Rule based diagnostic system with dynamic alteration capability

• US 4,642,782 A
• Filed: 07/31/1984
• Issued: 02/10/1987
• Est. Priority Date: 07/31/1984
• Status: Expired due to Fees

• Current Assignee: Westinghouse Electric Company LLC (Brookfield Asset Management Incorporated)
• Original Assignee: Westinghouse Electric Company LLC (Brookfield Asset Management Incorporated)
• Inventors: Lowenfeld, Simon, Kemper, Christian T., Fox, Mark S.
• Primary Examiner(s): NOT, DEFINED
• Application Number:
  US06/636,196
• Time in Patent Office: 924 Days
• Field of Search: 364/431.01, 364/492, 364/494, 364/550, 364/551, 364/554, 364/556
• US Class Current: 706/53
• CPC Class Codes: G05B 23/0278 Qualitative, e.g. if-then rules; Fuzzy logic; Lookup tables; Symptomatic search; FMEA
• CPC Class Codes: G05B 23/0297 Reconfiguration of monitoring system, e.g. use of virtual sensors; change monitoring method as a response to monitoring results
• CPC Class Codes: Y10S 706/911 Nonmedical diagnostics

Abstract

Sensor based diagnostic apparatus for performing on-line realtime monitoring of an industrial or other operating system. The diagnostic apparatus has a rule network for propagation of belief leading to one or more possible malfunctions of equipment in the operating system. The propagation of belief is dynamically altered as a function of the operating conditions of the sensors themselves.

Claims

1. A method of diagnosing an operating system to malfunctions comprising the steps of:

(A) storing a rule base pertinent to the particular operating system being diagnosed and being operable to reach one or more conclusions relative to the condition of said system and being comprised of a plurality of schema, each said schema being defined by a data structure having a plurality of slots in which particular attributes of the schema are stored; and

(B) changing the contents of any predetermined slot of any predetermined other schema upon the occurrence of a predetermined event.

2. A method according to claim 1 which includes the step of:

(A) providing said plurality of schema with a plurality of rules each linking evidence to a consequent hypothesis.

3. A method according to claim 2 which includes the step of:

(A) providing each said hypothesis with an update slot which must be provided with a true value before said hypothesis will participate in said diagnosis; and

wherein

(B) said step of changing provides said true value to said update slot of a predetermined hypothesis upon the occurrence of said event.

4. A method according to claim 2 which includes the steps of:

(A) providing each of said rules with a context slot, designating a context, where the context may be true or false, and said rules will participate in said diagnosis only if their respective context is true; and

(B) said step of changing provides at least one selected rule with a context of a known true or false state.

5. A method of diagnosing an operating system subject to malfunctions, comprising the steps of:

(A) providing a plurality of sensors throughout said operating system to obtain respective signals indicative of predetermined operating parameters of said system;

(B) storing a rule base pertinent to the particular operating system being diagnosed;

(C) providing said sensor signals as inputs to said rule base to propagate belief leading to at least one particular consequent malfunction of said operating system;

(D) diagnosing by said rule base, possible malfunctions of said sensors themselves; and

(E) modifying said propagation of belief as a function of any sensor malfunction.

6. A method of diagnosing an operating system subject to malfunctions and having a plurality of sensors providing signals related to system parameters, comprising the steps of:

(A) storing a rule base pertinent to the particular operating system being diagnosed;

(B) providing said sensor signals as inputs to said rule base to diagnose malfunctions of said operating system, based upon said sensor signals said rule base being comprised of a plurality of schema, each said schema being defined by a data structure having a plurality of slots into which particular attributes of the schema are stored; and

(C) changing the contents of at least one slot in a predetermined other schema upon the occurrence of a predetermined condition of at least one sensor of said plurality.

7. A method according to claim 6 which includes the step of:

(A) providing said plurality of schema with a plurality of rules each linking evidence to a consequent hypothesis.

8. A method according to claim 7 which includes the steps of:

(A) providing each of said rules with an associated predetermined sufficiency factor numerically defining the degree of belief in the rule'"'"'s hypothesis when the rule'"'"'s evidence is true.

(B) storing said sufficiency factor of the rule'"'"'s data structure; and

wherein

(C) said step of changing effectively changes the value of another rule'"'"'s sufficiency factor upon said predetermined malfunction.

9. A method according to claim 7 which includes the steps of:

(A) providing each of said rules with an associated predetermined necessity factor numerically defining to what degree the presence of the evidence is necessary for the hypothesis to be true;

(B) storing said necessity factor in a slot of the rule'"'"'s data structure; and

wherein

(C) said step of changing effectively changes the value of another rule'"'"'s necessity factor upon said predetermined malfunction.

10. A method of on-line diagnosing of an operating system comprising the steps of:

(A) connecting a plurality of sensors to said system to obtain sensor signals relating to predetermined parameters of said system, while said system is in operation;

(B) providing indications relating to the overall operating condition of said system, based upon said sensor signals;

(C) monitoring the operating condition of said sensors themselves;

(D) modifying the value of a sensor'"'"'s input contribution to said diagnosis if the particular sensor itself is in an abnormal operating condition.

11. A method according to claim 10 which includes the step of:

(A) utilizing a rule-based expert system to provide said indications relating to the overall operating condition of said system.