JP2018073316A - Plant diagnostic system and diagnostic method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a plant diagnostic system and diagnostic method capable of understanding which of multiple correlation abnormalities is the root cause.SOLUTION: A plant diagnostic system comprises: an operation data storage unit 12 for storing temporal changes of a plurality of operation parameters; a correlation change analysis unit 18 for analyzing correlation changes between operation parameters, and determining correlation abnormality when a correlation change is greater than a preset threshold; a correlation abnormality graph generation unit 19 for generating correlation abnormality graphs; a qualitative model storage unit 13 for storing plant qualitative models; a qualitative graph generation unit 20 for generating qualitative graphs based on the qualitative models; a root cause estimation unit 21 for estimating which of multiple links comprising the correlation abnormality graphs is the root cause by comparing the correlation abnormality graphs with the qualitative graphs; and a display device 16 for displaying the estimation results from the root cause estimation unit 21.SELECTED DRAWING: Figure 2

Description

  The present invention relates to a diagnostic system and a diagnostic method for diagnosing a power plant or a chemical plant, for example.

  Patent Document 1 discloses an operation management apparatus that is connected to a plurality of computers via a network and manages the plurality of computers. This operation management apparatus acquires a change over time in performance values (specifically, for example, CPU usage rate, remaining memory capacity, etc.) of each computer. Then, the change in the correlation between the performance values is analyzed, and when the change in the correlation is outside the error range, it is determined that the correlation is abnormal. Then, a correlation graph is created in which the performance value is indicated by the node and the correlation is indicated by the link between the nodes, and the correlation graph is displayed by making the correlation abnormal link distinguishable from the normal correlation link.

  Non-Patent Document 1 discloses a change detection method (algorithm) based on direct estimation of a density ratio as an analysis method of a change in correlation.

JP 2014-238852 A

Takeshi Ide, Masaru Sugiyama, “Abnormality Detection and Change Detection”, Kodansha, August 2015, p. 164-165

  In a diagnosis system for diagnosing a plant, for example, it is conceivable to employ the technique described in Patent Document 1 or Non-Patent Document 2. That is, changes with time of a plurality of operation parameters are stored as plant operation data. Then, a change in the correlation between the operation parameters is analyzed, and when the change in the correlation is larger than a preset threshold, it is determined that the correlation is abnormal. Then, a correlation anomaly graph is generated and displayed with the operation parameters indicated by the nodes and the correlation anomalies indicated by links between the nodes.

  However, in the plant, in addition to the correlation abnormality that is the root cause, a large number of correlation abnormalities are derived. Therefore, it cannot be estimated which of the plurality of correlation anomalies is the root cause, and it is difficult to determine the response.

  The present invention has been made in view of the above circumstances, and an object thereof is to provide a plant diagnosis system and a diagnosis method capable of estimating which of a plurality of correlation anomalies is a root cause. It is in.

  In order to achieve the above object, a representative present invention provides a diagnosis system for diagnosing a plant, wherein as the operation data of the plant, an operation data storage unit that stores changes over time of a plurality of operation parameters; A correlation change analysis unit that analyzes a change in correlation and determines that the correlation is abnormal when the change in correlation is greater than a preset threshold; and based on an analysis result of the correlation change analysis unit, the operation parameter is set as a node And a correlation anomaly graph creation unit for creating a correlation anomaly graph indicating the correlation anomaly by a link between nodes, and a causal relationship between operation parameters associated with normal operation of the plant as a qualitative model of the plant. Based on a qualitative model storage unit and a qualitative model involving the operation parameters constituting the correlation abnormality graph A qualitative graph creation unit that creates a qualitative graph that indicates the operation parameter by a node and indicates a causal relationship by a link between the nodes, and a plurality of components that constitute the correlation anomaly graph by comparing the correlation anomaly graph with the qualitative graph A root cause estimation unit that estimates which of the links is a root cause, and a display device that displays an estimation result of the root cause estimation unit.

  According to the present invention, it is possible to estimate which of a plurality of correlation anomalies is a root cause.

