JP2010272023A - Device and method for maintenance of plant - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To readjust the maintenance of equipment highly reliably and efficiently, even when the number of equipment installed in a plant is huge. <P>SOLUTION: The plant maintenance device 1 for use in management of faults, deteriorated states, and maintenance histories of equipment provided in the plant includes: a maintenance result storage means 3 wherein at least identification information of equipment to be maintained and abnormal condition information for indicating faults and degrees of deterioration per parameter of these equipment are stored as device data; a data compression means 4 which ranks the device data to obtain compression data in accordance with abnormal condition information per parameter of the equipment which the device data has; a maintenance readjustment object device extraction means 10 which refers to ranks of compression data to extract the equipment to which ranks equal to or higher than a reference rank are assigned, as maintenance readjustment object equipment; and a maintenance review propriety determination means 15 which determines whether abnormalities of parameters of the maintenance readjustment object equipment are accidental or definite to determine whether maintenance readjustment is proper or not. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a plant maintenance technology that supports management of failure / deterioration status and maintenance history of equipment provided in a plant, and more particularly to a plant maintenance apparatus and a plant maintenance method for efficiently reviewing a maintenance method.

  Maintenance of equipment installed in nuclear power plants, thermal power plants, combined cycle power plants, etc. is based on time-planned maintenance (TBM), condition-based maintenance (CBM), and subsequent maintenance (BDM) according to standardized standards (such as JIS). Three maintenance methods are used. TBM is a maintenance method that periodically checks the equipment. CBM is a maintenance method that monitors the equipment status (vibration and wear) and checks the equipment according to the equipment status. This is a maintenance method that repairs equipment when a failure or deterioration is discovered. In plant maintenance, it is important to appropriately combine various maintenance methods according to the importance and characteristics of equipment to keep the plant sound and maintain repair costs and production loss costs.

  In maintenance for nuclear power plants, maintenance effectiveness evaluation (JEAG4210) related to the TBM cycle and CBM method is required. In the maintenance effectiveness evaluation, for example, when maintenance is recognized as good and it can be evaluated that the maintenance cycle is sufficiently safe in consideration of the deterioration rate of the equipment, the maintenance cycle can be extended.

  Conventionally, as a plant maintenance technology that enables maintenance effectiveness evaluation, a plant maintenance technology that uses equipment charts that record the state of equipment (vibration and wear) in the plant and the results of various inspections in overhauls has been proposed. (See Patent Document 1). In the maintenance effectiveness evaluation based on the equipment chart, a huge amount of data stored in the equipment chart is processed to determine whether the maintenance is good or bad. Therefore, as a technology to support maintenance effectiveness evaluation based on this equipment chart, a plant maintenance technology is proposed that reduces the power of manpower by screening the quality of maintenance using a threshold value with the equipment status as an index. (See Patent Document 2). Furthermore, a plant maintenance technique for setting a priority order related to maintenance review according to the importance of equipment has been proposed (see Patent Documents 3 and 4).

JP 2000-330627 A JP 2006-244334 A JP 2008-191900 A JP 2006-252311 A

  In the maintenance effectiveness evaluation, whether or not the maintenance is performed effectively is determined based on information such as that no failure has occurred in the device and that deterioration has not progressed more than expected. That is, instead of targeting each component that constitutes a device, a device that includes many components is comprehensively determined as one unit. Such maintenance effectiveness evaluation in units of equipment can be said to be a good measure from the viewpoint of efficiency in maintenance performed on a large-scale plant having a huge number of devices such as a nuclear power plant.

  However, when it is evaluated that maintenance is effective in the maintenance effectiveness assessment, for example, when reviewing maintenance methods such as extending the period of overhaul and inspection or extending the period of state monitoring, it is necessary to check the individual parts of the equipment. Characteristics must be considered. This is because the TBM cycle differs depending on the material of the component, the usage environment (temperature, pressure, etc.), and other conditions, and must be set based on a fast degradation rate. It is also necessary to estimate the replacement time of the part using information indicating the deterioration state of each part.

  Furthermore, when a failure or deterioration occurs in equipment, it should be immediately evaluated whether maintenance was appropriate and maintenance should be reviewed as necessary. At this time, it is necessary to determine whether or not the failure / deterioration is accidental such as unexpected stress due to human error or earthquake. This is because if the failure / deterioration is accidental, shortening the TBM cycle will adversely affect the maintenance cost of the plant.

