Plant management system

Abstract

To provide a plant management system capable of supporting from identification of a cause for occurrence of an abnormality to recovery, and recovering the failure at an early stage.SOLUTION: A plant management system comprises: an abnormality signal storage unit which searches for process change amount data associated with a plant device when a failure occurs in the plant device, and stores the data in an abnormality signal database as image data: an abnormality history storage unit which stores an alarm history when a failure occurs and the movement and operation history of the device where the failure has occurred; and a failure analysis unit which extracts similar events of failure occurrence from the data when past device failures occurred. The failure analysis unit subjects the image data stored in the abnormality signal database and the image data stored when past device failures occurred or the history patterns stored in the abnormality history database and the history patterns stored when past device failures occurred to pattern matching and extracts similar events of failure occurrence.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a plant management system for managing the operation of a plant. [Background technology]

[0002] During normal operation of a nuclear power plant, plant monitoring by operators is performed in real time using the system monitoring screen of the plant monitoring system. In the unlikely event that a malfunction occurs in plant equipment (pumps, valves, etc.), it is necessary to quickly identify the cause of the malfunction (abnormality), recover from it, and return the plant to a normal operating state.

[0003] When a failure occurs in plant equipment (pumps, valves, etc.), it is necessary to quickly identify the cause of the failure (abnormality), recover the equipment, and return it to a normal operating state.Currently, operators identify the cause of the failure by comparing and contrasting various historical information and process value trends stored in different locations and investigating the sequence of occurrence, etc.

[0004] Depending on the circumstances of the failure, an investigation may be required at the site where the plant equipment is installed, which takes time. Therefore, in order to shorten the time until recovery, it is necessary to quickly identify the cause of the failure. In addition, based on the identified cause of the failure, we are considering recovery methods and taking recovery measures.

[0005] For example, in Patent Document 1, in order to monitor whether a plant has gone into an abnormal state, similar events are searched for using principal component scores obtained by performing principal component analysis on process amounts and types of alarms. That is, the purpose is to reduce the process amounts to be monitored and to facilitate monitoring whether a plant has gone into an abnormal state by converting a large number of process amounts, regardless of whether they are normal or abnormal, into principal component scores. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] JP 2001-75642 A Summary of the Invention [Problem to be solved by the invention]

[0007] In the technology of Patent Document 1, when identifying the cause of an abnormality after it occurs, the operation history is referred to in order to know the operation procedure for returning the plant to a normal state. However, the operation history is only used to display the operation history of the equipment that has influenced the change in the process amount, and is not included in the direct analysis target.

[0008] On the other hand, in the technology of Patent Document 1, when the amount of change in the process quantity becomes small (for example, when the device is chattering and repeatedly turning on and off), the principal component score also becomes constant, so the cause cannot be identified from the principal component score. In this case, it is considered that it is not possible to trace the operation history when the past abnormality occurred.

[0009] In consideration of the above-mentioned problems, an object of the present invention is to provide a plant management system that can effectively detect events that have a high degree of consistency with past abnormal conditions when identifying the cause of an abnormality, rather than at the stage of monitoring abnormal conditions in a plant, and that can provide support from identifying the cause to recovery, thereby enabling early recovery from a malfunction.

[0010] The above and other objects of the present invention and novel features of the present invention will become apparent from the description of this specification and the accompanying drawings. [Means for solving the problem]

[0011] The plant management system of the present invention is a plant management system for managing a plant, and includes a process signal input unit that stores process change data of process signals in equipment of the plant in a process change database; an abnormality signal storage unit that, when a fault occurs in the equipment of the plant, searches the process change data of the process signal linked to the equipment from the process change database and stores the data in an abnormality signal database as image data; an abnormality history storage unit that, when a fault occurs in the equipment of the plant, stores a history of issued alarms and the operation and manipulation history of the equipment in which the fault occurs in chronological order in the abnormality history database as history patterns; and a fault analysis unit that, when a fault occurs in the equipment of the plant, extracts events similar to the occurrence of the fault from data of past equipment faults. The fault analysis unit then extracts similar fault occurrences by pattern matching the image data stored in the abnormal signal database with image data stored when a past equipment fault occurred, or by pattern matching the history patterns stored in the abnormality history database with history patterns stored when a past equipment fault occurred. Effect of the Invention

[0012] According to the plant management system of the present invention, the fault analysis unit performs pattern matching between image data stored in the abnormal signal database and image data stored when a past equipment fault occurred, or performs pattern matching between a history pattern stored in the abnormality history database and a history pattern stored when a past equipment fault occurred, thereby extracting similar events of fault occurrence. As a result, when a fault occurs in plant equipment, past events that are highly similar can be extracted, providing support from identifying the cause of the fault to recovery, enabling early recovery from the fault.

[0013] Problems, configurations and effects other than those described above will become apparent from the following description of the embodiments. [Brief description of the drawings]

[0014] [Figure 1] 1 is a block diagram showing an overall process in an embodiment of a plant management system according to the present invention; DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0015] Hereinafter, the embodiments of the present invention will be described with reference to text and drawings. However, the structures, materials, and other specific configurations of the present invention are not limited to the embodiments described here, and can be appropriately combined or improved without departing from the spirit of the present invention.

