System and method for identifying manufacturing data deviations

• US 7,460,917 B1
• Filed: 09/30/2005
• Issued: 12/02/2008
• Est. Priority Date: 09/30/2005
• Status: Active Grant

• Current Assignee: Rockwell Automation Technologies Incorporated (Rockwell Automation, Inc.)
• Original Assignee: Rockwell Automation Technologies Incorporated (Rockwell Automation, Inc.)
• Inventors: Reichard, Douglas J., Dorgelo, Eric G., Gordon, Kevin G., Virji, Shafin A., Bromley, Clifton H., Semkow, Marc D.
• Primary Examiner(s): DeCady; Albert
• Assistant Examiner(s): Garland; Steven R
• Application Number:
  US11/240,865
• Time in Patent Office: 1,159 Days
• Field of Search: 700/47, 700/33, 700/174, 702/182, 706/904, 706/912
• US Class Current: 700/47
• CPC Class Codes: G05B 19/05 Programmable logic controllers, e.g. simulating logic interconnections of signals according to ladder diagrams or function charts
• CPC Class Codes: G05B 2219/33002 Artificial intelligence AI, expert, knowledge, rule based system KBS
• CPC Class Codes: G05B 2219/34145 Bezier interpolation, spline
• CPC Class Codes: G05B 2219/37542 Curve fitting measured points, predict, extrapolate dimension in time

Abstract

Systems and methods of industrial control processes that employ a data matching component associated with a programming logic controller (PLC), to substitute a plurality of collected data points with a data pattern (e.g., a curve). Such data matching component can facilitate data trending analysis, wherein a running industrial process can be compared with a predetermined criteria (industrial process with optimal/desired performance). A graphical tool (e.g., an on-screen) can be provided as part of the matching component, to enable a user to interactively set deviation thresholds from a predetermined criteria (e.g., optimum performance of an industrial operation.)

Claims

1. A programmable control system, comprising:

a matching component that fits a pattern thru a plurality of data points that are gathered via a collection device associated with an industrial process;

an artificial intelligence component that facilitates curve fitting comprising at least one of an implicit classifier or an explicit classifier; and

a programmable logic controller that controls the industrial process via the pattern fitted thru the data points.

2. The programmable control system of claim 1, the matching component further comprising a Bezier curve fitting system that fits a Bezier curve thru the data points.

3. The programmable control system of claim 2, the matching component further comprising a graphical tool to interactively set deviations thresholds from a predetermined criteria.

4. The programmable control system of claim 3 further comprising a notification component that supplies a notification regarding a deviation from deviation thresholds.

5. The programmable control system of claim 4, the notification component includes an audio herald.

6. A method of identifying deviation of an industrial process comprising:

collecting data from an industrial process to program a programmable logic controller,

passing a curve thru collected data,

employing an artificial intelligence component comprising employing one of an explicit classifier or an implicit classifier to facilitate data trending analysis;

storing the curve to facilitate data trend analysis, and

performing a data trending analysis via the curve.

7. The method of claim 6 further comprising discarding the collected data.

8. The method of claim 6 further comprising assigning a threshold value deviation from a desired operation for the industrial process.

9. The method of claim 6 further comprising computing a plurality of dependencies among data collected via the curve.

10. The method of claim 6, performing the data trending analysis act further comprising setting a threshold deviation to compare the running industrial process with a predetermined optimal performance.

11. The method of claim 10 further comprising changing parameters that control the industrial process to obtain desired performance and results.

12. The method of claim 10 further comprising notifying a user upon a deviation from the threshold deviation.

13. The method of claim 12 further comprising employing an audio herald.

14. The method of claim 6 further comprising excluding a data point from the collected data for curve fitting.

15. A programmable control system, comprising:

means for fitting a Bezier curve thru a plurality of data points gathered via a collection device associated with an industrial process, said means comprising an artificial intelligence means that facilitates curve fitting comprising at least one of an implicit classifier or an explicit classifier;

one or more means controlling the industrial process via the Bezier curve.