Fault diagnosis using simulation

• US 5,528,752 A
• Filed: 11/22/1994
• Issued: 06/18/1996
• Est. Priority Date: 01/11/1991
• Status: Expired due to Term

• Current Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Original Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Inventors: Murata, Makoto, Masegi, Koichi, Kise, Takashi
• Primary Examiner(s): Beausoliel, Jr., Robert W.
• Assistant Examiner(s): LE, DIEU MINH T
• Application Number:
  US08/347,174
• Time in Patent Office: 574 Days
• Field of Search: 371/23, 371/5.1, 371/15.1, 371/16.1, 371/16.2, 371/16.4, 371/16.5, 371/25.1, 371/29.1, 371/29.5, 364/221.2, 364/232.3, 364/237.2, 364/264.3, 395/183.02, 395/183.03, 395/183.05, 395/183.09
• US Class Current: 714/29
• CPC Class Codes: G05B 17/02 electric
• CPC Class Codes: G06F 11/2273 Test methods
• CPC Class Codes: G06F 11/261 by simulating additional hardware, e.g. fault simulation

Abstract

In a fault diagnosis or a process simulation arrangement, data is held on first interactions between attributes of a part and between attributes of a parent part and a child part according to object parts in a parts data base. Data on an order for executing processes and data on a process model describing each process operation as second interactions between attributes of a medium and a part used in the process operation are held in a process model memory. In the fault diagnosis, fault statuses of attributes of parts and media used in each process are determined as candidates for causes of effecting statuses of attributes corresponding to the fault of the following process in reverse order of the execution order held in the process model memory using model based reasoning by referring to the first and second interaction relationships and statuses of attributes of the parts and the media determined in the following process. The attributes of the parts and the media may be expressed as a tree structure. The simulation is performed by determining statuses of attributes of parts and media used in each process sequentially in the execution order held in the process model memory using model based reasoning by referring to the first and second interactions and statuses of the parts and media determined in a preceding process.

Claims

1. A simulation method for simulating operations of a plurality of processes sequentially executed in an object, said method comprising the steps of:

holding rules for first interactions between an attribute of a part and another attribute of the part or an attribute of a child part of the part according to each part of the object, in a part data base;

holding an execution order for the plurality of processes, a knowledge of relationships between a part used in each process and another part, and rules for second interactions between attributes of a medium and a part used in the operation in each process as a process model in a process model memory; and

performing a simulation of the operations by determining statuses of attributes of parts and media after each of the processes, sequentially in the execution order held in the process model memory, using a model-based reasoning by referring to the first and second interactions, the knowledge of the relationship between the part used in each process and the another part and statuses of attributes of the parts and the media determined in a preceding process.

2. A simulation method according to claim 1, wherein the attributes of the medium are expressed in a tree structure and are updated in accordance with both the data on the process and the data on the parts.

3. A fault diagnosis method for detecting a cause of the fault occurring in an object executing a plurality of processes comprising the steps of:

holding rules for first interactions between an attribute of a part and another attribute of the part or an attribute of a child part of the part according to each part of the object, in a part data base;

holding an execution order for the plurality of processes, a knowledge of a relationship between a part used in each process and another part, and rules for second interactions between attributes of a medium and a part used in the operation in each process as a process model in a process model memory; and

performing the fault diagnosis by determining fault statuses of attributes of parts and media after each of the processes, said fault statuses being candidates for causes of effecting statuses of attributes corresponding to the fault in a following process, sequentially in reverse order of the execution order held in the process model memory, using model-based reasoning by referring to the first and the second interaction relationships, the knowledge of the relationship between the part used in each process and the another part and statuses of attributes of the parts and the media determined in the following process.

4. A fault diagnosis method according to claim 3, wherein the fault diagnosis is executed using a tree which expresses relationships between causes and effects of faults and which has been created on a basis of both the attributes of the parts and the attributes of the media.

5. A fault diagnosis method according to claim 4, wherein both the attributes of the parts and the attributes of the media are expressed in a tree structure, each of the attributes having a degree of reliability indicating that each part is normal.

6. A fault diagnosis method according to claim 5, wherein a plurality of diagnosis levels are provided by a degree of reliability of each of the attributes which indicates that the attribute is normal.