  And even when maintenance based on CBM is implemented, whether the maintenance is appropriate using inspection results obtained by overhauling inspection, etc., or whether there is a failure / deterioration that cannot be handled by maintenance based on CBM. It is also important to always check.

  However, in the maintenance performed for a large-scale plant, the review of the maintenance takes time and cost, coupled with the huge number of devices to be maintained.

  The present invention has been made in view of the above circumstances, and there is provided a plant maintenance apparatus and a plant maintenance method capable of reviewing maintenance of equipment with high reliability and efficiency even when the number of equipment provided in the plant is enormous. The purpose is to provide.

  In order to achieve the above-described object, in the plant maintenance apparatus according to the present invention, at least identification information of equipment to be maintained in the plant maintenance apparatus used for management of failure / deterioration status and maintenance history of equipment provided in the plant. And maintenance result storage means for storing abnormality information indicating the degree of failure / degradation for each parameter of the device as device data, and ranking the device data according to the abnormality information for each parameter of the device held by the device data. Data compression means to be compressed data, a maintenance review target device extraction means for referring to the rank of the compressed data and extracting a device assigned a rank higher than a standard rank as a maintenance review target device, and the maintenance review target Judge whether equipment parameter abnormalities are accidental or deterministic and maintain Characterized in that it comprises a protection review suitability determination means for determining the appropriateness of re.

  Further, in the plant maintenance method according to the present invention, in the plant maintenance method for managing the failure / deterioration status and maintenance history of the equipment provided in the plant, at least the identification information of the equipment to be maintained and the parameters of this equipment. The step of storing abnormality information indicating the degree of failure / degradation as device data, the step of ranking the device data into compressed data according to the abnormality information for each parameter of the device held by the device data, A process for extracting a device assigned with a rank higher than the standard rank as a maintenance review target device with reference to the rank of the compressed data, and a deterministic whether the parameter abnormality of the maintenance review target device is accidental And determining whether or not maintenance review is appropriate.

  According to the present invention, even if the number of devices provided in the plant is enormous, it is possible to review the maintenance of the devices with high reliability and efficiency.

The functional block diagram which shows embodiment of the plant maintenance apparatus which concerns on this invention. The functional block diagram in connection with the data compression means of FIG. Explanatory drawing of the data compression performed by the data compression means of FIG. The functional block diagram in connection with the maintenance review object apparatus extraction means of FIG. The functional block diagram in connection with the maintenance performance detailed evaluation means of FIG. The functional block diagram in connection with the maintenance review suitability determination means of FIG.

  Embodiments of a plant maintenance apparatus and a plant maintenance method according to the present invention will be described with reference to the accompanying drawings.

  The plant maintenance apparatus 1 according to the present embodiment includes an input unit 2, a maintenance result storage unit 3, a data compression unit 4, a comprehensive evaluation result storage unit 5, a failure / deterioration generation device extraction unit 6, a failure / deterioration generation device storage unit 7, Maintenance method storage means 8, equipment importance degree storage means 9, maintenance review target equipment extraction means 10, maintenance review target equipment storage means 11, maintenance performance detail evaluation means 12, evidence data storage means 13, failure / degradation characteristic storage means 14, Maintenance review suitability determination means 15 and maintenance review execution means 16 are provided.

  The input means 2 is used to give necessary instructions to the plant maintenance apparatus 1 and is configured using, for example, an operation device such as a keyboard or a mouse.

  The maintenance result storage means 3 stores various pieces of equipment data related to plant maintenance such as overhaul inspection data, state monitoring data, failure history data, and operation time data of equipment provided in the nuclear power plant.

  The overhaul data includes the results of various inspections such as dimensional inspection, visual inspection, ultrasonic inspection, and trial operation inspection performed in the overhaul. The state monitoring data includes information indicating vibration, temperature, lubricating oil, flow rate, pressure, and noise during operation of the device. The failure history data includes information indicating a failure or deterioration that has occurred in the device. The operation time data includes information indicating the operation time of the device and the usage period of the parts constituting the device. Various types of data are stored in the maintenance result storage means 3 using the input means 2, for example.

  As shown in FIG. 2, the data compression unit 4 includes a period setting unit 17, a data comparison unit 18, a rank index storage unit 19, and a first graph display unit 20.

  The period setting means 17 sets the data evaluation period.

