JPH11175144A - Maintenance supporting device for plant equipment - Google Patents

Abstract

PROBLEM TO BE SOLVED: To make surely and correctly estimable the cause of a trouble by a simple operation at the time of a failure generated at a plant equipment by specifying the cause generating the pertinent state of the plant equipment acceding to the strength of a relation by sorting by an event-based cause sorting part and the length of time by sorting by a time-based cause sorting part. SOLUTION: A relating degree storing part 101 stores a relation between the candidate of the cause of the failure and an event which can be generated by each cause of a failure. A failure interval storing part 102 stores the time interval of generating the failure for every cause of the failure. The event-based cause sorting part 103 sorts causes based on the strength of a relation among the candidates of the failure cause of the part 101 with respect to an observed event. The time-based cause sorting part 104 estimates the cause based on the length of the interval of the failures. A cause estimating part 105 specifies the cause of the generation of the pertinent state of the device according to the strength of the relation by sorting by the part 103 and the length of time by sorting by the part 104.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus for providing support information when performing maintenance work for restoration of a plant equipment in a power plant or the like when a failure occurs, and in particular, to quickly identify items that may cause a failure. The present invention relates to a maintenance support device for plant equipment, which notifies maintenance personnel and provides information for instructing a procedure of a fault recovery work.

[0002]

2. Description of the Related Art In thermal and nuclear power plants, inspections are carried out several times a day to maintain stable operation, and periodic inspections after plant operation is stopped are basically performed for two to four years. It is implemented at annual intervals. The number of devices subject to these patrols and periodic inspections is enormous.Recovery work once a failure has occurred is not limited to identification of the cause, but also provisional recovery measures, planning of permanent measures, parts for countermeasures. A great deal of labor is required for arranging, replacing, and the like. In addition, if such a plant is stopped for a long time, the economic loss becomes extremely large. On the other hand, particularly in these operations, estimating the cause requires skillful and experienced knowledge, and inexperienced maintenance personnel may make a wrong decision.

For this reason, it is very important that, when a failure occurs, provisional measures are taken to enable the rapid narrowing down of the cause of the failure and the continuation of the plant operation, and the maintenance person estimates the cause by a simple operation. There is a need for a device that can provide support information for planning measures and full-scale recovery.

For example, a conventional maintenance support device has a configuration shown in FIG. 23 for inference by a knowledge base. In other words, this device comprises a knowledge storage unit 11 and a matching unit 12
And a cause estimating unit 13, which is “IF to THEN
It is possible to specify the cause of a malfunction by a so-called rule base that describes knowledge for diagnosis in accordance with "rules". In some cases, the cause of the failure is specified by a relatively simple Tedel base as shown in FIG. 24 without using a full-scale database.

[0005]

However, when a rule-based method is used for the purpose of plant restoration, a large amount of time is required until a final diagnosis result is obtained, such as spending time in specifying input items. However, there is a drawback that quick response is not possible. In addition, the structure of the knowledge base is complicated, and it is difficult to intuitively grasp the causal relationship between the cause and the event,
There is a problem that it is difficult to apply according to the situation on site. Maintenance of the knowledge base also requires knowledge engineering experts, and although the results obtained by actual operation are effective later, there is a drawback that they cannot be used effectively.

Further, the conventional table-based method enables a maintenance person to search for the cause by a simple and quick operation, but quantitatively reflects the operation history of the plant equipment operated continuously. It has no function and is used only in a limited part of the scene. Especially,
When the same type of plant is operated, a method of reflecting the operation history on an integrated knowledge base has not yet been put to practical use.

[0007] An object of the present invention is to provide a plant equipment capable of accurately estimating a cause by a simple operation in the case of a trouble occurring in a huge amount of plant equipment constituting a power plant, and obtaining support information for planning a countermeasure and for full-scale restoration. A maintenance support device is provided.

[0008]

In order to achieve the above object, an invention according to claim 1 stores a degree of association which stores numerical values of a candidate of a cause of a defect and a relationship between an observed event and the candidate of the cause of a defect. Unit, a failure interval storage unit that stores a time interval at which a failure occurs for each failure cause, and a failure cause candidate given from the relevance storage unit with respect to an observed event, based on the strength of the relationship. Event-based cause classification unit for classifying, time-based cause classification unit for classifying causes based on the length of the failure interval provided from the failure interval storage unit, and the strength and time-based cause of the relationship based on the classification by the event-based cause classification unit And a cause estimating unit that specifies the cause of the occurrence of the state of the plant equipment according to the length of time according to the classification by the classification unit.

