Apparatus and method for analyzing chemical system data

• US 6,591,166 B1
• Filed: 08/31/1999
• Issued: 07/08/2003
• Est. Priority Date: 01/18/1996
• Status: Expired due to Term

• Current Assignee: Electric Power Research Institute Incorporated
• Original Assignee: Electric Power Research Institute Incorporated
• Inventors: Millett, Peter J., Brobst, Gary E., Burns, Gary D.
• Primary Examiner(s): Warden, Jill
• Assistant Examiner(s): QUAN, ELIZABETH S
• Application Number:
  US09/387,230
• Time in Patent Office: 1,407 Days
• Field of Search: 700/266, 422/62, 422/67, 702/22, 702/30, 702/32
• US Class Current: 700/266
• CPC Class Codes: G01N 33/18 Water
• CPC Class Codes: G06F 18/24765 Rule-based classification
• CPC Class Codes: G06F 2219/00 Indexing scheme relating to application aspects of data processing equipment or methods

Abstract

A power plant water chemistry analysis apparatus and method relies upon water chemistry sensors to obtain power plant water chemistry data characterizing the chemical activity of a power plant water system. An analytical model processor is used to generate model predictions for the power plant water system. A statistical data fitting processor selects screened data from the power plant water chemistry data that corresponds to the model predictions. The screened data is processed by an artificial intelligence processor to derive power plant water chemistry diagnostic information. The artificial intelligence processor includes an expert system, rule base, plant water chemistry system simulator, and pattern recognition module.

Claims

1. A physical system analysis apparatus, comprising:

sensors to obtain physical system data characterizing the physical activity of a physical system;

an analytical model processor to generate model predictions for said physical system independent of said physical system data obtained from said sensors; and

an artificial intelligence processor to process said physical system data and said model predictions, which in combination constitutes over specified physical system data with more known physical system data than unknown model predictions, said artificial intelligence processor generating diagnostic information from said over specified system data.

2. The apparatus of claim 1 wherein said artificial intelligence processor includes a physical system simulator to process said over specified system data.

3. The apparatus of claim 1 further comprising:

a statistical data fitting processor to select screened data from said over specified system data.

4. The apparatus of claim 3 wherein said statistical data fitting processor determines the statistical best fit of said physical system data that satisfy the constraints given by said model predictions.

5. The apparatus of claim 1 wherein said artificial intelligence processor includes a rule base to process said over specified system data.

6. The apparatus of claim 1 wherein said artificial intelligence processor further includes a pattern recognition module to process said over specified system data.

7. A computer readable memory to direct a computer to function in a specified manner comprising:

physical data characterizing the physical activity of a physical system;

model predictions for said physical system independent of said physical system data obtained from said sensors; and

executable instructions stored in said memory, said executable instructions including instructions to process said physical data and said model predictions, which in combination constitutes over specified system data with more known physical data than unknown model predictions, to generate diagnostic information from said over specified system data.

8. The apparatus of claim 7 wherein said executable instructions include instructions defining a physical system simulator to process said over specified system data.

9. The apparatus of claim 7 wherein said executable instructions include instructions defining a statistical data fitting processor to select screened data from said over specified system data.

10. The apparatus of claim 7 wherein said executable instructions include instructions defining a rule base to process said over specified system data.

11. The apparatus of claim 7 wherein said executable instructions include instructions defining a pattern recognition module to process said over specified system data.

12. A method of analyzing physical data, said method comprising the steps of:

accumulating physical data characterizing the physical activity of a physical system;

generating model predications for said physical system independent of said physical system data obtained from said sensors; and

deriving diagnostic information from said physical data and said model predictions, which in combination constitute over specified system data with more known physical data than unknown model predictions.

13. The method of claim 12 further comprising the step of selecting screened data from said over specified system data.

14. The method of claim 12 wherein said deriving step include the step of processing said over specified system data with a rule base.

15. The method of claim 12 wherein said deriving step includes the step of processing said over specified system data with a physical system simulator.

16. The method of claim 12 wherein said deriving step includes the step of processing said over specified system data with a pattern recognition module.