  The data comparison unit 18 performs abnormality evaluation of various device data stored in the maintenance result storage unit 3 in the evaluation period set by the period setting unit 17 and performs data compression based on the evaluation result.

  FIG. 3 is an explanatory diagram of data compression performed by the data compression means 4 and relates to data compression for the status monitoring data of the device A. FIG.

  First, the data comparison means 18 has abnormal information (good, caution, required repair) for each parameter (for example, pump bearing vibration, impeller wear, etc.) stored in the state monitoring data of the device A (see FIG. 3 (upper)). Refer to follow-up observation).

  Next, the data comparison means 18 compares the abnormality information of each parameter with a predetermined rank index stored in the rank index storage means 19, and the state monitoring data of the device A is based on the abnormality information of each parameter. Add a rank. That is, the data comparison unit 18 ranks (compresses) the state monitoring data of the device A itself using the abnormality information recorded for each parameter, and adds the date (evaluation date) on which the abnormality evaluation is performed. Compressed data consisting of

The relationship between the rank index and rank used in data compression is as follows, for example. (Rank) (Rank index)
I: All parameters are good.
II: There are parameters for elimination of abnormalities and deterioration (observation required).
III: There are parameters that require signs of abnormality or deterioration (attention required).
IV: There are parameters that require anomalies and deterioration that require repair (repair required).
V: There are parameters that cause abnormalities and deterioration that can cause loss of device functionality.

  In the data compression shown in FIG. 3, since all the parameters of the device A are “good” on the “evaluation date 1” when the abnormality evaluation is performed, the rank “I” is assigned to the state monitoring data of the device A. Next, since the parameter “Needs Caution” was included in the device A in the “evaluation date 2” in which the abnormality evaluation was performed, the rank “III” is assigned to the state monitoring data of the device A. Since the parameter “requires repair” is included in the device A in “evaluation date 3” in which the abnormality evaluation is performed, the rank “IV” is assigned to the state monitoring data of the device A. Then, since the parameter “Necessary follow-up” was included in the device A in “evaluation date 4” in which the abnormality evaluation was performed, the rank “V” is assigned to the state monitoring data of the device A. Note that the relationship between the rank index and the rank is stored in the rank index storage means 19.

  The first graph display means 20 refers to, for example, the status monitoring data of the device A stored in the maintenance result storage means 3, displays the status monitoring data in a graph or a table and displays it on a display unit (not shown). And let the user determine the rank of the state monitoring data. Then, when the user makes a request for adding a corresponding rank using the input unit 2, the first graph display means 20 includes a rank requested by the user and a date (evaluation date) when the rank is requested to be added. Is used to generate compressed data.

  The comprehensive evaluation result storage unit 5 stores the compressed data of the data comparison unit 18 or the first graph display unit 20 as a comprehensive evaluation result. The comprehensive evaluation result storage means 5 stores the systematically for each device and each device data.

  The failure / deterioration occurrence device extraction unit 6 evaluates the compressed data stored in the comprehensive evaluation result storage unit 5, and extracts devices in which failure or deterioration of various devices provided in the plant has occurred.

  The failure / deterioration occurrence device storage means 7 stores information (for example, the name and ID of the device) that can identify the device in which the failure or deterioration has been extracted, extracted by the failure / deterioration occurrence device extraction means 6.

  The maintenance method storage means 8 stores various types of information related to maintenance such as the model, parts, inspection / repair method, and the cycle thereof for each device.

  The device importance storage unit 9 stores information indicating the importance corresponding to the device model for each device.

  As shown in FIG. 4, the maintenance review target device extraction means 10 needs to review the maintenance method by referring to the data stored in the comprehensive evaluation result storage means 5, the maintenance method storage means 8 and the device importance degree storage means 9. Extract compatible devices (hereinafter, devices subject to maintenance review). The maintenance review target device extraction unit 10 includes a determination period storage unit 21, a determination criterion storage unit 22, a rank determination unit 23, a maintenance method comparison unit 25, and a model classification unit 24.

  The determination period storage unit 21 stores information indicating a determination execution period. The execution period of this determination is set based on the device importance stored in the device importance storage means 9 from the viewpoint of ensuring plant safety. For example, the higher the importance, the shorter the determination execution period is set and the determination is made more frequently.

  The criterion storage means 22 stores a criterion rank for distinguishing between necessity of maintenance review and absence of necessity. This reference rank is set based on the equipment importance stored in the equipment importance storage means 9 from the viewpoint of ensuring plant safety.