7. A fault diagnosis apparatus for diagnosing a fault occurring in an object comprising:

parts data base means for storing rules for first interactions between an attribute of a part and another attribute of the part or an attribute of a child part of the part according to each part of the object, in a part data base;

process model storage means for storing an execution order for the plurality of processes, a knowledge of a relationship between a part used in each process and the another part, and rules for second interactions between attributes of a medium and a part used in the operation in each process as a process model; and

diagnosis means for performing the fault diagnosis by determining fault statuses of attributes of parts and media after each of the processes, said fault statuses being candidates for causes of effecting statuses of attributes corresponding to the fault in a following process, sequentially in an order reverse to the execution order in the process model memory using a model-based reasoning by referring to the first and second interaction relationships, the knowledge of the relationship between the part used in each process and the another part and the statuses of attributes of the parts and the media determined in the following process.

8. A fault diagnosis apparatus for diagnosing a fault occurring in an object comprising:

parts data base means for storing rules for first interactions of parts of an object to be diagnosed in a form of relationships between attributes of the parts in a part data base;

process model storage means for storing knowledge of relationships between parts used in each of processes of the object and for storing rules for second interactions between an attribute of each of the parts and an attribute of a medium in an operation of each of the processes; and

diagnosis means for performing the fault diagnosis by using a model based reasoning by referring to the first and second interactions, to the knowledge of the relationship between the part used in each process and the another part and to inference relationships between attributes of parts not stored in said parts data base means nor stored in said process model storage means, said inference relationships being inferred through the second interactions between attributes of a medium and attributes of a part stored in the process model storage means,

said diagnosis means having a tree which expresses relationships between causes and effects of faults which has been created on the basis of both the attributes of the parts and the attributes of the media and which is updated by results of an inspection.

9. A fault diagnosis apparatus for diagnosing a fault occurring in an object comprising:

parts data base means for storing rules for first interactions of parts of an object to be diagnosed in a form of relationships between attributes of the parts in a part data base;

process model storage means for storing knowledge of relationships between parts used in each of processes of the object and for storing rules for second interactions between an attribute of each of the parts and an attribute of a medium in an operation of each of the processes; and

diagnosis means for performing the fault diagnosis by using a model based reasoning by referring to the first and second interactions, to the knowledge of the relationship between the part used in each process and the another part and to inference relationships between attributes of parts not stored in said parts data base means nor stored in said process model storage means, said inference relationships being inferred through the second interactions between attributes of a medium and attributes of a part stored in the process model storage means, said diagnosis means having a tree which expresses relationships between causes and effects of faults which has been created on the basis of both the attributes of the parts and the attributes of the medium and which is updated by results of an inspection,

wherein both the attributes of the parts and the attributes of the media are expressed in a tree structure in which each of the attributes has a degree of reliability indicating that they are normal, and wherein said diagnosis means has level selection means for making a selection of diagnosis level in accordance with the degree of reliability.

10. A fault diagnosis method for detecting a cause of the fault occurring in an object executing a plurality of processes comprising the steps of:

storing first interactions of each of parts of an object to be diagnosed in a form of relationships between attributes of the parts;

storing data on relationships between parts used in each of processes of the object and for storing data on a cause-and-effect relationship between an attribute of one of the parts and an attribute of a medium in an operation of each of the processes; and

performing the fault diagnosis by using a model-based reasoning by referring to said stored data on the operation and the relationships and by referring to inference relationships between attributes of parts not stored, said inference relationships being inferred through the second interactions between attributes of a medium and attributes of part stored, said stored data having a tree which expresses relationships between causes and effects of faults which has been created on the basis of both the attributes of the parts and the attributes of the medium and which is updated by results of an inspection.

11. A fault diagnosis method for detecting a cause of the fault occurring in an object executing a plurality of processes comprising the steps of:

storing rules for first interactions of parts of an object to be diagnosed in a form of relationships between attributes of the parts in a part data base;

storing knowledge of relationships between parts used in each of processes of the object and for storing rules for second interactions between an attribute of each of the parts and an attribute of a medium in an operation of each of the processes; and

performing the fault diagnosis by using a model-based reasoning by referring to the first and second interactions and to inference relationships between attributes of parts data on which is not stored, said inference relationships being inferred through the second interactions between attributes of a medium and attributes of part stored, said stored rules having a tree which expresses relationships between causes and effects of faults which has been created on the basis of both the attributes of the parts and the attributes of the media and which is updated by results of an inspection, storing data on an operation and attributes of each of the parts of an object to be diagnosed;

wherein both the attributes of the parts and the attributes of the media are expressed in tree structure in which each of the attributes has a degree of reliability indicating that they are normal, and further making a selection of a diagnosis level in accordance with the degree of reliability.