JPS6129901A - Control method of thermal power plant - Google Patents

Abstract

PURPOSE:To decrease the effect of a fault of a controller given on the plant operation by arranging a sub-loop controller to a control module of a plant equipment, a master controller to a high-order position and a supervisory controller to a further high-order position. CONSTITUTION:A plant is classified by taking notice on plant characteristics, arranging a drive control module DCM and a signal conditioning module SGM corresponding to a device 40 as the lowest-order control, arranging the sub-loop controller SLC corresponding to a small process to the high-order, arranging a master controller MC corresponding to a process/sub-process to the further high- order to attain control while taking mutual cooperation between the process and the sub-process. Then the control system is closed by providing a supervision controller SVC to the highest position. Thus, the redundancy due to multiplexing is eliminated and the effect of a fault of controller given on the plant operation is reduced to the utmost.

Description

[Detailed Description of the Invention] [Field of Application of the Invention] The present invention analyzes the characteristics of the plant in a thermal power plant, and optimally arranges control devices in a hierarchical manner in accordance with the characteristics, and controls each control device. The present invention relates to an automatic control method that allows automatic operation to continue without stopping the plant even in the event of an abnormality.

[Background of the invention]

Conventionally, control devices in a thermal power plant are generally arranged in groups of auxiliary machines that make up the plant. For example, it is a water supply control device that controls a group of water supply pumps that supply water to a boiler. In this case, failure of the control device will make it difficult to continue automatic operation of the plant, so
Techniques such as duplication of control devices have been adopted as a measure to improve reliability.

Below, we will discuss the concept of system division based on plant characteristic analysis, the conventional method of arranging control equipment within it, and the problems associated with it.

Figure 6 shows system division focusing on the plant characteristics of a thermal power plant.

Thermal power plants involve a complex combination of various functions.
Combustion process 1. is a collection of complex processes, as shown in FIG. It can be organized into four processes: water/steam process (n), generation process (2), and cooling process (2).

The purpose of controlling a thermal power plant is to combine these four processes and cooperate with each other to stably supply power in response to external power demand.

Each process shown in FIG. 6 can be further classified into a plurality of sub-processes by focusing on its functions. Figure 2 shows the further classification of the water/steam process (■) into sub-processes. , In the figure, [10 is the condensate subprocess, ■1
1 is a low-pressure feed water heating/deaeration sub-process, ■12 is a water supply sub-process, [3 is a high-pressure feed water heating sub-process, ■14
is the evaporation/superheating subprocess, and ■15 is the reheating subprocess.

The sub-processes include multiple small processes of equal capacity as necessary to strengthen the plant's capacity.

Furthermore, small processes can be further subdivided into units of equipment.

Figure 3 shows the division of the water supply subprocess of I[12 into small processes and equipment. [12A to 12c each indicate a water pump small process. The equipment in the small process is the water supply pump u12Al, the water supply pump/pull transmission [12A2, the water supply pump motor n12A3, the water supply booster pump I [
12A4, water booster pump motor l112A5, water pump outlet valve 112A6, water pump recirculation valve [12
Enter A7.

To summarize the above, the configuration of a power generation plant can be divided into process, subprocess, small process, and equipment levels as shown in FIG. If a failure occurs at the process level or subprocess level of the plant, it is impossible to continue the plant operation, and the plant operation is often shut down. However, if a failure occurs at the small process level or equipment level, it should be possible to continue operation by reducing the plant output.

However, in the past, the above-mentioned plant analysis was insufficient and the results were not reflected in the control device configuration. As shown in FIG. 5, an example of water supply sub-processes spans a plurality of small processes n12A to II'12G.
and water supply control sequencer 31 for 0N-OFF control equipment.
are placed.

In this case, a failure of the control device 30 or 31 makes it impossible to continue automatic operation of the small processes 1112A to 112C. This indicates that it is impossible to continue the automatic operation of the full rate process of the water supply sub-process 12, and poses a serious hindrance to the continuation of the zigrant operation. Therefore, in order to strengthen reliability, the system configuration of the control device was complicated, requiring redundancy such as duplication of the control device and nil backup. An example of these is the Special Publication No. 51-3774.
There are many such as No. 4.

[Purpose of the invention]

In light of the above, an object of the present invention is to provide a method for controlling a thermal power plant that can reduce the influence of control device abnormalities on plant operation as much as possible without duplicating control devices as much as possible. It is in.

