JPH0650778A - Apparatus and method for control of plant preventive maintenance information - Google Patents

Abstract

PURPOSE:To prevent the input error or the omission of data and to make the accuracy of the data uniform. CONSTITUTION:Preventive maintenance information which has been input from an information input/output means 1 is registered, by an information processing means 2, in a preventive maintenance database 4 in a hierarchically developed form. Common data for every item registered in the preventive-maintenance database 4 or common data between a plurality of items is extracted by using a statistical treatment by a feature extraction means 3. The data which has been extracted is related to its extraction criterion and registered in a feature database 5. By using the data in the feature database 5, new data can be estimated, and it is possible to prevent the input error or the omission of the data. In addition, the accuracy of the data can be made uniform.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a preventive maintenance information management apparatus and management method for supporting preventive maintenance of a large-scale plant, and in particular, information capable of reducing data maintenance work and making data accuracy uniform. The present invention relates to a management device and an information management method.

[0002]

2. Description of the Related Art Generally, in a large-scale plant such as an atomic plant, preventive maintenance is an elemental technology (eg, FM) in each stage of design, operation, and maintenance including monitoring and diagnosis, which are important.
Analysis / evaluation support systems such as EA and FTA)
Contributes to preventive maintenance of the plant.

For example, driving operation support, maintenance work and planning support, design improvement formulation support, abnormal event monitoring /
Diagnostic support, reliability evaluation of system equipment, etc.

By the way, many of the elemental technologies (systems) are developed and applied individually according to the above-mentioned individual problems. Therefore, even if commonly usable information, various devices developed by application of each technology. It is individually acquired and managed by a database and a database.

The information managed here is, for example, information on the equipment and configuration of the plant, specialized technical information specific to each technology, and knowledge information such as expert judgment criteria and values.

[0006]

However, regarding the information managed by the above-mentioned devices and databases, the necessary information is extracted from the actual plant according to the individual application, and an analysis / evaluation model is created. It is tuned to suit each problem. For this reason, this information is not versatile, and if the required specifications differ even a little, it is necessary to prepare the data again or recreate the model from the beginning, which requires a lot of labor, time, and cost. Especially in the field of diagnosis / maintenance in which appropriate measures are taken according to the state of the plant, there is a risk that the safety evaluation will be delayed.

Further, when the information is extracted or when the analysis / evaluation model is created, variations in data or model accuracy, omission of data, or misjudgment of data may occur due to differences in values obtained from the creator's viewpoint or experience. Etc.

The present invention has been made in consideration of such a point, and supports the preventive maintenance work of a plant design engineer, an operator, and a maintenance / repair staff to prevent data input errors and omissions. It is also an object of the present invention to provide a plant preventive information management device and a plant preventive information management method capable of performing more appropriate preventive maintenance by reducing data maintenance work and making data accuracy uniform.

[0009]

In order to achieve the above object, a plant preventive maintenance information management apparatus according to the first invention of the present application comprises an information input / output means for inputting / outputting preventive maintenance information of a large-scale plant. A preventive maintenance database that stores this preventive maintenance information; a characteristic database that stores a causal relationship between items related to the preventive maintenance information; and the information input from the information input / output means is a component of plant equipment. Hierarchical development is performed around the structure, the function that is the required function information for each structure, and the failure mode that is the total failure mode information for each function that causes loss or deterioration of the function, and the expanded information is stored in the preventive maintenance database. Information processing means for registering, deleting, and searching data in the preventive maintenance database based on the contents of the characteristic database ; Common data for each item such as structure and function registered in the preventive maintenance database or common data between multiple items is extracted over all items using statistical processing, and the extracted data is extracted. In addition to storing in the feature database in association with the reference,
A feature extraction means for analyzing the preventive maintenance database based on the contents of the feature database and performing addition, deletion and correction of data through the information processing means is provided.

The plant preventive maintenance information management apparatus according to the second invention of the present application is an information input / output unit for inputting / outputting preventive maintenance information of a large-scale plant; A word database that stores in advance, for each word, a function that is required function information for each and a failure mode name that is all failure mode information for each function that causes loss or deterioration of the function; and the preventive maintenance information is stored. Preventive maintenance database; a characteristic database for storing causal relationships between items related to preventive maintenance information; information input from the information input / output unit is classified based on a word database and stored in the preventive maintenance database, and a user Addition of data upon request,
Information processing means for deleting and searching; based on the classification result by the word database, data common to items of structure, function and failure mode are extracted and stored in the feature database, and preventive maintenance is performed based on the contents of the feature database. And a feature extracting means for analyzing the database and adding, deleting and correcting data through the information processing means.