In the maintenance support apparatus having the above-described configuration, when a failure occurs in a huge amount of plant equipment constituting a power generation plant, the operation history of the plant is quantitatively diagnosed while ensuring the simplicity of the operation in estimating the cause. It can be used as knowledge and the accuracy of cause classification can be improved.

Further, the invention according to claim 2 is characterized in that the malfunction interval storage unit stores an operation time storage unit that stores the total operation time of the plant, and a malfunction interval correction unit that corrects the malfunction interval stored in the malfunction interval storage unit. In addition, when the true cause is found, the malfunction interval of the cause stored in the malfunction interval storage unit is corrected based on the operation time given from the operation time storage unit.

In the maintenance support device having the above-described configuration, the fault interval of the cause stored in the fault interval storage unit is corrected based on the operation time in the operation time storage unit. A failure interval reflecting the characteristics can be obtained, and even in a plant where quantitative data on the failure interval has not been obtained in advance, it is possible to sequentially improve diagnostic accuracy.

According to a third aspect of the present invention, when the cause estimating unit determines the true cause, the cause and event history storage unit stores the cause and the observed event, and the cause and event history storage unit stores the cause and the observed event. A history-based cause classification unit that searches for and classifies the cause of the coincidence between the current event and the event observed in the current plant is provided.When the cause is estimated by the cause estimation unit, the relationship based on the classification by the event-based cause classification unit is determined. The cause that caused the state of the plant equipment to be specified is specified according to the length of time according to the classification by the strength and time based cause classification unit or the presence or absence of the cause according to the classification by the history based cause classification unit.

In the maintenance support apparatus having the above-described configuration, whether or not the event of the fault that occurred in the past and the event observed in the current plant match can be added to the criterion of the cause classification, and the diagnostic accuracy can be further improved. Can be improved.

Further, in the invention according to claim 4, the relevance storage unit calculates the product of the relevance and the failure interval for each event and each cause combination to obtain the maximum value for each event, and the maximum value is large. An event input order determination unit that determines the order as the order of event input is provided, and the relevance stored in the relevance storage unit is two-dimensionally related to a plant state that can be caused by each failure cause for a failure cause candidate. It is characterized in that a matrix table is replaced with numerical values and stored.

In the maintenance support apparatus having the above-described configuration, events having a large influence on the classification of the cause of a failure can be sequentially input, and when a quick cause classification is required, the events can be efficiently input. Become.

The invention according to claim 5 is characterized in that the event input order judging section has a primary investigation number storage section for storing the minimum number of event inputs as a primary investigation number, and the primary input is performed in the order determined by the event input order judging section. Primary diagnosis is performed based on the content of the events input by the cause estimating unit by batch inputting up to the number of investigations, and then inputting and estimating the cause of the events having the input order after the number of primary investigations are performed one item at a time. It is characterized by the following.

In the maintenance support apparatus having the above configuration, when quick cause classification is required, the cause can be provisionally classified even during the process of inputting all events.

Further, the invention of claim 6 is characterized in that when the number of events input to the cause estimating unit is less than all events, the product of the sum of the relevance of the observed event and the defect rate with respect to the sum of the relevance is not input. The upper limit value of the degree of inconvenience provided by the upper limit value of the inferiority degree provided by the upper limit of the inferiority degree in the estimation by the cause estimating section is provided with an inferiority degree upper limit value calculating section for calculating a numerical value divided by the sum of all the degrees of relevance including , The cause of the large value is identified at this point in time.

In the maintenance support apparatus having the above configuration, when quick cause classification is required, the user can quantitatively determine whether or not the classification operation has been substantially determined in the middle of the process of inputting all events. Thus, it is possible to determine whether or not it is necessary to continue the event input.

According to a seventh aspect of the present invention, there is provided an assertive rule storage unit for storing an event sufficient for the cause estimating unit to determine the cause of the failure and a failure cause name which can be determined when the event is observed. A rule-based cause classification unit that classifies the cause based on whether or not the event satisfies the rules of the assertive rule storage unit. The cause that caused the state of the plant equipment is specified in accordance with the length of the time according to the classification by the strength and time based cause classification unit or the pass / fail of the rule according to the classification by the rule based cause classification unit.

