JPH09288512A - Plant state visualized system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To detect the minor abnormality occurrence event even in an operation situation where an abnormality alarm is not generated, to supply a process state corresponding to the situation and to support the situation judgment of an operator by providing a specified abnormality judgment processing part, a knowledge data base and a transmission route estimation part. SOLUTION: An input processing part 3 inputs an observation signal in a plant 1. The abnormality judgment processing part 4 periodically judges whether the observation signal is deviated from a plan value or not. The knowledge data base 5 stores a constant causal relation between the observation signals and information whether the causal relation is the relation through the characteristics of various control systems and interlock in the plant 1 and the various units single bodies constituting the plant 1 or not. The transmission route estimation part 6 estimates an abnormal transmission route by using stored causal relation information when the observation signal is judged to be deviated from the plan value. A state display part 7 displays the estimated transmission route on a display means.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention automatically recognizes the operating state of a large-scale plant such as a nuclear power plant, a thermal power plant, and aroha chemical plant, and provides the operating state to the operator as information. Plant visualization system.

[0002]

2. Description of the Related Art For a large-scale plant such as a nuclear power plant, in order to continue stable and safe operation, various process states are provided to an operator not only during normal operation but also when some failure occurs. A computer-based monitoring and control system is provided.

Further, the operator judges whether or not the plant is operating as planned based on the process state provided by the supervisory control system, and if a plant failure occurs, the operator can quickly determine the occurrence status. It is required to judge and take appropriate measures.

For this reason, it is of the utmost importance for the supervisory control system to always provide the operator with appropriate information according to the operating conditions of the plant. Especially when an abnormality occurs in the plant and the magnitude of the abnormality is large, or when the transient phenomenon propagates quickly, the burden on the operator becomes extremely large, which may lead to misjudgment. come.

Therefore, when a failure occurs in the plant, it is important to ensure the safe operation of the plant by promptly detecting the failure and preventing the failure from spreading at a stage where the failure is minor. Also, preventing the expansion of anomalies that have occurred is naturally demanded from the viewpoint of improving the plant operating rate, and it is necessary to provide operators with appropriate information in accordance with the operating conditions of the plant. A plant monitoring and control system for the purpose is being developed.

[0006]

Conventionally, as a plant supervisory control system, the operating state of the plant is judged based on the occurrence state of an alarm due to an abnormality detection, and the process information related to the alarm that requires the most attention is given. The offering is being developed. However, such a supervisory control system automatically recognizes an event that occurs after some failure is detected in the plant and presents a process state according to the situation, and no alarm is issued. It was up to the operator to judge the plant operation status in each situation.

In a situation where no alarm is issued,
Using a dynamic characteristic model that simulates the dynamic characteristics of a plant,
By matching with the process state that changes from moment to moment,
Development of a system for measuring the transition of a transient state is in progress. However, even if the cause of the transient state is unspecified, there is a demand that it should be simulated in synchronization with the plant behavior, but it was difficult to put it into practical use.

The object of the present invention is to detect an occurrence event at a minor abnormal stage due to the causal relationship of process variation even in an operating situation in which an alarm due to abnormality detection is not issued, and to detect the situation. An object of the present invention is to provide a plant state visualization system that supports the operator's situation determination by providing a process state corresponding to the information as information that the operator can easily recognize.

[0009]

In order to achieve the above object, a plant state visualization system according to the invention of claim 1 has an input processing unit for inputting an observation signal in a plant, and the observation signal deviates from a planned value. An abnormality determination processing unit that determines whether or not the observation signal and the knowledge database that stores the relationship between the observation signals, and the observation signal and the knowledge database when the abnormality determination processing unit determines an abnormal state It is characterized by comprising a propagation path estimation unit for estimating an abnormal propagation path and a state display unit for displaying the estimation result of the propagation path estimation unit on a display means.

Using a knowledge database that stores the relationship between the observed signals, it is determined that the observed signal from each part of the plant deviates from the planned value to be an abnormal state, and the propagation path of the abnormality is estimated, and this is estimated. It is provided to the operator by the display means of the status display section.

According to a second aspect of the present invention, a plant state visualization system is provided with an event transition prediction unit that predicts event transitions and process behaviors using the observation signals and a knowledge database, and the abnormality determination processing unit When it is determined that the state is abnormal, the determination result is displayed on the display means by the state display unit to present the propagation route of the abnormality and the transition prediction of the abnormality when the abnormality occurs.

Based on the observation signal and the knowledge database, the change state of the observation signal is determined as an abnormal state, and the transition of the abnormal event and the behavior of the process are predicted by the event transition prediction unit, which is displayed on the display means of the state display unit. To the operator via

A plant state visualization system according to a third aspect of the present invention determines whether or not the sensor value is normal by comparing and evaluating the behavior of the process predicted by the event transition prediction unit and the behavior of the observation signal. Provide a sensor diagnostic unit to
The determination result of the sensor diagnosis unit is displayed on the display unit by the state display unit.

Based on the observation signal and the knowledge database, it is determined that the change state of the observation signal is an abnormal state, and the behavior of the process predicted by the event transition prediction unit and the behavior of the observation signal are compared and evaluated by the sensor diagnosis unit, It is determined whether or not the sensor is functioning normally, and this is provided to the operator by the display means of the status display unit.

According to a fourth aspect of the present invention, in the plant state visualization system, the knowledge database stores the relationship between the observation signals via a control system and the control system in the event transition prediction unit. A control system diagnostic unit for determining whether or not the control system is operating normally by comparing and evaluating the event propagation prediction result and the observed signal is provided, and the determination result in the control system diagnostic unit is displayed in the status display unit. Is displayed on the display means.

Based on the observation signal and the knowledge database, the change state of the observation signal is determined to be an abnormal state, and the relationship between the observation signals stored in the knowledge database via the control system and the observation signal are related to the control system diagnostic unit. In step S1, the control system is compared and evaluated to determine whether or not the control system is operating normally, and this is provided to the operator through the display means of the state display unit.