It is a figure showing the specific example of a structure of the plant which is the diagnostic object of the 1st Embodiment of this invention. It is a block diagram showing the structure of the diagnostic system in the 1st Embodiment of this invention. It is a figure showing the specific example of the structure information of the plant stored in the structure data storage part of the 1st Embodiment of this invention. It is a figure showing the specific example of the structure image of the plant stored in the structure data storage part of the 1st Embodiment of this invention. It is a figure showing the specific example of the correlation abnormality graph created in the correlation abnormality graph creation part of the 1st Embodiment of this invention. It is a figure showing the specific example of the qualitative graph created in the qualitative graph creation part of the 1st Embodiment of this invention. It is a figure showing the specific example of the correlation abnormality graph displayed with the display apparatus of the 1st Embodiment of this invention. It is a figure showing the specific example of the structural image of the plant displayed with the display apparatus of the 1st Embodiment of this invention. It is a flowchart showing the procedure of the diagnostic method in the 1st Embodiment of this invention. It is a block diagram showing the structure of the diagnostic system in the 2nd Embodiment of this invention. It is a figure showing the specific example of the corresponding | compatible operation information displayed with the display apparatus of the 2nd Embodiment of this invention. It is a flowchart showing the procedure of the diagnostic method in the 2nd Embodiment of this invention. It is a block diagram showing the structure of the diagnostic system in the 3rd Embodiment of this invention. It is a figure showing the specific example of the simulation result displayed with the display apparatus of the 3rd Embodiment of this invention. It is a flowchart showing the procedure of the diagnostic method in the 3rd Embodiment of this invention.

  A first embodiment of the present invention will be described with reference to the drawings.

  FIG. 1 is a diagram illustrating a specific example of a configuration of a plant that is a diagnosis target of the present embodiment.

  This plant includes a boiler device 1 that heats water, a boiler control device 2 that controls the boiler device 1, a power generation device 4 that generates electric power using hot water supplied from the boiler device 1 via the pump systems 3A and 3B, And a power generation control device 5 that controls the power generation device 4.

  The pump system 3 </ b> A includes a pump 6. The pump 6 has a mechanical seal S, seal chambers R1 and R2, and a working chamber R3.

  Purge water is supplied from the boiler device 1 to the seal chamber R1 of the pump 6 through the pipes t1 and t2 and the valve v1. The purge water supplied to the seal chamber R1 flows out to the seal chamber R2 through the mechanical seal S, and further flows out through the pipes t3 and t4 and the valve v2. A pipe t5 for sampling purge water is connected to the downstream side of the connection part of the pipe t3 in the seal chamber R2.

  The pipe t1 is provided with a measuring instrument 7a for measuring the purge water state quantity P1 (specifically, for example, temperature or pressure, the same applies hereinafter), and the pipe t2 is provided with a measuring instrument 7b for measuring the purge water state quantity P2. Is provided. The pipe t3 is provided with an instrument 7c for measuring the purge water state quantity P3, and the pipe t4 is provided with an instrument 7d for measuring the purge water state quantity P4. The pipe t5 is provided with a meter 7e for measuring the purge water quantity P5.

  Hot water is supplied from the boiler device 1 to the working chamber R3 of the pump 6 through the pipes t7 and t8 and the valve v3. And working room R3 pressurizes warm water and supplies it to power generator 4 via piping t6. The pipe t6 is provided with an instrument 7f for measuring the hot water state quantity P6.

  The pump system 3B has the same configuration as the pump system 3A, and a description thereof is omitted.

  The boiler device 1 adjusts the heat output on the pump system 3A side according to the heat output command value A1 from the boiler control device 2, and the heat on the pump system 3B side according to the heat output command value A2 from the boiler control device 2. Adjust the output. The boiler device 1 is provided with an instrument 7g for measuring the state quantity P7.

  The power generation device 4 adjusts the power generation output in accordance with the power generation output command value A3 from the power generation control device 5. The power generation device 4 is provided with a meter 7h that detects the state quantity P8.

  FIG. 2 is a diagram showing the configuration of the diagnostic system in the present embodiment.