In the maintenance support apparatus having the above-described configuration, when there is a definite cause for the content of the event, the cause is completely specified regardless of the degree of relevance and the failure interval or the history data. It becomes possible.

Further, the invention according to claim 8 is a satellite data totalizing section for totalizing each cause event history and operation time recorded as a result of operation of a plurality of plants of the same type,
A failure interval initial value storage unit that holds the initial values of the failure interval and the operation time of the plant in the corresponding format, and a total number of the cause event histories of all plants added to a value obtained by dividing the operation time initial value by the failure interval initial value. The sum of the total operation time of the same type of plant and the initial value of the operation time totaled in the satellite data tabulation unit as the number of defects is defined as the total operation time, and the total operation time divided by the total number of defects is calculated as the new Higher-level system including a defect interval correcting unit that determines a defective interval, a corrected data distribution unit that distributes a corrected defect interval provided from the defect interval correcting unit to each plant, and a lower system including a maintenance support device of the present invention. And characterized in that:

In the maintenance support apparatus having the above configuration, the operation history of a plurality of the same kind of plants located in remote places is totaled in a host system, so that the trouble interval can be calculated with high accuracy based on a long plant operation time. Can be.

[0024]

Embodiments of the present invention will be described below with reference to the drawings.

(First Embodiment) FIG. 1 shows the configuration of the present embodiment, in which a relevance storage unit 101, a failure interval storage unit 102, an event-based cause classification unit 103, a time-based cause A classification unit 104 and a cause estimation unit 105 are provided.

The association degree storage unit 101 stores a relationship between a candidate for a cause of a failure and an event that can be caused by each cause of the failure. The storage items here are, for example, as shown in FIG. 2, a relationship between a cause of a failure and an event that can occur due to the cause of the failure in the table-type knowledge base. The strength of the relationship is not limited to a symbol, but may be represented by a numerical value.

Further, the fault interval storage unit 102 stores a time interval at which a fault occurs for each fault cause. This is, for example, as shown in FIG.
The time interval at which the malfunction occurs is stored.

The event-based cause classifying unit 103 classifies the observed event based on the strength of the relationship among the fault cause candidates in the fault relevance storage unit. Here, a score is assigned to each symbol, and the cause is classified based on the strength of the relationship.

Further, the time-based cause classification unit 104 classifies the cause based on the length of the defect interval. This causes the cause to be classified based on the length of the defect interval, as shown in FIG. 3B, for example. In this example, classification is performed in the order of shortest failure interval.

The cause estimating unit 105 specifies the cause of the state of the device in accordance with the strength of the relationship based on the classification by the event-based cause classifying unit 103 and the length of time based on the classification by the time-based cause classifying unit 104.

FIG. 4 shows an example in which the cause estimating unit 105 determines that the cause of the state has occurred. In this example, the event-based cause classification unit 103 scores each defect based on a table-type knowledge base. In addition, the cause estimating unit 105 obtains the reciprocal of the time stored in the defect interval storage unit 102, and determines that the defect in which the product of the reciprocal and the score obtained here is the maximum is the cause of the state. doing.

As described above, according to the maintenance support apparatus for plant equipment of the first embodiment, in the event of a problem occurring in a huge amount of plant equipment constituting a power generation plant, the operation for estimating the cause is ensured with ease. The operation history of the plant can be used as quantitative diagnostic knowledge, and the accuracy of cause classification can be improved.

(Second Embodiment) FIG. 5 shows another embodiment of the present invention. This embodiment is directed to correcting a trouble interval on the basis of a trouble experience and a plant operating time. Is what makes it possible. This apparatus includes a relevance storage unit 101, a failure interval storage unit 102, an event-based cause classification unit 103, a time-based cause classification unit 104, a cause estimation unit 105, a driving time storage unit 106, a failure interval correction unit 107. Among the above configurations, the relevance storage unit 101, the failure interval storage unit 102, the event-based cause classification unit 103, the time-based cause classification unit 104, and the cause estimation unit 10
5 operates in the same manner as in the first embodiment.

The operation time storage unit 106 stores the total operation time of the plant. Here, the operation time of the plant is stored as shown in FIG.

Further, when the true cause is determined, the trouble interval correction unit 107 determines the trouble interval of the cause stored in the trouble interval storage unit using the operation time of the operation time storage unit based on the following equation. Fix it.

