JPH04352001A - Operation supporting device - Google Patents

Abstract

PURPOSE:To detect a completion condition or error/omission operation due to operation start and to show guidance for misoperation even when a signal directly indicating individual operation is not obtained. CONSTITUTION:An operation supporting device for inferring a scenario to be a procedure sequence necessary for the execution of an applied operation target and outputting the scenario to guide an operation procedure to an operator is provided with a differentiating device 1 for inputting the process variable of a plant 10 through an input device 20 and outputting the differential value of the process variable and the 1st memory 2 for storing a konwledge base for inferring the scenario by using the differential value for the description of a prescribed condition, inputs the process variable of the plant 10 and the differential value outputted from the device 1 and infers the scenario based upon knowledge data stored in the 1st memory 2.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an operation support device which is applied to support the operation of a plant, and which not only outputs operation guides to operators, but also outputs the operation itself as a plant operation signal via an actuator.

0002

2. Description of the Related Art In large-scale plants such as power plants and chemical plants, manuals are prepared for operating the plant in normal and abnormal situations, and standard operations are possible. However, what is written in the operating manual is only the standard operating procedure for typical cases, and operators deal with each individual case while considering the meaning and effects of the operation. That is, it is human ability to perform appropriate driving operations according to the momentary situation, such as changing the execution order of a series of driving operation items, adding or omitting them, etc.

[0003] Therefore, a system has been developed that computerizes the driving instructions and outputs a driving guide according to the situation.In addition to a series of standard driving operations, driving operation items are hierarchized by purpose and function, In some cases, improvements have been made to enable output of a certain degree of flexible driving guidance by allowing for conditional branching and conditional cancellation.

[0004] Patent regarding improved example (Japanese Patent Application No. 110-1983
034, Japanese Patent Application No. 63-163913), an overview of the prior art will be explained. In the following, we will refer to "fulfillment of operational objectives" such as "starting up the plant", "achievement of functions" such as "establishment of filling/extraction flows", and "execution of operational operations" such as "adjustment of filling/extraction source valves". collectively referred to as "execution of procedures".

FIG. 4 is an explanatory diagram of a flexible operating procedure. In other words, the startup operation procedure for a PWR plant is as follows:
Step A: Plant start-up is placed at the top, and this is performed by executing steps A1: Establishment of filling/extraction flow, A2: Start-up of primary coolant pump, etc., which are one step lower. Procedure A
1: To establish the filling/extraction flow, the specific operating steps a: Adjust the filling/extraction flow source valve, b: Open the filling high pressure injection pump source valve, c.
This is accomplished by: starting the filling high-pressure pump, d: adjusting the filling flow rate, and e: adjusting the extraction line pressure. A2: The primary coolant pump is achieved by performing f: startup check, g: sequential startup of the primary coolant pump, and h: extraction line pressure adjustment.

However, at this time, the most standard method is to perform in the order of abcdefgh, but depending on the case, it may be performed in the order of basedfgh, or e
As a result of the extraction line pressure adjustment in h, if that state is maintained, the extraction line pressure adjustment in h may be omitted. In Figure 4, the symbol "→" indicates something that must be executed in the order of execution, such as A1→A2 or f→g.
The symbol "=" is used to indicate that either one can be executed first, such as a=b or d=e. What is shown in FIG. 5 is an example of a data format when the driving operation procedure illustrated in FIG. 4 is expressed as knowledge data to be handled by a computer. (1) The procedure name is the name of the procedure enclosed in a rectangle in FIG. (2) The start condition is data for determining the situation in which execution of the procedure should be started. (3) Cancellation conditions are data for determining the situation in which execution of the procedure should be canceled. (4) Completion conditions are data for determining that execution of the procedure has been completed. (5) Incorrect operation condition means that the procedure was executed incorrectly (=
This data is used to determine whether the start condition was started even though the start condition was not met. (6) Standard required time is the standard required time from when the start condition of the procedure is met until the completion condition is met. (7) The upper procedure is the name of a procedure one level higher than the procedure. (8) The lower procedure is the name of a procedure one level lower than the procedure. However, the lower steps can be selected according to the situation using branch conditions. However, *() indicates that the data may be omitted.
()+ indicates that any number of the data may be written. The inference algorithm used to perform inference using the data format shown in FIG. 5 is as follows. (a) Set driving goals. The settings may be input by an operator, or may be automatically set by another operating state grasping device or abnormality diagnosis device, and any method may be used. (b) Select a knowledge base.

[0007] Search for the one that matches the driving goal set in (a) from among the plurality of knowledge bases created for each purpose, add the name of the highest level procedure to the working memory, and add status information ``Waiting for execution''. Set. (c) Determine cancellation conditions.

[0008] Regarding the knowledge data corresponding to the procedure name stored in the working memory with state information of "waiting to be executed", a cancellation condition is determined, and if the condition is true, the state information of the procedure is set to "cancelled". Change to (d) Determine starting conditions.

