Process monitoring technique and related actions

• US 8,032,328 B2
• Filed: 04/13/2007
• Issued: 10/04/2011
• Est. Priority Date: 04/14/2006
• Status: Active Grant

• Current Assignee: Dow Global Technologies LLC (Dow, Inc.)
• Original Assignee: Dow Global Technologies LLC (Dow, Inc.)
• Inventors: Samples, Paul K., Parrish, John R., McNeil, Thomas J., Rath, Debbie D.
• Primary Examiner(s): Bhat; Aditya
• Application Number:
  US12/226,291
• Publication Number:
  US 20090177439A1
• Time in Patent Office: 1,635 Days
• Field of Search: 702/179
• US Class Current: 702/179
• CPC Class Codes: G05B 23/024 Quantitative history assessment, e.g. mathematical relationships between available data; Functions therefor; Principal component analysis [PCA]; Partial least square [PLS]; Statistical classifiers, e.g. Bayesian networks, linear regression or correlation analysis; Neural networks
• CPC Class Codes: G05B 23/0272 Presentation of monitored results, e.g. selection of status reports to be displayed; Filtering information to the user

Abstract

The invention provides a novel method of monitoring a process. The method also has the ability to take predetermined actions based on the monitored data. These actions avoid or mitigate process abnormalities or upsets that might impact product quality, production, and/or process efficiencies. The method includes the steps of: obtaining at least one input process variable; determining a comparative process value based on the at least one input process variable using a first method having a first time-based weighting function; determining an expected process value based on the at least one input process variable using a second method having a second time-based weighting function; determining a first deviation value based on the at least one input process variable or historical data; calculating a limit range having a maximum limit and a minimum limit using the expected process value and the first deviation value; and comparing the comparative process value to the limit range.

Claims

1. A periodic or intermittent chemical process monitoring method comprising

a) obtaining at least one input process variable;

b) determining a comparative process value based on at least one input process variable using a first method having a first time-based weighting function;

c) determining an expected process value based on the at least one input process variable using a second method having a second time-based weighting function;

d) determining a first deviation value based on the at least one input process variable or historical data;

e) calculating a limit range having a maximum limit and a minimum limit using the expected process value and the first deviation value; and

f) comparing the comparative process value to the limit range,

wherein the at least one input process variable is temperature, pressure, voltage, flow, weight, switch position, current, power, height, frequency, model prediction error, concentration or physical product property measurement; and

wherein at least one of steps a), b), c), d), e) or f) is accomplished using a computer system.

2. The method of claim 1, wherein the first deviation value is based on the at least one input process variable using a third method having a third time-based weighting function.

3. The method of claim 2, wherein the first deviation value determines a high deviation value and a low deviation value.

4. The method of claim 2, further comprising a second deviation value based on the at least one input process variable or historical data.

5. The method of claim 4, wherein the second deviation value is based on the at least one input process variable using a fourth method having a fourth time-based weighting function.

6. The method of claim 5, wherein the first deviation value determines a high deviation value and the second deviation value determines a low deviation value.

7. The method of claim 5, wherein the fourth method uses a measurement value, moving average, recursive moving average, moving standard deviation, recursive moving standard deviation, moving variance, recursive moving variance, predictive modeling, or frequency analysis of the at the least one input process variable.

8. The method of claim 2, wherein at least one of the methods is different from the remaining methods.

9. The method of claim 2, wherein at least one of the time-based weighting functions produces a value different from the remaining time-based weighting functions.

10. The method of claim 2, wherein the third method uses a constant, measurement value, moving average, recursive moving average, moving standard deviation, recursive moving standard deviation, moving variance, recursive moving variance, predictive modeling, or frequency analysis of the at the least one input process variable.

11. The method of claim 1, wherein the at least one input process variable is measured, calculated or conditioned.

12. The method of claim 11, wherein the conditioned process variable is an average, variation or determined using a predictive model calculation.

13. The method of claim 1, wherein the first or second method uses a moving average, a recursive moving average, a moving standard deviation, a recursive moving standard deviation, a moving variance, a recursive moving variance, or frequency analysis of the at least one input process variable.

14. The method claim 1, wherein the at least one input process variable is a calculated intermediate value.

15. The method of claim 14, wherein the calculated intermediate value comprises a sum, difference, product, quotient, ratio or estimation of process variables.

16. The method of claim 1, further comprising taking a predetermined action if the comparative process value is outside the limit range.

17. The method of claim 16, further comprising determining if at least one process indicator is provided.

18. The method of claim 17, wherein the at least one process indicator comprises process operating modes, known equipment malfunctions, or alarm conditions.

19. The method of claim 18, further comprising providing a predetermined action based on the process indicator.