According to a fifth aspect of the present invention, in the plant state visualization system, the knowledge database stores the relationship between the observation signals via an interlock and the event transition prediction section uses an interlock. An interlock diagnostic unit for determining whether or not the interlock is operating normally by comparing and evaluating the prediction result of the event propagation and the observed signal is provided, and the determination result in the interlock diagnostic unit is displayed as the status. It is characterized in that it is displayed on the display means by the section.

An interlock diagnostic unit determines the abnormal state from the changed state of the observed signal based on the observed signal and the knowledge database, and determines the relationship between the observed signals through the interlock stored in the knowledge database and the observed signal. In the above, the soundness of the interlock is determined by comparison and evaluation, and this is provided to the operator through the display means of the state display unit.

According to a sixth aspect of the present invention, there is provided a plant state visualization system in which the knowledge database stores the relationship between the observed signals depending on the characteristics of the individual devices.
A device characteristic diagnosis unit that determines whether or not the characteristic of the device itself is changed by comparing and evaluating the event propagation prediction result due to the characteristic of the device single unit in the event transition prediction unit, and It is characterized in that the determination result in the device characteristic diagnosis section is displayed on the display means by the status display section.

Based on the observation signal and the knowledge database, the change state of the observation signal is determined to be an abnormal state, and the prediction result of the event propagation due to the characteristic of the single device stored in the knowledge database and the observation signal are analyzed by the device characteristic diagnosis unit. The soundness of the device is judged by comparing and evaluating and judging whether the characteristic of the device itself is changed, and this is provided to the operator through the display means of the status display unit.

According to a seventh aspect of the present invention, in the plant state visualization system, the abnormality determination processing unit performs logarithmic averaging processing for each observation signal and determines whether or not there is an abnormality based on the deviation between the logarithmic average value and the current value. Is characterized in that it is determined early. The abnormal state determination in the abnormality determination processing unit is performed by logarithmic averaging processing for each observation signal, and is performed based on the deviation between the logarithmic average value and the current value, so that a gradual change state can be detected relatively early. The result is provided to the operator through the display means of the status display section.

According to the eighth aspect of the present invention, in the plant state visualization system, the state display section displays the estimation result of the propagation path estimation section on the display means in a block diagram together with the time axis, and displays the time series transitions. It is characterized by presenting a propagation route. In the display means for providing the processing result to the operator in the status display unit, a time axis is provided on the display screen, a horizontal display position of each change state and each change state are shown as a block diagram, and as a time-series transition. Display the propagation route.

According to a ninth aspect of the present invention, there is provided a plant state visualization system in which the state display section displays the prediction result of the event transition prediction section on a display means in a block diagram together with a time axis for time-series transition. It is characterized in that the prediction is presented together with the estimation result in the propagation path estimation unit. In the display means for providing the processing result to the operator in the status display unit, a time axis is provided on the display screen, a horizontal display position of each change state and each change state are shown as a block diagram, and as a time-series transition. Display the propagation route and prediction result.

According to a tenth aspect of the present invention, there is provided a plant state visualization system in which the state display section displays the estimation result of the propagation path estimation section on a display means in a block diagram on the plant configuration diagram and displays it in the plant. It is characterized by presenting propagation paths as spatial propagation.
In the display means for providing the operator with the processing result in the status display unit, a plant configuration diagram showing the arrangement of target devices on the display screen, an observation position of each change state on this plant configuration diagram and a block diagram of each change state Display the propagation route as.

According to a 11th aspect of the present invention, there is provided a plant state visualization system, wherein the state display section displays the prediction result of the event transition prediction section on a display means in a block diagram on a plant configuration diagram and displays it in the plant. The spatial transition prediction of is presented together with the estimation result in the propagation path estimation unit.

In the display means for providing the operator with the processing result in the status display unit, a plant configuration diagram showing the arrangement of the target devices on the display screen, the observation position of each variation state on this plant configuration diagram and each variation state Is a block diagram, and the propagation path and prediction results are displayed.

According to a twelfth aspect of the present invention, there is provided a plant state visualization system, wherein in the state display section, the sensor diagnosis section, the control system diagnosis section, the interlock diagnosis section and the equipment characteristic diagnosis section diagnose that an abnormality or characteristic has changed. The sensor or control system and the interlock or the device that has been operated, the color or shape is changed on the plant configuration diagram and displayed on the display means, and the plant component whose abnormality or characteristic has changed is displayed in the event transition prediction unit. The present invention is characterized in that both the prediction result and the estimation result in the propagation path estimation unit are presented.

In the display means for providing the operator with the processing result in the status display section, a plant configuration diagram showing the arrangement of the target devices on the display screen, the observation position of each variation state on this plant configuration diagram and each variation state A block diagram is used to display the propagation path and the prediction result, and the color or shape of the part where the abnormality or the characteristic has changed is displayed.

[0029]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention will be described with reference to the drawings. The first embodiment relates to claim 1, and as shown in the functional configuration diagram of FIG. 1, a plant state visualization system 2 used for a large-scale plant 1 such as a nuclear power plant is formed by a computer.

The plant state visualization system 2 includes an input processing unit 3 for inputting a plurality of observation signals from a plant 1 by sensors arranged at a plurality of locations (not shown), planned values for all the observation signals and their allowable values. Is stored, and an abnormality determination processing unit 4 is provided for periodically determining whether or not the observed signal deviates from the planned value based on the stored information.

Further, the qualitative causal relationship between the observation signals and the causal relationship are relationships through various control systems and interlocks in the plant 1 and the characteristics of various equipments constituting the plant 1. Information stored in the knowledge database 5, and the causal relationship information stored in the knowledge database 5 when the observation signal is determined by the abnormal state determination unit 4 to deviate from the planned value. A propagation path estimation unit 6 that estimates an abnormal propagation path is provided.

Further, for example, a display means such as a CRT is provided, and is composed of a status display section 7 for displaying the propagation path estimated by the propagation path estimation section 6 on the display means.

Next, the operation of the above configuration will be described. As shown in the process flow diagram of FIG. 2, a plurality of observation signals M (not shown) obtained from the plant 1 are input to the input processing unit 3 of the plant state visualization system 2, and the input processing unit 3 performs the first step 8 As ", read the observation signal periodically" from the plant 1.

