Method and apparatus for diagnosing faults

• US 4,964,125 A
• Filed: 08/19/1988
• Issued: 10/16/1990
• Est. Priority Date: 08/19/1988
• Status: Expired due to Term

• Current Assignee: Raytheon Company (Rtx Corporation)
• Original Assignee: Hughes Aircraft Company (Rtx Corporation)
• Inventors: Kim, Sachol E.
• Primary Examiner(s): Fleming, Michael R.
• Application Number:
  US07/234,434
• Time in Patent Office: 788 Days
• Field of Search: 364/200, 364/900, 364/513, 364/579, 371/20.1, 371/24, 371/26, 371/11.3, 371/11.1, 371/15.1, 371/29.1
• US Class Current: 714/26
• CPC Class Codes: G06F 11/2257 using expert systems
• CPC Class Codes: Y10S 706/911 Nonmedical diagnostics
• CPC Class Codes: Y10S 706/916 Electronic or computer, internal or network, circuit

Abstract

The present invention is directed to an expert system for diagnosing faults. The system comprises a knowledge base containing information regarding a conceptual model of the equipment under diagnosis. The system also comprises an inference engine for generating a set of symptom diagnostic rules from the information contained within the knowledge base in conjunction with user-generated inputs.

Claims

1. An expert system for diagnosing faults in equipment, said system including a computer having a memory for storing diagnostic routines and a user interface that permits communication between the expert system and an operator, said system comprising:

a knowledge base coupled to the computer for storing data that is defined by a plurality of files that characterize the physical, functional and operational relationships of the components and between components of the equipment and information concerning selected operating conditions of the equipment, said files containing data indicative of the following attributes of the equipment;

(1) the function of each component, (2) a wire list indicating outputs and interconnection of a component to other components of the equipment, (3) a level jump condition indicator which indicates whether a component may be decomposed into a plurality of other components, (4) a test message representing a stimulus applied to generate outputs that indicates whether a component is faulty, and (5) a listing of normal operating state values for a component indicating when that component is not faulty when the stimulus is applied thereto; and

an inference engine that is storable in said memory that interfaces with the knowledge base and the operator by way of the user interface and which comprises recursive diagnostic routines that operate on the data contained in the knowledge base to generate symptom diagnostic rules in response to operator entered symptoms which are indicative of the faults present in the equipment.

2. The expert system of claim 1 which further comprises:

a blackboarder coupled between said knowledge base and said inference engine, for accessing the data stored in the knowledge base and retrieving data necessary for the inference engine to perform its diagnosis.

3. The expert system of claim 1 wherein said knowledge base further comprising files containing enablement conditions required for normal operation of the components of the equipment.

4. The expert system of claim 1 wherein said knowledge base further comprises files containing maintenance actions that may be applied to selected components.

5. The expert system of claim 1, wherein the information stored in said knowledge base comprises a conceptual model of said equipment.

6. The expert system of claim 5, wherein said conceptual model of said equipment comprises the determinant of said equipment.

7. The expert system of claim 6, wherein said determinant is represented as a plurality of nodes.

8. The expert system of claim 7, wherein said nodes comprise a plurality of knowledge cells.

9. The expert system of claim 8, wherein each of said knowledge cells is of a type selected from the group consisting of a blank cell, a regular cell, a simple cell, a composite cell, a logical cell and a boundary cell.

10. The expert system of claim 9, wherein said determinant is further represented by a plurality of connectors disposed between said knowledge cells.

11. The expert system of claim 10, wherein said connectors are selected from the group consisting of a link connector, a leveler connector, an excursion connector, and a blank edge connector.

12. The expert system of claim 11, wherein said determinant is divided in such a manner as to be globally leveled.

13. The expert system of claim 12, wherein said determinant comprises a plurality of taxonomical pieces each containing at least one knowledge cell, each of said taxonomical pieces being connected to another taxonomical piece by a leveler, the knowledge cells in each of said taxonomical pieces being connected to at least one other knowledge cell by a link.

14. The expert system of claim 8, wherein each of said knowledge cells is uniquely identified by a cyber, and said inference engine includes means for generating a cyber transition sequence.

15. The expert system of claim 14, wherein said means for generating a cyber transition sequence comprises a global control heuristic.

16. The expert system of claim 15, wherein said expert system further comprises a blackboard and means for constructing a portion of said determinant on said blackboard.

17. The expert system of claim 16, wherein said means for constructing a portion of said determinant on said blackboard comprises a blackboarder-taxonomy routine.

18. The expert system of claim 17, wherein said means for constructing a portion of said determinant on said blackboard further comprises a blackboarder-knowledge cell routine.

19. The expert system of claim 16, wherein said inference engine further comprises means for isolating a faulty knowledge cell from the information contained on said blackboard.

20. The expert system of claim 19, wherein said means for isolating a faulty knowledge cell is operable to decompose said knowledge cell when said knowledge cell is composed of other knowledge cells.

21. The expert system of claim 20, wherein said means for constructing a portion of said determinant on said blackboard comprises a blackboarder, said inference engine comprises means for indicating to said blackboarder which portion of said conceptual model to retrieve.

22. The expert system of claim 21, wherein said means for indicating to said blackboarder which portion of said conceptual model to retrieve comprises a leveler instruction routine.

23. The expert system of claim 6, wherein said determinant is represented as a lattice Klein surface.

24. A process for operating a digital computer to enable a user to correct a fault in electronic equipment having a plurality of components, said computer having a memory for storing a knowledge base having a plurality of knowledge cells representing a functional portion of said electronic equipment and comprising a set of normal operating state values associated with said knowledge cells, and means for indicating whether said knowledge cells are decomposable into other knowledge cells, said knowledge cells being grouped into a plurality of taxonomical pieces that form a determinant of said electronic equipment, said process comprising the steps of:

(a) locating a fault in said knowledge cells in said determinant by comparing said normal operating state values of said knowledge cells with test data derived from applying selected inputs to said electronic equipment to generate outputs therefrom;

(b) indicating an ignorance condition and terminating said process if said knowledge cell determined to be faulty has been previously located;

(c) attempting to isolate said fault in said knowledge cell which has been determined to be faulty;

(d) indicating an ignorance condition and terminating said process if said fault is not isolated after repeatedly attempting to isolate said fault a predetermined number of times;

(e) indicating the existence of a fault isolation condition and terminating said process if said fault is isolated; and

(f) repeating steps (a) through (e);

whereby the fault in said electronic equipment is isolated by said digital computer to thereby permit the user to correct the fault in said electronic equipment.

25. The process of claim 24, wherein said knowledge cells comprise files containing data indicative of the following attributes:

(1) the function of each component of the equipment, (2) a wire list indicating outputs and interconnection of a component of the equipment to other components thereof, (3) a level jump condition indicator which indicates whether a component of the equipment may be decomposed into a plurality of other components, and (4) a test message representing a stimulus applied to generate outputs that indicates whether a component is faulty.

26. The process of claim 25, wherein said step of examining said knowledge cells in said determinant comprises the step of instructing a blackboarder to obtain a portion of a composite model of said electronic equipment contained in said knowledge base and place said portion of said composite model on a blackboard.

27. The process of claim 26, wherein said step of instructing a blackboarder to obtain a portion of said conceptual model comprises the step of executing a leveler instruction routine.

28. The process of claim 27, wherein said step of examining said knowledge cells in said determinant comprises the step of accessing sad knowledge base and constructing a taxonomical piece on said blackboard.

29. The process of step 28, wherein said step of accessing said knowledge in said knowledge base and constructing a taxonomical piece on said blackboard comprises the step of executing a blackboarder-taxonomy routine.

30. The process of claim 29, wherein said step of examining said knowledge cell in said determinant comprises the step of generating a cyber transition sequence.

31. The process of claim 30, wherein said step of generating a cyber transition sequence comprises the step of executing a global control heuristic.

32. The process of claim 31, wherein said step of attempting to isolate said fault in said knowledge cell comprises the step of accessing said knowledge base to construct said knowledge cell on said blackboard.

33. The process of claim 32, wherein said step of accessing said knowledge base to construct said knowledge cell on said blackboard comprises the step of executing a blackboarder-knowledge cell routine.

34. The process of claim 32, wherein said step of attempting to isolate said fault in said knowledge cell comprises the step of decomposing said knowledge cell when said knowledge cell is composed of other knowledge cells.

35. The process of claim 34, wherein said step of decomposing said knowledge cell comprises the step of executing a cell internal control heuristic.