K = i / t1 (1) inew = (t1 + t2) / (k + 1) (2) where k: failure frequency calculation value inew: corrected defect interval t1: lastly updated defect interval Plant operation time t2: Time from the last update until the true cause is found i: Fault interval before correction FIG. 6 (b) shows an example in which the fault interval is corrected based on equations (1) and (2). Is shown in This example is an example in which a cabin bend has been found as a true cause after a lapse of 1000 hours from the last update of the trouble interval. There is also a method of correcting a malfunction interval using the operation time from the start of operation of the apparatus. further,
It is also possible to update the defect interval on the assumption that no cause other than the cause has occurred by the time of updating the defect interval.

As described above, according to the maintenance support apparatus for plant equipment of the second embodiment, the failure interval of the cause stored in the failure interval storage unit 102 is determined based on the operation time of the operation time storage unit 106. I decided to fix it,
Failure intervals that gradually reflect the characteristics of the actual machine can be obtained with the operation, and even in a plant for which quantitative data on the failure intervals has not been obtained in advance, the diagnostic accuracy can be gradually improved.

(Third Embodiment) FIG. 7 shows a third embodiment of the present invention.
In the present embodiment, whether or not an event of a fault that has occurred in the past coincides with an event observed in the current plant is added to the judgment criteria for cause classification.

This apparatus includes a relevance storage unit 101, a failure interval storage unit 102, an event-based cause classification unit 103, a time-based cause classification unit 104, a cause estimation unit 105, a cause event history storage unit 108, a history Standing cause classification unit 109
And Among the above configurations, the relevance storage unit 10
1, defect interval storage unit 102, event-based classification unit 103,
The time-based cause classification unit 104 and the cause estimation unit 105
The operation is the same as that of the embodiment.

The cause event history storage unit 108 stores the cause and the observed event when the true cause is found.
Here, for example, assuming that the aging of the gap in the radial direction has been found as the true cause, the event group observed at that time is stored separately for each item as shown in FIG. In this example, whether or not observation has been performed is stored. However, if unknown, it may be possible to store that it is unknown.

The history-based cause classification unit 109 searches for and classifies the cause in which the event already stored in the cause event history storage unit 108 matches the event observed in the current plant.

Further, in the cause estimating unit 105, the strength of the relationship based on the classification in the event based cause classifying unit 103 and the length of time based on the classification in the time based cause classifying unit 104 or the cause based on the classification in the history based cause classifying unit 109 The cause that caused the state of the plant equipment according to the presence / absence is specified. FIG. 9 shows an example in which the event observed in the current plant matches the event stored in the history-based cause classification unit 109, and is determined to be the cause of the failure.

As described above, according to the plant equipment maintenance support apparatus of the third embodiment, the cause classification judgment is made based on whether or not the event in the past trouble and the event observed in the current plant coincide. It can be added to the standard, and the diagnostic accuracy can be further improved.

(Fourth Embodiment) FIG. 10 shows a fourth embodiment of the present invention. This embodiment is particularly quick by inputting events in order from the one having the greatest influence on the classification of the cause of a defect. This makes it possible to efficiently input an event when a specific cause classification is required.

The apparatus includes a relevance storage unit 101, a failure interval storage unit 102, an event-based cause classification unit 103, a time-based cause classification unit 104, a cause estimation unit 105, and an event input order determination unit 110. Have. In the above configuration, the failure interval storage unit 102, the event-based cause classification unit 10
3. The time-based cause classification unit 104 and the cause estimation unit 105 operate in the same manner as in the first embodiment.

The relation storage unit 101 stores the relation stored in a two-dimensional matrix table such as that shown in FIG. Replace and save.

The event input order judging section 110 calculates the product of the relevance and the reciprocal of the defect interval for each event and each cause combination, and obtains the maximum value for each event. The order of the largest value is determined as the order of event input. FIG. 12 shows an example in which the order of event input is determined by the above method. In this example, “operating state length” and “small aging change”
Since the maximum values are equal, it is determined that the "operating time length" having the larger total value for each event is input first.

As described above, according to the maintenance support apparatus for plant equipment of the present embodiment, it is possible to sequentially input events having a large influence on the classification of the cause of a failure. It becomes possible to input efficiently.