A second step 9 is a process in the abnormal state judging section 4, which "determines whether or not the state is abnormal" based on the stored planned value for each observation signal and its allowable value. For example, if the planned value for the observed value Mm is Md and the permissible value is Me, it is determined that an abnormality has occurred when the condition represented by the following equation (1) is satisfied.

| Mm-Md |> Me (1) Note that this process is periodically performed on all the observed signals, and whether the observed signals are smaller is abnormal.
The direction of the larger one is also determined.

The third step 10 is a process in the propagation path estimation unit 6, which uses the causal relationship information between the observation signals separately stored in the knowledge database 5 and the determination result of the abnormal state determination unit 4 to Estimate the path of anomaly propagation. " Further, the fourth step 11 is a state display process in the state display unit 7, and the state display unit 7 "displays the propagation route on the display means" which is the estimation result of the propagation route estimation unit 6.

Further, FIG. 3 is a block diagram of the knowledge database, showing an example, in which the causal relationship between the observation signals is described in the frame format in units of the change state of the observation signals. Here, the change state indicates two ways of deviating in the direction of a large value or deviating in the direction of a small value with respect to the planned value.

For example, it is described that the change conditions for changing an observation signal to the change state A include the change states B to D. The change states B to D represent different change states of the observation signal. Also,
The format registered in the knowledge database 5 is a fragmentary one of the change state units of one observation signal, but by connecting the change states according to the change conditions, it has a network structure showing a causal relationship between the change states. A propagation sequence representing the propagation path of the event is generated.

As an example, the estimation method will be specifically described by assuming the propagation path shown in the relationship diagram of the change state of FIG.
Note that FIG. 4 shows the causal relationships between the change states A to I. In the knowledge database 5, the following individual information is described in frame format.

The changing conditions of A include B, C and D. B
There is H in the changing condition of. G is a change condition of C. E is a change condition of D. The changing conditions of E include
There are F and G.

D is a change condition of F. The changing condition of G is I. The change condition of H is I. The changing condition of I is F.

Here, if it is determined by the abnormal state determination unit 4 that the changed states A, C, D, E, and F have actually changed, the following processing is performed for these changed states.
First, as shown in the relational diagram of the changed states in the process of FIG. 5, a flag (stripe) indicating “at the most downstream of the propagation sequence” for all changed states
(Hereinafter, for convenience of explanation with the drawings, the most downstream is distinguished as a stripe flag and the most upstream is distinguished as a white flag).

Next, all the conditions for changing itself for each changed change state are investigated, and if any are satisfied, the path indicating whether the conditions are satisfied is saved, and the path Defeat the most downstream flag (stripe) in the upstream change state.
If the condition for generating oneself is not satisfied, the most downstream flag (stripe) is defeated and a flag (white) indicating "at the most upstream of the propagation sequence" is set.

By applying this processing to all the changed states, as shown in the relation diagram of the changed state showing the most downstream and the most upstream in the processing process of FIG. 6, the most downstream and the most upstream are changed. It is possible to identify the states and then recognize the paths on the propagation sequence, as shown in the changing state path diagram of FIG. Further, the propagation path is generated by searching the route in the changed state using the generated path.

This processing procedure is as shown in the processing flow chart of FIG. 8. When "M = 1" 12 is input, "change state (P
From the start of the investigation routine of the change state upstream of V (i) ”13, the PV (i) investigation list (LIST-
Go to “N = 0?” 16 via “Enter 1)” 14 and “N ← number of change states upstream of PV (i)” 15. If “N = 0?” 16, Yes is “Return” 17
To No, go to “L ← 1” 18 and then to “PV
PV (j), which is the Lth upstream of (i), is LIST-1
Is it in? Go to 19.

If the answer in 19 is Yes, "LIST
A change state from PV (j) to PV (i) in -1 is set as a race state group. If PV (j) belongs to the previously defined race state group, the change state in that race state group shall be included. "20, and then from" L ← L + 1 "21 to" L≤N? "22.
If Yes, the process returns to 19 again, and if No, the process returns to "Return" 17.

If No in 19 above, "PV
Is (j) already registered as a survey completed state? "
Proceeding to 23, the process proceeds to “L ← L + 1” 21 via “Record the survey completion status ID of PV (j) registered as upstream of PV (i)” 24 in Yes.

If No in 23, "PV
Via “start of change state investigation routine upstream of (j)” 25, “is PV (j) in the same race group as PV (i)?” 26, where Ye
If s, proceed to "L ← L + 1" 21.

In the case of No in 26, "PV
Does (j) belong to a race group? 27, and if Yes in 27, “Register all PVs in the conflict state group to which PV (j) belongs to the survey completion state COND (M) with ID M” ”28, “The change status registered in the survey completion status COND (M) is LIST.
Delete from -1 ”29.

Further, from the above 29, "register all the IDs of all the investigation completed states recorded as upstream of the changed state belonging to COND (M) as upstream of COND (M)" 30.
Then, “record M as an ID of the survey completed state upstream of PV (i)” 31, and further proceed to “L ← L + 1” 21 via “M ← M + 1” 32. When the answer is No in 27, the process proceeds to 29 through “Register PV (j) in the survey completion state COND (M) with ID M” 33.

Therefore, in the case of FIG. 7 as an example, it is recognized that the change state at the most downstream is A, and the investigation point is A, and the route search is started from this A. First, enter A in the survey list and see C, which is upstream of A. Since C has not been entered in the survey list and has not been registered as a survey completed state, the survey point is moved to C, where C is entered in the survey list and G is the upstream of C. I see.

This G is not entered in the survey list,
Also, because it is not registered as a survey completed state,
Next, the survey point is moved to G, where G is entered in the survey list, but since there is no upstream of G, the survey point is returned to C. At this time, C and G are not in the conflicting state group, and since G itself does not belong to the conflicting state group, only G is registered in the investigation completed state (COND (1)) with ID = 1.

Next, G is deleted from the survey list, and the ID of the survey completed state, upstream of C, is recorded. Also,
Since there is nothing but G upstream of C, the survey point is A
Return to At this time, A and C are not in the race group,
Also, since C itself does not belong to the race condition group, C
Completed state with only ID 2 (COND (2))
Register with.