In the plant preventive maintenance information management method according to the third invention of the present application, when the preventive maintenance information is classified based on the word database, the word data relating to the structure, function and failure mode is converted into the plant, system, The data is divided into four hierarchical levels of equipment and parts and registered, and the data regarding the structure, function, and failure mode of the input preventive maintenance information is expressed hierarchically by the combination of words in the word database, It is characterized in that the set is automatically expanded from the set of structure-function-failure mode, and this set is stored in the preventive maintenance database together with the input preventive maintenance information.

Further, the plant preventive maintenance information management method according to the fourth invention of the present application is a set of words automatically classified by the word database when extracting data common to the items of structure, function and failure mode. On the basis of the above, data regarding a causal relationship common among items including at least one of structure, function and failure mode is extracted, and the extracted data is stored in the feature database in association with the extraction criterion and the appearance frequency ratio. In addition, the preventive maintenance database is analyzed according to the extraction criteria and the appearance frequency ratio in the feature database, and data is added, deleted, or modified.

Further, the plant preventive maintenance information management method according to the fifth invention of the present application includes format information relating to tabular text data created by a word processor or the like among preventive maintenance information existing in various forms. It is provided with a table format database in which information about the initial character string set by the user is stored in advance. Based on this table format database, the information input / output means reads the table format text and the information processing means prevents it. It is characterized by being stored in a maintenance database.

Further, the plant preventive maintenance information management method according to the sixth invention of the present application relates to the vertical ruled lines of the table in the table format database when reading the text data in the table format. Starting from the read position based on the information and regarding the horizontal ruled line, the preceding line is a blank line, or even if it is not a blank line, the beginning of the line is a bullet character string previously registered in the table format database. The feature is that it automatically reads the relations of table items, starting from the fact that they match.

[0015]

In the plant preventive maintenance information management apparatus according to the first aspect of the present invention, the information input from the information input / output means is hierarchically expanded in the preventive maintenance database with the structure, function and failure mode as axes. Remembered.

The common data for each item such as the structure and function registered in the preventive maintenance database or the common data among a plurality of items are extracted and extracted for all items by using statistical processing. The data is stored in the feature database in association with the extraction criteria. Then, the preventive maintenance database is analyzed based on the contents of the feature database, and data is added, deleted, or modified.

Therefore, by using a common causal relationship between data items, new data can be estimated, data input errors and omissions can be checked, and terms in the database can be unified. Also, it becomes possible to delete the work related to data maintenance.

Further, in the Punrat preventive maintenance information management apparatus according to the second aspect of the invention, names relating to the structure, function and failure mode are preliminarily stored in the word database separately for each word, and information is stored using this word database. Processing is performed by the processing means and the feature extraction means.

Therefore, information necessary for preventive maintenance can be provided immediately and from various viewpoints, labor saving of maintenance work and improvement of reliability, optimization of regular inspection work, and reliability of plant can be achieved. It will be possible.

In the plant preventive maintenance information management apparatus according to the third aspect of the invention, the preventive maintenance database contains:
The information input from the information input / output means is automatically classified into a tree-like data format that is hierarchically expanded as a combination of words relating to the structure, function, and failure mode and is easy to search.
Remembered. As a result, the keyword used when searching the database is automatically generated. Therefore, the search is facilitated, and the data maintenance labor such as adding detailed keywords for the search is reduced. Further, it becomes easy to manage keywords when adding data.

Further, in the plant preventive maintenance information management apparatus according to the fourth aspect of the present invention, data relating to a common causal relationship between items including at least one of the structure, function and failure mode registered in the preventive maintenance database. Are extracted over all of the above items using a statistical process, and stored in the feature database in association with the criterion at the time of extraction and the appearance frequency ratio obtained from the result of the statistical process. The analysis of the preventive maintenance database is performed based on the common causal relationship and appearance frequency ratio among the data items stored in the feature database. From this common causal relationship and appearance frequency ratio,
It is possible to estimate the relationship between new data, check data input errors and omissions, and unify terms in the database.

Further, in the plant preventive maintenance information management apparatus according to the fifth aspect of the present invention, the information about the tabular text data and the information about the initial character string set by the user are stored in advance in the table format database. There is. Therefore, it becomes possible to automatically read and process tabular text data.

Further, in the plant preventive maintenance information management apparatus according to the sixth aspect of the present invention, out of the forecast maintenance information existing in various media forms, the tabular text data created by a word processor or the like is automatically read. Stored in the preventive maintenance database. This makes it possible to effectively use the existing data and further reduce the data maintenance labor.