(Fifth Embodiment) FIG. 13 shows a configuration in which a primary survey number storage unit 111 is added to the fourth embodiment. That is, this device is associated with the relevance storage unit 1
01, a failure interval storage unit 102, an event-based cause classification unit 103, a time-based cause classification unit 104, a cause estimation unit 105, an event input order determination unit 110, and a primary investigation number storage unit 111. I have.

The event input order determining unit 110 collectively inputs up to the number of primary investigations in the determined order,
The primary diagnosis is performed based on the content of the event input in step 5, and subsequently, the input of the event having the input order after the number of primary investigations and the estimation of the cause are performed one item at a time.

As described above, according to the present embodiment, when quick cause classification is required, the cause can be provisionally classified even during the process of inputting all events.

(Sixth Embodiment) FIG. 14 shows an apparatus similar to the fourth embodiment, which is configured together with the defect degree upper limit value calculation unit 112. That is, the apparatus includes an association degree storage unit 101, a failure interval storage unit 102, an event-based cause classification unit 103, a time-based cause classification unit 104, a cause estimation unit 105, an event input order determination unit 110, a failure degree An upper limit value calculation unit 112.

The defect degree upper limit value calculation unit 112 compares the results of the event standing cause classification unit 103 and the time standing cause classification unit 104 with the observed events relative to the sum of the relevance when the number of input events is less than all the events. The product of the sum of the degrees of relevance and the defect rate is divided by the sum of all the degrees of relevance including the uninput event, and a numerical value calculated as the upper limit of the degree of defect.

In the cause estimating unit 105, the cause of the large value is specified as the cause of the state of the plant equipment at the present time based on the upper limit of the degree of malfunction given by the upper limit of degree of malfunction calculation unit 112. The above determination example is shown in FIG.
Is shown in This example is for the case where three events are currently input. Since the score changes according to the input of "axial contact", the upper limit value of the defect degree is calculated assuming that a state that increases the score of the defect cause is observed, and finally "radial gap aging" is calculated. Judge as the cause.

As described above, according to the present embodiment, when quick cause classification is required, the user can determine whether or not the classification operation has been substantially determined in the middle of the process of inputting all events. This makes it possible to determine whether it is necessary to continue the event input.

The fourth, fifth and sixth aspects of the present invention described above
In the maintenance support device for plant equipment of the embodiment of the present invention, when a quick cause classification is required by performing a provisional cause classification at an intermediate stage in the process of inputting all the events for the cause classification, the effect is effective. It becomes possible to input it.

(Seventh Embodiment) FIG. 16 shows a seventh embodiment of the present invention, in which a function of cause classification based on a rule base is added. This device has an association degree storage unit 101 and a failure interval storage unit 102
, Event-based cause classification unit 103, time-based cause classification unit 104, cause estimation unit 105, assertive rule storage unit 11
3 and a rule-based cause classification unit 114.

The assertive rule storage unit 113 stores an event sufficient to determine the cause of the malfunction and the name of the cause of the malfunction that can be determined when the event is observed. FIG. 17 shows a storage example of the above-mentioned trouble cause name. In this example, the cause of the failure is determined to be affirmative, but the cause of the failure may be determined to be negative.

The rule-based cause classification unit 114 classifies the cause based on whether the observed event satisfies the rule in the assertive rule storage unit 113. FIG. 18 shows an example in which the observed event satisfies the rule shown in FIG. In this example, the event of "operating time length" and "radial contact" match the rule, and the event of "axial contact" and "small aging change" satisfies the rule because it does not depend on the observation on the rule You have determined that.

In the cause estimating unit 105, the strength of the relationship based on the classification in the event based cause classifying unit 103 and the length of time based on the classification in the time based cause classifying unit 104, or the pass / fail of the rule based on the classification in the rule based cause classifying unit 111. According to the above, the cause of the occurrence of the state of the plant equipment is specified.

As described above, according to the maintenance support apparatus for plant equipment of the present embodiment, irrespective of the degree of relevance and the failure interval, or the presence or absence of history data, the content of the event is deterministic. When a cause exists, the cause can be specified without omission.