[0009] With respect to the knowledge data stored in the working memory with state information of "waiting for execution" added, the start condition is determined, and if it is true, the state information of the procedure is set as "waiting for completion"; (a) Outputs guidance to the operator to carry out the procedure. (b) Furthermore, the branch condition is determined, and for those that are satisfied, the lower steps are extracted and written into the working memory together with the "waiting to execute" status information. (e) Determine completion conditions. Regarding the procedure stored in the working memory with status information of "waiting for completion" added, the completion condition is determined, and if the condition is true, the status information of the procedure is changed to "completed". (f) Determine erroneous operation conditions. Regarding the procedure stored in the working memory, an erroneous operation condition is determined, and if the condition is true, guidance is output to the operator to the effect that the operation was performed in error. (g) Determine a missing operation.

[0010] For a procedure in which the completion condition is not met even after the standard required time has passed after the state information changes from "waiting for execution" to "waiting for completion," guidance is output to the effect that the operation has not been performed.

FIG. 6 shows an example of the contents of the working memory when inference is made using the above knowledge base and inference algorithm. An example of this is the plant start-up shown in Figure 4.
This is the state when step d has been completed. The reason why A and A1 are waiting for completion is because lower procedures A2 and e have not been completed, respectively. Furthermore, f, g, and h are not extracted because of the start condition (=A1
This is because the requirements (completion of

0012

[Problem to be solved by the invention] In the above-mentioned determination of completion conditions, determination of erroneous operation conditions, and determination of missing operation, a signal that directly indicates whether or not the operator has performed (started) each operation. However, it is not always possible to obtain all of them due to economic and technical reasons. In such cases,

(1) Since no clear evidence indicating completion of the procedure can be obtained, it is not known whether the operator actually performed the operation or not. Also, due to this, in order to progress the scenario, the completion condition has to be unconditionally completed,
The correct driving progress cannot be grasped.

(2) Even if the operator performs an incorrect operation or forgets to perform an operation, it cannot be detected, and therefore no guidance can be provided regarding the erroneous or missing operation. There was a problem.

[0015] For this reason, the mental burden on operators involved in monitoring and making judgments is large, or operators with little experience may perform redundant operations or standby, or in extreme cases, this may lead to unnecessary plant trips. There was a risk that the plant's operating rate would decrease.

The present invention has been developed in view of the above points, and even when a signal directly indicating an individual operation cannot be obtained, it is possible to detect completion conditions and erroneous/missing operations due to the start of an operation, The purpose of the present invention is to provide a driving support device that can provide guidance for erroneous operations.

[0017]

[Means for Solving the Problems] The present invention provides a driving support system that infers a scenario, which is a sequence of procedures that need to be carried out for a given driving target, and outputs the scenario to guide the driving procedure to an operator. In the device, a differentiator that inputs the process amount of the plant and outputs the differential amount, and a knowledge base for inferring the above scenario, which stores the knowledge base in which the differential amount is used to describe the predetermined conditions. and inference means for inputting process quantities of the plant and differential quantities output from the differentiating means and inferring the scenario based on knowledge data stored in the memory means. .

[0018]

[Operation] According to the above configuration, when inferring a scenario,
By using the differential amount output from the differentiating means, it is possible to detect and judge whether or not an operation has been initiated or not, even if a signal directly indicating the initiation of an operation cannot be obtained.

[0019]

DESCRIPTION OF THE PREFERRED EMBODIMENTS A driving support system according to an embodiment of the present invention will be described below with reference to the drawings. (1) Equipment configuration

FIG. 1 shows an example of the device configuration of the present invention. In FIG. 1, the differentiator 1 inputs the process amount of the plant 10 via the input device 20, and outputs the differential amount to the inference mechanism 3. The inference mechanism 3 receives the outputs of the input device 20 and the differentiator 1 as input, infers the scenario and determines erroneous/missing operations based on the knowledge base stored in the first memory 2, and stores the results in the working memory 4. The output device 30 further provides driving guidance through the output device 30. The working memory 4 is a memory for storing data temporarily required for processing. (2) Differentiator 1 The differentiator 1 is configured by, for example, arranging incomplete differential elements (d/dt)X=[s/(1+Ts)]X in a ladder shape. An example of the configuration of the differentiator 1 is shown in FIG. In the example shown in Figure 2, by connecting incomplete differential elements in series, 2
Order differential coefficient (d2 /dt2)X

[0021] is output. By using the output signal of this differentiator 1, even when a signal directly suggesting the start of an operation cannot be obtained, the start of the operation and the validity can be detected.
can be judged. The simplest example will be the opening and closing of a valve. Even if the valve opening signal cannot be directly measured, if the downstream flow rate can be measured, the validity of the operator's initiation of the operation can be determined as follows. Let the flow rate be X, and an arbitrarily determined threshold e1,
Regarding e2, within the standard required time after issuing the guidance, (a) If d2 X/dt2 > e1, the flow rate increases → the valve opening operation is completed. (a) If d2 X/dt2 < e2, the flow rate decreases → valve close operation (incorrect operation). (c) If the above (a) and (b) do not appear within the standard required time, it is an incomplete operation. (3) Contents of first memory 2

[0022] The first memory 2 stores knowledge data for expressing the driving operation procedure described above as a prior art. An example of the storage format of the knowledge data is the same as that already shown in FIG. However, differential quantities can be used to describe completion conditions and error conditions.