Next, C is deleted from the survey list and CO
The survey completion state ID: 2 upstream of ND (2) is registered, and the survey completion state ID: 2 upstream of A is recorded. Then look at D, which is upstream of A. Since D is neither entered in the survey list nor registered as the survey completed state, the survey point is moved to D next. Therefore, enter D in the survey list and see E, which is upstream of D. Since E is neither entered in the survey list nor registered as the survey completed state, the survey point is moved to E next.

Next, E is entered in the survey list to see F which is upstream of this E. Since F is neither entered in the survey list nor registered as the survey completed state, the survey point is moved to F next. Therefore, enter F in the survey list and see D, which is upstream of F. D has already been entered in the survey list, and D to F in the survey list are set as a race condition group and the survey point is moved to E.

Here, since E and F are in the same race state group, G, which is the upstream of E, is examined without generating the check completion state. Although G is not in the survey list, since it is registered as the survey completion state of ID: 1, ID: 1 is registered as the survey completion state upstream of E, and the survey point is moved to D.

Since D and E are in the same race condition group, the investigation point is moved to A without generating the investigation completion state. Since A and D are not in the same race state group, the race state group (D, E, F) has ID: 3
Registered as the survey completion status (COND (3)) of
Delete D, E, F from the survey list.

In addition, the survey completed state (CO
ND (1)) is registered as the upstream investigation completed state of COND (3). Further, at the stage of returning to A, which is the search start point, only A is in the investigation completed state (CON
I registered as D (4)) and recorded as upstream of A
The DND 2 and ID: 3 survey completion status is COND
Register as the upstream survey completion status of (4).

Through the series of processes described above, the propagation sequence shown in the propagation path diagram of FIG. 9 in which the competitive relationship and the position of each change state are clarified is recognized. Further, the processing procedure diagram of FIG. 10 shows processes such as the survey list, the race state group, and the survey completed state formed by the process of the above procedure.

In the above example, the most downstream change state is recognized, but the most downstream change state may compete with each other and the most downstream change state may not be recognized. In that case, the propagation path can be generated by starting the investigation from an arbitrary change state and repeating until all the change states have been searched. This propagation sequence represents the transition of the event, and the propagation path of the abnormality is estimated by the above processing.

The fourth step 11 is a display process on a display means (not shown) in the status display section 7, and the propagation path estimated by the propagation path estimation section 6 is provided in the status display section 7, for example, a CRT or the like. Is displayed on the screen of the display means in various predetermined forms and provided to the operator.

As described above, according to the first embodiment,
It is possible to automatically estimate the propagation path of anomalies based on the changes and the directions of multiple observation signals in the plant 1, and to appropriately and quickly visualize the events that occurred in the plant 1 and present them to the operator. By doing so, it is possible to support the situation determination work of the plant and reduce the burden on the operator.

The second embodiment relates to claim 2, and as shown in the functional block diagram of FIG. 11, a plant state visualization system 34
Is an input processing unit 3, an abnormal state determination unit 4, a knowledge database 5, a propagation route estimation unit 6, and an event transition prediction unit 35.
And a status display section 7.

Further, when the abnormal state judging section 4 judges that the observation signal from the plant 1 deviates from the planned value, the event transition predicting section 35 uses the causal relationship information stored in the knowledge database 5 It is used to estimate the observed signal that is predicted to deviate from the planned value in the future. The other configurations and operations of the respective parts are the same as those in the first embodiment, and therefore description thereof will be omitted.

Next, the operation of the above configuration will be described. The process flow diagram of FIG. 12 shows the process flow of the event transition prediction unit 35.
First, by using the knowledge database 5 to "search upstream of all unchanged change states Zni (i = 1, 2, ...)" Investigate all the conditions that change you.

Next, the change state Zn that has not changed is investigated for "whether the conditions for changing Zni are satisfied", and if there is one that is satisfied from the result, "Zni is changed to Z
"Register as ai" to register as a change state Za that will change in the future. From the result of this investigation, if there is no change state Zn that changes in the future, it is assumed that the future plant state is normal.

Next, in the status display section 7, the change status Z
A is displayed on the screen of the display means together with the propagation path estimated by the propagation path estimation unit 6 as a change state that changes in the future, and is provided to the operator. Thus, the second book
According to the embodiment, in the future, it is possible to automatically predict the observed signal that deviates from the planned value, including the direction of change, and appropriately present the transition of abnormalities in the future plant to the operator. Thus, it is possible to support the operator to judge the situation of the plant 1.

The third embodiment relates to claim 3, and as shown in the functional block diagram of FIG. 13, a plant state visualization system 36
Is composed of an input processing unit 3, an abnormal state determination unit 4, a knowledge database 5, a propagation route estimation unit 6, an event transition prediction unit 35, a sensor diagnosis unit 37, and a state display unit 7. In addition to the causal relationship information between the change states of the observed signal from the plant 1, the knowledge database 5 stores the propagation delay time.

Further, in the sensor diagnosis section 37, when the abnormal state determination section 4 determines that the observation signal of the plant 1 deviates from the planned value, the causal relationship information stored in the knowledge database 5 and the information The soundness of the sensor related to the observation signal is diagnosed using the change state information predicted by the image transition prediction unit 35. The other configurations and operations of the respective parts are the same as those of the first and second embodiments, and therefore description thereof will be omitted.

Next, the operation of the above configuration will be described. The process flow diagram of FIG. 14 shows the process flow of the sensor diagnosis unit 37. First, the change state Za predicted by the event transition prediction unit 35 and the change state Za described in the knowledge database 5 are shown. The delay time Ta until reaching is registered. Next, based on "whether Ta time has elapsed" and "only Za changes," it is investigated whether the abnormal state judgment unit 4 has judged that the changed state Za has changed after the Ta time has elapsed.

If the change state Za has not changed, it is diagnosed that "the sensor corresponding to Za is abnormal". In addition, the change state Za
If is changed, it means that the sensor is normal.

Next, in the state display unit 7, information indicating that the sensor corresponding to the changed change state Za is abnormal is displayed as the propagation path estimated by the propagation path estimation unit 6 and the event transition. It is displayed on the screen of the display means together with the future plant state predicted by the prediction unit 35 and provided to the operator.