(Eighth Embodiment) FIG. 19 shows an eighth embodiment of the present invention. In this embodiment, the operation histories of a plurality of similar plants existing in remote locations are totaled in a host system. By doing so, the malfunction interval is corrected based on the long plant operation time. The apparatus according to the tenth embodiment is arranged as a lower system 115 in each of the remote similar plants. The host system includes a satellite data totaling unit 116 and a fault interval initial value storage unit 11.
7, the defect interval correction unit 118, and the correction data distribution unit 1
19

The satellite data totaling section 116 totalizes the history of each causal event and the operation time recorded as a result of the operation of the plurality of plants of the same type. Here, the history for each plant is collected, for example, as shown in FIG. The failure interval initial value storage unit 117 stores the failure interval and the initial value of the operation time of the type plant.

FIG. 21 shows an example of the initial value of the trouble time interval. As shown in the figure, an initial value of a trouble interval and an operation time is stored in the host system for each plant type. The failure interval correction unit 118 divides the initial operation time by the failure interval initial value and adds the number of cause event histories of all plants to the total number of failures. The sum of the total time and the operation time initial value is defined as the total operation time, and a value obtained by dividing the total operation time by the total number of defects is determined as a new defect interval of the type plant.

FIG. 22 shows an example of a corrected defect interval obtained by dividing the total operation time by the total number of defects.
The host computer recalculates the defect interval for each format.
The correction data distribution unit 119 distributes the corrected defect interval to each plant through a network or a storage medium such as a CD-ROM.

As described above, according to the maintenance support apparatus for plant equipment of the present embodiment, the operation history of a plurality of similar types of plants existing in remote locations is totaled in the host system, so that the operation history can be calculated based on the long plant operation time. The defect interval can be calculated with high accuracy.

[0067]

As described above, according to the present invention, in the event of a problem occurring in plant equipment, it is possible to maintain the simplicity of the operation in estimating the cause without relying on a complicated knowledge base and to accurately capture the operation history of the plant. The cause of the failure can be specified. Therefore, according to the present invention, it is possible to obtain support information for planning and restoring measures in a short time.

[Brief description of the drawings]

FIG. 1 is a functional configuration diagram showing an embodiment of a maintenance support device for plant equipment according to the present invention.

FIG. 2 is a diagram showing an example of storage in an association degree storage unit in FIG. 1;

3A is a diagram showing a storage example in a trouble interval storage unit in FIG. 1, and FIG. 3B is a diagram showing a classification example in a time standing cause classification unit.

FIG. 4 is a diagram showing an example of determination by a cause estimating unit in FIG. 1;

FIG. 5 is a functional configuration diagram showing another embodiment of the present invention.

6A is a diagram illustrating a storage example in an operation time storage unit in FIG. 5, and FIG. 6B is a diagram illustrating a correction example in a trouble interval correction unit.

FIG. 7 is a functional configuration diagram showing another embodiment of the present invention.

8 is a diagram showing a storage example in a cause event history storage unit in FIG. 7;

FIG. 9 is a diagram illustrating an example of determination by a cause estimating unit in FIG. 7;

FIG. 10 is a functional configuration diagram showing another embodiment of the present invention.

FIG. 11 is a diagram showing a storage example in a relevance storage unit in FIG. 10;

FIG. 12 is a diagram showing an example of determination by an event input order determination unit in FIG. 10;

FIG. 13 is a functional configuration diagram showing another embodiment of the present invention.

FIG. 14 is a functional configuration diagram showing another embodiment of the present invention.

FIG. 15 is a diagram showing an example of determination by the cause estimating unit in FIG. 14;

FIG. 16 is a functional configuration diagram showing another embodiment of the present invention.

FIG. 17 is a diagram showing a storage example in the assertive rule storage unit of FIG. 16;

FIG. 18 is a diagram illustrating an example of determination by the cause estimating unit in FIG. 16;

FIG. 19 is a functional configuration diagram showing another embodiment of the present invention.

FIG. 20 is a diagram showing a modification example in the satellite data collection unit in FIG. 19;

FIG. 21 is a diagram showing an example of storage in a failure interval initial value storage unit in FIG. 19;

FIG. 22 is a diagram showing an example of correction by a defect interval correction unit in FIG. 19;

FIG. 23 is a diagram showing an example of a conventional maintenance support device.

FIG. 24 is a diagram showing another example of a conventional maintenance support device.