[0024]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below with reference to the drawings. FIG. 1 shows a plant preventive maintenance information management apparatus according to the first embodiment of the present invention. This apparatus is an information input / output unit 1.
The information processing unit 2, the feature extraction unit 3, the preventive maintenance database 4, and the feature database 5 are provided.

The information input / output unit 1 inputs / outputs preventive maintenance information necessary for a plurality of elemental technologies such as design, operation, maintenance, monitoring / diagnosis and reliability evaluation that support preventive maintenance of a large-scale plant. The data from the user U is input.

That is, the data handled by the information input / output means 1 includes preventive maintenance technical information expressed by words, numerical values, rank values, etc., and data relating to the judgment criteria of extraction predefined by the user U. , These are CRT devices,
Input / output is performed with the user U via a keyboard, a mouse device, a floppy disk drive device, a Kanji graphic printer device, or the like.

The preventive maintenance technical information is information specific to each technology such as information common to each technology such as plant equipment or configuration, information that can be used only between certain specific technologies, and expert judgment criteria and values. It is composed of various technical information, and the inputted information is registered in the preventive maintenance database 4.

That is, the input information is, as shown in FIG. 2, the component information of the plant equipment (hereinafter, "structure").
And the required function information for each "structure" (hereinafter,
"Function" and all failure mode information for each "Function" that causes loss or deterioration of function (hereinafter "Failure Mode")
(Referred to as “)” as an axis, and is hierarchically expanded and registered / saved in the preventive maintenance database 4 in this state.

Here, the "structure" corresponds to information such as a system name, a function name, and a part name, and the "function" means, for example, in the case of a pump, a water supply function, a pressure resistance function,
It corresponds to information such as environmental safety maintenance function, and "failure mode" means, for example, when "structure" is a part,
It corresponds to information such as friction, corrosion, and cracks.

Further, as additional information for each item, "structure" is, for example, in the case of a device, a serial number, performance specification,
It has information such as drawing number, related number, model, maintenance condition, and "function" is, for example, required specifications in the case of equipment,
It has information such as interface conditions with other "functions", and the "failure mode" is, for example, in the case of equipment,
It has information on failure impact, occurrence frequency / probability, failure distribution form, failure monitoring / diagnosis support, maintenance conditions, etc.

The development of such information and the registration of the developed information in the preventive maintenance database 4 are performed by the information processing means 2 as shown in FIG. Further, according to a request from the user U and an analysis result by the feature extracting means 3 which will be described later, the data registered in the preventive maintenance database 4 are corrected, added, and searched.

On the other hand, the feature extraction means 3 inspects the similarity of technical information for terms, numerical values, rank values, etc. for each single item or multiple items of the preventive maintenance database 4, and information common to the items to be inspected. To extract. For example, the cause of failure of each "structure" of pumps, valves, etc., the cause of failure of each "function" of pressure / water supply function, or the cause of failure of pressure function of XXX model YYY pump. A process of extracting a causal relationship common to each or a plurality of "functions" or a plurality of items by statistical processing is performed. Then, the causal relationship of the obtained data is registered in the feature database 5 as a set of the information regarding the item of the extracted data and the causal relationship thereof.

As for the combination of data items, assuming that the number of data items is n, the total number of combinations A that can be assumed below or the number of times according to the items defined by the user U is performed.

[0034] When registering a new causal relationship, both the set of data items registered at the time of the previous extraction and the causal relationship are compared with the current one, and only the newly obtained causal relationship is presented to the user U for registration. Feature database 5 only for those requested
You are supposed to register at.

The feature extracting means 3 also analyzes the preventive maintenance database 4 including new input information by using the causal relationship between items registered in the feature database 5.
Necessary data changes or additions are performed via the information processing means 2.

This is based on, for example, the "failure mode" data of the "heat resistant function" of the "reactor coolant circulation pump for nuclear power generation equipment" registered in the feature database 5 "Failure mode" for "pressure resistance function" of "lubricating oil circulation pump for pump", or from "failure mode" for "pressure resistance function" of "Reactor coolant circulation pump" This is an analogy to the "failure mode" data for the "pressure resistance function" of the reactor coolant circulation pump outlet valve.

As described above, since new data can be estimated by using a common causal relationship between data items, data input errors and omissions can be checked to unify terms in the database. Moreover, the work required for data maintenance can be reduced.

The processing relating to the analysis using the feature database 5 uses the method of obtaining a new feature extraction database after registering all the data in the preventive maintenance database 4 and the feature extraction database obtained by the previous analysis. And registering the data in the preventive maintenance database 4 and obtaining a new feature extraction database.
Two methods can be considered, but either method can be performed according to the request of the user U.

