JP2925286B2 - Fault diagnosis system - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Purpose of the Invention] (Industrial application field) The present invention is intended to identify the cause of an abnormality or an abnormal sign in an industrial device or system and determine a corresponding measure. The present invention relates to an abnormality diagnosis system for diagnosing the abnormality.

(Prior art) When identifying the cause of an abnormal event that has occurred based on a knowledge base that indicates the degree of association between the cause of the abnormality and the survey item, the conventionally used knowledge base is that the survey item is unified, The degree of association was fixed.

(Problems to be Solved by the Invention) For this reason, when there are quite a lot of items to be investigated for abnormal events that have occurred, the efficiency of diagnosis has become extremely poor and time has been required.

The present invention has been made in view of the above circumstances, and has as its object to provide an abnormality diagnosis system that can efficiently identify the cause of an abnormality that has occurred in industrial equipment or a system.

[Structure of the Invention] (Means for Solving the Problems) That is, the abnormality diagnosis system of the present invention is a system in which the correlation between the investigation item and the cause of the abnormality with respect to the diagnosis target is converted into data. A diagnostic knowledge matrix that classifies the initial investigation item groups into easy-to-understand and detailed investigation item groups that require a lot of investigation, and displays the strength of association between each investigation item and each cause of the abnormality by the degree of effectiveness; An initial investigation guide module for presenting an effective initial investigation item based on the diagnostic knowledge matrix for an event, and the diagnostic knowledge matrix only when a result of an investigation conducted on the presented initial investigation item is valid. Is determined, the sum of the determined effectiveness is calculated for each cause of abnormality, and the sum of each effectiveness is defined as the sum of the effectiveness of the specific cause of abnormality. An initial diagnosis module for diagnosing an abnormal cause by multiplying a standardized value for each abnormal cause, which is a diagnosis result of the initial diagnostic module, by a value of the corresponding effectiveness of the detailed investigation item group of the diagnostic knowledge matrix A detailed research guide module that corrects the corresponding validity of the detailed research item group of the diagnostic knowledge matrix and presents a valid detailed research item,
It is characterized by comprising a detailed diagnosis module for identifying a cause of abnormality of a diagnosis target based on a research result performed on the presented detailed research item and the corrected diagnosis knowledge matrix.

(Operation) According to the abnormality diagnosis system of the present invention, the initial diagnosis guide module presents valid investigation items among the initial investigation item groups that are easy to investigate, and the results of the investigation performed on the presented initial investigation items and each of the investigation results. An initial diagnosis of an abnormal cause is performed by an initial diagnosis module based on a diagnostic knowledge matrix indicating the effectiveness between the investigation item and each abnormal cause, and the detailed investigation guide module is used to validate the diagnostic knowledge matrix based on the initial diagnosis result. The validity of the detailed survey item group, which requires time-consuming investigations, is presented in accordance with the degree of revision, and based on the revised diagnostic knowledge matrix described above, based on the results of the survey conducted on the presented detailed survey item Thus, the cause of the abnormality to be diagnosed is identified by the detailed diagnosis module. Therefore, it is possible to efficiently identify the cause of an accident or abnormality that has occurred in an industrial device or system.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings.

FIG. 1 shows an embodiment of the abnormality diagnosis system of the present invention. In the figure, reference numeral 1 denotes a knowledge base for storing a diagnostic knowledge matrix 2 in which the correlation between the investigation item and the cause of the abnormality is converted into data for the diagnostic object as shown in FIG. 2, and the diagnostic knowledge matrix 2 investigates the investigation item. It is classified into an initial investigation item group that is easy to perform and a detailed investigation item group that requires time and effort such as data measurement, and the degree of association between each investigation item and each cause of abnormality is displayed by the degree of effectiveness.