FIG. 3 shows an example of knowledge data. In the example of FIG. 3, the procedure "A valve open" may be started unconditionally, the completion condition is that the second derivative value of the outlet flow rate of A valve becomes equal to or greater than the threshold value e1, and the cancellation condition of this procedure is without (unconditionally false),
If the second derivative value of the A valve outlet flow rate is less than the threshold value e2, it can be determined that this procedure has been operated by mistake, and the standard time required for the procedure due to the incorrect operation is 10 seconds, and the upper step of this procedure is "securing the flow rate of system A", indicating that there are no lower steps. (4) Contents of Working Memory 4 The working memory 4 stores the progress status of operating procedures in the form of individual procedure names and their status information. The status information includes the following: "Waiting for execution": State where the start conditions are not met "Waiting for completion"
: Outputs guidance to the operator when the start conditions are met,
The state of waiting for the completion of that procedure to occur. "Completed": A state in which the completion conditions for the procedure are met and completion is confirmed. "Cancelled": The cancellation condition has been met and the execution of the procedure has been canceled. A storage example of the working memory 4 is similar to that introduced in FIG. 6 as a conventional technique. (5) Inference algorithm The algorithm outlined in the prior art is used. The algorithm is shown below. (a) Set driving goals. Here, for the sake of simplicity, it is assumed that the driving target is set by the operator. The operator inputs the driving goal to the inference mechanism 3 via the input device 20. (b) Select a knowledge base.

Search the first memory 2 for a knowledge base that matches the driving goal set in (a) from a plurality of knowledge bases created for each purpose, and store the name of the highest level procedure in the working memory 4 as status information. Add and set "Execution Waiting". (c) Determine cancellation conditions.

Regarding the knowledge data corresponding to the procedure name stored in the working memory 4 with status information of ``waiting to execute'', the cancellation condition is determined, and if it is true, the status information of the procedure is changed to ``cancelled''. ”. (d) Determine the starting conditions.

Regarding the knowledge data stored in the working memory 4 with state information of "waiting for execution", the start condition is determined, and if it is true, the state information of the procedure is set as "waiting for completion", and (a ) Output guidance to the operator to carry out the procedure. (b) Further, the branch condition is determined, and for those that are satisfied, the lower steps are taken out and written into the working memory 4 together with the "wait for execution" status information. (e) Determine the completion conditions.

Regarding the procedure stored in the working memory 4 with status information of ``waiting for completion'' added, the completion condition is determined, and if the condition is true, the status information of the procedure is changed to ``completed''.
Change to (f) Determine erroneous operation conditions. Regarding the procedure stored in the working memory 4, an erroneous operation condition is determined, and if the condition is true, guidance is output to the operator to the effect that the operation has been performed in error. (g) Determine missing operations.

[0028] For a procedure in which the completion condition is not met even after the standard required time has passed after the state information changes from "waiting for execution" to "waiting for completion," guidance is output to the effect that the operation has not been performed.

[0029]

As described above, according to the present invention, even if a signal directly suggesting an individual operation cannot be obtained for economical or technical reasons, it is possible to prevent errors or omissions from occurring due to completion conditions caused by the start of an operation. It becomes possible to detect operations and provide guidance for erroneous operations. As a result, the mental burden on operators involved in monitoring and making decisions is reduced, and there is a possibility that inexperienced operators may perform redundant operations or standby, or lead to unnecessary plant trips, reducing the plant's operating rate. is reduced.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing a configuration according to an embodiment of the present invention.

FIG. 2 is a diagram showing the configuration of a differentiator according to the same embodiment.

FIG. 3 is a diagram showing an example of knowledge data storage in the same embodiment.

FIG. 4 is a diagram for explaining a conventional driving operation procedure.

FIG. 5 is a diagram showing an example of a conventional knowledge base expression format.

FIG. 6 is a diagram showing the contents of a conventional working memory.

[Explanation of symbols] 1... Differentiator, 2... First memory, 3... Inference mechanism, 4...
working memory, 10...plant, 20...input device,
30...Output device.

Claims (1)

[Claims] Claim 1: An operation support system that infers a scenario, which is a sequence of procedures that need to be carried out for a given operation target, and outputs the scenario to guide operating procedures to an operator. a differentiating means for inputting the input and outputting the differential amount thereof; a memory means for storing a knowledge base for inferring the above scenario in which the differential amount is used to describe the predetermined condition; A driving support device comprising: inference means for inputting a process quantity and a differential quantity output from the differentiating means, and inferring the scenario based on knowledge data stored in the memory means.