JP2006099298A - Fault diagnosis method for device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method for easily specifying the cause of a fault in a short period of time in view of the fact that, if a fault has occurred in a controller or a device, it takes a great amount of time to specify its cause because the creation of an FT chart for use in the investigation of the cause requires each factor that constitutes the FT chart to be investigated on every occasion, as by asking experienced expert engineers. <P>SOLUTION: The fault diagnosis method includes a step for creating the FT chart on the basis of first factors where respective parts of a device configuration are functionally classified, second factors where the respective parts are subdivided for each device, and a second chart having a comparison chart in which know-hows are associated with the first factors. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention is an FT diagram created when investigating the cause using an FT (Fault Tree Analysis) diagram when a failure occurs in equipment installed in a large plant facility such as a control device of a thermal power plant. This invention relates to a device failure diagnosis method for identifying the cause of a failure.

Conventionally, FTA (Fault Tree Analysis) is a primary factor for the occurrence of a failure when a failure occurs in a control device or equipment as a method for investigating the cause, Develop secondary factors and lower factors into an FT diagram based on the functions of the control device or equipment in which the failure occurred and the expertise or experience (know-how) of experts related to the control device or equipment. The FT diagram was created each time according to the failure occurrence event.
For this reason, the form and terminology become inconsistent due to differences between FT diagram creators, and the cause is troublesome. In addition, investigation of each event constituting the FT diagram is conducted each time, and it takes a lot of time for diagnosis. However, there is a problem that the generation trouble period is prolonged due to the delay in investigating the cause.
Therefore, in order to solve the above problems, the FT diagram of the target equipment is obtained in a unified form, and as a diagnostic device that unifies the investigation method of each event, an FT diagram database storage device, FT diagram creation from the database, etc. A technique including a terminal device to perform is disclosed (for example, Patent Document 1).

Japanese Patent Laid-Open No. 02-161567 (FIG. 1)

  However, although the thing shown by the said patent document 1 is obtained in the form with which the FT figure was unified, and the investigation method of each event has been presented, the investigation of each factor which constitutes the FT figure, etc. Each time, it is necessary to make an inquiry by consulting with past experience and experience holders, and it takes a lot of time to identify which factor is the cause. As a rapid cause investigation method required when a failure occurs in a large-scale plant, there is a problem that it cannot be sufficiently dealt with.

  The present invention solves the above-mentioned problems. FT knows various know-how concerning operation of equipment and countermeasures in case of conventional troubles possessed by specialists such as plant equipment and equipment operation engineers. An object of the present invention is to provide a method for easily identifying a cause of a failure in a short time without necessarily relying on an expert by displaying in advance on a format for creating a diagram.

The device failure diagnosis method according to the present invention includes the following steps.
Step 1: As factors for estimating the failure location, the first and second factors are listed,
The first factor is to classify each part of the device by function,
The second factor is obtained by further subdividing each part of the first factor for each component device,
Creating a first table displaying at least first and second factors;
Step 2: A step of creating a second table having a know-how owned by a professional engineer and a correspondence diagram corresponding to the first factor.
Step 3: A step of creating an FT diagram in which know-how is displayed in the first factor column connected to the failure detection items shown on the screen based on the first and second tables.
Step 4: A step of identifying a failure location based on the FT diagram.

  Since the device failure diagnosis method according to the present invention creates an FT diagram based on a table describing know-how to be identified in correspondence with each first factor for estimating a failure location, the failure analysis time can be reduced, Since know-how is displayed in the FT diagram, there is an effect that the cause of the failure can be investigated easily and in a short time even if it is not an expert engineer for equipment operation, maintenance or the like.

Embodiment 1 FIG.
A first embodiment of the present invention will be described below with reference to FIGS.
In FIG. 1, a first table 100 is a table in which, for example, first to fourth factors 1 to 4 are input and displayed in order to create an FT (Fault Tree Analysis) diagram. In the first embodiment, as the first factor 1, each part of the target device that performs failure diagnosis is classified by function. As can be seen from the first table 100, for example, a control unit, a drive unit, Examples include a detection unit and a main body unit.
The second factor 2 is obtained by further subdividing each part of the first factor 1 for each component device. For example, factors corresponding to the control unit include malfunction of the fault display circuit, analog operation unit Other failures are mentioned. The third factor 3 is obtained by subdividing the second factor 2 into each device level. In the first table 100 shown here, the fourth factor 4 is a factor that is expanded and classified into the lowest part level that has received the immediately preceding third factor.
In the first embodiment, the first to fourth factors are used. However, the present invention is not limited to this, and the first to nth factors greater than the first to fourth factors may be used in a complicated system device.

  In the second table 200, for example, a plurality of other know-how 5 “the main turbine actual rotational speed does not increase even though the TV valve is open” is associated with the first factor 1 in the first table 100. This is a comparison table. The plurality of know-hows 5 are obtained by accumulating experiences and knowledge possessed by specialists such as operation and maintenance of the target device in advance, and the first to fourth of the assumed failure phenomenon (know-how 5). FIG. 6 is a correspondence diagram 6 showing how the factors correspond.

The FT diagram automatic generation tool 7 creates an FT diagram 300 shown in FIG. 2 on the screen based on the first table 100 and the second table 200.
In the FT diagram 300 shown in FIG. 2, the matter 8 detected during operation of the equipment, that is, a failure phenomenon, for example, “large speed deviation” is the highest, and the first to first items in the first table that lead to this are shown. The fourth factors 1 to 4 are displayed in a tree shape, and each know-how 5 of the second table 200 is displayed in the first factor 1 column to form an FT diagram.

Next, an operation for creating the FT diagram 300 will be described.
First, the factors 1 to 4 are input to the first table 100 shown in FIG. 1 corresponding to the first format. The input is made so that the development becomes finer as the first factor 1 → the second factor 2 → the third factor 3 → the fourth factor 4. As a specific example of this input, as described above, the function of each unit → the device of each component → the component of the device → the failure of the component. In this case, at least the first factor 1 and the second factor 2 need to be input. If there is no input item equal to or less than the third factor 3, it is not necessary to input.
Next, in the second table 200 in FIG. 1 corresponding to the second format, a plurality of know-hows 5 for specifying the first factors 1 to 4 are input. This know-how 5 inputs the expert's past know-how, that is, what has been investigated and accumulated in advance for failure phenomena.
Next, in correspondence with the failure phenomenon (know-how) 5, FIG. 6 corresponding to the first to fourth factors is input. In other words, this second table 200 shows which functional part of the device the failure phenomenon (know-how) 5 experienced by a professional engineer in the past, and a technician with little experience identifies the failure part. In some cases, this is valuable data.

  After completing the input of the first table 100 and the second table 200, the FT diagram automatic generation tool 7 creates the FT diagram 300 shown in FIG. In the FT diagram 300, the first to fourth factors input to the first table 100 connected in a tree shape are displayed on the screen with the detection item 8 as the highest level. Further, the know-how 5 input in the second table 200 is shown in the first factor 1 column. As described above, in the FT diagram 300 according to the first embodiment, the know-how of the experienced person is described under the first factor. Therefore, the failure factor can be identified quickly and easily even if it is not a specialized technique.

Embodiment 2. FIG.
Next, a second embodiment will be described with reference to FIGS.
3 is different from FIG. 1 shown in the first embodiment described above in that the factor correspondence diagram of the second table 200 which is a comparison table including the know-how 5 and the factor correspondence diagram 6 of the first embodiment. A weighting point is added to FIG. 6 to obtain a correspondence diagram 6a, and the result is a second table 200a which is a comparison table. Similarly, the question screen call button 9 is added to the FT diagram 300 as shown in FIG. 4 to obtain the FT diagram 300a, and the rest is the same as in the first embodiment. The FT diagram 300a shown in FIG. 4 is the screen display shown in the FT diagram 300b of FIG. 6 after the creation of the FT diagram according to the second embodiment. The question screen call button 9 is a button for calling a question screen 10 to be described later from the FT diagram 300a displayed on the screen using the know-how 5 of the second table 200a as a question contrast item.
In FIG. 5, the question screen 10 displays the know-how of the second table 200a as the question contrast item 5a, and further includes a check box 11 for answering the question contrast item 5a and a button 12 for determining the result. is doing. The point calculation logic 13 calculates the weighting points in the second table 200a. In the FT diagram 300b shown in FIG. 6, the result of the point calculation performed on the question contrast item 5a is displayed in the first factor 1 column by color change. FIG. 7 shows a flowchart of the point calculation logic.

Next, the creation operation of the FT diagram 300b shown in FIG. 6 will be described.
Each factor 1 to 4 is entered in the first table 100. In this case, at least the first factor 1 and the second factor 2 need to be input.
The know-how 5 is entered in the second table 200a. In FIG. 6a, weighting points (numerical values) are described for each functional part, which part of the first factor 1 is highly likely to cause a failure when know-how 5 is established. Here, as a restriction when inputting a numerical value, it is assumed that only one maximum value is input. Other input methods are free.
After the input of the first table 100 and the second table 200a is completed, the FT diagram automatic generation tool 7 creates the FT diagram 300a shown in FIG.
In the created FT diagram 300a, the display of the know-how 5 for specifying the first factor 1 is that of the first factor 1 in which the maximum value is described among the weighting points described in 6a of the second table 200a. Displayed in the column.
When the button 9 for calling the question screen is pressed from the created FT diagram 300a, the question screen 10 for performing the specific support of the first factor 1 is displayed as shown in FIG. Here, when a question is established for each question contrast item 5a, the input operator checks the check box 11 for answering the question contrast item. When the button 12 for judging the result is pushed after each input is completed, the value inputted in the check box 11 for answering the question contrast item and the weighting point described in the correspondence diagram 6a of the second table 200a are displayed. Use the point calculation logic 13 for calculation. The point calculation logic 13 is as described in the flowchart of FIG. 7. Based on this result, the FT diagram 300b shown in FIG. 6 is obtained by changing the color of the first factor 1 of the FT diagram 300a on the screen. Created. In this system, the possibility that each factor is the cause is distinguished by color change. For example, the large possibility is represented by red, the middle is yellow, and the low is represented by cyan.
In this way, in the second embodiment, weighting points are added to the know-how of the experienced person under the first factor column, and the identification of the cause of failure can be performed more easily by distinguishing and displaying by color change. It becomes.

Embodiment 3 FIG.
Next, a third embodiment will be described with reference to FIGS.
The third embodiment is different from the second embodiment described above in that a second diagram shown in FIG. 8 is obtained by substituting the correspondence diagram 6a of the second table 200a shown in FIG. Table 200b is the same as that of the second embodiment except that the accompanying explanation is shown in FIGS.
In FIG. 8, the second table 200b is for inputting know-how 5 for specifying the first factor 1, and the corresponding FIG. 6b is a part indicating which part of the first factor 1 the know-how 5 corresponds to. , Indicate weighting points. Others are the same as in the second embodiment.
In FIG. 10, a question screen 10a shows an option button 11a for answering a question and a point calculation logic 13a, and is the same as in the second embodiment except for the created FT diagram 300c. FIG. 12 is a flowchart of the point calculation logic.

Next, the creation operation of the FT diagram 300c shown in FIG. 11 will be described.
Each factor 1 to 4 is input to the first table 100 as in the second embodiment.
Next, know-how 5 is input to the second table 200b. In the correspondence diagram 6b, when know-how 5 is established (in the case of “Yes”) and not established (in the case of “NO”), which part of the first factor 1 is likely to be the cause, Write the weighting points (numerical values) for each factor.
As a restriction when inputting a numerical value, it is assumed that only one maximum value is always entered in the field when know-how 5 is established (in the case of “Yes”). Other input methods are free.
After the input of the first table 100 and the second table 200b is completed, the FT diagram automatic generation tool 7 creates the FT diagram 300a. In the created FT diagram 300a, the display of the know-how 5 for specifying the first factor 1 is the first factor column in which the maximum value in the “Yes” column is described among the points described in the corresponding diagram 6b. Is displayed below.
When a button 9 for calling a question screen from the FT diagram is pressed from the created FT diagram 300a, a question screen 10a for performing specific support for the first factor 1 is displayed as shown in FIG. Here, the input operator answers “Yes” when the question is established, “No” when the question is established, “Unknown” when the question is not established, and “Unknown” when the question is not well understood. Check the option button 11a. When the button 12 for judging the result is pressed after each input is completed, the point calculation is performed by using the value inputted by the option button 11a for answering the question item and the weighting point described in the corresponding FIG. 6b. . The point calculation logic 13a is as described in the flowchart of FIG. 12. Based on this result, the color change of the first factor 1 of the FT diagram 300a, that is, the possibility is red, medium is yellow, and low is low. By performing the cyan display, the FT diagram 300c shown in FIG. 11 is created.
As described above, in the third embodiment, by adding “Yes” and “No” points to the know-how of the experienced person under the first factor column, it is possible to more easily and accurately identify the cause of the failure. It becomes.

  Embodiments 1 to 3 of the present invention can be used for failure diagnosis of large Punrat equipment such as a thermal power plant.

It is a FT figure preparation point figure of Embodiment 1 of this invention. It is an FT figure of Embodiment 1 of this invention. It is a FT figure preparation point figure of Embodiment 2 of this invention. It is an FT figure of Embodiment 2 of this invention. It is a FT figure preparation point figure of Embodiment 2 of this invention. It is an FT figure of Embodiment 2 of this invention. It is a point calculation flowchart figure of Embodiment 2 of this invention. It is a FT figure preparation point figure of Embodiment 3 of this invention. It is an FT figure of Embodiment 3 of this invention. It is a FT figure preparation point figure of Embodiment 3 of this invention. It is an FT figure of Embodiment 3 of this invention. It is a point calculation flowchart figure of Embodiment 3 of this invention.

Explanation of symbols

1 First factor, 2 Second factor, 5, 5a Know-how (question contrast),
6,6a, 6b Corresponding diagram, 7 FT diagram automatic generation tool, 8 Detection items,
9 Question screen call button, 10 Question screen, 100 First table,
200, 200a, 200b second table,
300, 300a, 300b, 300c FT diagram.

Claims (4)

A device failure diagnosis method comprising the following steps, wherein a failure location is specified.
Step 1: As factors for estimating the failure location, the first and second factors are listed,
The first factor categorizes each part constituting the device for each function,
The second factor is obtained by further subdividing each part of the first factor for each component device,
Creating a first table displaying at least the first and second factors;
Step 2: A step of creating a second table having know-how possessed by a professional engineer and a correspondence diagram corresponding to the first factor.
Step 3: A step of creating an FT diagram in which the know-how is displayed in the first factor column connected to the failure detection items shown on the screen based on the first and second tables.
Step 4: A step of identifying a failure location based on the FT diagram. A device failure diagnosis method comprising the following steps, wherein a failure location is specified.
Step 1: As factors for estimating the failure location, the first and second factors are listed,
The first factor categorizes each part constituting the device for each function,
The second factor is obtained by further subdividing each part of the first factor for each component device,
Creating a first table displaying at least the first and second factors;
Step 2: A step of creating a second table including a know-how owned by a professional engineer and a correspondence diagram in which weighting points are attached to the first factor.
Step 3: Based on the first and second tables, create an FT diagram in which the know-how is displayed in the first factor column that leads to the failure detection items shown on the screen, and As a question contrast item, the weighting point of the second table is calculated by operating the question screen call button on the screen, and the first factor column on the FT diagram may cause a failure based on the calculation result. A step of distinguishing and displaying by color change according to the degree of sex.
Step 4: A step of identifying a failure location based on the FT diagram. A device failure diagnosis method comprising the following steps, wherein a failure location is specified.
Step 1: As factors for estimating the failure location, the first and second factors are listed,
The first factor categorizes each part constituting the device for each function,
The second factor is obtained by further subdividing each part of the first factor for each component device,
Creating a first table displaying at least the first and second factors;
Step 2: A know-how possessed by a professional engineer and a correspondence diagram corresponding to the first factor are provided, and the know-how is used as a question contrast item, and the question matter is YES and NO respectively. Creating a second table in which weighting points are attached to the first factor;
Step 3: Based on the first and second tables, create an FT diagram in which the know-how is displayed in the first factor column that leads to the failure detection items shown on the screen, and further on the screen The question screen call button is operated to calculate the weighting points in the second table when the question item is YES or NO, and the first factor column on the FT diagram fails based on the calculation result. A step of distinguishing and displaying by color change according to the degree of occurrence possibility.
Step 4: A step of identifying a failure location based on the FT diagram. The apparatus failure diagnosis method according to any one of claims 1 to 3, wherein step 1 of claims 1 to 3 is replaced by the following steps.
Step 1: As factors necessary for estimating the failure location, the first, second, third,.
The first factor categorizes each part constituting the device for each function,
The second factor is obtained by further subdividing each part of the first factor for each component device,
The third factor is obtained by classifying the second factor into each device level.
The nth factor is developed and classified into the lowest part level that has received the factor immediately before the nth factor,
Creating a first table displaying the first to nth factors;

Images