[0016] The plant management system of the present invention is a plant management system that manages a plant. The plant management system of the present invention includes a process signal input unit, an abnormality signal storage unit, an abnormality history storage unit, and a fault analysis unit. The process signal input unit stores process variation data of the process signal in the equipment of the plant in a process variation database. When a fault occurs in a device of the plant, the abnormal signal storage unit searches the process variation data of a process signal associated with the device from the process variation database, and stores the data in the abnormal signal database as image data. When a fault occurs in a plant device, the fault history storage unit stores the history of the alarm issued and the operation and manipulation history of the device in chronological order as a history pattern in an fault history database. When a failure occurs in a plant device, the failure analysis unit extracts similar failure events from data on past equipment failures. Furthermore, the fault analysis unit extracts similar fault occurrences by pattern matching the image data stored in the abnormal signal database with image data stored when a past equipment fault occurred, or by pattern matching the history patterns stored in the abnormality history database with history patterns stored when a past equipment fault occurred.

[0017] According to the plant management system of the present invention, the fault analysis unit extracts similar fault occurrences by pattern matching image data stored in the abnormal signal database with image data stored when a past equipment fault occurred, or by pattern matching history patterns stored in the abnormality history database with history patterns stored when a past equipment fault occurred. As a result, when a fault occurs in plant equipment, past events that are highly similar can be extracted, providing support from identifying the cause of the fault to recovery, enabling early recovery from the fault.

[0018] The above plant management system may further include an equipment operation data editing unit that stores a history of operations and equipment operation in an equipment operation / operation history database when an operation is performed on equipment of the plant, and the fault analysis unit determines whether or not there is an operation history of the equipment from the history data stored in the equipment operation / operation history database before the occurrence of an equipment fault, and if there is an operation history of the equipment, increases the priority of pattern matching based on the abnormality signal database, and if there is no operation history of the equipment, increases the priority of pattern matching based on the abnormality history database. This configuration can further increase the matching rate between similar past events and currently occurring failures. Also, by setting priorities, the time required for pattern matching in abnormality history patterns that do not involve device operation can be reduced, and the time required to calculate the overall matching rate can be shortened.

[0019] The plant management system of the present invention is suitable for application to a nuclear power plant, but is not limited to application to a nuclear power plant and can be applied to various types of plants.

[0020] Hereinafter, a specific embodiment of the plant management system of the present invention will be described.

[0021] FIG. 1 is a block diagram showing the overall processing in an embodiment of a plant management system according to the present invention.

[0022] The plant management system 100 of this embodiment manages the operation of a plant 101 . The plant management system 100 shown in FIG. 1 has a process signal input unit 102, a process signal database 103, a process change amount storage unit 104, a process change amount database 105, an abnormal signal storage unit 106, an abnormal signal database 107, an abnormal signal determination unit 108, an alarm history database 109, an abnormality history storage unit 110, an abnormality history database 111, an equipment information database 112, an equipment operation data editing unit 113, an equipment operation / operation history database 114, a fault analysis unit 115, and a recovery procedure database 116.

[0023] In addition, in FIG. 1, the plant management system 100 has a display device 117 for displaying data and the like. This display device 117 is configured to include a touch panel or the like so that the user can operate the device.

[0024] A plant management system 100 shown in FIG. 1 manages a plant 101, for example, as described below.

[0025] During operation of the plant 101 , process signals of the plant 101 are stored in a process signal database 103 by a process signal input unit 102 . The data stored in the process signal database 103 is stored in a process variation database 105 by a process variation storage unit 104 together with the acquisition time of the process signal. During operation of the plant 101, an abnormality signal determination unit 108 monitors whether or not an abnormality has occurred in the plant 101.

[0026] When an equipment operation is performed from the display device 117, the equipment operation data editing unit 113 obtains equipment information from the equipment information database 112, outputs an equipment operation command, and stores the equipment operation history and the equipment operation history in the equipment operation / operation history database 114.

[0027] When a fault occurs in equipment of the plant 101, an abnormal signal determination unit 108 that monitors the abnormal state of the process signal stores information about the generated alarm together with the time of the fault in an alarm history database 109. The abnormality signal storage unit 106 converts the process change amount data stored in the process change amount database 105 up to the time the fault occurred into image data, and stores the image data in the abnormality signal database 107 together with the time the fault occurred and device information of the device in which the fault occurred, obtained from the device information database 112. The abnormality history storage unit 110 stores the equipment operation and equipment operation history data stored in the equipment operation / operation history database 114 up to the time the failure occurred and the alarm history data stored in the alarm history database 109 as an abnormality history pattern in the abnormality history database 111, together with the time the failure occurred and the equipment information of the equipment in which the failure occurred, obtained from the equipment information database 112.

[0028] Thereafter, the process up to the point of failure recovery is carried out based on information obtained by the process control system 100, such as process variation data and device information. Specifically, the condition of the equipment in which the failure occurred is confirmed, and an on-site investigation is conducted. The location of the failure is then identified, the cause of the failure is investigated, and a recovery method is considered. Recovery work is then carried out according to the recovery method that has been decided upon.

[0029] After recovery from the failure, the process control system 100 stores the procedure used for recovery in the recovery procedure database 116 together with the time of occurrence of the failure and information on the device in which the failure occurred.

[0030] As described above, for each occurrence of a fault, information about the fault (the time the fault occurred, alarm information, process change data up until the time of the occurrence, historical data on equipment operation and equipment operation, information about the equipment where the fault occurred, etc.), the procedure manual used for recovery, etc. are saved.

[0031] Next, the operation of the process control system 100 in the case where, when a fault occurs, information on a past event in which the fault occurred is used will be described.

[0032] When using information about past events in which a fault occurred, an analysis request is made from the display device 117 to the fault analysis unit 115 of the process control system 100 based on the user inputting an analysis request onto a touch panel or the like of the display device 117.

[0033] The fault analysis unit 115 receives an analysis request from the display device 117 and extracts events similar to the fault that has occurred from the data stored in the abnormal signal database 107 and the abnormality history database 111 by using pattern matching of process variation or abnormality history. More specifically, the pattern matching of the process variation is performed by performing pattern matching using image recognition with image data of past events stored in the abnormal signal database 107, and calculating the coincidence rate. The pattern matching of the abnormality history is performed by matching with the abnormality history patterns of past events stored in the abnormality history database 111, and the matching rate is calculated.

[0034] Next, the failure analysis unit 115 searches for a recovery procedure stored in the recovery procedure database 116 based on the extracted failure occurrence time of the past event and the device information, and displays the procedure on the display device 117 .

[0035] In addition, when extracting similar events, if there is no operation history of the faulty device for a certain period of time prior to the time the fault occurs, the priority of pattern matching of the process change amount is increased, and if there is an operation history for a certain period of time prior to the time the fault occurs, the priority of pattern matching of the abnormality history pattern is increased, and the match rate is calculated.

[0036] In the case of an abnormality that does not involve device operation, the amount of data for the abnormality history pattern will be small, and it is expected that pattern matching cannot be performed with sufficient accuracy. In this case, by increasing the priority of pattern matching based on the amount of process change, it is possible to increase the matching rate for the occurrence of anomalies not involving equipment operation. Furthermore, by setting the priority, the time required for pattern matching in abnormality history patterns that do not involve device operation can be reduced, and the time required for calculating the overall matching rate can be shortened.

[0037] According to the plant management system 100 of the present embodiment described above, the fault analysis unit 115 extracts similar fault occurrence events by performing pattern matching between image data stored in the abnormal signal database 107 and image data stored when a past equipment fault occurred, or by performing pattern matching between history patterns stored in the abnormality history database 111 and history patterns stored when a past equipment fault occurred. As a result, when a fault occurs in equipment of plant 101, fault analysis unit 115 extracts past events with high similarity, thereby providing support from identifying the cause of the fault to recovery, thereby enabling early recovery from the fault.

[0038] The present invention is not limited to the above-described embodiments, but includes various modified examples. For example, the above-described embodiments are described in detail to easily explain the present invention, and the present invention is not necessarily limited to those including all of the configurations described. [Explanation of symbols]

[0039] 100 Plant management system, 101 Plant, 102 Process signal input section, 103 Process signal database, 104 Process change amount storage section, 105 Process change amount database, 106 Abnormal signal storage section, 107 Abnormal signal database, 108 Abnormal signal determination section, 109 Alarm history database, 110 Abnormal history storage section, 111 Abnormal history database, 112 Equipment information database, 113 Equipment operation data editing section, 114 Abnormal operation / operation history database, 115 Fault analysis section, 116 Recovery procedure database, 117 Display device

Claims

1. A plant management system that manages a plant, a process signal input unit that stores process variation data of a process signal in an apparatus of the plant in a process variation database; an abnormality signal storage unit that, when a failure occurs in a device of the plant, searches the process variation database for the process signal associated with the device and stores the searched data in an abnormality signal database as image data; an abnormality history storage unit that stores, when a fault occurs in a device of the plant, a history of issued alarms and a history of the operation and manipulation of the device in which the fault occurred in chronological order as a history pattern in an abnormality history database; a failure analysis unit that, when a failure occurs in equipment of the plant, extracts similar events of the failure from data of past equipment failures, The failure analysis unit performs pattern matching between image data stored in the abnormal signal database and image data stored when a past equipment failure occurred, or performs pattern matching between a history pattern stored in the abnormality history database and a history pattern stored when a past equipment failure occurred, thereby extracting similar events of failure occurrence. A plant management system characterized by:

2. The plant further includes an equipment operation data editing unit that stores a history of the operation and the operation of the equipment in an equipment operation / operation history database when the equipment of the plant is operated, The failure analysis unit determines whether or not there is an operation history of the device from history data stored in the device operation / operation history database before the device failure occurs, and if there is an operation history of the device, increases the priority of pattern matching based on the abnormal signal database, and if there is no operation history of the device, increases the priority of pattern matching based on the abnormality history database.

2. The plant management system according to claim 1.

Images