As described above, according to the third embodiment,
In addition to the operation of the second embodiment, it is possible to automatically perform the soundness diagnosis of the sensor arranged in the plant, and present it to the operator including the sensor failure information. Assist operators in judging the status of plant 1.

The fourth embodiment relates to claim 4, and as shown in the functional block diagram of FIG. 15, a plant state visualization system 38
Is composed of an input processing unit 3, an abnormal state determination unit 4, a knowledge database 5, a propagation route estimation unit 6, an event transition prediction unit 35, a control system diagnosis unit 39, and a state display unit 7. Further, in the knowledge database 5, in addition to the causal relationship information between the change states of the observed signals from the plant 1, if the causal relationship propagates through the operation of the control system,
The control system and the propagation delay time are stored.

Further, the control system diagnosing section 39, when the abnormal state judging section 4 judges that the observation signal of the plant 1 deviates from the planned value, the causal relationship information stored in the knowledge database 5, Using the change state information predicted by the event transition prediction unit 35, the soundness of the control system in the plant 1 is diagnosed. The other configurations and operations of the respective parts are the same as those of the first and second embodiments, and therefore description thereof will be omitted.

Next, the operation of the above configuration will be described. The processing flow diagram of FIG. 16 shows the processing flow of the control system diagnosis unit 39. First, the change state Za predicted by the event transition prediction unit 35 and the change state Za described in the knowledge database 5 are described. The delay time Ta up to is registered.

Next, the knowledge database 5 is investigated, and when the change state Za is propagated through the operation of the control system Ca, the control system Ca is also registered and "Ta
"Time has passed" and "Z which should change due to the relationship of the control system"
Based on "whether a has changed", it is investigated whether or not the abnormal state determination unit 4 determines that the changed state Za has changed after Ta time.

Here, when the change state Za has not changed, it is diagnosed that the control system Ca is abnormal by "assuming that the corresponding control system Ca is abnormal". In addition,
If the change state Za that should change from the result of the investigation has changed, the control system Ca is assumed to be normal.

Next, in the state display section 7, information indicating that the control system Ca corresponding to the changed change state Za is abnormal is provided with the propagation path estimated by the propagation path estimation section 6 and the information. It is displayed on the screen of the display means and provided to the operator together with the future plant state predicted by the image transition prediction unit 35.

As described above, according to the fourth embodiment,
In addition to the operation of the second embodiment, it is possible to automatically diagnose the soundness of the control system at the time of transition in the plant 1, and present it to the operator including the information of the control system failure. It is possible to support the operator to judge the situation of the plant 1.

The fifth embodiment relates to claim 5, and as shown in the functional block diagram of FIG. 17, a plant state visualization system 40
Is composed of an input processing unit 3, an abnormal state determination unit 4, a knowledge database 5, a propagation route estimation unit 6, an event transition prediction unit 35, an interlock diagnosis unit 41, and a state display unit 7. Further, in the knowledge database 5, in addition to the causal relationship information between the change states of the observed signals from the plant 1, if the causal relationship propagates through the interlock operation, the interlock and the delay of the propagation. The time and is remembered.

Further, the interlock diagnosing section 41, when the abnormal state judging section 4 judges that the observation signal of the plant 1 deviates from the planned value, the causal relationship information stored in the knowledge database 5, The soundness of the interlock is diagnosed using the change state information predicted by the event transition prediction unit 35. The other configurations and operations of the respective parts are the same as those of the first and second embodiments, and therefore description thereof will be omitted.

Next, the operation of the above configuration will be described. The process flow diagram of FIG. 18 shows the process flow of the interlock diagnosis unit 41. First, the event transition prediction unit 35
The change state Za predicted in [4] and the delay time Ta until the change state Za described in the knowledge database 5 are registered.

Next, the knowledge database 5 is examined, and when the change state Za is propagated through the operation of the interlock Ia, the interlock Ia is also registered and "Ta time has elapsed. Or “Is Za that should have changed due to the interlock relationship?”
After a time, it is investigated whether the abnormal state determination unit 4 determines that the changed state Za has changed.

Here, when the change state Za has not changed, it is diagnosed that the interlock Ia is abnormal by "assuming that the corresponding interlock Ia is abnormal". If the change state Za, which should change according to the result of the investigation, has changed, the interlock Ia is assumed to be normal.

Next, in the state display unit 7, the information indicating that the interlock Ia corresponding to the changed change state Za is abnormal, and the propagation route estimated by the propagation route estimation unit 6 and the information It is displayed on the screen of the display means and provided to the operator together with the future plant state predicted by the image transition prediction unit 35.

As described above, according to the fifth embodiment,
In addition to the operation in the second embodiment, it is possible to automatically diagnose the soundness of the interlock that operates during a transient period in the plant 1, and to inform the operator including information about the failure of the interlock. It can be presented to assist the operator in judging the situation of the plant 1.

The sixth embodiment relates to claim 6, and as shown in the functional block diagram of FIG. 19, a plant state visualization system.
42 includes an input processing unit 3, an abnormal state determination unit 4, a knowledge database 5, a propagation route estimation unit 6, an event transition prediction unit 35, a device characteristic diagnosis unit 43, and a state display unit 7. In addition to the causal relationship information between the changing states of the observed signal from the plant 1, if the causal relationship propagates through the characteristic of the device that originally does not change, the knowledge database 5 The device and the propagation delay time are stored.

Further, the equipment characteristic diagnosing section 43, when the abnormal state judging section 4 judges that the observation signal of the plant 1 deviates from the planned value, the causal relationship information stored in the knowledge database 5, The device characteristics are diagnosed using the change state information predicted by the event transition prediction unit 35. The other configurations and operations of the respective parts are the same as those of the first and second embodiments, and therefore description thereof will be omitted.

Next, the operation of the above configuration will be described. The process flow diagram of FIG. 20 shows the process flow of the device characteristic diagnosis unit 43. First, the change state Za predicted by the event transition prediction unit 35 and the change state Za described in the knowledge database 5 are described. The delay time Ta up to is registered.

Next, the knowledge database 5 is examined, and when the change state Za is propagated through the operation of the characteristic of the device, the device Ea is also registered and the "Ta time has elapsed?" And "Z should change depending on the device characteristics."
Whether or not the changed state Za has been changed by the abnormal state determination unit 4 after Ta time is investigated by "whether a has changed".

Here, if the change state Za has not changed, it is diagnosed that the characteristic of the device Ea has changed, by saying "the corresponding device Ea characteristic has changed". If the change state Za that should change according to the result of the investigation has changed, the characteristic of the device Ea corresponding to the change state Za is assumed to be normal.

Next, in the state display unit 7, information indicating that the characteristic of the device Ea corresponding to the changed change state Za is abnormal is displayed on the propagation route estimated by the propagation route estimation unit 6 and the previous information. It is displayed on the screen of the display means and provided to the operator together with the future plant state predicted by the event transition prediction unit 35.

As described above, according to the sixth embodiment,
In addition to the operation in the second embodiment described above, it is possible to automatically diagnose the device characteristics in the plant 1, and present it to the operator, including information on changes in the device characteristics,
It is possible to assist the operator in judging the situation of the plant 1.

The seventh embodiment relates to claim 7, and the plant state visualization system 34 has the same function configuration as shown in the function configuration diagram of FIG. The state determination unit 4, the knowledge database 5, the propagation route estimation unit 6, the event transition prediction unit 35, and the state display unit 7.
It is composed of However, the abnormal state determination unit 4 performs logarithmic averaging processing for each observation signal from the plant 1 and performs processing to determine whether or not there is an abnormality based on the deviation between the logarithmic average value and the current value.

Next, the operation of the above configuration will be described. Most of the changes in the gradual state during normal operation in the plant 1 have a first-order lag characteristic. For example, changes in the reactor outlet temperature when pulling out the control rods from the reactor core in a nuclear power plant, or changes in the outlet temperature of the intermediate heat exchanger, evaporator, and superheater when changing the coolant flow rate. The change in the process state corresponding to one operation, such as, can be approximated by the first-order lag characteristic.

In particular, during normal operation, the process state due to the completion of one operation settles down and then the next operation is started. Therefore, in the range of normal operation, the process state at a certain point in time goes back to the past. The value becomes exponentially distant from the value. Therefore, in detecting the change in the process state during normal operation, the deviation D from the current value obtained by logarithmic averaging deviates from the permissible value determined in advance by the observation signal unit, as shown in the following equation (2). If it does, the direction of change will be captured.

[0098]

[Equation 1]

Further, P is an observation signal, T is a constant, t is time,
τ indicates a sampling period. Here, when the deviation D is larger than a predetermined positive value, it changes in the + (plus) direction, and when the deviation D is smaller than the predetermined negative value, it changes in the − (minus) direction. Determine that

As described above, according to the seventh embodiment,
In the future, the observation signal deviating from the planned value will be detected at the time when the change state including the changing direction thereof is detected. Therefore, by providing it to the operator earlier than in the seventh embodiment, the change of the state Prediction and countermeasures can be done with sufficient time.

The eighth embodiment relates to claim 8, and the plant state visualization system 34 has the same functional configuration as shown in the functional configuration diagram of FIG. It is composed of an abnormal state determination unit 4, a knowledge database 5, a propagation route estimation unit 6, an event transition prediction unit 35, and a state display unit 7.

The abnormal state determination section 4 transfers the time at which this change is detected to the state display section 7 for each change state. Next, in the state display unit 7, while inputting the propagation sequence estimated by the propagation path estimation unit 6, the change time of the change state in this propagation sequence is input from the abnormal state determination unit 4, The propagation sequence is displayed on the display means as a time series transition.

Next, the operation of the above configuration will be described. As shown in the display screen diagram of FIG. 21, the display screen 44 is a display example on the display means of the status display unit 7, and the connection of the change states A, C, D, E, F displayed on the display screen 44 is as follows. This is the result estimated by the propagation route estimation unit 6.

A time axis is provided in the horizontal direction of the display screen 44, the time when each change state is changed is aligned with this time axis, the horizontal display position of each change state is determined, and each change state is determined. Display as a block diagram. As described above, according to the eighth embodiment, by representing the propagation route along with the time axis, the operator can easily understand the propagation route as a time-series transition.

The ninth embodiment relates to claim 9, and this plant state visualization system 34 has the same functional configuration as shown in the functional configuration diagram of FIG. It is composed of an abnormal state determination unit 4, a knowledge database 5, a propagation route estimation unit 6, an event transition prediction unit 35, and a state display unit 7.

The event transition predicting unit 35 transfers the predicted time at which the change state changes to the state display unit 7. Next, the state display unit 7 causes the display means to display the change state predicted by the event transition prediction unit 35 in addition to the propagation sequence displayed along with the time axis as a time-series transition.

Next, the operation of the above configuration will be described. As shown in the display screen diagram of FIG. 22, the display screen 45 is an example of display on the display means of the status display unit 7. The display screen 45 has a propagation sequence according to the change states A, C, D, E, and F. Is displayed.

Further, in addition to this propagation sequence, the change state V predicted to change due to the change state A is a time Tv at which the change state V predicted on the time axis provided in the horizontal direction changes. Together, the horizontal display position of the predicted change state V is determined and displayed in a block diagram. As described above, according to the ninth embodiment, by expressing the predicted change state together with the time axis,
The operator can easily understand the time-series transition of future events.

The tenth embodiment relates to claim 10, and the plant state visualization system 34 has the same functional configuration as shown in the functional configuration diagram of FIG. It is composed of an abnormal state determination unit 4, a knowledge database 5, a propagation route estimation unit 6, an event transition prediction unit 35, and a state display unit 7. The state display unit 7 stores in advance a drawing showing the configuration of the plant and observation position information indicating which part of the plant configuration diagram the observation signals corresponding to all the change states are observed.

Next, the operation of the above configuration will be described. As shown in the display screen diagram of FIG. 23, the display screen 46 is a display example in the display means of the status display unit 7, and on this display screen 46, together with the schematic view and the arrangement of the respective devices 50 to 55 that configure the plant, Changed state A, C, D, E,
F is displayed. The change states A, C, D, E, F
Is the result estimated by the propagation path estimation unit 6, and the change state is displayed at the observation position such as a sensor on the plant configuration.