  The diagnostic system of the present embodiment includes an operation data acquisition device 11, an operation data storage unit 12, a qualitative model storage unit 13, a configuration data storage unit 14, an arithmetic processing device 15, a display device 16, and an input device 17. The arithmetic processing unit 15 includes a correlation change analysis unit 18, a correlation abnormality graph creation unit 19, a qualitative graph creation unit 20, and a root cause estimation unit 21 as functional configurations. The input device 17 is for a user to input, and is configured with, for example, a keyboard and a mouse.

  The configuration data storage unit 14 stores plant configuration information and configuration images. FIG. 3 shows the configuration information of the pump system 3 </ b> A as a specific example of the plant configuration information stored in the configuration data storage unit 14. “CONN (part A, part B)” in FIG. 3 indicates the connection between part A and part B, and “DTCT (state quantity, part C)” in FIG. It is shown that. FIG. 4 shows a configuration image of the pump system 3 </ b> A as a specific example of the plant configuration image stored in the configuration data storage unit 14.

  The operation data acquisition device 11 periodically inputs signals from the plant (specifically, signals from the above-described meters 7a to 7h and the control devices 2 and 5), and outputs them to the operation data storage unit 12 in association with the time. To do. As a result, the operation data storage unit 12 stores changes over time of N operation parameters (specifically, the state quantities P1 to P8 and the command values A1 to A3 described above) as plant operation data. Yes.

  The correlation change analysis unit 18 of the arithmetic processing device 15 analyzes the change in the correlation between the operation parameters by, for example, a change detection method based on direct estimation of the density ratio described in Non-Patent Document 1.

More specifically, first, the past time range T1 to T2 is input by the input device 17, and the correlation change analysis unit 18 performs the operation data x ^{(n) at} point N in the past time range T1 to T2 (however, n = 1, 2,..., N) are read from the operation data storage unit 12. Similarly, the current time range t1 to t2 is input by the input device 17, and the correlation change analysis unit 18 operates the operation data x ′ ^{(n ′) at} point N ′ in the current time range t1 to t2 (however, n = 1, 2,..., N ′) are read from the operation data storage unit 12. Then, an optimum accuracy matrix Θ (11 × 11 square matrix in this embodiment) is calculated under the condition of the following formula (1). X ⁽ⁿ⁾ in the formula is a matrix (P1, P2, P3, P4, P5, P6, P7, P8, A1, A2, A3), and x ^{(n) T} means the transposed matrix. Similarly, x ′ ^{(n ′)} in the equation is a matrix (P1, P2, P3, P4, P5, P6, P7, P8, A1, A2, A3), and x ′ ^{(n ′) T} is It means the transpose matrix. Also, || Θ || ₁ in the equation means the sum of absolute values of all components of the accuracy matrix Θ, and R is a constant.

  As a result, the change in the correlation between the i-th (where i = 1, 2,..., 11) operation parameter and the j-th (where j = 1, 2,..., 11) operation parameter is expressed as an accuracy matrix Θ. To store. As the accuracy matrix Θ, a difference between the accuracy matrix Λ of the past operation data and the accuracy matrix Λ ′ of the current operation data or a difference of the partial correlation matrix may be used. When the absolute value of the component a (i, j) of the accuracy matrix Θ is larger than a preset threshold value, it is determined that the correlation between the i-th operation parameter and the j-th operation parameter is abnormal.

  Based on the analysis result of the correlation change analysis unit 18, the correlation abnormality graph creation unit 19 creates a correlation abnormality graph (sparse structure graph) indicating the operation parameters by nodes and indicating the correlation abnormality by links between the nodes. Yes.

  FIG. 5 shows a specific example of the correlation abnormality graph created by the correlation abnormality graph creation unit 19. The correlation abnormality graph shows the operation parameters P1 to P5 as nodes, the correlation abnormality between the operation parameters P1 and P3, the correlation abnormality between the operation parameters P1 and P5, the correlation abnormality between the operation parameters P2 and P3, and the operation parameters P2 and P4. The correlation abnormality, the correlation abnormality between the operation parameters P2 and P5, and the correlation abnormality between the operation parameters P3 and P5 are shown by links.