FIG. 3 is a flow chart showing the flow of processing of plant preventive maintenance information management according to this embodiment. The operation of this embodiment will be described below with reference to FIG.

First, in step S1, it is determined whether the preventive maintenance technical information is newly input, the input data is changed / deleted, or the input preventive maintenance technical information is retrieved.

In the case of new input of preventive maintenance technical information,
Through the CRT device, keyboard and mouse device, floppy disk drive device, Kanji graphic printer device and the like, in step S2, plant equipment information, technical information represented by words, numerical values, rank values, etc.,
Preventive maintenance technical information such as expert knowledge information is hierarchically developed and input to the preventive maintenance database 4 with the "structure-function-failure mode" as the axis.

If the input data is changed / deleted, the data in the preventive maintenance database 4 is changed or deleted in step S3. Then, after the new input or change / deletion processing, the data registered in the preventive maintenance database 4 is analyzed by statistical processing in step S4, and the new causal relationship is set in the feature database 5 in step S5. register.

On the other hand, in the case of the search for preventive maintenance technical information, after setting the search conditions in step S6, the data of the preventive maintenance database 4, that is, the search contents are displayed in step S7.

Here, since the data required by the user U is not always registered, step S8
In step S1, it is determined whether or not to evaluate the data based on the feature database 5, and if it is to be evaluated, step S
At 9, the feature database 5 is used to provide new data. Then, in step 10, the obtained result is displayed to the user U and registered in the preventive maintenance database 4 in response to the request.

If it is determined in step S8 that the data does not need to be evaluated, or after the registration processing in step S5 is completed, and further after the processing in step S10, it is determined in step S11 whether or not the processing is continued. , In case of continuation, the process returns to step S1 and
If there is no continuation, the process ends.

Thus, even if the data is incomplete,
Since it is possible to estimate data that has not been input, it is possible to reduce the data maintenance work and make the data accuracy uniform (to suppress variations).

By the way, the terms of data handled as preventive maintenance technical information are not necessarily unified, and sometimes have different words with the same meaning. For example, other words having the same meaning as “damage” indicating a failure mode include “damage” and “deficiency”. For this reason, there are cases where the type of input data is limited due to the constraint conditions at the time of inputting the preventive maintenance database, and the essential relationships that should be present cannot be found so much.

FIG. 4 shows a second embodiment of the present invention made in order to deal with these problems. The user U prepares and sets judgment criteria in advance so that only the inputted data can be used. It is designed to guide common relationships that cannot be obtained.

That is, in step S21, it is determined whether the data is input, edited, or displayed. If the data is input or edited, these processes are performed in step S22. If the data is displayed, the step is performed. Data display is performed in S23. Then, new data is registered in the feature database 5 as needed. Then, in step S24, it is determined whether or not the process is continued. If the process is continued, the process returns to step S21, and if the process is ended, all the processes are ended.

As described above, by performing this process as a preprocess of the process shown in FIG. 3, when the common data is extracted by the feature extracting means 3, the judgment standard prepared and set by the user U in advance is used. Can be performed using.

FIG. 5 shows a plant preventive maintenance information management apparatus according to the third embodiment of the present invention. This apparatus comprises an information input / output unit 1, an information processing unit 2, and a feature extraction unit 3.
, Preventive maintenance database 4 and feature database 5
And a word database 6.

The information input / output means 1 inputs / outputs preventive maintenance information necessary for a plurality of elemental technologies such as design, operation, maintenance, monitoring / diagnosis and reliability evaluation that support preventive maintenance of a large-scale plant. The data from the user U is input.

That is, the data handled by the information input / output means 1 includes preventive maintenance technical information represented by words, numerical values, rank values, etc., and data concerning the judgment criteria of extraction predefined by the user U. , These are CRT devices,
Input / output is performed with the user U via a keyboard, a mouse device, a floppy disk drive device, a Kanji graphic printer device, or the like.

Preventive maintenance technical information is information specific to each technology such as information common to each technology such as plant equipment and configuration, information that can be used only between certain specific technologies, and expert judgment criteria and values. It is composed of various technical information, and the inputted information is registered in the preventive maintenance database 4.

That is, the input information is hierarchically developed with the "structure", "function", and "failure mode" as axes, as in the case of the first embodiment. In this state, the preventive maintenance database is created. 4 will be registered.

The development of this information and the registration of the developed information in the preventive maintenance database 4 are performed by the information processing means 2 as shown in FIG.