  The rank determination unit 23 refers to the determination execution period stored in the determination period storage unit 21 and refers to the comprehensive evaluation result stored in the comprehensive evaluation result storage unit 5 for each execution period. Then, the rank determination unit 23 reviews, from the compressed data that is the comprehensive evaluation result, a device to which device data equal to or higher than the predetermined reference rank recorded in the determination criterion storage unit 22 is assigned as a review target device.

  Here, the standard rank is classified into, for example, a maintenance mitigation rank and a maintenance strengthening rank, and rank “II” is set as the maintenance mitigation rank and rank “IV” is set as the maintenance strengthening rank under the rank index described above. The In this case, the device to be reviewed to ease the maintenance is a device related to device data to which rank “II” or lower is assigned. On the other hand, the devices to be reviewed that strengthen the maintenance are devices related to device data to which rank “IV” or higher is assigned.

  The model classification means 24 refers to various maintenance-related information stored in the maintenance method storage means 8 (equipment model, parts, inspection / repair method and its cycle, etc.), and checks the maintenance review target equipment model and inspection・ Classify by repair method and cycle.

  The maintenance method comparison unit 25 gives priority to the maintenance review with respect to the maintenance review target device extracted by the rank determination unit 23, for example, for each model classified by the model classification unit 24 and with a shorter inspection / repair cycle. Is allocated.

  The maintenance review target device storage unit 11 stores the review target devices of the maintenance method extracted by the maintenance review target device extraction unit 10 and given priority.

  As shown in FIG. 5, the maintenance result detailed evaluation unit 12 includes a second graph display unit 26, a display period setting unit 27, a graph setting unit 28, a graph preset storage unit 29, and a graph preset call unit 30.

  The second graph display means 26 refers to the maintenance result storage means 3 at a predetermined timing, and also part of various equipment data related to plant maintenance such as overhaul inspection data, condition monitoring data, failure history data, and operation time data, or the like. Display everything as a graph. The display period storage unit 27 stores a period to be displayed on the second graph display unit 26.

  The graph setting unit 28 refers to the maintenance review target device storage unit 11 and identifies a maintenance review target device extracted as having a need for review among devices on the graph displayed by the second graph display unit 26. Then, generate evidence data to which information indicating the necessity of review is added. Further, the second graph setting unit 28 refers to the failure / deterioration occurrence device storage unit 7, identifies the device in which failure / deterioration has occurred among the devices on the graph displayed by the second graph display unit 26, and Generate evidence data to which information (for example, the name of a parameter) indicating a parameter in which deterioration has occurred is added.

  The graph preset storage unit 29 stores preset information such as the type of graph to be displayed (for example, a trend graph) and items.

  The graph preset calling unit 30 calls the preset information stored in the graph preset storage unit 29 and sends it to the graph setting unit 28. The graph setting means 28 sets the type and item of the graph to be displayed based on the received preset information and sends it to the second graph display means 26. The preset of the graph can be set or changed manually by the user using the input unit 2.

  The evidence data storage unit 13 stores the evidence data generated by the maintenance result detailed evaluation unit 12.

  The failure / deterioration characteristic storage means 14 stores the failure / deterioration characteristics of the component parts for each model of the device, a general maintenance method or inspection / repair method for the same, and a period thereof as a database.

  Here, “failure / deterioration characteristics” are classified according to the mechanism or process leading to the phenomenon of failure / deterioration. For example, failure / deterioration always (deterministically) occur over time. Or the one that cannot occur without accidents such as human error or stress not considered in the design, the former is called time degradation characteristics and the latter is called accident characteristics . If the failure / deterioration has an accidental characteristic, it can be said that it is a mistake to perform maintenance reviews such as changing the cycle of overhaul inspection just because a failure / deterioration has occurred. Further, if the failure / deterioration has time deterioration characteristics, it is necessary to change the TBM cycle to an appropriate one or to perform CBM when the failure / deterioration occurs.

  The maintenance review propriety determination means 15 compares the device data stored in the evidence data storage means 13 with the failure / degradation characteristics stored in the failure / degradation characteristics storage means 14, and is true for the maintenance review target equipment. Determine whether a maintenance review is required. The maintenance review suitability determination means 15 includes a data collating means 31, an evaluation data set storage means 32, a device state determination means 33, a deterioration parameter extraction means 34, a maintenance review final determination means 35, a time degradation parameter extraction means 36, and a period extension determination. Means 37 are included.

  The data collating unit 31 includes information indicating the importance level of the device stored in the device importance level storage unit 9, failure / degradation characteristics stored in the failure / degradation characteristic storage unit 14, and evidence data storage unit 13. The correspondence between the data is specified by collating with the evidence data stored in the data.

  The evaluation data set storage unit 32 stores information indicating the importance level of the device stored in the device importance level storage unit 9 and the failure / degradation characteristic storage unit 14 based on the correspondence specified by the data matching unit 31. The stored failure / deterioration characteristics and the like and the evidence data stored in the evidence data storage unit 13 are collectively set as an evaluation data set.

  The device state determination unit 33 refers to the evaluation data set stored in the evaluation data set storage unit 32 and determines whether or not various devices have failed or deteriorated based on the rank of the evidence data.

  The deterioration parameter extraction unit extracts a deterioration parameter for a device that is determined to have a failure / deterioration by the device state determination unit 33. The deterioration parameter is defined as a parameter whose rank has become larger than a predetermined value within a predetermined period.

  The maintenance review final determination means 35 is intended for, for example, a device having a degradation parameter, and it is really necessary to review the maintenance of the device in consideration of the failure / degradation characteristics of the parameter (whether it is accidental characteristics or time degradation characteristics). It is determined whether or not. In addition, in this determination, appropriate maintenance evaluation such as whether inspection / repair by TBM has not been performed based on failure / deterioration of accidental characteristics, whether the TBM cycle setting is not too long or too short, etc. Quantitative assessment of maintenance is also performed.

  The time degradation parameter extracting unit 36 extracts a parameter having a time degradation characteristic (time degradation parameter) related to the device, which is targeted for the device determined by the device state determination unit 33 that no failure / deterioration has occurred.

  The cycle extension determination unit 37 uses the time degradation parameter extracted by the time degradation parameter extraction unit 36 and a preset value for the time degradation parameter to determine whether the TBM cycle can be extended. In this determination, the date of the intersection of the extrapolation curve and the design value is predicted by, for example, extrapolation using the least square method, and it is determined whether or not the TBM cycle can be extended.

  The maintenance review execution means 16 updates the maintenance method of the equipment stored in the maintenance method storage means 8 based on the result generated by the maintenance review suitability determination means 15.

  Next, the effect of the plant maintenance apparatus 1 will be described.

In the plant maintenance device 1,
(1) Using the state (parameters) of parts constituting the device as an index, device data (status monitoring data, etc.) is first ranked according to the degree of failure / deterioration of each parameter. Then, the presence / absence of maintenance is determined for the ranked device data, and the maintenance review target device is extracted. For this reason, it is possible to efficiently extract the maintenance review target devices. Next, it is determined whether or not the maintenance review is really necessary by individually evaluating the parameters related to the extracted maintenance review target device. As a result, even if the number of devices provided in the plant is enormous, it is possible to review the maintenance of the devices with high reliability and efficiency.

  (2) Since the necessity of review of maintenance is determined based on various information, it is possible to review maintenance with higher reliability.

  (3) The device data is not ranked based on a uniform standard, but can reflect the user's experience.

  (4) Since the compressed data holds information that can be said to be the minimum necessary for determining whether or not the maintenance needs to be reviewed, the maintenance can be reviewed more efficiently.

  (5) Since maintenance review target devices are extracted at an extraction frequency corresponding to the importance of the device, maintenance can be reviewed with higher reliability.

  (6) The extracted maintenance review target devices are classified according to their models, and information that prioritizes maintenance review is assigned to each model as the inspection / repair cycle is shorter. For this reason, maintenance can be reviewed with higher reliability.

  (7) Since evidence data is generated by adding information indicating that the device is subject to maintenance review, the user can examine the necessity of maintenance with reference to detailed device data, so maintenance can be performed with higher reliability. Can be reviewed.

  (8) Identify the equipment subject to maintenance review based on the evidence data, determine whether or not the equipment subject to maintenance review has failed / deteriorated, and select the equipment subject to maintenance review that has been judged to have failed / deteriorated. By determining whether the parameters of the target device for maintenance review are accidental or deterministic, the user can examine the necessity of maintenance while referring to detailed device data. The maintenance can be reviewed with higher reliability and ease.

  (9) A maintenance review target device whose rank has become larger than a predetermined value as time passes is specified, and a parameter that causes the rank to be increased is extracted, and whether the abnormality of the parameter is accidental or deterministic Therefore, it is possible to easily determine whether or not a review of maintenance is really necessary.

  As described above, the plant maintenance apparatus and the plant maintenance method according to the present invention have been described based on one embodiment. However, the specific configuration is not limited to this embodiment, and the invention described in the claims. Design changes and additions are allowed without departing from the gist of the present invention.

  For example, although the example which applies a plant maintenance apparatus and a plant maintenance method to a nuclear power generation plant etc. was shown, it is applicable also to transport machines, such as a chemical plant, an airplane, or a ship.

  DESCRIPTION OF SYMBOLS 1 ... Plant maintenance apparatus, 2 ... Input means, 3 ... Maintenance result storage means, 4 ... Data compression means, 5 ... Comprehensive evaluation result storage means, 6 ... Failure / deterioration apparatus extraction means, 7 ... · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · · ·····································································. Results detailed evaluation means, 13 …… Evidence data storage means, 14 …… Failure / degradation characteristic storage means, 15 …… Maintenance review suitability determination means, 16 …… Maintenance review execution means, 17 …… Period setting means, 18 …… Data comparison means, 19 ... rank index storage means, 20 ... first graph display means, 21 ... determination period storage means, 22 ... determination criteria storage means, 23 ... rank determination means 24 …… Model classification means, 25 …… Maintenance method comparison means, 26 …… Second graph display means, 27 …… Display period storage means, 28 …… Graph setting means, 29 …… Graph preset storage means, 30 …… Graph preset calling means, 31... Data collating means, 32... Evaluation data set storage means, 33... Equipment state judging means, 34 .. deterioration parameter extracting means, 35. Degradation parameter extraction means, 37... Period extension determination means.

Claims (6)

In plant maintenance equipment used for managing failure / deterioration status and maintenance history of equipment installed in the plant,
Maintenance result storage means for storing at least identification information of a device to be maintained and abnormality information indicating a degree of failure / deterioration for each parameter of the device as device data;
According to the abnormality information for each parameter of the device held by the device data, the data compression means that ranks the device data and sets it as compressed data;
Maintenance review target device extraction means for referring to the rank of the compressed data and extracting a device assigned a rank higher than the standard rank as a maintenance review target device;
A maintenance review suitability judging means for judging whether the abnormality of the parameter of the maintenance review target device is accidental or deterministic and judging the suitability of the maintenance review;
A plant maintenance device comprising:   2. The plant maintenance according to claim 1, wherein the data compression unit ranks the equipment data using a rank index indicating a corresponding rank in accordance with a degree of failure / deterioration of each parameter of the equipment. apparatus.   The plant maintenance apparatus according to claim 1 or 2, wherein the maintenance review target device extraction unit extracts a maintenance review target device at an extraction frequency corresponding to the importance of the device. Maintenance result detailed evaluation means for generating evidence data formed by adding information indicating that the device is a maintenance review target device to the device data stored in the maintenance result storage unit,
The maintenance review suitability determining means identifies the maintenance review target device based on the evidence data, determines whether the maintenance review target device has a failure / deterioration, and determines that the failure / degradation has occurred. 4. A device according to any one of claims 1 to 3, wherein, for the device subject to maintenance review, it is determined whether the parameter of the device subject to maintenance review is accidental or deterministic. The plant maintenance apparatus as described in the item.   The maintenance review suitability determination means identifies a maintenance review target device whose rank has become larger than a predetermined value as time passes, extracts a parameter that causes the rank to increase, and the abnormality of the parameter is accidental. The plant maintenance device according to claim 4, wherein it is determined whether the device is deterministic or not. In the plant maintenance method that manages the failure / degradation status and maintenance history of equipment installed in the plant,
Storing at least identification information of a device to be maintained and abnormality information indicating a degree of failure / degradation for each parameter of the device as device data;
According to the abnormality information for each parameter of the device held by the device data, the device data is divided into a plurality of ranks and compressed data, and
Referring to the rank of the compressed data, and extracting a device assigned a rank higher than the standard rank as a maintenance review target device; and
Determining whether the abnormality of the parameter of the maintenance review target device is accidental or deterministic, and determining the suitability of the maintenance review;
A plant maintenance method comprising:

Images