  The qualitative model storage unit 13 stores a causal relationship between operating parameters accompanying the normal operation of the plant as a qualitative model of the plant. The qualitative graph creation unit 20 reads from the qualitative model storage unit 13 a qualitative model related to the operation parameters constituting the above-described correlation abnormality graph. And based on the read qualitative model, the qualitative graph (sparse structure graph) which shows a driving | running parameter with a node and shows a causal relationship with the link between nodes is produced.

  FIG. 6 shows a specific example of the qualitative graph created by the qualitative graph creation unit 20. Since the correlation abnormality graph of FIG. 5 shows the operation parameters P1 to P5 as nodes, the qualitative graph of FIG. 6 also shows the operation parameters P1 to P5 as nodes.

  In the qualitative model storage unit 13, as a qualitative model, for example, according to the If then rule, “if the purge water state amount P 1 increases when the valve V 1 is not closed, the purge water state amount P 2 increases”, “valve If the purge water state amount P1 decreases while V1 is not closed, the purge water state amount P2 decreases "is stored. The qualitative graph of FIG. 6 shows the causal relationship between the operation parameters P1 and P2 by a link based on such information.

  Further, in the qualitative model storage unit 13, as a qualitative model, for example, according to the If then rule, “if the purge water state amount P 3 increases when the valve V 2 is not closed, the purge water state amount P 4 increases”. “If the purge water state amount P3 decreases when the valve V2 is not closed, the purge water state amount P4 decreases.” “If the purge water state amount P3 decreases, the purge water state amount P5 decreases. “Decrease” and “If the purge water state amount P3 decreases, the purge water state amount P5 decreases” are stored. The qualitative graph of FIG. 6 shows the causal relationship between the operation parameters P3 and P4 and the causal relationship between the operation parameters P3 and P5 based on these pieces of information.

  The qualitative model may be created by the user, or may be automatically created by the arithmetic processing unit 15 based on the plant configuration information stored in the configuration data storage unit 14. Further, it may be described in another format such as a table format instead of being described in the If then rule.

  The root cause estimating unit 21 compares the correlation abnormality graph described above with the qualitative graph described above, thereby estimating which of the plurality of links constituting the correlation abnormality graph is the root cause. .

  More specifically, first, the correlation abnormality graph and the qualitative graph are decomposed into an island set (a set of operation parameters connected to each other via a link). The correlation abnormality graph of FIG. 5 is an island set 1 {P1, P2, P3, P4, P5}, and the qualitative graph of FIG. 6 is an island set 2 {P1, P2} and an island set 3 {P3, P4, P5. }become. Then, in order to create an island set of the correlation abnormality graph, a link that connects the island sets of the qualitative graph to each other and that has the smallest number of links and a small component interval corresponding to the link is searched. In this example, any one of the links P1-P3, the links P1-4, the links P1-P5, the links P2-P3, the links P2-P4, and the links P2-P5 is a candidate, and the corresponding component interval is the smallest. The links P2-P3 are selected as things. Thereby, it is estimated that the link P2-P3 is the root cause. The part interval corresponding to the link (specifically, for example, the number of parts existing between parts, for example) is determined using the plant configuration information stored in the configuration data storage unit 14.

  The root cause estimation unit 21 further extracts an abnormality occurrence site corresponding to the link estimated as the root cause based on the plant configuration information stored in the configuration data storage unit 14. For example, the abnormality occurrence sites corresponding to the links P2-P3 are the mechanical seal S and the seal chambers R1, R2.

  For example, as shown in FIG. 7, the display device 16 is configured to display the correlation abnormality graph by making the links P <b> 2 to P <b> 3 estimated as the root cause distinguishable from other links. Alternatively, in accordance with the input of the input device 17, for example, the display is switched to the display of the correlation abnormality graph shown in FIG. 3 and the graph (not shown) showing only the links P2-P3 estimated as the root cause and the nodes on both sides thereof. You may do it.