As described above, according to the tenth embodiment,
Since the arrangement of each target device in the plant and the propagation path on the plant configuration diagram showing the observation position are represented, the operator can obtain knowledge about the plant configuration that the plant propagation path is owned by the operator. While collating, it is possible to easily grasp the estimation of the cause of the abnormality, the prediction of the development of the abnormality, the method for dealing with the abnormality, and the like as spatial propagation.

The eleventh embodiment relates to claim 11, and the plant state visualization system 34 has the same functional configuration as that shown in the functional configuration diagram of FIG. It is composed of an abnormal state determination unit 4, a knowledge database 5, a propagation route estimation unit 6, an event transition prediction unit 35, and a state display unit 7.

The event transition predicting unit 35 transfers the predicted change time of the change state to the state displaying unit 7, and the state displaying unit 7 displays the spatial transition on the display unit together with the plant configuration diagram. In addition to the gating sequence, the change state predicted by the event transition prediction unit 35 is displayed.

Next, the operation of the above configuration will be described. As shown in the display screen diagram of FIG. 24, the display screen 47 is a display example on the display means of the status display unit 7, and displays a schematic diagram and arrangement of the respective devices 50 to 55 that configure the plant. Furthermore, in addition to the propagation sequence by the change states A, C, D, E, and F, the change state V predicted to change due to the change state A is changed to the change state V predicted on the plant configuration diagram. Display at the observation position of the corresponding observation signal.

Thus, according to the eleventh embodiment,
The predicted change state is displayed along with the location of each target device in the plant and the propagation path on the plant configuration diagram showing the observation position, so the operator can easily grasp the spatial transition of future events. It becomes possible.

The twelfth embodiment relates to claim 12, and the plant state visualization system 48 includes an input processing section 3, an abnormal state judging section 4, and a knowledge database 5 as shown in the functional block diagram of FIG. Propagation path estimation unit 6, event transition prediction unit 35, sensor diagnosis unit 37, control system diagnosis unit 39, interlock diagnosis unit 41, device characteristic diagnosis unit 43, and status display unit 7
It is composed of

It should be noted that the status display section 7 includes a drawing showing the configuration of the plant in advance, a sensor diagnosis section 37, a control system diagnosis section 39,
The observation signals corresponding to all the changed states in the interlock diagnosis unit 41 and the device characteristic diagnosis unit 43 store observation position information of a sensor or the like indicating which part on the plant configuration diagram is observed.

Further, in the status display section 7, the propagation means displayed together with the plant configuration diagram on the display means (not shown) as a spatial transition and the predicted change status, as well as the sensor diagnosis section 37 and the control system diagnosis. The section 39, the interlock diagnosis section 41, and the device characteristic diagnosis section 43 are configured to display, by shape or color, the portion diagnosed as having a failure or a change in characteristics.

Next, the operation of the above configuration will be described. As shown in the display screen diagram of FIG. 26, the display screen 49 is a display example in the display means of the status display unit 7, and on this display screen 49, together with the schematic view and the arrangement of the respective devices 50 to 55 constituting the plant, Changed state A, C, D, E,
F and the predicted change state V are displayed.

Here, for example, when the device characteristic diagnosis unit 43 diagnoses that the characteristic of the device 53 has changed, the display screen
The device 53 in 49 is displayed in a different color. Also,
You may distinguish by changing the shape of the schematic of the said apparatus. Note that, here, for the sake of convenience, a shaded area indicates that the color has changed. Thus, according to the twelfth embodiment,
By displaying the propagation route together with the plant configuration diagram via the display means of the state display unit 7, the operator can easily grasp the propagation route as spatial propagation.

In the present invention, the qualitative causal relationship between observed signals is used to detect the propagation path of an abnormality that has occurred, even when the alarm is not issued even in a slight abnormal state. , It is possible to predict the transition of events. Therefore, it is possible to diagnose the soundness of the sensor, the control system, and the interlock and to diagnose the device characteristics by using the information on what is the causal relationship of the observed signals.

By displaying these estimation result, prediction result, and diagnosis result from a time-series viewpoint or a spatial viewpoint, accurate plant information according to the plant state can be easily recognized. Since it can be provided automatically,
The burden on operators to monitor the plant is reduced.

[0123]

As described above, according to the present invention, the propagation path of an anomaly is automatically and early estimated based on the changing direction of the observed signal in the plant, and the observed signal deviating from the planned value in the future is included in the changing direction. Since it is possible to make automatic predictions, the soundness of sensors and control systems, interlocks, and device characteristics can be automatically diagnosed.

In addition, the propagation route and future abnormal transition are displayed as time-series and spatial information, and the part where the abnormality or the characteristic has changed is also displayed on the system diagram. Since the corresponding process state is provided to the operator as easily recognizable information, the burden on the operator is reduced and the stability and safety of plant operation are improved.

[Brief description of drawings]

FIG. 1 is a functional configuration diagram of a plant state visualization system according to a first embodiment of the present invention.

FIG. 2 is a process flow chart of the first embodiment according to the present invention.

FIG. 3 is a configuration diagram of a knowledge database according to the first embodiment of this invention.

FIG. 4 is a related diagram of a changed state of the first embodiment according to the present invention.

FIG. 5 is a related diagram of a change state in a processing process of the first embodiment according to the present invention.

FIG. 6 is a related diagram of change states showing the most upstream and the most downstream in the processing process of the first embodiment according to the present invention.

FIG. 7 is a route diagram of a changed state of the first embodiment according to the present invention.

FIG. 8 is a processing flowchart of a propagation route estimation unit according to the first embodiment of this invention.

FIG. 9 is a propagation path diagram of a determination result by the propagation path estimation unit according to the first embodiment of the present invention.

FIG. 10 is a processing procedure diagram in the propagation route estimation unit of the first embodiment according to the present invention.

FIG. 11 is a functional configuration diagram of a second embodiment according to the invention.

FIG. 12 is a process flow diagram of an event transition prediction unit according to the second embodiment of the present invention.

FIG. 13 is a functional configuration diagram of a third embodiment according to the invention.

FIG. 14 is a processing flowchart of the sensor diagnosis unit according to the third embodiment of the present invention.

FIG. 15 is a functional configuration diagram of a fourth embodiment according to the invention.

FIG. 16 is a process flow chart of a control system diagnosis unit according to the fourth embodiment of the present invention.

FIG. 17 is a functional configuration diagram of a fifth embodiment according to the invention.

FIG. 18 is a processing flowchart of the interlock diagnosis unit according to the fifth embodiment of the present invention.

FIG. 19 is a functional configuration diagram of a sixth embodiment according to the invention.

FIG. 20 is a processing flowchart of a device characteristic diagnosis unit according to the sixth embodiment of the present invention.

FIG. 21 is a display screen diagram of the eighth embodiment according to the present invention.

FIG. 22 is a display screen diagram of the ninth embodiment according to the present invention.

FIG. 23 is a display screen diagram of the tenth embodiment according to the invention.

FIG. 24 is a display screen diagram of the eleventh embodiment according to the present invention.

FIG. 25 is a functional configuration diagram of a twelfth embodiment according to the invention.

FIG. 26 is a display screen diagram of the twelfth embodiment according to the present invention.

[Explanation of symbols]

1 ... Plant, 2, 34, 36, 38, 40, 42, 48 ... Plant state visualization system, 3 ... Input processing unit, 4 ... Abnormal state determination unit, 5 ... Knowledge database, 6 ... Propagation path estimation unit, 7
... Status display section, 8 ... First step, 9 ... Second step,
10 ... Third step, 11 ... Fourth step, 12-33 ... Flow step in propagation path estimation unit, 35 ... Event transition prediction unit, 37 ... Sensor diagnosis unit, 39 ... Control system diagnosis unit, 41 ... Interlock diagnosis Part, 43 ... Device characteristic diagnosis part, 44 to 47, 49 ...
Screen, 50-55 ... Equipments that make up the plant.

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI Technical display location G21D 3/00 G21D 3/00 B

Claims (12)

[Claims] 1. An input processing unit for inputting an observation signal in a plant, an abnormality determination processing unit for determining whether or not the observation signal deviates from a planned value, and a knowledge database storing the relationship between the observation signals. A propagation route estimating unit that estimates a propagation route of the abnormality using the observation signal and the knowledge database when the abnormality determining processing unit determines an abnormal state; and a display unit that displays the estimation result in the propagation route estimating unit. A plant status visualization system, comprising: a status display unit for displaying on a plant. 2. The plant state visualization system includes an event transition prediction unit that predicts event transitions and process behaviors using the observation signals and a knowledge database, and the abnormality determination processing unit determines an abnormal state. The plant state visualization system according to claim 1, wherein the determination result is displayed on the display unit by the state display unit to present a propagation path of the abnormality and a prediction of transition of the abnormality when the abnormality occurs. 3. In the system for visualizing a plant state, a sensor diagnostic unit for judging whether or not a sensor value is normal by comparing and evaluating the behavior of the process predicted by the event transition prediction unit and the behavior of the observation signal. 3. The plant state visualization system according to claim 2, wherein the plant state visualization system is provided, and the determination result of the sensor diagnosis unit is displayed on the display means by the state display unit. 4. In the plant state visualization system, the knowledge database stores the relationship between the observation signals via a control system, and a prediction result of event propagation via the control system in the event transition prediction unit. And a control system diagnostic unit for determining whether or not the control system is operating normally by comparing and evaluating the observed signal with the observation signal, and the determination result in the control system diagnostic unit is displayed on the display means by the state display unit. The plant state visualization system according to claim 2, wherein 5. The plant state visualization system stores a relationship between the observation signals via an interlock in the knowledge database, and predicts a result of event propagation via the interlock in the event transition predicting unit. An interlock diagnostic unit for determining whether or not the interlock is operating normally by comparing and evaluating the observed signal with the observation signal is provided, and the determination result in the interlock diagnostic unit is displayed on the display unit by the status display unit. The plant state visualization system according to claim 2, which is displayed. 6. In the plant state visualization system, the knowledge database stores the relationship between the observed signals due to the characteristics of the individual devices, and the event propagation prediction result due to the characteristics of the individual devices in the event transition prediction unit. By providing a device characteristic diagnosis unit that compares and evaluates the observed signal to determine whether the characteristics of the device itself have changed,
3. The plant state visualization system according to claim 2, wherein the determination result of the device characteristic diagnosis unit is displayed on the display means by the state display unit. 7. The abnormality determination processing unit of the plant state visualization system performs logarithmic averaging processing for each observation signal, and early determines whether or not there is an abnormality based on the deviation between the logarithmic average value and the current value. The plant state visualization system according to claim 2. 8. The state display unit of the plant state visualization system displays the estimation result of the propagation route estimation unit on a display means in a block diagram together with a time axis, and presents the propagation route as a time-series transition. The plant state visualization system according to claim 2. 9. The state display unit of the plant state visualization system displays a prediction result in the event transition prediction unit on a display means in a block diagram together with a time axis, and a time-series transition prediction is performed by the propagation path estimation. The plant state visualization system according to claim 8, which is presented together with the estimation result in the department. 10. The state display unit of the plant state visualization system displays the estimation result of the propagation path estimation unit on a display means in a block diagram on a plant configuration diagram and propagates as spatial propagation in the plant. The plant state visualization system according to claim 2, wherein a route is presented. 11. The state display unit of the plant state visualization system displays a prediction result in the event transition prediction unit on a display means in a block diagram on a plant configuration diagram and displays spatial transition prediction in the plant. 11. The plant state visualization system according to claim 10, wherein is presented together with the estimation result in the propagation path estimation unit. 12. A sensor or control system in the status display section of the plant status visualization system, which is diagnosed as abnormal or changed in characteristics by the sensor diagnostic section, control system diagnostic section, interlock diagnostic section, and device characteristic diagnostic section. The interlock or equipment is displayed on the display means by changing the color or shape on the plant configuration diagram, and the plant component whose abnormality or characteristic has changed is predicted by the event transition prediction unit and the propagation path. The plant state visualization system according to any one of claims 3 to 6 and 11, wherein the estimation result of the estimation unit is presented together.