The computer main body 3 that reads the information of the knowledge base 1 and diagnoses abnormalities includes an initial investigation guide module 4 programmed to present effective initial investigation items based on the diagnostic knowledge matrix 2, and an initial investigation result. An initial diagnosis module 5 programmed for diagnosing the cause of abnormality from the diagnostic knowledge matrix 2 and a program for presenting valid detailed investigation items by rewriting the validity of the diagnostic knowledge matrix 2 based on the initial diagnosis results. The detailed investigation guide module 6 and the detailed diagnosis module 7 programmed to identify the cause of the abnormality to be diagnosed by the diagnostic knowledge matrix 2 'rewritten with the detailed investigation result are incorporated.

In FIG. 1, reference numerals 8, 9, and 10 denote a display device, a keyboard, and a mouse for displaying and operating inputs and outputs of the computer main body 3, respectively.

Next, the operation of the abnormality diagnosis system having the above configuration will be described for each component.

The diagnostic knowledge matrix 2 is a numerical value representing the degree of effectiveness of the degree of correlation between the results of the initial investigation item and the detailed investigation item with respect to the cause of each abnormal event. This degree of effectiveness is a maximum value of 1 and a minimum value of −1, and corresponds to “this cause item is the cause of the abnormal event” and “this cause item is not the cause of the abnormal event”. I have. Therefore, the absolute value of effectiveness indicates the strength of association between cause and investigation.

The initial investigation guide module 4 reads out the diagnostic knowledge matrix 2 corresponding to the abnormal event that has occurred, refers to the initial investigation items in the matrix 2, and lists the investigation items corresponding to the related abnormality causes. For the listed survey items, the sum of the absolute values of the effectiveness for each cause is calculated for each item. For example, in FIG. 2, the investigation 1 of the initial investigation item is as follows.

| a | + | b | + | b | + | c | + ‥‥ In this way, the sums obtained for the respective survey items are compared, and the one with the larger value is displayed preferentially. In this case, an arbitrary number of survey items are selected from those having a large value as needed, and the survey is performed.

The initial diagnosis module 5 inputs the results of the initial research performed on the items presented by the initial research guide module 4 and calculates the numerical value on the research item that is determined to be valid as a result of the research in the diagnostic knowledge matrix 2 as validity. To be determined. Please note that the survey results are not valid,
The values on the survey items for which no survey was performed are all zero. The sum of the degrees of effectiveness is calculated for each cause item, the larger the sum is, the higher the possibility of the cause is, and the maximum value of the sum is set to 1 for normalization. For example, in FIG. 2, the surveys 1, 2, and 3 are presented as initial survey items by the initial survey guide module 4, and when all the surveys are determined to be valid as a result of these surveys, the column of the survey 1 Are determined as the validity.
Similarly, for Survey 2, the numerical values c, c, c, a ‥‥ are:
For Survey 3, numerical values b, c, a, c, c, b} are determined. Next, the sum of the determined validity for each cause item is (a + 0 + b) for cause 1 and (0 + 0) for cause 2
+ C + c). When (c + a) of the cause 4 is the maximum value among the sums of the degrees of effectiveness of the respective cause items obtained in this way, this is normalized so that it becomes 1, and the degree of the possibility of the cause and Become. That is, the degree of possibility of cause 1 is (a + b) / (c + a), similarly, for cause 2 (c + c) / (c + a), and for cause 3 (b + c).
/ (C + a).

The detailed investigation guide module 6 calculates the product of the degree of the possibility of the cause determined by the initial diagnosis module 5 and the corresponding validity in the detailed investigation item of the diagnostic knowledge matrix 2, and newly calculates the product of the item. The corrected diagnostic knowledge matrix 2 ′ is created by replacing the diagnostic knowledge matrix with the validity. Based on the diagnostic knowledge matrix 2 ′, detailed investigation items to be investigated are presented in the same manner as the initial investigation guide module 4. For example, to explain the continuation of the above-described example, in FIG. 2, the degree of effectiveness (c + c) / possibility of the cause 2 obtained by the initial diagnosis module 5 is added to the effectiveness b of the investigation 1 and the cause 2 of the detailed investigation item. (C + a) is multiplied, and the obtained product b ′ b ′ = b × (c + c) / (c + a) is set as a new effectiveness b ′.