  Further, for example, as shown in FIG. 8, the display device 16 makes the mechanical seal S and the seal chambers R <b> 1 and R <b> 2 extracted as an abnormality occurrence part distinguishable from other parts and displays a configuration image of the pump system 3 </ b> A. It is like that. The correlation abnormality graph and the configuration image of the pump system 3A are displayed and linked simultaneously. That is, if an arbitrary operation parameter is selected on the correlation abnormality graph, the measurement parameter measurement part blinks on the configuration image of the pump system 3A.

  Next, the diagnosis method and operational effects of this embodiment will be described. FIG. 9 is a flowchart showing the procedure of the diagnostic method in the present embodiment.

  First, in step S <b> 101, the operation data storage unit 12 stores changes with time of a plurality of operation parameters as plant operation data. Proceeding to step S102, the correlation change analysis unit 18 of the arithmetic processing unit 15 analyzes the change in the correlation between the operation parameters, and determines that the correlation is abnormal when the change in the correlation is greater than a preset threshold. In step S 103, the correlation abnormality graph creation unit 19 creates a correlation abnormality graph based on the analysis result of the correlation change analysis unit 18.

  And it progresses to step S104 and the qualitative graph preparation part 20 produces a qualitative graph based on the qualitative model in which the operation parameter which comprises a correlation abnormality graph is concerned. Proceeding to step S105, the root cause estimating unit 21 compares the correlation abnormality graph with the qualitative graph to estimate which of the plurality of links constituting the correlation abnormality graph is the root cause. Furthermore, an abnormality occurrence site corresponding to the link estimated as the root cause is extracted based on the plant configuration information.

  And it progresses to step S106 and the display apparatus 16 displays the estimation result of the root cause estimation part 21. FIG. Specifically, for example, the correlation abnormality graph of FIG. 7 is displayed. This makes it possible to estimate which of the plurality of links (that is, a plurality of correlation anomalies) is the root cause. Further, for example, a configuration image of the plant in FIG. 8 is displayed. Thereby, the abnormality occurrence part can be estimated.

  In the first embodiment, for example, the case where both the correlation abnormality graph of FIG. 7 and the plant configuration image of FIG. 8 are displayed has been described as an example. However, the present invention is not limited to this. Modifications are possible without departing from the scope of the invention. That is, for example, only one of the correlation abnormality graph in FIG. 7 and the plant configuration image in FIG. 8 may be displayed.

  A second embodiment of the present invention will be described with reference to FIGS. Note that in this embodiment, the same parts as those in the first embodiment are denoted by the same reference numerals, and description thereof will be omitted as appropriate.

  FIG. 10 is a block diagram showing the configuration of the diagnostic system in the present embodiment.

  The diagnosis system of this embodiment includes an operation data acquisition device 11, an operation data storage unit 12, a qualitative model storage unit 13, a configuration data storage unit 14, a corresponding operation information storage unit 22, an arithmetic processing unit 15A, a display device 16, and an input. A device 17 is provided. The arithmetic processing unit 15A includes a correlation change analysis unit 18, a correlation abnormality graph creation unit 19, a qualitative model creation unit 20, a root cause estimation unit 21, and a corresponding operation information extraction unit 23 as functional configurations.

  The corresponding operation information storage unit 22 stores a plurality of corresponding operation information for a plurality of abnormality occurrence sites. As one specific example, when the abnormality occurrence part is a part of the pump system 3A, as the first corresponding operation information, in order to reduce the heat output of the boiler device 1 to 1/2, the heat of the boiler control device 2 The fact that the output command value A1 is 0% is stored. Further, as the second corresponding operation information, it is stored that the valve v2 is closed in order to separate the pump system 3A. Moreover, it is stored as 3rd corresponding | compatible operation information that the electric power generation output E of the electric power generating apparatus 4 shall be 50%. In addition, the electric power generation output E of the electric power generating apparatus 4 is represented by following formula (2). In Formula (2), F1 is the heat output on the pump system 3A side, F2 is the heat output on the pump system 3B side, and T0 is a response time constant.

  The corresponding operation information extraction unit 23 extracts the corresponding operation information for the abnormality occurrence unit extracted by the root cause estimation unit 21 from the plurality of corresponding operation information stored in the corresponding operation information storage unit 22. .

  The display device 16 displays the corresponding operation information extracted by the corresponding operation information extraction unit 23 together with, for example, the correlation abnormality graph of FIG. 7 and the plant configuration image of FIG. That is, for example, when the abnormality occurrence part is a part of the pump system 3A, as shown in FIG. 11, as the corresponding operation information, the heat output command value A1 of the boiler control device 2 is set to 0%, and the valve v2 is closed. This indicates that the power generation output E of the power generation control device 5 is adjusted to 50%.

  Next, the diagnosis method and operational effects of this embodiment will be described. FIG. 12 is a flowchart showing the procedure of the diagnostic method in the present embodiment.

  In the present embodiment, steps S101 to S106 are performed as in the first embodiment. Then, the process proceeds to step S107, and the corresponding operation information extraction unit 23 extracts the corresponding operation information for the abnormality occurrence site extracted by the root cause estimation unit 21 from the plurality of corresponding operation information. The display device 16 displays the corresponding operation information extracted by the corresponding operation information extraction unit 23. Thereby, it can help a user's judgment.

  A third embodiment of the present invention will be described with reference to FIGS. In the present embodiment, the same parts as those in the first and second embodiments are denoted by the same reference numerals, and description thereof will be omitted as appropriate.

  FIG. 13 is a block diagram showing the configuration of the diagnostic system in the present embodiment.

  The diagnosis system of the present embodiment includes an operation data acquisition device 11, an operation data storage unit 12, a qualitative model storage unit 13, a configuration data storage unit 14, a corresponding operation information storage unit 22, a simulation model storage unit 24, an arithmetic processing unit 15B, A display device 16 and a command input device 17 are provided. The arithmetic processing unit 15B includes, as functional configurations, a correlation change analysis unit 18, a correlation abnormality graph creation unit 19, a qualitative graph creation unit 20, a root cause estimation unit 21, a corresponding operation information extraction unit 23, and a simulation execution unit 25. Have.

  The simulation model storage unit 24 stores a plant simulation model. This simulation model is described in the function block of the component part, and the function and time delay with respect to the input are set. For example, the heat output F1 on the pump system 3A side and the heat output F2 on the pump system 3B side of the boiler device 1 are expressed by the following equation (3). The total heat output F of the boiler device 1 is the sum of the heat outputs F1 and F2. The power generation output E of the power generation device 4 is expressed by the above formula (2).

  The simulation execution unit 25 uses the current operation data and the like (specifically, the state quantities P1 to P8 and the command values A1 to A3 and the opening amounts of the valves v1 to v3) as initial values for the simulation model described above. The simulation is executed on the assumption that the corresponding operation extracted by the root cause estimating unit 21 is executed while inputting.

  The display device 16 displays the simulation result. FIG. 14 shows a specific example of the simulation result displayed on the display device 16. This example shows temporal changes in the heat output F1 on the pump system 3A side, the heat output F2 on the pump system 3B side, the total heat output F, and the power generation output E. However, although not shown, a change with time of the operating parameter may be indicated.

  Next, the diagnosis method and operational effects of this embodiment will be described. FIG. 15 is a flowchart showing the procedure of the diagnostic method in the present embodiment.

  In the present embodiment, steps S101 to S107 are performed as in the second embodiment. And it progresses to step S108 and the simulation execution part 25 performs a simulation supposing the case where the corresponding | compatible operation information extracted by the root cause estimation part 21 was performed. In step S109, the display device 16 displays a simulation result. Thereby, it can help a user's judgment.