[Explanation of symbols]

 101 Relevance storage unit 102 Failure interval storage unit 103 Event-based cause classification unit 104 Time-based cause classification unit 105 Cause estimation unit 107 Failure interval correction unit 109 History-based cause classification unit 110 Event input order determination unit 112 Failure degree upper limit calculation unit 114 Rule-based cause classification unit

Claims (8)

[Claims] 1. A failure degree candidate, a degree-of-association storage unit for storing numerical values of the relationship between an observed event and the failure cause candidate, and a failure interval for storing a failure time interval for each failure cause. An interval storage unit, an event-based cause classification unit that classifies the cause based on the strength of the relationship among the failure cause candidates given from the relevance storage unit for the observed event, A time-based cause classification unit for classifying the cause based on the length of the trouble interval to be performed, and plant equipment according to the strength of the relationship by the classification by the event-based cause classification unit and the length of time by the classification by the time-based cause classification unit. And a cause estimating unit for specifying the cause of the state. 2. The trouble interval storage unit includes an operation time storage unit that stores a total operation time of a plant, and a trouble interval correction unit that corrects a trouble interval stored in the trouble interval storage unit. 2. The method according to claim 1, wherein when the cause is determined, the trouble interval of the cause stored in the trouble interval storage unit is corrected based on the operation time given from the operation time storage unit. Plant equipment maintenance support equipment. 3. A cause event history storage unit for storing a cause and an observed event when the cause estimating unit determines a true cause, an event already stored in the cause event history storage unit and a current A history-based cause classification unit that searches for and classifies a cause that coincides with an event observed in the plant, and performs the estimation by the cause estimating unit. And determining the cause of the state of the plant equipment in accordance with the length of time according to the classification by the time-based cause classification unit or the presence or absence of the cause according to the classification by the history-based cause classification unit. The maintenance support device for plant equipment according to claim 1. 4. The relevance storage unit calculates the product of the relevance and the failure interval for each event and each cause combination to determine the maximum value for each event, and determines the maximum value of the event input in descending order of the maximum value. An event input order determination unit that determines the order is provided,
The relation stored in the relation storage is stored in the form of a two-dimensional matrix as a table of a two-dimensional matrix with respect to a plant state that can be caused by each of the trouble candidates for the cause of the trouble. The maintenance support device for plant equipment according to any one of claims 1 to 3, characterized in that: 5. The event input order determination unit further comprises a primary investigation number storage unit for storing a minimum number of event inputs as a primary investigation number, and a primary investigation number in an order determined by the event input order determination unit. To perform a primary diagnosis based on the contents of the event input by the cause estimating unit, and to successively perform the input and the cause estimation of the event having the input order after the number of primary investigations one by one. The maintenance support device for plant equipment according to claim 4, wherein 6. The product of the sum of the relevance of the observed event and the failure rate with respect to the sum of the relevance at the time when the event input to the cause estimating unit is less than all the events is the total relation including the uninput event. A defect degree upper limit value calculating unit that calculates a numerical value divided by the sum of degrees as a defect degree upper limit value, and in the estimation by the cause estimating unit, based on a defect degree upper limit value given from the defect degree upper limit value calculating unit. 6. The maintenance support apparatus for plant equipment according to claim 4, wherein the cause of the large value is identified at the present time as the cause of the state of the plant equipment. 7. An assertive rule storage section for storing an event sufficient for the cause estimating section to determine the cause of the error and a fault cause name that can be determined when the event is observed. A rule-based cause classification unit for classifying the cause based on whether or not a rule in the rule storage unit is satisfied, and in the estimation by the cause estimation unit, the strength of the relationship by the classification by the event-based cause classification unit And determining the cause of the state of the plant equipment in accordance with the length of time by the classification by the time-based cause classification unit or the success or failure of the rule by the classification by the rule-based cause classification unit. The maintenance support device for plant equipment according to any one of claims 1 to 6. 8. The result of the operation of a plurality of plants of the same type,
A satellite data totalizing unit that totals the recorded cause event histories and operation times, a defect interval initial value storage unit that holds a defect interval and an initial value of the operation time of the plant in question, and a defect interval that stores the operation time initial value. The number obtained by adding the number of cause event histories of all plants to the numerical value divided by the initial value is the total number of defects, and the total of the operation time of the same type of plant and the initial value of the operation time of the same type plant totaled by the satellite data tabulation unit is the total operation number A failure interval correction unit that determines a value obtained by dividing the total operation time by the total number of failures as a new failure interval of the type plant, and distributes the corrected failure interval given from the failure interval correction unit to each plant. And a lower-level system comprising the maintenance support device according to any one of claims 1 to 7. Maintenance support system of the plant apparatus including a door.