The detailed diagnosis module 7 inputs the result of the detailed research performed on the item presented by the detailed research guide module 6, and the numerical value of the column of the research item determined to be valid as a result of the research in the corrected diagnostic knowledge matrix 2 '. Is determined as the validity. Next, the sum of the degrees of effectiveness determined for each cause item is calculated, and the larger the sum is, the higher the possibility of the cause is, and presents it as the identified abnormal cause. For example, in the continuation of the above-described example, if the investigations 2 and 3 of the presented detailed investigation items are examined, and the investigation 2 and the investigation 3 are determined to be valid, the modified diagnostic knowledge matrix 2 ', The numbers a', b ', b', b '‥‥ in the column of survey 2 and the numerical values b',
c ′, a ′} are determined as the validity. Next, the sum of the validity determined for each cause item is obtained. In the case of cause 1, (a '+ b'), and in the case of cause 3, (b '+
c '). In this way, it is presented as an abnormal cause identified in the descending order of the sum.

As is clear from the above description, in the abnormality diagnosis system of the present invention, the investigation items are divided into an initial investigation item to be investigated initially and a detailed investigation item to be investigated after the result of the initial investigation, and each investigation item and abnormality First, select the initial investigation to be performed on the diagnosis target using the diagnostic knowledge matrix, which displays the degree of effectiveness between each cause item that lists the cause of the event by item and expresses the strength of the mutual relationship numerically. Then, from the results of this initial investigation, the magnitude of the possibility of an abnormal cause for each cause item is determined, and based on this possibility, the effectiveness of the detailed investigation items on the diagnostic knowledge matrix is reviewed, and a more effective By selecting detailed investigation items and identifying more probable causes of abnormalities based on the results of this effective detailed investigation, efficient and reliable abnormality diagnosis can be performed. It can be.

[Effects of the Invention] As described above, according to the present invention, it is possible to efficiently identify the above causes that have occurred in industrial equipment and systems,
It is extremely useful in industry.

[Brief description of the drawings]

FIG. 1 is a diagram showing the configuration of an embodiment of the abnormality diagnosis system of the present invention, and FIG. 2 is a diagram showing the contents of the diagnostic knowledge matrix shown in FIG. 1 ... Knowledge base, 3 ... Computer body, 8 ... Display device, 9 ... Keyboard, 10 ... Mouse

──────────────────────────────────────────────────続 き Continued on front page (58) Field surveyed (Int.Cl. ⁶ , DB name) G06F 9/44

Claims (1)

(57) [Claims] The present invention is a computerized data of a correlation between a survey item and a cause of an abnormality with respect to a diagnosis target, and is classified into an initial survey item group in which the survey item can be easily investigated and a detailed survey item group in which the investigation is troublesome. Presenting a diagnostic knowledge matrix indicating the strength of association between each investigation item and each cause of abnormalities by the degree of effectiveness, and an effective initial investigation item based on the diagnostic knowledge matrix for abnormal events that have occurred. The initial investigation guide module and the result of the investigation conducted on the presented initial investigation item, the validity displayed in the diagnostic knowledge matrix is determined only when the investigation is valid, and the determined validity is determined for each abnormality cause. An initial diagnosis module for calculating the sum of the degrees and standardizing the sum of the respective degrees of effectiveness with the sum of the degrees of validity for a specific abnormality cause to diagnose the abnormality cause; By multiplying the value of the corresponding effectiveness of the detailed investigation item group of the diagnostic knowledge matrix by the standardized value for each abnormality cause, which is the diagnosis result, the corresponding effectiveness of the detailed investigation item group of the diagnostic knowledge matrix A detailed investigation guide module that corrects and presents an effective detailed investigation item, and a detailed diagnosis module that identifies a cause of abnormality of a diagnosis target based on a result of the investigation performed on the presented detailed investigation item and the corrected diagnostic knowledge matrix. An abnormality diagnosis system comprising: