Distributed maintenance system based on causal networks

• US 6,208,955 B1
• Filed: 06/12/1998
• Issued: 03/27/2001
• Est. Priority Date: 06/12/1998
• Status: Expired due to Term

• Current Assignee: Teledyne Scientific & Imaging, LLC (Teledyne Technologies Incorporated)
• Original Assignee: Rockwell Science Center LLC
• Inventors: Provan, Gregory M., Sitter, Charles J.
• Primary Examiner(s): Teska, Kevin J.
• Assistant Examiner(s): PHAN, THAI Q
• Application Number:
  US09/097,206
• Time in Patent Office: 1,019 Days
• Field of Search: 703/16, 703/2, 703/27, 703/20, 706/47, 706/20, 706/60, 706/45, 706/52, 706/26, 714/739, 714/26, 707/104, 712/28, 701/14, 701/35, 709/213, 709/201
• US Class Current: 703/20
• CPC Class Codes: G05B 23/0245 based on a qualitative model, e.g. rule based; if-then decisions
• CPC Class Codes: G06F 11/2257 using expert systems
• CPC Class Codes: G06F 11/261 by simulating additional hardware, e.g. fault simulation
• CPC Class Codes: G08G 1/165 for passive traffic, e.g. including static obstacles, trees

Abstract

The present invention relates to a diagnostic system for complex systems such as avionics systems. The diagnostic system acquires a description of the system from a variety of design tools and creates a causal network model as an intermediate step. From the causal network model, the diagnostic system builds and compiles a Q-DAG model of the system, which is then embedded in the central maintenance computer of the aircraft. The present invention integrates two elements, a graphical user interface (GUI), which acts as a data capture tool and graphical display of the avionics system and an inference system, which acts as a diagnostic tool with a presenter. The presenter permits diagnosis of faulty sub-systems and a report may be relayed to remotely located maintenance crews to minimize repair time upon arrival of the aircraft.

Claims

1. In an avionics system having a plurality of interrelated sub-systems, a diagnostic system for diagnosing a failing sub-system, said diagnostic system comprising:

a communication network incorporating an object oriented communication paradigm for transmitting objects;

a data collector, coupled to said sub-systems and to said communication network, for deriving state variables representative of the operation of said sub-systems;

an evaluator, coupled to said communication network for receiving said state variables, having a set of query directed acyclic graph files associated therewith, said evaluator adapted to access said files to generate a diagnostic inference in response to erroneous operation of at least one of said sub-systems as represented by said state variables; and

a presenter, coupled said communication network, for combining said inference from said evaluator with a schematic representation of said avionic system to graphically identify at least one sub-system causing said erroneous operation.

2. The diagnostic system of claim 1 wherein said network communication means further comprising an object request broker for transferring objects.

3. The diagnostic system of claim 2 further comprising at least one additional presenter for displaying said diagnostic inference, said at least one additional presenter coupled to evaluator and said database by said communications network.

4. The diagnostic system of claim 1 further comprising:

means for building a schematic model of said avionics system;

a causal network generator for receiving said schematic model and for generating a causal network system model;

said causal network system model providing a ranked set of diagnoses for identifying failures of said avionics system;

a compiler for generating a compiled representation of said avionics system, said representation being an expression of possible diagnoses for said avionics system; and

a database coupled to said schematic model building means, said causal network generator, said compiler and said communication network;

said database having information describing said avionics system stored therein.

5. The causal network system model of claim 4 wherein said causal network system model is based on behavioral characteristics of said avionics system.

6. A diagnostic system for identifying faults in an electro-mechanical system comprising:

means for building a model-based description of said avionics system based on normal operating modes and potential failure modes of said avionics system;

a database for storing said model-based description;

means for acquiring state variables of said system and for determining an inference regarding the source of a fault in said system;

means for displaying a graphical representation of said system together with a fault indication;

communication network for coupling said building means, said acquiring means and said displaying means to said database, said communications network using the Internet Inter-ORB Protocol; and

a CORBA interface, associated with said model building means, said database, said acquiring means and said display means, for brokering and communicating objects over said communications network.

7. The diagnostic system of claim 6 wherein said means for building a model-based description of said system further comprises:

means for developing a schematic model of said system, said schematic developing means comprising a plurality of abstract model blocks and a Boolean rules set describing the behavior of said abstract model blocks;

each of said abstract model blocks having a monitor and generator block associated therewith;

means for developing a causal network model from said schematic model, said causal network model defining behavioral operation of said abstract model blocks in terms of said Boolean rules set;

means for applying causal inference to said causal network model to develop a query directed acyclic graph model representing said system; and

a compiler for compiling said query directed acyclic graph.

8. In an avionics system having a plurality of interrelated sub-systems, a method for diagnosing system faults comprising the steps of:

building a schematic model describing the interconnection of said sub-systems;

converting said schematic model to a causal network model describing normal functioning and faulty functioning of said avionics system;

converting said causal network model to a query directed acyclic graph (Q-DAG) representation of said causal network;

compiling said Q-DAG representation;

storing said compiled Q-DAG representation in a memory means associated with a computer, said computer electrically associated with said plurality of sub-systems;

monitoring observable variables of said avionics system in accordance with said compiled Q-DAG representation to correlate said observable variables with flight deck effects; and

graphically identifying at least one faulty sub-system and the cascade effect caused by said faulty sub-system on said avionics system.

9. in an electro-mechanical system, a method for diagnosing electro-mechanical system faults comprising the steps of:

generating a compiled a query directed acyclic graph (Q-DAG) representation of said electro-mechanical system

using said Q-DAG representation, diagnosing faults occurring in said electro-mechanical system based on a plurality of state inputs received from said electro-mechanical system;

transporting, via a communications network, a diagnostic inference of failures of said electro-mechanical system to a viewer;

combining said diagnostic inference with a schematic representation of said electro-mechanical system;

generating a graphical display, on said viewer, of said combined diagnostic inference and said electro-mechanical system.

10. The method of claim 9 wherein said generating step further includes the step of animating the propagation of faults through said electro-mechanical system.

11. The method of claim 9 wherein said generating step further comprises the step of graphically illustrating the cascade effect of said fault.

12. The method of claim 9 wherein said transporting step further includes the step of transferring objects over a communications network using the Internet Inter-ORB Protocol.

13. The method of claim 12 wherein said transporting step further includes the step of providing a communications standard for brokering and communicating objects over said communications network.

14. The method of claim 13 wherein said transporting step further includes the step of providing a CORBA interface for brokering and communicating objects over said communications network.

15. The method of claim 9 further including the step of storing said Q-DAG representation of said electro-mechanical system in a database.

16. The method of claim 15 further including the step of accessing said database to acquire said schematic representation of said electro-mechanical system prior to said combining step.

17. In an electrical system having a plurality of interrelated sub-systems, a diagnostic method comprising the steps of:

modeling said system;

generating a compiled representation of said system, said representation comprising a set of diagnostic inference algorithms based on causal networks;

generating diagnostic inferences, using said compiled representation of said system, in response to a detected state change;

transporting objects over a communications network using the Internet Inter-ORB Protocol, said objects comprising at least said diagnostic inference and said system model to a viewer; and

graphically displaying said diagnostic inference on said viewer.

18. The diagnostic method of claim 17 wherein said transporting step further includes the step of providing a CORBA interface for brokering and communicating objects over said communications network.

19. The method of claim 18 further including the step of accessing a common database to acquire a schematic representation of said electro-mechanical system prior to said modeling step.

20. The method of claim 19 wherein said modeling step further comprises the step of selecting at least one of a plurality of primitives from a library of primitives to define the function of said electrical system, said selected primitive having both a generator and a monitor block associated therewith.

21. The method of claim 20 wherein said modeling step further comprises the step of defining the behavioral characteristics of said electrical system.

22. The method of claim 17 further comprising the steps of:

transmitting said diagnostic inference to a remote viewer; and

combining said diagnostic inference with a schematic representation of said electro-mechanical system.

23. The method of claim 22 further including the step of accessing said database to acquire said schematic representation of said electro-mechanical system prior to said combining step.

24. An electro-mechanical system diagnostic system comprising

a schematic capture and viewer system for developing a model of said electro-mechanical system;

a database for capturing and storing system information;

a system model generator, coupled to said database, for generating a causal network model of said electro-mechanical system; and

a Q-DAG compiler, coupled to said database, for generating a set of table-loadable files representative of a Q-DAG model of said electro-mechanical system.

25. The diagnostic system of claim 24 further comprising a communication network for coupling said schematic capture and viewer system, said generator and said compiler with said database.

26. The diagnostic system of claim 25 wherein said communication network comprises an Internet Inter-ORB Protocol (IIOP) compliant communication network.

27. The diagnostic system of claim 26 further comprising a plurality of CORBA interface modules, each of said plurality of modules associated with said schematic capture and viewer system, said database, said generator and said compiler.

28. The diagnostic system of claim 27 further comprising:

means for acquiring state variables of said electro-mechanical system and for determining an inference regarding the source of a fault in said electro-mechanical system; and

means for displaying a graphical representation of said electro-mechanical system together with a fault indication.

29. The diagnostic system of claim 28 wherein said display means comprises a graphical viewer, coupled to said communication network, for developing hierarchical models and for providing a graphical interface between system models generated by said schematic capture and said inference means.

30. The inference means of claim 29 further comprising:

a data collector for monitoring bus activity of said electro-mechanical system and for generating state changes representative of the state of said electro-mechanical system;

an evaluator for evaluating said state variables and generating a diagnostic inference upon detection of an error in the operation of said electro-mechanical system; and

a presenter for graphically displaying a combined system level model of said electro-mechanical system.