DESCRIPTION OF SYMBOLS 12 Operation | movement data storage part 13 Qualitative model storage part 14 Configuration data storage part 16 Display apparatus 18 Correlation change analysis part 19 Correlation abnormality graph creation part 20 Qualitative graph creation part 21 Root cause estimation part 22 Corresponding operation information storage part 23 Corresponding operation information extraction Section 24 Simulation model storage section 25 Simulation execution section

Claims (10)

In a diagnostic system for diagnosing a plant,
As the operation data of the plant, an operation data storage unit that stores temporal changes of a plurality of operation parameters;
A correlation change analysis unit that analyzes a change in correlation between operation parameters and determines a correlation abnormality when the change in correlation is greater than a preset threshold;
Based on the analysis result of the correlation change analysis unit, a correlation abnormality graph creating unit that creates a correlation abnormality graph that indicates the operation parameter by a node and indicates the correlation abnormality by a link between the nodes;
As a qualitative model of the plant, a qualitative model storage unit that stores a causal relationship between operation parameters accompanying normal operation of the plant,
Based on a qualitative model involving the operation parameters constituting the correlation abnormality graph, a qualitative graph creation unit that creates the qualitative graph that indicates the causal relationship with the nodes while indicating the operation parameters with the nodes;
By comparing the correlation abnormality graph with the qualitative graph, a root cause estimation unit that estimates which one of a plurality of links constituting the correlation abnormality graph is a root cause;
And a display device for displaying an estimation result of the root cause estimation unit. The diagnostic system according to claim 1,
The said display apparatus displays the link estimated as a root cause by the said root cause estimation part on the said correlation abnormality graph so that identification with another link is possible. The diagnostic system according to claim 1,
As the plant configuration data, a configuration data storage unit for storing the plant configuration information and a configuration image is provided,
The root cause estimation unit is extracting an abnormality occurrence site corresponding to the link estimated as a root cause in the root cause estimation unit based on the configuration information of the plant,
The said display apparatus displays the abnormality generation site | part extracted by the said root cause estimation part on the structure image of the said plant so that identification of other site | parts is possible. The diagnostic system according to claim 3,
A corresponding operation information storage unit for storing a plurality of corresponding operation information;
A plurality of corresponding operation information stored in the corresponding operation information storage unit, a corresponding operation information extraction unit that extracts the corresponding operation information for the abnormal site extracted by the root cause estimation unit,
The diagnostic system, wherein the display device displays the corresponding operation information extracted by the corresponding operation information extraction unit. The diagnostic system according to claim 4,
A simulation model storage unit for storing a simulation model of the plant;
Using a simulation model of the plant, comprising a simulation execution unit that executes a simulation assuming that the corresponding operation extracted by the corresponding operation information extraction unit is executed,
The display system displays a result of simulation. In a diagnostic method for diagnosing plant operation data,
As the operation data of the plant, a change with time of a plurality of operation parameters is stored,
Analyzing changes in correlation between operating parameters and determining that the correlation change is greater than a preset threshold value,
Based on the analysis result, create a correlation abnormality graph indicating the operation parameters with nodes and indicating the correlation abnormality with links between nodes,
Based on a qualitative model involving the operation parameters constituting the correlation abnormality graph, create a qualitative graph indicating the operation parameters with nodes and a causal relationship with a link between the nodes,
By comparing the correlation abnormality graph with the qualitative graph, it is estimated which of the plurality of links constituting the correlation abnormality graph is the root cause, and the estimation result is displayed. Method. The diagnostic method according to claim 6, wherein
A diagnostic method, wherein a link estimated as a root cause is displayed on the correlation abnormality graph so as to be distinguishable from other links. The diagnostic method according to claim 6, wherein
Based on the configuration information of the plant, extract the abnormality occurrence site corresponding to the link estimated as the root cause,
A diagnostic method characterized by displaying the abnormality occurrence part on the plant configuration image so as to be distinguishable from other parts. The diagnostic method according to claim 8, wherein
A diagnostic method, comprising: extracting and displaying corresponding operation information for the abnormality occurrence site from among a plurality of corresponding operation information. The diagnostic method according to claim 9, wherein
Using the simulation model of the plant, executing the simulation assuming the case where the extracted corresponding operation is executed,
A diagnostic method characterized by displaying a result of a simulation.

Images