At the time of this registration, the information processing means 2 uses the word database 6 of the structure, function, and failure mode, which is registered by dividing into four levels of plant, system, equipment, and parts as shown in FIG. It has become. That is,
The input information is analyzed and automatically converted into a tree-like data format having a hierarchical connection from the plant to the parts for each of “structure”, “function”, and “failure mode”. For example, "nuclear power generation demo plant" as the plant-level word data of the structure, "reactor recirculation system" as the system level, and "pump" as the equipment level.
, "Motor", "casing", "bearing", etc. as the component level, and input data "nuclear power demo plant reactor recirculation system pump" and "nuclear power demo plant reactor recirculation system" Pump casing ",
"Nuclear power generation demo plant Reactor recirculation system pump motor", "Nuclear power generation demo plant"-"Reactor recirculation system"-"Pump" and "Nuclear power generation demo plant"-"Reactor recirculation system"-""Pump"-"Casing","Nuclear power demo plant"-"Reactor recirculation system"-"Pump"-"Motor" is automatically converted into a hierarchical tree-like data format, and "Nuclear power demo" Plant "-
Search keywords such as "reactor recirculation system"-"pump" are automatically created. That is, the converted data "nuclear power generation demo plant"-"reactor recirculation system"-"pump" and the like can be used as they are as search keywords.

Here, for example, the word data such as "bearing" is used a plurality of times such as "pump"-"bearing" and "motor"-"bearing", so that the reusability is high,
Management of word data becomes easy.

The terms of data handled as preventive maintenance technical information are inconsistent, and may be omitted or written in different words with the same meaning. For example, “nuclear power generation demo plant” is simply described as “demo plant”, “reactor recirculation system” is abbreviated as “PLR system”, and “damage” indicating a failure form has the same meaning. It may be described by the words "damage" or "loss".

Therefore, there are cases in which the type of input data is limited due to the constraint conditions when the preventive maintenance database 4 is input, and the essential relationship that should be present cannot be found so much.

In view of this, in the second embodiment, the user U prepares and sets judgment criteria in advance so that a common relationship can be derived. In the present embodiment, for example, "nuclear power" is used. Prepare word data for thesaurus that collects synonyms and synonyms such as "power generation demo plant" = "demo plant", "reactor recirculation system" = "PLR system", "damage" = "damage", "deficiency". By
"Demo plant PLR pump" as "nuclear power generation demo plant"-"reactor recirculation system"-"pump" and "damaged"
It has become possible to correctly recognize that "and" is the same as "damage".

As described above, the information processing means 2 develops the inputted information centering on the "structure-function-failure mode", and registers it in the preventive maintenance database 4 in pairs with the obtained search keyword. In accordance with the request from the user U and the result of the analysis by the feature extracting means 3, the registered data is corrected, added, or searched.

On the other hand, the feature extracting means 3 determines the similarity of the terms, numerical values, rank values, etc. of the above-mentioned technical information based on the automatically created keyword for retrieval, and a single item or a plurality of items in the preventive maintenance database 4 are selected. It is designed to conduct inspections and extract information common to the items to be inspected. For example a pump,
"Structure" and "function" such as the cause of failure of "structure" such as valves, the cause of failure of "function" such as pressure resistance and water supply, and the structural characteristics of "failure mode" such as vibration and leakage ], "Cause failure mode", or a common causal relationship for each of a plurality of items is extracted by statistical processing. And the relationship of the obtained data is
The reference at the time of extraction and the appearance frequency ratio are associated with each other and registered in the feature database 5.

Regarding the combination of data items, assuming that the number of data items is n, the following envisionable total number A of combinations is performed. Regarding the appearance frequency ratio, if the number of data items that match the extraction criteria is B out of the total number of combinations A, the appearance frequency ratio S
Is represented by the following formula. S = B / A When registering a new causal relationship, whether it is a completely new relationship or whether only the appearance frequency ratio is updated is feature database 5
If it is completely new, the new extraction criterion and the appearance frequency ratio are registered together with the data. If not, the appearance frequency ratio is updated and the data relation is added. .

The feature extraction means 3 also uses the search keyword in the preventive maintenance database 4 and the causal relationship between the items registered in the feature database 5 and the frequency of appearance to prevent the new input information. The maintenance database 4 is analyzed and necessary data changes or additions are performed via the information processing means 2.

This is, for example, "reactor recirculation system pump of a nuclear power plant" registered in the feature database 5 (expressed as a set of word data which is a keyword for retrieval, "nuclear power plant"-"reactor" Recirculation system "-
From the "Failure mode" data for the "pressure resistance function"("pressureresistance") of the "pump", the "reactor recirculation pump lubricant oil circulation pump of the same plant"("nuclear power plant"-"reactor Circulation system-Lubricant-Pump,
Note that the word data of "lubricating oil" and "circulation pump" and "pump" have the same meaning. It is assumed in advance that the "failure mode" for "pressure resistance function" or the above "atom" From the "failure mode" data for the "pressure resistance function"("pressureresistance") of the "reactor recirculation pump", the "pressure resistance function" of the "reactor recirculation pump outlet valve of the plant"
It is an analogy to the data of "failure mode" regarding ("breakdown voltage"). In this case, the former is "pump", "pressure resistant"
For the word data such as, the latter analogizes only "breakdown voltage" and uses it as a key at the time of evaluation.

At the time of this analogy, the correction is automatically performed or the user U is inquired based on the appearance frequency ratio obtained from the result of the statistical processing. For example, the processing corresponding to the appearance frequency ratio as shown in FIG. 7 is determined, and if the ratio is 70% or more, the data is automatically updated and added without inquiring the user U.
If it is more than 70% and less than 70%, the user U is inquired.

In this way, by using a common causal relationship and appearance frequency ratio among data items, new data is estimated by automatically inquiring those with a high appearance frequency and inquiring the user U for those with a low appearance frequency. Therefore, it is possible to check data input errors and omissions, unify the terms in the database, and reduce the work required for data maintenance.

FIG. 8 is a flow chart showing the flow of processing of plant preventive maintenance information management according to this embodiment. The operation of this embodiment will be described below with reference to FIG.

First, in step S31, it is determined according to a request from the user U whether the preventive maintenance technical information is newly input, the input data is changed / deleted, or the input preventive maintenance technical information is retrieved.

In the case of new entry of preventive maintenance technical information,
Through a CRT device, a keyboard or a mouse device, a floppy disk drive device, a Kanji graphic printer device, etc., plant equipment information, technical information, and specialized knowledge information expressed in terms of phrases, numerical values, rank values, etc. in step S32. The preventive maintenance technical information such as is hierarchically expanded and input to the preventive maintenance database 4 with "structure-function-failure mode" as the axis.

If the input data is changed / deleted, the data in the preventive maintenance database 4 is changed or deleted in step S33. Then, after new input or change / deletion processing, in step S34,
The data registered in the preventive maintenance database 4 is analyzed by statistical processing, and the data is estimated and evaluated in step S35 based on the causal relationship between the obtained data and the appearance frequency ratio. Then, in step S36,
The result is displayed to the user U, and those having a high appearance frequency are automatically registered, and those having a relatively low appearance frequency are registered in the preventive maintenance database 4 in response to a request. Further, step S3
In step 7, the obtained new causal relationship is registered in the feature database 5, and the relationship caused by the data change or deletion is deleted or modified.

On the other hand, in the case of a search for preventive maintenance technical information, after the search conditions are set in step S38,
In step S39, the data of the preventive maintenance database 4, that is, the search content is displayed. Then, after that, in step S40, it is determined whether or not the process is continued. If the process is continued, the process returns to step S31, and if the process is not continued, the process ends.

In FIG. 9, the word data of "Structure", "Function", and "Failure mode" are divided into four layers of plant, system, equipment, and parts, and registered and edited in the word database 6.
This shows the process of deleting, which will be described below.

First, in step S41, it is determined by inquiring the user U whether the data is input, edited, or displayed. If the data is input or edited, step S42 is performed.
In step S43, these processes are executed, and in the case of data display, the data is displayed in step S43. Then, new data is registered in the word database 6 as needed. After that, in step S44, it is determined whether or not the process is continued. If it is continued, the process returns to step S41, and if it is finished, all the processes are terminated.

By thus performing this process as a preprocess of the process shown in FIG. 8, it is not necessary to set word data for each data input. Also, the word data can be individually saved and changed according to the read data.

By the way, since the amount of information registered in the preventive maintenance database 4 is enormous, if the data evaluation processing by the feature database 5 is performed for each data input as in the processing shown in FIG. There is a possibility that it will be large.

FIG. 10 shows a fourth embodiment of the present invention made to deal with this problem, in which data evaluation is performed within a limited range.

That is, after displaying the search contents in step S39, the user U is inquired whether or not to perform data evaluation in step S51. Then, in step S52, the data evaluation is executed only for the search condition determined in advance according to the request of the user U, and the result is displayed in step S53. After the display, step S37 is executed.

As described above, if necessary data is registered, it can be used, and if not, it can be carried out by using the data estimated and evaluated by the feature database 5, thus shortening the processing time. be able to.

FIG. 11 shows a plant preventive maintenance information management apparatus according to the fifth embodiment of the present invention. This apparatus has a structure in which a table format database 7 is added to the apparatus shown in FIG. There is.

That is, the information input / output means 1 uses the table format database 7 corresponding to the tabular text data d to be read.
The contents of the items in the table are read based on. Here, in the table format database 7, only information about vertical ruled lines of the table is registered in advance. For example,
As shown in FIG. 12, assuming that the ranges of table items relating to “structure”, “function”, and “failure mode” are P, Q, and R, respectively, the “structure” is based on these P, Q, and R. "Function" is set as the subordinate information of "," and "Failure mode" is set as the subordinate information of "Function," so that the relationship between the items in the table can be understood and read. Since the process of converting the "PLR pump" into the "reactor recirculation pump" is performed by the information processing means 2 after that, the process is passed to the next process without being converted as shown in the figure. Has become.

At the time of this reading, the vertical ruled lines are read based on the table format database 7, and the horizontal ruled lines are blank lines before reading, or even if they are not blank lines, the first character string of the line is read. Is read when it matches the character string as shown in FIG.

Further, in the table, it may be omitted by "same as above" or the like. Therefore, in the present embodiment, it is checked whether or not there is this "same as above", and if there is, the data for the previous time, which is the same as above in the table, is copied and expanded. This allows the contents to be correctly expanded even if there is "same as above".

FIG. 14 is a flow chart showing the flow of processing of plant preventive maintenance information management according to this embodiment.
The operation of this embodiment will be described below with reference to FIG.

First, in step S61, the user U
If the user selects to input data, in step S62, the tabular text data d is read, so that the corresponding tabular format data is first read from the tabular format database 7.

Next, in step S63, it is determined whether or not all the data have ended, and in step S64, it is determined whether or not the previous line is blank or the initial character strings match, and in step S65, the table is displayed. Read the format text data d.

If it is determined in step S64 that the previous line is a blank line or matches the initial character string,
In step S66, expansion of the table item is executed.

The above operation is executed until it is determined in step S63 that the end of all data has been completed.

FIG. 15 shows the process of registering, editing, and deleting the data relating to the table format and the initial character string, which will be described below.

First, in step S71, the user U is inquired to determine whether data is input, edited or displayed. If data is input or edited, step S7 is performed.
In step 2, the processing is executed, and in the case of data display, the data is displayed in step S73. Then, new data is registered in the table format database 7 as needed. Then, in step S74, it is determined whether or not the process is continued. If the process is continued, the process returns to step S71, and if it is finished, all the processes are terminated.

Thus, the tabular text data d can be read as it is, and the existing information can be effectively used.

[0093]

As described above, according to the first aspect of the present invention, it is possible to prevent data input error and omission.
Further, even if the data is incomplete, it is possible to estimate the uninput data from the input data. Therefore, the data maintenance work can be reduced and the data accuracy can be made uniform, and more accurate preventive maintenance can be performed.

According to the second invention, information necessary for preventive maintenance is provided immediately and from various viewpoints, labor saving and improvement of reliability of maintenance work, optimization of regular inspection work and plant The reliability of can be improved.

Further, according to the third invention, the search is facilitated, the data maintenance labor such as the detailed keyword addition for the search is deleted, and the keyword management in the data addition is facilitated.

Further, according to the fourth invention, it is possible to estimate the relationship between new data based on the common causal relationship and the appearance frequency ratio, and it is possible to check data input errors and omissions. In addition, the terms in the database can be unified.

Further, according to the fifth invention, it is possible to automatically read and process tabular text data.

Further, according to the sixth invention, it is possible to effectively utilize the existing data and reduce the data maintenance labor.

[Brief description of drawings]

FIG. 1 is an overall configuration diagram showing a plant preventive maintenance information management device according to a first embodiment of the present invention.

FIG. 2 is an information management configuration diagram in the plant preventive maintenance information management method.

FIG. 3 is a flowchart showing the flow of processing of plant preventive maintenance information management.

FIG. 4 is a view corresponding to FIG. 1 showing a second embodiment of the present invention.

FIG. 5 is a view corresponding to FIG. 1 showing a third embodiment of the present invention.

FIG. 6 is a configuration diagram showing a registration state of data in a word database.

FIG. 7 is an explanatory diagram showing the relationship between the appearance frequency ratio and the correspondence at the time of data analogy.

FIG. 8 is a flowchart showing the flow of processing of plant preventive maintenance information management.

FIG. 9 is a flowchart showing the flow of processing for registering, editing, and reducing hierarchically divided data in the word database.

FIG. 10 is a flowchart showing a fourth embodiment of the present invention.

FIG. 11 is an overall configuration diagram showing a plant preventive maintenance information management device according to a fifth embodiment of the present invention.

FIG. 12 is an explanatory diagram showing a method of reading tabular text data.

FIG. 13 is an explanatory diagram showing an example of a leading character string of a line.

FIG. 14 is a flowchart showing the flow of processing of plant preventive maintenance information management.

FIG. 15: Registering data related to table format and bullet character,
The flowchart which shows the flow of the process which edits and deletes.

[Explanation of symbols]

 1 information input / output means 2 information processing means 3 feature extraction means 4 preventive maintenance database 5 feature database 6 word database 7 table format database U user d table format text data

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Yu Morishige 8 Shinsita-cho, Isogo-ku, Yokohama-shi, Kanagawa Incorporated company Toshiba Yokohama Works

Claims (6)

[Claims] 1. An information input / output unit for inputting / outputting preventive maintenance information of a large-scale plant, a preventive maintenance database for storing the preventive maintenance information, and a feature database for storing a causal relationship between items related to the preventive maintenance information. And information that is input from the information input / output unit is a structure that is a component of plant equipment, a function that is required function information for each structure, and a failure that is all failure mode information for each function that causes loss or deterioration of the function. And an information processing means for hierarchically expanding the mode as an axis, storing the expanded information in the preventive maintenance database, and registering, deleting, and searching data in the preventive maintenance database based on the contents of the characteristic database. , Common data for each item such as structure and function registered in the preventive maintenance database, or sharing between multiple items. General data is extracted over all items using statistical processing, the extracted data is stored in the feature database in association with the extraction criteria, and the preventive maintenance database is analyzed based on the contents of the feature database, A plant preventive maintenance information management device, comprising: a feature extraction unit that adds, deletes, and corrects data via an information processing unit. 2. Information input / output means for inputting / outputting preventive maintenance information of a large-scale plant, a structure which is a component of plant equipment, a function which is required function information for each structure, and a function which causes loss or deterioration of the function. A word database that stores in advance the names related to the failure modes, which are all failure mode information for each, separately for each word, a preventive maintenance database that stores the preventive maintenance information, and a causal relationship between items related to the preventive maintenance information. And a feature database and information input from the information input / output unit are classified based on a word database and stored in a preventive maintenance database, and information processing unit that adds, deletes, and retrieves data according to a user's request And a structure based on the classification result by the word database,
Data common to the items of function and failure mode are extracted and stored in the feature database, the preventive maintenance database is analyzed based on the contents of the feature database, and data is added, deleted, or modified through the information processing means. A plant preventive maintenance information management device comprising: a feature extraction means. 3. When classifying preventive maintenance information based on a word database, word data relating to a structure, a function, and a failure mode are registered by dividing them into four hierarchical levels of a plant, a grid, equipment, and parts. Data regarding the structure, function, and failure mode of the entered preventive maintenance information is hierarchically expressed by a combination of words in a word database, and automatically expanded from the structure-function-failure mode set. The plant preventive maintenance information management method according to claim 2, wherein the set is stored in the preventive maintenance database together with the entered preventive maintenance information. 4. When extracting data common to items of structure, function and failure mode, at least one of structure, function and failure mode is included based on a set of words automatically classified by a word database. Data related to the causal relationship common to the items is extracted, the extracted data is stored in the feature database in association with the extraction criteria and the appearance frequency ratio, and the preventive maintenance database is created based on the extraction criteria and the appearance frequency ratio in the feature database. The plant preventive maintenance information management method according to claim 2, wherein the analysis is performed and data is added, deleted, or modified. 5. Pre-stored format information regarding tabular text data created by a word processor and the like among preventive maintenance information existing in various forms and information regarding a bullet character string set by a user. 3. The plant according to claim 2, further comprising a table format database, wherein the information input / output means reads the table format text based on the table format database and stores the table format text in the preventive maintenance database by the information processing means. Preventive maintenance information management device. 6. When reading the text data in the tabular format, regarding the vertical ruled lines of the table, the reading position based on the information about the vertical ruled lines of the table in the table format database is used as the starting point and the horizontal ruled lines are Even if the previous line is a blank line, or even if it is not a blank line, the relationship between table items is automatically read starting from the fact that the beginning of the line matches the initial character string registered in the table format database in advance. The plant preventive maintenance information management method according to claim 5, further comprising: