JP2003216923A - Fault diagnostic apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a fault diagnostic apparatus reducing the cost of maintaining a knowledge base, while enhancing the success rate of fault search. <P>SOLUTION: When a maintenance person inputs a trouble phenomenon, an inferring mechanism part 14 generates a hypothesis through interference using knowledge data stored in an FTA (fault tree analysis) type knowledge data storage part 16b, an FMECA (failure mode effect and criticality analysis) type knowledge data storage part 16c, and a diagnosis example type knowledge data storage part 16a. When a hypothesis is generated through FTA inference, it is used as the result of search. If a hypothesis is generated not through FTA inference but through FMECA inference, it is used as the result of search. Further, if only a hypothesis based on diagnosis example inference is generated, it is used as the result of search. The input of an answer to whether or not the trouble phenomenon has been solved is requested by a technique information collecting part 15, and if there is an input indicating that the trouble phenomenon has been solved, the result of search is stored in the diagnosis example type knowledge data storage part 16a while being associated with the current trouble phenomenon. <P>COPYRIGHT: (C)2003,JPO

Description

Description: BACKGROUND OF THE INVENTION The present invention relates to a so-called expert
The system is used to lead to malfunctions in vehicles, aircraft, etc.
In connection with a failure diagnosis device that investigates the cause of the attached failure,
Automatic accumulation of technical data by accumulating results as diagnosis cases
The present invention relates to a failure diagnosis device that collects information. [0002] In recent years, in various fields such as medicine, architecture, chemistry, etc.,
Extracts that use computer to help solve problems
Part system is adopted. This expert
Stem compiles expert knowledge in a particular field.
Computer and use it to address complex issues with experts.
A system that can be solved at an equivalent level
It is. Conventionally, this expert system has been used in vehicles.
For example, Japanese Patent Application Laid-Open No. Sho 62
As disclosed in Japanese Patent Publication No.
The root cause of the failure that is causing the phenomenon
(Failure location) is stored in the knowledge data stored in the knowledge base.
It is known to search by inference using data. [Patent Document 1] Japanese Patent Application Laid-Open No. 62-6846 [0005] Problems to be Solved by the Invention
Conventional fault diagnosis systems use rule-based knowledge as a knowledge base.
Use independent knowledge data such as knowledge data and case type knowledge data.
To investigate the cause of the failure, or
As shown in the technology, rule-based knowledge data and examples
Failure cause using different knowledge data such as type knowledge data
There is something to explore, but in any case, knowledge
A clear hypothesis if not fully captured
I can't. However, in the field of vehicles, aircraft, etc., control systems
Technology has been remarkably advanced, and the content of
There is a tendency for complexity to increase, and in such fields,
If you try to build the knowledge base without leaking,
Not only is the cost enormous, but the interviews become redundant
It is not easy to use. On the other hand, in the fields of vehicles, aircraft, etc.,
Building a sequential knowledge base in response to complexity is enormous
It is not feasible because it takes a lot of time and maintenance costs. [0008] The present invention has been made in view of the above circumstances.
While keeping down the maintenance costs required for renovating the knowledge base,
Effectively complicate the diagnostic content with the existing knowledge base.
Providing easy-to-use and easy-to-use failure diagnosis equipment
It is intended to be. [0009] In order to achieve the above object,
The failure diagnosis device according to the present invention has a failure phenomenon and a failure cause.
Analysis of the causal relationship of a tree and tree-like as a set of hypotheses
Tree that stores the fault tree analysis type knowledge data represented in
Analysis-type knowledge data storage unit
Failure modes organized in a connection matrix by confidence
Failure mode effect type knowledge data that stores effect type knowledge data
Data storage unit and the past diagnosis cases as diagnosis case type knowledge data.
Diagnostic case type knowledge data storage unit that stores
Tree analysis type knowledge data and the above failure mode effect type knowledge data
Fault phenomena entered with the data and the diagnosis case type knowledge data
Investigate the cause of failure by inference using character strings containing
Inspection and replacement or assembly of components
A manual data storage unit storing the procedure, and the inference machine
When the results of verifying the hypotheses generated by the
The above-mentioned diagnosis case type knowledge data collected in association with the condition phenomenon
A technical information collection unit that accumulates in a storage unit;
In the structural part, a hypothesis similar to the input character string
Extracted from the hypothesis of tree analysis type knowledge data and extracted
Using these hypotheses as the root of a fault tree analysis type knowledge base reasoning
Join along and apply preset selection criteria
Perform processing to generate the first temporary cause of the malfunction phenomenon, and
Input for failure phenomenon of failure mode effect type knowledge data
Confidence for a malfunction similar to the above string
The cause of failure that is combined via
Apply the selected selection criteria to create a second temporary cause of the failure phenomenon
Perform the same process as the one entered above.
Examples are extracted from the above diagnosis case type knowledge data and
Applying the preset selection criterion, the third temporary
A process for generating a cause is performed, and the first to third temporary causes are performed.
Is performed, and the first temporary cause is generated.
The first temporary cause is output as a search result and the search is performed.
And the first temporary cause is not generated and the second
If a cause is generated, the second temporary cause is searched for
And the search is terminated, and only the third temporary cause is generated.
If so, the third temporary cause is output as a search result.
To end the search, and the above technical information
Replacement procedure for components when the investigation at the logical unit is completed
Or read the inspection procedure from the manual data storage
Output and output whether the problem has been resolved.
You are required to enter a response and enter that the problem has been resolved.
When there is power, link the above search results with the problem phenomenon.
And store it in the diagnosis case type knowledge data storage unit.
Features. [0010] In such a configuration, the maintenance technician is in trouble.
, The inference mechanism firstly enters the input character string
A hypothesis similar to that in the hypothesis of the fault tree analysis type knowledge data
And extract the extracted hypotheses from each other.
Join along the route of base inference and pre-set
The first tentative cause of the failure phenomenon by applying the selected selection criteria
Is performed. In addition, the failure mode effect type knowledge data
Failure similar to the above character string input in the condition phenomenon
The cause of failure that is coupled to the
Error and apply a preset selection criterion
A process for generating a second temporary cause of the elephant is performed. In addition,
Diagnostic case-type knowledge data
From the list and apply the preset selection criteria.
Then, a process for generating a third temporary cause of the malfunction phenomenon is performed.
Next, a process of matching these first to third temporary causes is performed.
When the first temporary cause is generated, the first temporary cause is generated.
The cause is output as an inquiry result and the inquiry is terminated. on the other hand,
The first temporary cause has not been generated and the second temporary cause has been generated.
In this case, the second temporary cause is output as a search result and the search is performed.
To end. If only the third temporary cause is generated,
Output the third temporary cause as the inquiry result and end the inquiry
I do. [0011] Then, the technical information collecting unit, the inference machine
There is a procedure for replacing components when the exploration is completed
Or read the inspection procedure from the manual data storage
And answer whether the problem has been resolved.
Request for power and there is no input that the malfunction has been resolved.
When the result of the diagnosis is
It is stored in the type knowledge data storage unit. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings.
An embodiment will be described. Note that, in this embodiment,
Of the failure phenomena of the fuel system of the machine as an example of the failure diagnosis target
It will be described while describing. [0013] As shown in FIG.
The fault diagnosis device A is composed of a device main body 1 and an input pen 2.
The main body 1 uses the input pen 2 to store information.
Is provided on the touch screen 1a. [0014] The above-mentioned failure diagnosis device A is used in an airport maintenance department or the like.
It is provided, and the above touch
When a trouble phenomenon is input to the touch screen 1a,
Computer can identify the cause of the failure or the location of the failure.
Investigate and provide necessary information such as the results of
The information is displayed on the touch screen 1a to assist the maintenance staff. As shown in FIG. 1, built in the apparatus main body 1.
Computers have a function to execute fault diagnosis.
Window processing unit 12, system control unit 13, inference mechanism unit
14, technical information collection unit 15, knowledge base unit 16,
The recording data section 17 is configured. Also, the above touch
The screen 1a includes an input unit 11a and a display unit 11b.
Has been established. The window processing unit 12 includes pre-processing and
Character recognition unit 12a and operation input unit 12b
And a display control unit 12c is provided as post-processing.
Have been killed. In the character recognition section 12a, the touch screen
Window for pen input displayed on the screen 1a (see FIG. 10).
Phenomena, etc. entered by the maintenance staff with the input pen 2
Converts handwritten characters to character codes
Is output to the system control unit 13. Operation input section
At 12b, the graphic displayed on the touch screen 1a is displayed.
Fix or menu selected with input pen 2
At the time, the commands corresponding to the
13 is output. In addition, the display control unit 12 c
Display based on the signal output from the stem control unit 13
The unit 11b displays characters, graphics, and the like. The system control unit 13 includes a man machine
Interface control unit 13a and operation mode control unit
13b and a system management unit 13c are provided.
In the man-machine interface controller 13a,
Execution processing according to commands from window processing unit 12
Work. Alternatively, the character data is sent to the display control unit 12c.
Data and graphics data. Operation mode
The controller 13b operates according to the use mode selected by the maintenance personnel.
Suspension / resumption of diagnostic processing, printing and ordering of maintenance records
Equipment including peripheral equipment such as data transmission of the management system
Controls the overall operating mode. In the system management unit 13c
Is the entire system, such as the operating status of the system, data management, etc.
Manage. The inference mechanism 14 has a character string search function.
Unit 14a, rule-based inference unit 14b, and case base
An inference unit 14c and an inference control unit 14d are provided. In the character string search section 14a, the maintenance staff
The character string that represents the trouble phenomenon that was
(“.” “,” “Ga” “ha” etc.)
Using the decomposed character string, a knowledge base
Stored in the respective knowledge data storage units 16a to 16c.
Is the same or similar to the above character string
Character string, word unit, or character
Searched in units and stored in each of the knowledge data storage units 16a to 16c
It is totaled for each stored knowledge data. The rule-based inference unit 14 b
Analysis of the fault tree provided in the knowledge base unit 16 (fault tree
Analysis; hereinafter abbreviated as "FTA") type knowledge day
Data stored in the data storage unit 16b and failures
Modal effect analysis (matrix with confidence, hereinafter referred to as “FME
CA ”) is stored in the type knowledge data storage unit 16c.
Cause or location of failure using knowledge data
By inference to generate hypotheses (see
Will be described later in the flowchart.) In the case-based inference unit 14c, the character string
Same or similar to the defect phenomenon searched by the search unit 14a
Similar events are stored in the knowledge base type
Search from knowledge data stored in data storage unit 16a
Then, a hypothesis of a failure cause or a failure location is generated. Then, the inference control unit 14d sets each of the above
Determine consistency of hypotheses generated by inference units 14b and 14c
And if the hypotheses are consistent, this hypothesis is considered the conclusion hypothesis.
And output. The technical information collecting unit 15 includes a data collecting unit 15
a and a case registration unit 15b. Data collection
In the section 15a, the maintenance staff inputs the
Input during maintenance work such as trouble items, inspection points, measured values, etc.
Replace parts according to the data and the hypothesis above
As a result of verifying the above hypothesis,
Is not resolved or the cause of the failure can be identified
And other input data. In the case registration section 15b
Is based on the data input to the data collection unit 15a.
Diagnose the results of this inquiry (both successful and unsuccessful cases)
As a case, a diagnosis case type knowledge data storage unit 16 described later is used.
The knowledge data of a is accumulated. The knowledge base unit 16 stores diagnosis case type knowledge.
Diagnostic case type knowledge data storage unit 16a for storing data;
FTA type knowledge data storage for storing FTA type knowledge data
Unit 16b and FMEC storing FMECA type knowledge data
A-type knowledge data storage unit 16c and digitized manual data
And an electronic manual data storage unit 16d for storing
Has been established. The above-mentioned diagnosis case type knowledge data is based on past failure sources.
A database that summarizes the results of exploring factors as examples,
As shown in FIG. 11, a failure phenomenon (A)
Or the location of the malfunction), the cause of the malfunction and
The actions, defective parts, and the type of knowledge source are stored.
Have been. In addition, as a knowledge source of this diagnostic case data
Is a defect record sheet, maintenance record sheet, and
There are views etc. As shown in FIG.
A column to record "defects and inspection points" found by staff
And a column to record the “treatment” performed by maintenance personnel
Have been. The maintenance record sheet contains the information
Elephant ”and“ disabled ”written by the department that performed the work
"Parts", "failure status", etc. are recorded. Also, interviews were conducted with maintenance personnel.
Know-how on maintenance procedures and troubleshooting that are not documented
Are collected and reflected in the diagnosis cases. Recorded in the FTA type knowledge data storage section 16b
The remembered FTA type knowledge data consists of design materials and skilled
Analyze the maintenance staff's experience, etc.
Analyzed logically for each component system and expressed as a set of rules
As shown in FIG. 14, rules for each component
(Rules 1 to 6 in the figure)
They are linked in a chain via hypotheses. For example, le
Rule 2 assumes that "booster / pump abnormality" is an intermediate hypothesis.
And the conclusion hypothesis (fuel leakage etc.)
Conclusion hypothesis when “fuel leakage” is assumed as an intermediate hypothesis (not tightened)
Good, bad inside the pump, etc.). On the other hand, FMECA type knowledge data is
It is displayed in a box-shaped table format, and shows the experience and
Analyze the reliability information of components and parts or systems
In this way, the failure phenomena and the cause of failure are separated for each part or system.
It is expressed in a tabular format as a so-called FMECA table.
Therefore, the failure mode of the part or system is
The failure, the effects of the failure, and how to detect the failure.
You. In addition, certainty is clearly indicated inside each grid.
You. This confidence is based on the MTBF (mean failure
Interval) or for the same malfunction
Consider the number of failure causes or the actual number of failure occurrences, etc.
The degree of certainty can be set as
The depth of the causal relationship with the cause of failure is shown. Also, the digitized manual data includes component parts.
And text describing inspection, replacement or assembly procedures, etc.
And graphics data, for example, as shown in FIG.
As described above, on the touch screen 1a of the device main body 1,
The booster pump graphic and the booster pump
Data such as the inspection procedure of the pump
Data and display it in the window. The operation record data section 17 stores a maintenance record.
Record data storage unit 17a and work progress temporary storage unit 17b.
And is stored in the maintenance record data storage unit 17a.
Are the actions taken for the failure, the inspection results, and the
The maintenance result such as the situation is stored for each case. Also,
For example, the work progress temporary storage unit 17b may be replaced during maintenance.
Suspended to order parts, then resumed diagnostics
In such cases, continue from the interrupted maintenance work.
In order to be able to
The maintenance progress or maintenance status is sequentially stored. Next, the failure diagnosis procedure will be described with reference to FIGS.
7 will be described. Note that this implementation
In the form, description will be given by exemplifying a malfunction of the fuel system of the aircraft.
You. This failure diagnosis is performed according to the inference processing routine shown in FIG.
Therefore, it is executed. First, the maintenance technician operates the device diagnosis failure device A.
From the menu displayed on the touch screen 1a of the body 1
When “Start diagnosis” is selected, as shown in FIG.
The window for inputting the condition and the
A pen input window for inputting elephants in sentence expression
It is displayed and the inference process is started. In step S1, the mechanic checks that the hand
Refer to the forms such as the condition record slip 21 (see FIG. 9) and
Necessary information such as the operational state of the aircraft system, or the occupants
Input pen 2 to input the malfunction phenomenon (symptom) reported from
Input to the touch screen 1a. It should be noted that the processing of the defect record slip 21 shown in FIG.
After maintenance is completed by the maintenance staff,
Maintenance record data such as driving status can be written
(See FIGS. 27 and 30). On the other hand, in step S2, a later-described step is performed.
Is obtained by a tester connection according to the medical inquiry displayed in step S12.
Enter symptoms such as measured values directly. Then, an error on the touch screen 1a is detected.
Complete the necessary items in the condition phenomena input window, etc.
Then, select the “input completed” window with the input pen 2
Then, the entered information (date, mission, contents, etc.) is entered.
Output as force data and stored in memory in step S3
Is done. Next, in step S4, in step S1
Failure phenomena entered and stored in step S3
Knowledge stored in knowledge data storage units 16a to 16c
Calculate the similarity between the data and the event. The calculation of the similarity is performed by the character string search unit 14a.
This is performed, for example, in the following procedure. (1) First, each of the knowledge data storage units 16a
From which the similarity is calculated from each of the knowledge data
Extract each character string (sentence). Note that each of these strings
Are stored in the respective knowledge data storage units 16a to 16c.
ID (self) for associating with the knowledge data
Certificate) number. (2) Next, the sentence to which this ID number is assigned
String, string, word, and character units
Each knowledge data is stored in the working memory for each knowledge data.
To pay. Note that the data stored in each working memory is
The “decomposition character string” has the same ID number as that before decomposition.
Has been granted. Here, a "string" is a character string
"Word" is a blank character, registered in advance
Separating characters (“.”, “,”, “Ga”, “wa”
)), For example, the string "Engine does not start"
Is disassembled into "engine" and "not started". [0045] A character is a character string that is formed by a predetermined number of characters.
Refers to disassembly, for example, the string "
"No", "Eng", "N-begin", "No movement", or
Can be “E”, “Engine”, “Start”, “No”, or “E”.
Every three characters, such as "n", "gin", "start", "zu", etc.
Decompose. (3) Next, store in the working memory
"Ignored word" registered in advance from the "decomposed character string"
Phrase ("but", "I am", etc.) ". (4) After that, the “synonyms” registered in advance
("Failure" is "failure" etc.)) "
Replace "Unsolved string". (5) On the other hand, the trouble phenomenon input by the maintenance staff
Strings of (symptoms) can be changed to strings as in (1) to (4) above.
For each word, word, and character. (6) Then, the character of the broken down phenomenon
"Decomposed" stored in the working memory using the
Character string ". For example, when the engineer enters the
"Doesn't work" in the knowledge data storage units 16a to 16c.
Similar to character strings extracted from stored knowledge data
The similarity is as shown in Table 1 below. [Table 1] (7) Next, a string, word, or
Is the ID number of the data that matches exactly in each character unit
Aggregated for each issue, and the aggregated result is used as the similarity
Output to the corresponding inference units 14b and 14c. In addition, total
String, word, or character
This is performed using a weighting coefficient for each. As a result, the malfunction phenomena “fuel loss” “left”
The tank is not reduced and is consumed only from the right. "
Knowledge data stored in the knowledge data of 6a to 16c
Is similar to the following. Stored in the diagnosis case type knowledge data storage section 16a
Of similarity calculation in the knowledge data
Diagnosis cases are as shown in FIG.
The similarity has the value shown in FIG. Among them, Case -1048
"Cruising at 3000 ft""No left tank fuel consumption"
Is the highest similarity, that is, the entered defect phenomenon
This is the case with the most similar symptoms. Further, the information is stored in the FTA type knowledge data storage section 16b.
Among the stored knowledge data, the similarity
The tree shown in FIG.
Tree consisting of a chain of rules with
The similarity of each hypothesis is as follows. 1) Rule 5 “Internal failure of pump itself” →
3, 2) Intermediate hypothesis between rule 2 and rule 4 "fuel leakage" →
40, 3) Rule 3 "fuel leak" → 40, 4) Intermediate hypothesis between Rule 1 and Rule 2 "Boostapon
Anomaly ”→ 3, 5)“ Transf is an intermediate hypothesis between Rule 1 and Rule 3.
6) Rule 1 i) "Fuel shut-off valve error" → 35, ii) "Fuel sub-tank error" → 36, iii) "Incorrect indication of fuel indicating system" → 38 ,. On the other hand, the FMECA type knowledge data storage 16
c in the knowledge data stored in c.
Elephant phenomena are as shown in FIG.
Is the highest value of “fuel loss” as shown in FIG.
Is shown. At step S4, each knowledge data storage unit 1
6a to 16c and the fault data
When the similarity with the elephant is calculated, the following steps S5 and S5
6 and S7 are executed in parallel (concurrent processing).
In steps S5, S6, and S7, the respective knowledge data storage units
Based on knowledge data stored in 16a to 16c
Then, the cause of the failure is investigated by inference to generate a hypothesis. That is, in step S5, the diagnosis
Knowledge data stored in the example knowledge data storage unit 16a
Case-based estimation, which is a preset selection criterion,
The cause of failure is investigated by the theory, and the hypothesis which is the third temporary cause is
Generate. In step S6, the FTA type knowledge data
Using the knowledge base stored in the storage unit 16b,
The rule-based inference, which is a selection criterion
The cause of the fault is searched for, and a hypothesis that is the first temporary cause is generated.
Further, in step S7, the FMECA type knowledge data
Using the knowledge base stored in the storage unit 16c,
Failure due to set selection criteria, rule-based inference
Investigate the cause and generate a second hypothesis, the hypothesis. First, the diagnosis case inference in step S5 is performed.
Generation of a hypothesis will be described. This diagnosis case inference is
This is executed by the example-based inference unit 14c.
Is a hypothesis generation routine based on the diagnosis case inference shown in FIG.
It is performed according to. In step S31, the above inference processing routine
Diagnosis case type knowledge data storage unit calculated in step S4
The similarity for each case of the knowledge data stored in 16a
The hypothesis is narrowed down based on the hypothesis. Refine this hypothesis
Depends on whether the similarity is equal to or greater than a preset threshold.
to decide. For example, if the “threshold value” is 10, then FIG.
1. As shown in Fig. 12, "Case-1048" is a diagnosis case
It is a hypothesis generated by inference. 11 and FIG.
In the knowledge data shown in Fig. 12, a case exceeding the "threshold"
There was only one, but a class higher than this "threshold"
If there are multiple cases with similarity,
This is a hypothesis based on a case inference. Thereafter, in step S32, the above steps
The reliability of the hypothesis narrowed down in S31 is calculated. This message
The reliability is based on other inferences (FTA inference or FMECA inference).
Criteria for resolving conflicts with the hypotheses obtained by
In step S4 of the above inference processing routine,
It is calculated by multiplying the calculated similarity by a coefficient. In addition,
In the present embodiment, as shown in FIG.
In this case, the coefficient is set to 1. Then, in the step S31, the narrowing down is performed.
Inference control unit
14d, and outputs to step S8 of the above inference processing routine.
Return to Next, a provisional calculation based on the FTA inference in step S6 is performed.
Generation of the theory will be described. This FTA inference is based on the rules
-This is executed by the base inference unit 14b, and specifically
Is a hypothesis generation routine based on FTA inference shown in FIG.
This is done accordingly. First, in step S41, the above inference processing
FTA type knowledge data calculated in step S4 of the routine
“Fuel” in the knowledge data stored in the storage unit 16b
Consists of a chain of rules with "system failure" as the top event
Based on the similarity of each hypothesis in the tree
Select hypotheses that are equal to or greater than the set "threshold" and narrow down the hypotheses
Perform For example, if the “threshold value” is 15,
In the FTA type knowledge data of FIG. 14, a) “fuel leakage” which is an intermediate hypothesis between rule 2 and rule 4 b) “fuel leakage” of rule 3 c) i) “fuel shut off” of rule 1 Valve abnormalities ii) "Fuel sub-tank abnormalities" iii) "Insufficient fuel indication system indication" is narrowed down. Then, in step S42, the above steps
Search for a route linking each hypothesis narrowed in S42
I do. In FIG. 14, the top event, "Failure of fuel system"
"" Is the "fault phenomenon" this time.
The roots extending to the left of the figure have a causal relationship to this phenomenon.
, That is, a “hypothesis”. Above
Link the hypothesis selected in step S41 with the above-mentioned problem
As shown in FIG.
It becomes street. Then, in step S43, the above steps
The completeness of the route searched in S42 is calculated. This complete
The degree is the connection (causality) between the failure phenomenon and the cause of the failure.
The higher the value, the higher the value. Of the route searched in FIG.
The completeness is as follows. 1) Route 1 → 50 2) Route 2 → 50 3) Route 3 → 50 4) Route 4 → 30 5) Route 5 → 30 Next, in step S44, based on completeness,
The hypothesis is based on whether all degrees are equal to or greater than a preset threshold.
Perform a refinement. For example, if the “threshold” is set to 20,
In this case, all of the routes 1 to 5 are targeted. Next, in step S45, the above steps
Calculate the reliability of the hypotheses narrowed down in S44,
Each of these hypotheses is referred to above as "hypothesis based on FTA inference".
Output to the control unit 14d, and execute the inference processing routine.
Return to step S8. The reliability is obtained by multiplying the perfection by a coefficient.
For example, the above “threshold value” is set to 20,
When the coefficient is 1, as shown in FIG.
A hypothesis based on A inference ”covers all routes 1 to 5
You. Next, step S in the inference processing routine is performed.
Generation of hypothesis by FMECA inference performed in 7
explain. This FMECA inference is the same as the above FTA inference.
As described above, the rule-based inference unit 14b executes
Specifically, generation of a hypothesis by FMECA inference shown in FIG.
It is performed according to a routine. First, in step S51, the FMECA type
Each of the knowledge data stored in the knowledge data storage unit 16c.
From the phenomena, narrow down the phenomena to be diagnosed this time.
U. The narrowing down of this phenomenon is a phenomenon corresponding to the malfunction phenomenon.
Is calculated in step S4 of the inference processing routine.
Based on the degree, the degree of similarity is determined by a predetermined threshold
This is done by selecting phenomena equal to or greater than the value. For example, "Shiki
When the “value” is 10, the so-called FM shown in FIG.
The phenomena displayed in the ECA table indicate that "fuel decrease"
Restricted to "Fuel depletion" and "Unable to correct fuel tank remaining amount"
(See FIG. 18). Next, in step S52, the above steps
The degree of certainty corresponding to the phenomenon narrowed down in S51 is calculated using the FM
Determined from the ECA table. For example, what is called a so-called
The FMECA table shows that the confidence level corresponding to “fuel reduction” is 1
6,21,63, confidence level corresponding to "fuel depletion" is 5
0, 31, 46, "Unable to correct fuel tank remaining amount"
Is 40,60. Then, in step S53, the above-mentioned certainty factors
Select a certainty factor that is equal to or greater than the preset "threshold value" from
Then, narrow down hypotheses. For example, the above "threshold"
Is set to 20, as shown in FIG.
In step S54, this confidence
Select the parts and causes corresponding to the degrees. Next, in step S55, the above steps
Calculate the reliability of the component or cause selected in S54
Then, the reliability and the hypothesis are referred to as “provisional by FMECA inference”.
For example, as shown in FIG.
Output to the inference control unit 14d,
It returns to step S8. In this embodiment, the above reliability is
The sum of the values obtained by multiplying the similarity and the confidence is multiplied by a coefficient.
It is as shown in the table below. [Table 2] Then, the flow of the inference processing routine shown in FIG.
Returning to step S8, the process proceeds to steps S5, S6, and S7.
Determine if your hypotheses conflict, and if so,
It is canceled according to a preset criterion. Competitive solution to this hypothesis
The deletion is executed by the inference control unit 14d shown in FIG.
Specifically, according to the hypothesis conflict resolution routine of FIG.
Done. First, in step S61, each of the above steps
The hypotheses generated in S5, S6 and S7 are summarized in a table
I do. Hypotheses generated in above steps S5, S6, S7
Is, for example, as shown in the table of FIG.
And are ranked according to the calculated reliability. Then, in step S62, each of the above steps is performed.
Check the consistency of the hypotheses generated in steps S5, S6, and S7.
From the target, attributes, etc. For example, in FIG.
Rank 1 of hypothesis by case inference and provisional by FMECA inference
According to the theory, the defective part is "Booster pump"
Consistent in some respects, but the order of hypotheses based on diagnostic case inference
Rank 1 and rank 1 of the hypothesis based on FTA inference
Are common in terms of "fuel system failure"
However, defective parts are competing. Similarly, FTA inference
And the hypothesis based on FMECA inference
Are competing for defective parts. Next, when the operation proceeds to step S63, the step
The consistency of the hypothesis calculated in step S62 is
To determine whether there is consistency.
Perform settings. In the present embodiment, the "consistency criterion"
"All the defects with the highest reliability of the hypothesis in each inference are
Match), and according to this definition,
Judge compatibility. As a result, for example, as shown in FIG.
Rank 1 hypotheses generated by each inference are not all the same
If there is no consistency in step S64,
Proceeding from S64 to step S65, the inference count measuring cow
Increments the count value C, and proceeds to step S66.
No. In step S66, the above inference frequency measurement
The count value C is compared with the set value N, and when C ≦ N,
It is determined that "diagnosis is not completed", and the process proceeds to step S67. Stay
In step S67, the hypothesis collected in step S61
Create data for re-inference according to certain criteria. This
The criterion is, for example, "The reliability of each hypothesis is the highest
If you define it as "select items,"
Is re-estimated based on the hypothesis of rank 1 by FMECA inference.
Rational data is created (see FIG. 21), and this data is
The information is output together with the information of “diagnosis not completed”, and step S9
Return to On the other hand, in step S66, it is determined that C> N.
Rejection, that is, re-inference
If the hypotheses do not match, the process proceeds to step S68,
Resolve conflicts according to preset priority criteria
You. The priority in this embodiment is the most logical
Hypotheses based on FTA-type knowledge data
Assuming the highest priority, a hypothesis is generated with this FTA type knowledge data.
If not, hypotheses based on FMECA type knowledge data
And both of these knowledge data
If no hypothesis is generated, the
Adopt a hypothesis. And compete according to this priority
As a result of exploring the hypothesis that has been resolved,
Output with the information and exit the routine. In addition, this knowledge
The cause of failure cannot be investigated in any of the knowledge data
In this case, in step S68, "the hypothesis is not satisfied"
The information of "diagnosis end" is output, and the process returns to step S9. On the other hand, in step S64, each of the above steps is performed.
When the hypotheses generated in steps S5, S6, and S7 match
Makes this hypothesis a search result and branches to step S69.
After clearing the inference count measurement count value C,
The above search results are output together with the information of "diagnosis end",
It returns to step S9. Then, step S9 of the inference processing routine
Returning to step, whether or not "inference is completed" is created in step S8.
Judgment is made based on the information obtained, and if
Proceed to step S10, and if "inference not completed"
Returns to step S4. Then, in the above step S9, "inference not completed"
If the process returns to step S4 when the
Failure phenomena stored in memory in S3 and predetermined measurement values
And the like, and the above-mentioned step S8 (step S6
Characters combined with "Re-inference data" created in 7)
Column, and stored in each of the knowledge data storage units 16a to 16c.
The similarity with the existing knowledge data is calculated again. In the re-inference data shown in FIG.
Fuel loss, left tank not reduced, only from right
Consumption, booster pump, own malfunction
The respective character strings and the respective knowledge data storage units 16
defect phenomena of knowledge data stored in a to 16c,
Comparing defective parts, failure causes, etc., and performing logical operations, etc.
Calculate the similarity. Then, according to the similarity of each knowledge data,
In steps S5, S6, and S7, the
Investigate the cause of the obstacle. As a result, diagnosis case inference in step S5
Then, even at the time of re-inference, based on the diagnosis case type knowledge data,
The generated hypothesis is not changed.
The inference result is output. On the other hand, the FTA estimation executed in step S6
In theory, as shown in FIG.
From the character string, “Failure in fuel system” and “Booster Po
Is abnormal, and the hypothesis that is the cause of the failure is
The causal relationship with the phenomenon is the strongest in route 5 shown in FIG.
Therefore, as shown in FIG.
Is a high value of 80, and rank 1 of the hypothesis by FTA inference
become. On the other hand, in the FMECA inference in step S7,
Is a sentence corresponding to “fuel loss” as shown in FIG.
There is a string, the hypothesis has not changed, and a similar result is output.
It is. As a result, FIG.
In step S64 of the hypothesis conflict resolution routine of FIG.
As shown in Table 22, the hypothesis ranked 1 in each inference
Because the parts match, the predefined
Is satisfied, and in step S64, it is determined that there is consistency.
When it is determined, the process branches to step S69, and the number of inferences is calculated.
After clearing the count value C for measurement,
Output together with the information of "diagnosis end" and proceed to step S9
Return. Then, in step S9, "the inference ends."
If it is determined to be "completed", the process proceeds to step S10, and other check items
Judge whether there is an eye, and if there is an inspection item,
Branching to S11, for the maintenance staff or the user
Narrow down the content of the interview. By this questionnaire, this conclusion tentative
Information for testing hypotheses and generating new hypotheses
And That is, for example, in the FTA type knowledge base,
As shown in FIG. 15, the intermediate hypothesis of Rule 2
"Stack / Pump Abnormality" Connected by Rules [Fuel Leakage]
Is linked to the conclusion hypothesis of Rule 4. Therefore,
Specify the location to identify the abnormality of the pump
By inspecting this location, the failure location can be further narrowed down.
Can be. Then, in step S12, the above steps
The contents of the medical inquiry narrowed down in S11 are
Display on the clean 1a, and at the same time,
The work contents and necessary information are, for example, as shown in FIG.
indicate. Then, returning to steps S1 and S2, the maintenance staff
Wait for input of inspection result from. Then, the maintenance person operates the touch screen 1
Input the inspection result from the input pen 2 or the tester into a
In step S3, this data is stored, and
At S4, the input character string and the previous "input data"
Data based on the character string of "
Is calculated again, and steps S5, S6, S7 and subsequent steps are performed.
Below, based on each knowledge data, infer again,
Investigate the cause. Then, the inference is completed, and from step S9
Proceed to step S10 to check items for verifying the hypothesis
If not, the process proceeds directly to step S13, and
Determine if the cause could be explored. If the cause of the failure can be found, step S
At 14, for example, as shown in FIG.
The reason is displayed on the touch screen 1a, and the maintenance person
Confirm the contents and enter the "OK" window with pen 2
Is selected in step S15.
Displays information on necessary procedures such as parts replacement procedures, etc.
Proceeding to step S16, the processing result from the
In other words, it waits for the input of the result as to whether the problem has been solved. The maintenance person operates the touch screen 1a
Replace the indicated parts or check necessary parts.
As a result, the problem has been resolved or
If can be specified, enter the verification details. Then, the inquiry result by this inference succeeds
In step S17, a confirmation screen for the result is displayed.
It is displayed on the touch screen 1a. The above touch screen
For example, as shown in FIG.
Display details, verification results, etc., and succeeded in this inquiry
And accumulate the results of this inquiry as successful cases.
Display a dialog box stating that The maintenance staff checks the contents of the dialog box.
And select the "OK" window with the input pen 2.
Then, in step S18, first, the FMECA type knowledge
Of the knowledge data stored in the data storage unit 16c
Phenomena and causes or parts selected through multiple searches
The success of this inquiry was to determine the degree of certainty (see Fig. 17)
Therefore, it is updated with a relatively high value. Then, in step S19, the diagnosis case type knowledge
Knowledge data stored in the knowledge data storage unit 16a,
Using the results of this inquiry as a successful case, for example,
Add and accumulate as in "Case-5721". Next, when the operation proceeds to step S20, the current
The details of the inquiry are shared with supplementary information such as the date of diagnosis and the name of the diagnoser.
In the maintenance record data storage unit 17a,
End the disconnection. The maintenance record data storage section 17a has a memory
Data can be retrieved via an external printer, etc.
Yes, for example, as shown in FIG.
And can be counted. On the other hand, the theory presented in step S14
Along with the procedure indicated in step S15
Therefore, replace parts or check necessary parts.
As a result, the failure phenomenon still persists or
When the location is not specified, in the above step S16
When the user inputs that fact, in step S17, the touch
On the screen 1a, for example, as shown in FIG.
In addition to displaying the history of the
Register that this investigation was unsuccessful and that it failed.
Confirmation screen appears. [0108] Then, the maintenance person operates the dialog box described above.
Check the contents of the entry and enter the "OK" window with pen 2
In step S18, the FMEC is selected.
Knowledge data stored in the A-type knowledge data storage unit 16c
The phenomena and causes selected in this inquiry
The reliability of connecting parts (see Fig. 17)
Since the result is a failure, update with a relatively low value. Then, in a step S 19, the present search
The result is used as a diagnosis case, and the diagnosis case type knowledge data storage unit 16 is used.
a, for example, as shown in FIG.
As such, it is added and stored as "Case-5721". By the way, in this failure diagnosis device A,
In steps S5, S6, and S7, a satisfactory hypothesis is generated.
If the cause of the failure cannot be determined, or
The conclusions drawn are clear based on my own experience.
Error, or this failure investigation failed.
In the event that a
It is set to be acquired. That is, in step S13, each of the above
Investigate the cause of failure as a result of inference using knowledge data
Failed, or with the failure
When the same inquiry result is derived, go to step S21
The branch branches, and the maintenance staff investigates the fault independently. Then, as a result of the investigation, the defect phenomenon was resolved.
In step S22, when the maintenance staff
The latitude is input to the touch screen 1a. Then,
In step S23, based on the input data, an
Knowledge to accumulate as a knowledge base is acquired and
Return to step S17, and the confirmation screen of the search result
It is displayed on the screen 1a. After confirming the display screen, the "OK"
When a window is selected by the input pen 2, step S18
And stored in the FMECA type knowledge data storage unit 16c.
Knowledge data that has been clarified through this unique search
Confidence between the failed cause or replacement part and the failure phenomenon
The degree (see FIG. 17) is updated with a relatively high value. Then, in a step S19, the diagnosis case type knowledge
Knowledge data stored in the knowledge data storage unit 16a,
Using the results of this inquiry as a successful case, for example,
Add and accumulate as in "Case-5722". As a result, as shown in “Case-5721”,
Even if the conclusion hypothesis sought by inference is a failure,
Example-5722 '', the diagnosis is made based on the knowledge
Accumulation in example knowledge data causes similar problems
The hypothesis similar to the failed case is generated again by inference
Without interpolating other knowledge data.
Wear. Then, the flow advances to step S20 to search for the current time.
The history of the investigation is stored in the maintenance record data storage unit 17a,
End fault diagnosis. [0117] It should be noted that the results were obtained by the independent search of the maintenance staff.
By connecting an external printing machine as described above,
As shown in FIG.
Can be taken out. As described above, according to the present invention, the information obtained by inference is obtained.
By verifying the hypothesis and inputting the results,
Whether the investigation of the cause of the failure was successful or unsuccessful in the knowledge base
Feedback and, if successful,
Knowledge stored in the disconnected case type knowledge data storage unit 16a
Since the data is updated automatically, use the fault diagnosis device A.
By gaining a lot of knowledge, the success rate gradually increases
Can be obtained. As a result, knowledge base maintenance costs
Not only can reduce
It becomes. [0119] Also, the results of the original search by the maintenance staff are used as knowledge.
The potential knowledge of maintenance personnel.
Relatively easy to acquire and refine your knowledge base.
Can be densely built. In addition, register failure cases
As a result, the same conclusions can be made during the next and subsequent failure diagnosis.
Theories were not derived and the maintenance work could only be made more efficient
However, this failure case and the above-mentioned unique search result input function and
The existing knowledge to complicate the diagnosis contents.
Can respond effectively on the basis of knowledge and trouble diagnosis
By using the device A, the failure diagnosis succeeds in the same way as described above.
Rate can be higher. In addition, failure cases and unique
Use the results of your inquiry as a reference when refining your knowledge base.
With this, it is possible to narrow down the search results by inference more precisely
it can. The present invention is limited to the above embodiment.
It is not a thing, for example, the target of failure diagnosis is limited to aircraft
Vehicles, railways and other vehicles, ships, etc.
Is also good. As described above, according to the present invention,
Verify the hypothesis obtained by inference and eliminate the problem phenomenon
If this is the case, associate the verification result of the hypothesis with the defect phenomenon.
To be stored in the diagnostic case type knowledge data storage unit
Therefore, the more the failure diagnosis device is used, the more convenient it is
You. As a result, the maintenance required for the renovation of the knowledge base, etc.
Not only can we reduce our costs, but we also get more complicated
Diagnosis contents can be effectively handled with the existing knowledge base.
And high reliability can be obtained. In generating a hypothesis, the most logical
Based on knowledge tree analysis type knowledge data which is constructed
The first temporary cause generated by
If a tentative cause is not generated in the analyzed knowledge data,
Second temporary data generated based on the mode-effect-type knowledge data
The cause is adopted, and in both of these knowledge data
If no temporary cause is generated even if the
The third temporary cause generated based on the data
As a result, the conflict of the temporary cause has been resolved and
You.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a functional block diagram of a failure diagnosis device. FIG. 2 is a flowchart showing an inference processing routine. FIG. 3 is a flowchart showing an inference processing routine (continued). FIG. 4 is a hypothesis based on diagnosis case inference. FIG. 5 is a flowchart showing a hypothesis generation routine based on FTA type knowledge base inference. FIG. 6 is a flowchart showing a hypothesis generation routine based on FMECA type knowledge base inference. FIG. 7 is a hypothesis conflict resolution routine. FIG. 8 is an external view of the failure diagnosis device. FIG. 9 is an explanatory diagram showing a defect record slip. FIG. 10 is an explanatory diagram showing an input screen of a defect phenomenon. FIG. 11 is an explanatory diagram showing diagnosis case type knowledge data. FIG. 12 is an explanatory diagram showing a calculation result of similarity in diagnosis case type knowledge base reasoning. FIG. 13 is a diagnosis case type knowledge base reasoning. FIG. 14 is an explanatory diagram showing FTA type knowledge data. FIG. 15 is an explanatory diagram of route search results in FTA type knowledge base inference. FIG. 16 is a hypothesis generation by FTA type knowledge base inference. FIG. 17 is an explanatory diagram of FMECA-type knowledge data. FIG. 18 is an explanatory diagram of a similarity calculation result by FMECA-type knowledge base inference. FIG. 19 is a diagram illustrating a hypothesis generation result by FMECA-type knowledge base inference. FIG. 20 is an explanatory diagram showing aggregation of hypotheses obtained by each inference. FIG. 21 is an explanatory diagram of data for re-inference. FIG. 22 is an explanatory diagram showing aggregation of hypotheses obtained by re-inference. An explanatory diagram showing a display screen at the time of an inquiry. [FIG. 24] An explanatory diagram showing a display screen of inference results and arguments. [FIG. 25] An explanatory diagram showing a display screen of confirmation of successful inquiry results. [FIG. 26] New FIG. 27 is an explanatory diagram of a diagnosis case type knowledge data in which a registered successful case has been registered. FIG. 27 is an explanatory diagram showing a failure record sheet describing the contents of a treatment. FIG. 28 is an explanatory diagram showing a display screen for confirming a failed inquiry result. FIG. 29 is an explanatory diagram of diagnosis case type knowledge data in which a new failure case and a successful case individually searched for are registered. FIG. 30 is an explanatory diagram showing a failure record form in which the contents of a treatment by a unique search are described. 14 Inference mechanism unit 15 Technical information collection unit 16a Disconnected case type knowledge data storage unit 16b Failure tree analysis type knowledge data storage unit 16c Failure mode effect analysis type knowledge data storage unit 16d Electronic manual data storage unit

   ────────────────────────────────────────────────── ─── Continuation of front page    (72) Inventor Satoshi Takahashi             Fuji, 1-7-2 Nishi Shinjuku, Shinjuku-ku, Tokyo             Heavy Industry Co., Ltd. (72) Inventor Yasuo Kagei             Fuji, 1-7-2 Nishi Shinjuku, Shinjuku-ku, Tokyo             Heavy Industry Co., Ltd. (72) Inventor Masaaki Koyama             Fuji, 1-7-2 Nishi Shinjuku, Shinjuku-ku, Tokyo             Heavy Industry Co., Ltd. (72) Inventor Kunihiro Abe             Fuji, 1-7-2 Nishi Shinjuku, Shinjuku-ku, Tokyo             Heavy Industry Co., Ltd. F term (reference) 5B048 AA15 DD12 DD13 FF02

Claims (1)

Claims: 1. A fault tree analysis type knowledge data in which a causal relationship between a failure phenomenon and a failure cause is theoretically analyzed and fault tree analysis type knowledge data expressed in a tree as a set of hypotheses is stored. A storage unit, a failure mode influence type knowledge data storage unit that stores failure mode influence type knowledge data arranged in a matrix by linking a failure phenomenon and a failure cause with a certainty factor, and a diagnosis case type knowledge data for past diagnosis cases. A diagnosis case type knowledge data storage unit, which stores the failure tree analysis type knowledge data, the failure mode influence type knowledge data, the diagnosis case type knowledge data, and a character string including the input failure phenomenon. An inference mechanism that searches for the cause by inference, a manual data storage that stores the inspection and replacement or assembly procedures of the components, and a generation by the inference mechanism described above. A technical information collection unit that collects the result of verifying the hypothesis that has been verified in association with the input failure phenomenon and accumulates the result in the diagnosis case type knowledge data storage unit. A similar hypothesis is extracted from the hypothesis of the faulty tree analysis type knowledge data, and the extracted hypotheses are connected along the route of the faulty tree analysis type knowledge base inference, and a failure is determined by applying a preset selection criterion. A process for generating a first temporary cause of the phenomenon is performed, and a failure cause that is coupled via a certainty factor to a failure phenomenon similar to the character string input in the failure phenomenon of the failure mode influence type knowledge data is determined. A process of extracting and applying a preset selection criterion to generate a second temporary cause of the malfunction phenomenon is performed, and a case similar to the input character string is diagnosed by the diagnosis case type knowledge. A process of extracting the data from the data and applying a preset selection criterion to generate a third temporary cause of the defect phenomenon is performed, and a process of matching the first to third temporary causes is performed. When the tentative cause is generated, the first tentative cause is output as a search result and the search is terminated. When the first tentative cause is not generated and the second tentative cause is generated, The second temporary cause is output as a search result, and the search is terminated. If only the third temporary cause is generated, the third temporary cause is output as a search result, and the search is terminated. In the above-mentioned technical information collecting unit, when the inquiry by the above-mentioned inference mechanism unit is completed, the replacement procedure or the inspection procedure of the component parts is read out from the above-mentioned manual data storage unit and output, and an answer as to whether the problem has been solved is input. Problem is solved A failure diagnosis apparatus characterized in that when there is an input indicating that the information has been erased, the search result is stored in the diagnosis case type knowledge data storage unit in association with the failure phenomenon.