[Summary of the invention]

The % characteristic of the present invention is to classify plants by focusing on their plant characteristics, and place a DCM ()"live control module) SGC (signal conditioning 7 joules) corresponding to the equipment as the lowest control, and a small A subloop control device is placed corresponding to the process, and the process is placed above it.

A master control device is placed corresponding to each sub-process to perform mutually coordinated control between processes and sub-processes.
A supervisory control device is installed at the top level, and all of these control devices are configured as a single system.

[Embodiments of the invention]

This basic idea is shown in Figure 1. 40 in the figure is equipment,
41 represents a small process, and 42 represents a subprocess. 5GC
Fi, which performs input signal processing of data from the process. The DCM is installed on a device-by-device basis and operates each device using control signals or manual commands. These 8 GCs and DCMs are the smallest units in the entire system and perform manual operations, equipment protection, and restrictions. The SLC is a sub-loop controller, which is arranged to correspond to a small process by putting together the above-mentioned DCM and SGC, and performs analog control and sequence control. MC is a master control device that supervises a plurality of subloop control devices MC and controls coordinated control between processes and subprocesses. S.C.
v is a monitoring and control device, comprehensive plant monitoring, man-machine interface, data logging, and document creation.

It performs high-performance control such as downloading control programs, modifying programs, and predictive control.

FIG. 7(a) shows an example of a control system for a drum boiler power generation plant configured in accordance with the above basics. The control device is composed of a supervisory control device SVC, a master control device MC, and 15 subloop control devices SLC. The master control device MO has a system frequency F, a main steam pressure MSF
, economizer outlet gas 02 GO2, 2 air flow rate AIR, furnace draft FD, drum level DL, main steam temperature MA
Control variables determined by mutual relationships among multiple processes and sub-processes, such as T and reheat steam temperature R8T, are controlled while coordinating with each other. Subloop control device S
LC power generation lid control (8LC1), mill control (SL0
2-8LC7), air amount control (SLC8, 9), water supply amount control (SLCIO-12).

It performs control corresponding to each small process such as desuperheater outlet temperature control (SLC13, 14) and gas distribution amount control (SLC15). FIG. 7(b) shows an example of a once-through boiler. As is clear from comparing these, the input signals and control circuits are somewhat different. However, in any case, the master controller MC is the subloop controller SL.
A target signal of C is given. Figure 7 The circuit can be easily explained in the same figure (
To explain using the example of a once-through boiler in b), the power generation command 1 from the central power supply command is corrected by fluctuations in the system frequency F, runback when a large auxiliary equipment trips, and the rate of change is limited by the plant. Command value 2 for the turbine is set to 5
Output to LCI. This command value 2 is also set as the boiler input amount command value 2, and a correction is made to this to make the main steam pressure MSP constant, and it is set as the water supply amount command value 3. The water supply amount command value 3 is compared with the water supply amount, the deviation is calculated by PI, and the resultant water supply amount command value 3 is outputted to 8LC10 to 8LC12 as the water supply amount command value 4 of each water pump. Furthermore, the ratio of the water supply amount and the fuel amount is adjusted so as to keep the main steam temperature MAT constant, and the command value 6 for the fuel amount is created, and the command value 8 for the air amount is 5LC8, Output to 9.

The coal feed amount command 6 is compared with the calorie-corrected total fuel amount, the deviation is calculated by PI, and is output to 5L02 to 5L07 as the mill master command value 7 for each mill.

Also, main steam temperature and furnace draft based on set values.

Set the main steam temperature command value 5 to keep the reheat steam temperature constant.
Transfer the furnace draft command value 9 to LC13 and 14 to 5LC8,
9 and outputs the reheat steam temperature command value 10 to 5LC15.

Above main turbine, water supply amount, fuel amount, air amount.

Main steam/reheat steam temperature, command value 1 for furnace draft
10 is the master controller MC.

The sub-loop controller 8LC is a part that outputs control commands to the operating end of the control pulp, Bontrol drive, etc. according to these command values 1 to 10. D
The CM outputs control commands from the subloop controller to the operating end, and also configures a manual operating circuit.
During manual operation, the output of the subloop controller is switched and a manual operation command is output. In the case of a drum boiler, a command value corresponding to the water supply amount command 3 in the case of once-through flow is output as the fuel amount command 6'. In addition, the water supply amount command 3' is output by adding correction signals for the main steam flow rate and main water supply flow rate to the drum level command signal 3'' that keeps the drum level constant.Other than that, it is the same as the once-through boiler. Output.

While FIG. 7 shows the system configuration with a focus on division of control functions, FIG. 8 shows the hardware configuration.

The symbols are the same as used in FIG.

50 is a serial data transmission loop, which includes a plurality of sub-loop control devices SLC, a master control device MC,
It organically connects the supervisory control device SVC.

In this figure, 200 in 8VC is the printer, 201°206
is a display, 202 is a keyboard, 203 is a floppy disk, 204 is a system console, 205 is a host controller used for predictive control, etc. $p, 20
7 is a monitoring controller. In addition, 208 is an SLC group, and 208-1 is an SLC group for water supply control, for example.
, 20B-2 is a SLC group for mill control, 208-3
is an SLC group for air volume control. One SLC is provided with multiple 8GCs and DCMs, and there is an I10 bus 6 between them.
Combined with 0. Note that the master control device MC does not have a DCM, and a plurality of 80Cs are coupled by a bus 60. The signal conditioning module 8GC detects state quantities of the power generation plant 100, and the drive control module DCM drives equipment in the power generation plant 10o. As described above, the control device of the present invention has a hierarchical structure as shown in FIG. 6, and as shown in FIG.
The operation amount of the m unit 40 is output.

FIG. 9 shows the circuit portion related to one 8LC. 5
0 is a serial data transmission loop, which connects 1 subloop control device SLC to other control devices MC and 8LC. 60il''tI10 bus with SLC and multiple SGCs.

It combines DCM. Subloop control device SL
C is a processor consisting of BPU, memory, etc.
3 SGCs and 44 DCMs are a kind of I10 device.

Details of these 80C and DCM will be described later.

What is important here is that even if the sub-loop controller 8LC, which is a processor, fails, the signal that the SGC takes in from the detection end of the plant 100 is transmitted to the dedicated protection circuit 61., which is provided separately, regardless of the failure. It is possible to protect the plant 100 by passing the data to the DCM via the DCM and operating the operating end as necessary. The upper subloop controller SLC has this protection function.

Even if the master control device MC is a single system, it is possible to construct a 7-stem system without damaging the plant.

Next, the details of the function of j5sGc will be explained with reference to FIG.

In the figure, the SGC performs disconnection detection on the signal from the detection end 70 in block 71, performs signal processing such as nonlinear correction, square root, addition and subtraction in block 72 as necessary, and performs signal processing in block 73.
performs signal distribution. Distribution destinations include those to the upper control device such as the subloop control device SLC via the I10 path 60, and those to the indicator, DCM, etc. without going through the I10 bus 60, and even in the event of a failure of the upper control device, the I10 path 6
The most important feature is that the process signal can be extracted, monitored, and controlled without going through 0. Similarly, when the host controller fails, the process signal is monitored at block 74 and I10
It is characterized by being able to output contact outputs for alarm and protection to the outside without going through the bus 60, and providing autonomy to the monitoring and protection device.

The details of the function of the DCM 17) for analog control will be explained with reference to Figure il1. A command signal from the host controller SLC or the like input from the I10 bus 60 normally drives the operating end 83 via a switch 80, an automatic/manual switch 81, and a voltage/current converter 82. The operation signal is from the abnormality diagnosis circuit 8
4, it is checked against the operating end position feedback signal 85. This check result, the automatic/manual status of the operating end 83, the operating end fully open/fully closed limit)S
The input from W86 is lamped by lamp logic circuit 87 to manual operation station 88A. 89 is automatic/
Manual switching circuit 90 is an open/close logic circuit, opening/closing operation by manual operation station 88A and protection circuit 61
The forced opening/closing signal 212 from the analog memory 92,
The operating end 831 can be opened and closed via the automatic/manual switch 81. This circuit allows manual operation even if the upper control device fails and control commands from the I10 bus 60 are no longer input.
D: The operating end can be opened and closed by the operation of the protection circuit.
This is a characteristic of commercials. Another feature of the DCM is the backup circuit 93. That is, in the event of a failure in the host controller 8LC, the process feedback signal 211 taken in via the SGC is held in the holder 94 and set as a set value.
It has a function of comparing the process feedback signal with the comparator 95 and controlling the process feedback amount to the value immediately before the occurrence of the upper fault via the proportional-integral calculator 96 and the switch 80.

The above SGC and DCM functions ensure manual operation even in the event of a failure of the upper control device (subloop control device 5LC).
This makes it possible to have a single system configuration that can ensure protection functions and a minimum process feedback control function.

Further, the details of the function of the DCM for sequence control will be explained with reference to FIG. The automatic start and start/stop commands from the higher-level control device 8LC given from the I10 bus 60 are condition-judged by the start/stop logic circuit 130, and the automatic/manual switching circuit 10.
Automatically when 1 is automatic, t;&PB station 88D
When manual is selected in
Perform ON10 FF of 04. Further, it is stopped by a command from the protection circuit 61. The diagnostic circuit 10 is operated according to the feedback signal 106 from the switchgear 104 and the operation command.
7 to perform abnormality diagnosis. The diagnostic results and feedback signals are displayed on the IB station 103 by the lamp logic circuit 108. D even for sequence control
A major feature of the CM is that even in the event of an abnormality in the upper control device (subloop control device), it is possible to ensure start/stop functions by manual operation and forced stop functions by protection logic. Note that in the figure, AND is a logical product, and 0 is a logical sum.

An autonomous control method will be described in which the reliability of the system is improved by causing the lower-level control device to degenerate and back up part of the control functions of the higher-level controller in the above configuration when the higher-level controller fails.

FIG. 13 shows a case where both the master MC and subloop SLC are normal. The details of the control functions of the master MC will be explained. The system master main control 110 is an essential control regarding the cooperative control of the sub-loop. System master advanced control 111 is related to the subloop control and is provided to improve its controllability.

The system master auxiliary control 112 is a restriction circuit related to the subloop control, a control circuit for starting and stopping, and the like. In the master, from these controls, load commands, etc., commands 113 for coordinating the subloop controls are created, and the subloop control device 5LC-A.

5LC-B, 5LC-C.

The details of each subloop control device 8LC can be classified into a subloop main control 114 which is essential as subloop control, a restriction 115 which is additional control, and a nonlinear correction 116.

In the present invention, the functions are divided as shown in FIG. 13, and the characteristic feature here is that the subloop control device SLC is a processing section 110' similar to the system master main control section 110 of the master control device MC in the upper system. The drive control module DCM has a subloop control device SL of its upper system.
It has a processing section 114' similar to the subloop main control 114 of C. 110' or 114' is always 11
It is only receiving the same input as 0 or 114, and more specifically, it is only waiting.

FIG. 14 is an explanatory diagram of functional degradation backup control by the subloop control device SLC when the master control device MC fails. 110'A has the same function as the system master main control 110 in FIG. 13 provided in the subloop control device 5LC-A. When the master control device MC fails, the subloop control device 5LC-A executes only the essential 110 functions among the control functions of the master control device MC at 110' and takes the place of the master MC, thereby subloop control 8LC-C.

5LC-B is also given a command, and subloop 5LC-A-8
Maintain coordination between LC-0. This is a functional degeneracy backup function by the subloop control device SLC when the master control device MC fails.

Next, FIG. 15 shows a function degradation backup by the DCM when the subloop control device 5LC-A also fails. 114' is a function equivalent to the main control function 114 in the subloop control device SLC provided in the DCM, and when a failure occurs in the subloop control device 8LC-A, the DCM can only execute the main control function of the subloop. It becomes possible.

The above is the basic principle of an autonomous control method using a highly reliable heavy-system control system based on the optimal hierarchical distributed arrangement of control devices based on plant characteristics and functional degradation backup.

An embodiment of the functional degradation backup function of the present invention will be described below, taking water supply subprocess control as an example.

FIG. 16 shows the water supply sub-process [12+7) control during normal operation. The detection signals of master control device M-C are each 5aC.
-@ is input through the I10 bus 60.

P IIIT is the pressure after the first stage of Kubin and is converted by the filter function generator 141 into υ main steam flow rate MSQ.

MST is the main steam temperature and the sb function generator 143 and multiplier 1
44, temperature correction is performed on the main steam flow rate ¥SQ, which is the output of the steamer 141. DL is compared with the setting value of 147 by the sb comparator 146 at the drum level, and after nonlinear gain correction is performed by the function generator 148, it is input to the proportional integrator 149. Main steam flow rate M
It is added to SQ and becomes the command value of the total water supply flow rate. M8F
is the total feed water flow rate and multiplier 15 is calculated by sb feed water temperature TSF.
3, the temperature is corrected, the comparator 154 compares it with the command value, and the function generator 155 and multiplier 156 calculate the main steam flow rate M8Q.
The gain correction is applied to the proportional integrator 157, and its output becomes a water supply flow rate command for each water supply small process. This command value is transmitted to each water supply subloop control device 8LC-A, 8LC- via the serial data transmission loop 50.
B.

5LC-C. ll0A' is a degenerate backup function provided within the subloop control device SLC.

Commands from the master MC are sent via 159 switchers to 16
51 through the p comparator 161A through the upper limit circuit for 0 people.
It is compared with the flow rate feedback signal of the water supply pump 163 detected at 62A, and subjected to gain correction by a function generator 164A and a multiplier 165A.The signal is input to a proportional integrator 166. The output is non-linearly corrected by the function generator 167 and given to the DCM-A via the I10 bus 60 as a pump rotation speed command.

114A' is a degeneration pull-up function provided in the DCM-A. The rotation speed command is provided by the switch 169A.

171A to the pump transmission 172 to control the speed.

FIG. 17 shows an example where the master control device MC fails. The water supply sub-loop controller 8LC-person's degenerate backup function 110A' operates and the sub-loop controller 5LC-B operates in place of the master.

A water supply command is output to 5LCL-C. 5LC - The person inputs the drum level signal 145, compares it with the set value 180A using the comparator 181, and sends the output to the proportional integrator 182, which output becomes the water supply command for each pump and the switch 15QA.
one for automatic water supply command, and the other for water supply subloop 8LC- via serial data transmission lunip 50.
B.

5LC-C water supply command. In this way, the water supply subloop 5LC-A cooperates with 8LC-B and 8LC-C to continue the citrus level control.

The switching of the switching device 159A is performed by constantly monitoring the water supply command signal from the master control device MC with the subloop controller 5LC-A, and when the signal disappears.

FIG. 18 shows a case where the water supply subloop 5LC-A also fails when the master control device MC fails.

In this case, the control function of the master MC is the water supply subloop 5.
The degenerate backup function 110B' in LC-B takes over the command value, and the command value is given to the water supply subloop 5LC-C via the serial data transmission loop 50.
It becomes possible to control the drum level in coordination with 5LC-C.

Further, the DCM-A disposed in the temporary water supply subloop 8LC-A can perform feedback control of the pump flow rate by operating the degenerate backup function 114A' provided therein. That is, the pump flow rate signal from the detector 162A, which is given from the SGC to the DCM 2A, is compared with the set value 190A by the comparator 191A, and given to the proportional integrator 192A, which then outputs the pump speed command via the switch 169A to 172A. It is applied to the transmission of the pump to control the speed.

Also, as in the case of the subloop controller, the switching device 169A is switched by constantly monitoring the water supply flow rate command signal for the water supply small process from the subloop controller SLC using the DCM, and when the signal stops coming. .

FIG. 19 shows an example of process input signals necessary for the subloop control device 8LC of each system to degenerate and back up a part of the control function of the upper MC when the master control device MC fails. By configuring the system master main control circuit in the subloop control device 8LC using this process signal as the control target signal, even if it is not possible to follow fast load changes like when the master control device MC is normal, it is possible to Slow load changes create a fist that can be used to follow. For example, in the case of a drum boiler, the process signals used to achieve the above objectives are the drum level for the water supply system, the main steam pressure for the fuel system, the gas 02 for the air system, the furnace draft for the gas system, and the reheat steam pressure for the rich circulation system. It is.

On the other hand, when the subloop control device SLC fails, cascade control using the command signal of the subloop control device is not possible, but the process signal of the operating end controlled by the DCM is taken into the DCM, and this process signal is used as the control target signal to be used as the subloop main control. By configuring the control circuit, fixed value feedback control based on the set value in the DCM can be performed, and mica process disturbance can be absorbed in each operating end. FIG. 20 shows an example of a process signal taken into the DCM at the time of 8LC failure.

For example, in the water subloop, the flow rate of each water pump, in the fuel subloop, the coal flow rate of each mill, the primary air flow rate,
In the air, gas, and recirculation subloop, each fan discharge pressure is the process signal to achieve the above objectives.

Furthermore, the normal sub-loop control device SLC is assigned to each of a plurality of similar operation end groups for each system.

(For example, 3 groups for water supply system, 6 groups for fuel system.

