the manipulated variable is determined by a proven proportional-integral calculation, it is possible to construct a highly reliable system. Further, even if the step is inputted by Iran, there is an advantage that the deviation between the set value and the predicted value of the controlled variable becomes zero in a steady state.

In each of the above embodiments, examples of thermal power generation plan 1 to 1 are shown, but the present invention is not limited thereto. It can also be applied to control other processes. For example, the present invention can be applied to a process in which there is at least one subprocess on the upstream side, and the subprocess has a state quantity of 5 units. In this case, the state quantity of the upstream subprocess may or may not be a controlled quantity.

Furthermore, it is also possible to construct a control model in which multiple processes each consisting of multiple subprocesses exist independently.

[Effects of the Invention] As explained above, according to the present invention, in predictive control of processes such as thermal power plants, near-future control is performed using models of sub-processes linked in the form of "first stream" and "downstream". Since the amount can be predicted, the model simulation of the process can be expanded and the accuracy of the predicted value can be improved.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an outline of the configuration of an embodiment of the present invention, FIG. 2 is a block diagram showing the configuration of an embodiment of thermal power generation plan 1 to which the present invention is applied, and FIG. 3 is a block diagram of the present invention. An explanatory diagram showing an example of a modeling range in a thermal power plant to which the invention is applied, FIG. 4 is an explanatory diagram showing a model of a subprocess in this embodiment, and FIG. An explanatory diagram showing another example of the modeling range, Fig. 6 is a flowchart showing the procedure for determining the manipulated variable by each system constituting this embodiment, and Fig. 7 is a thermal power generation plan to which the present invention is applied. FIG. 8 is a block diagram showing the overall system configuration of the control system. FIG. 8 is a block diagram showing the system configuration of the controller 1-roller hardware of each system. FIG. 9 is a block diagram showing the system configuration of the controller 1-roller hardware of each system. A block diagram showing the configuration of the control system,
Figure 10 is a block diagram showing the configuration of a conventional control system for a thermal power plant, Figure 11 is a block diagram showing the configuration of a control system with two predictive models, and Figure 12 is a block diagram showing the configuration of a control system with two predictive models. FIG. 2 is a block diagram showing the configuration of an embodiment when applied to integral control. 10...Thermal power generation plan h, 20...Target value prediction system, 30...Steam temperature prediction system, 40...Manipulated amount determination system, 51. ．． 52.53...P ■Con 1-
11-A, 1.08-Secondary superheater, 109-Secondary reheater, 110 Furnace water wall, 1111 Primary superheater,] 15.
・Water supply pump, 116 ・Main steam adjustment jt-5117・
・High pressure turbine, 118a・Intermediate pressure turbine, 118b・
...Low pressure turbine, 119... Generator, 120...
Condenser, 121...spray.

Claims (1)

[Claims] 1. A process predictive control method that includes a built-in process model, uses this model to predict a control amount in the near future, and determines a manipulated variable based on the prediction result. A process characterized by incorporating a model of a subprocess whose state quantity is a controlled quantity and at least one subprocess upstream of this subprocess, and using this model to predict the controlled quantity in the near future. predictive control method. 2. In a process predictive control method that incorporates a process model, uses this model to predict the control amount in the near future, and determines the manipulated variable based on this prediction result, A predictive control method for a process, characterized in that a manipulated variable is determined so that the deviation from a future target value is a predetermined value. 3. The predictive control method for a process according to claim 1 or 2, wherein the predictive control system including the process is a multi-input multi-output control system having a plurality of control variables and manipulated variables. 4. A process predictive control method according to claim 2 or 3, characterized in that the manipulated variable is determined so that the deviation between the predicted value of the controlled variable and a target value in the near future becomes zero. 5. A predictive control method for a process according to claim 1, characterized in that the manipulated variable is determined by performing proportional and integral calculations on the deviation between the predicted value of the controlled variable and a target value in the near future. 6. A control amount prediction system that predicts a control amount in the near future of a process from an operation amount and a control amount for a process to be controlled, a target value prediction system that predicts a target value in the near future of a process, and the predicted target. and a manipulated variable determination system that determines the manipulated variable of the process based on the deviation between the value and the controlled variable. 1. A predictive control system for a process, comprising a built-in model of at least one subprocess on the side, and means for predicting a control amount in the near future using this model. 7. A thermal power plant to be controlled, a steam temperature prediction system that predicts steam temperature as a controlled variable for the thermal power plant in the near future based on manipulated variables and controlled variables, and predicts target values for the thermal power plant in the near future. The steam temperature prediction system includes a target value prediction system and a manipulated variable determination system that determines the manipulated variable of the process based on the deviation between the predicted target value and the controlled variable. It is characterized by having a built-in model of a sub-process, one of which is steam temperature, and at least one sub-process upstream of this sub-process, and comprising means for predicting steam temperature in the near future using this model. thermal power plant system. 8. The predictive control of the process according to claim 6, wherein the manipulated variable determining system has a function of determining the manipulated variable so that the deviation between the predicted value of the controlled variable and a target value in the near future becomes a predetermined value. system, or the thermal power plant system according to claim 7. 9. The predictive control of the process according to claim 6, wherein the manipulated variable determining system has a function of determining the manipulated variable by performing proportional and integral calculations on the deviation between the predicted value of the controlled variable and a target value in the near future. system, or the thermal power plant system according to claim 7. 10. A steam temperature prediction system for predicting the steam temperature of a thermal power plant based on the manipulated variables and control variables for the thermal power plant, a subprocess in which at least one of the state variables is steam temperature, and this subprocess 1. A steam temperature prediction system comprising: a built-in model of at least one sub-process on the upstream side of the steam temperature prediction system; and means for predicting steam temperature in the near future using this model. 11. The thermal power plant system according to claim 7, 8 or 9, or the steam temperature prediction system according to claim 10, wherein the model of the sub-process is a heat exchanger in the plant. 12. The thermal power plant system or steam temperature prediction system according to claim 11, wherein the model of the sub-process is a secondary superheater and a primary superheater in the plant. 13. The thermal power plant system or steam temperature prediction system according to claim 11, wherein the model of the sub-process is a secondary reheater and a primary reheater in the plant.