For this reason, even if one SLC fails and the operating end controlled by the related lower DCM becomes fixed value feedback control based on the setting value in the DCM, the remaining normal SLC side The operating end is capable of cascade control, which allows it to follow even slight changes in load.

〔Effect of the invention〕

As described above, according to the present invention, even if the power plant has a hierarchically distributed/duplex configuration system, plant operation can be continued even in the event of a control device failure, and an economical and highly reliable control system can be achieved. It has the effect of being able to construct.

[Brief explanation of drawings]

Figure 1 is a diagram showing the basic concept of the control device arrangement in the present invention, Figure 2 is a diagram showing the configuration of the sub-process in the water/steam process, and Figure 3 is a diagram showing the configuration of the sub-process in the water/steam process. Figure 4 is a diagram showing the process configuration of a thermal power plant. Figure 5 is a diagram showing the relationship between conventional processes and control equipment arrangement. Figure 6 is a diagram showing the relationship between the conventional processes and the arrangement of control equipment. Diagram showing the overall process configuration of the plant,
Figure 7 is an example of a control system configuration based on the concept of the present invention, Figure 8 is its hardware configuration, Figure 9 is a diagram showing the basic system configuration in the present invention, and Figure 10 is a functional diagram of SGC C. Figure 11 is a functional diagram of the DCM for analog control, Figure 12 is a functional diagram of the DCM for sequence control, and Figures 13-1.
Figure 5 is a diagram showing the concept of a backup method in the event of a control device failure in the present invention, Figures 16 to 18 are diagrams showing an embodiment of the present invention using a water supply system as an example, and Figures 19 and 20 are FIG. 7 is a diagram illustrating an example of a process signal necessary for backing up a lower controller when a higher controller fails. 43...SGC, 44...DCM, 45...Sub loop control device, 46...Master control device, 47...
・Monitoring control device, 61...protection circuit, 60...I1
0 bus, 50... MΣNET, 110... System master main control, 111... System master advance) ilt
lJlgl, 112... System master auxiliary control, 113
...Coordination, 114...Subloop main control, 115.
...Restriction, 116...Nonlinear third mouth ■ -■

Claims (1)

[Claims] 1. Input the load request signal and a plurality of plant values from the thermal power plant, and calculate the power generation amount, water supply amount, fuel amount, air amount,
a master controller that outputs command signals for at least two of the recirculated gas quantities; a plurality of sub-loop controllers provided independently for each of said command signals; a plurality of sub-loop controllers that provide manipulated variables for controlling the operating end; a plurality of drive means provided for each manipulated amount from each sub-loop controller to drive the operating end; detecting each plant value and controlling the master control device or A method for controlling a thermal power plant comprising means for detecting a plurality of plant values to be delivered to a subloop controller, the subloop controller receiving one of the one or more plant values for determining a command signal. A method for controlling a thermal power plant, comprising: performing closed loop control using the plant value as a feedback value when a master control device is stopped. 2. Input the load request signal and multiple plant values from the thermal power plant, and calculate the power generation amount, water supply amount, fuel amount, air amount,
a master controller that outputs command signals for at least two of the recirculated gas quantities; a plurality of sub-loop controllers provided independently for each of said command signals; a plurality of sub-loop controllers that provide manipulated variables for controlling the operating end; a plurality of drive means provided for each manipulated amount from each sub-loop controller to drive the operating end; detecting each plant value and controlling the master control device or A subloop controller comprising a plurality of plant value detection means to be delivered to a subloop controller, a plurality of operating ends arranged in parallel to achieve the command signal, and independently provided to achieve the command signal. is a control method for a thermal power plant, in which at least as many sub-loop controllers are provided as the number of parallel operating terminals, and each sub-loop controller determines the operating amount of each operating terminal, and is provided in order to achieve one command signal. At least one subloop controller of the plurality of subloop controllers receives one of the one or more plant values for determining the command signal, and receives the plant value when the master controller is stopped. The sub-loop controller that receives the input performs closed-loop control using the plant value as a feedback value, and also provides a signal corresponding to the deviation between the plant value and a predetermined set value to other sub-loop controllers in place of the command signal. A method for controlling a thermal power plant characterized by: 3. In the method for controlling a thermal power plant according to paragraph 1 or 2, when the upper subloop controller is stopped, the drive means holds the operating end according to the amount of operation from the subloop controller immediately before the stoppage. A method for controlling a thermal power plant, characterized by: