METHODS AND APPARATUS TO PREDICT PROCESS QUALITY IN A PROCESS CONTROL SYSTEM

• US 20100318934A1
• Filed: 08/11/2009
• Published: 12/16/2010
• Est. Priority Date: 06/10/2009
• Status: Active Grant

• Current Assignee: Fisher-Rosemount Systems Incorporated (Emerson Electric Company)
• Original Assignee: Fisher-Rosemount Systems Incorporated (Emerson Electric Company)
• Inventors: Reiss, Randolf, Worek, Christopher, Muston, Paul Richard, Nixon, Mark, Blevins, Terrence Lynn, Wojsznis, Wilhelm K.
• Granted Patent: US 8,571,696 B2
• Time in Patent Office: Days
• US Class Current: 715/772
• CPC Class Codes: G05B 13/048 using a predictor
• CPC Class Codes: G05B 19/41875 characterised by quality surveillance of production
• CPC Class Codes: G05B 2219/32187 Correlation between controlling parameters for influence on quality parameters
• CPC Class Codes: G05B 2219/32194 Quality prediction
• CPC Class Codes: G05B 23/0235 based on a comparison with predetermined threshold or range, e.g. "classical methods", carried out during normal operation; threshold adaptation or choice; when or how to compare with the threshold
• CPC Class Codes: Y02P 90/02 Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]

Abstract

Example methods and apparatus to predict process quality in a process control system are disclosed. A disclosed example method includes receiving process control information relating to a process at a first time including a first value associated with a first measured variable and a second value associated with a second measured variable, determining if a variation based on the received process control information associated with the process exceeds a threshold, if the variation exceeds the threshold, calculating a first contribution value based on a contribution of the first measured variable to the variation and a second contribution value based on a contribution of the second measured variable to the variation, determining at least one corrective action based on the first contribution value, the second contribution value, the first value, or the second value, and calculating a predicted process quality based on the at least one corrective action at a time after the first time.

Claims

1. A method to predict process quality in a process control system, comprising:

receiving process control information relating to a process at a first time including a first value associated with a first measured variable and a second value associated with a second measured variable;

determining if a variation based on the received process control information associated with the process exceeds a threshold;

if the variation exceeds the threshold, calculating a first contribution value based on a contribution of the first measured variable to the variation and a second contribution value based on a contribution of the second measured variable to the variation;

determining at least one corrective action based on the first contribution value, the second contribution value, the first value, or the second value; and

calculating a predicted process quality based on the at least one corrective action at a time after the first time.

2. A method as defined in claim 1, further comprising:

displaying graphically the variation in a first graphic via a user interface;

displaying graphically the first contribution value and the second contribution value in a second graphic via the user interface, wherein the second graphic is selectable from a portion of the first graphic;

displaying the at least one corrective action via the user interface; and

displaying graphically the predicted process quality in a third graphic via the user interface, wherein the third graphic is selectable from a portion of the second graphic.

3. A method as defined in claim 2, further comprising:

displaying graphically the variation as an explained variation and an unexplained variation within the first graphic; and

displaying graphically the first contribution value and the second contribution value within the first graphic.

4. A method as defined in claim 2, further comprising:

predicting at least one of the predicted process quality or the variation based on at least one of the first value or the second value;

indicating a predicted fault if the at least one of the predicted process quality or the variation exceeds a second threshold; and

displaying the predicted fault within the second graphic and displaying the at least one corrective action within the third graphic.

5. A method as defined in claim 2, further comprising:

calculating a first unexplained variation and a first explained variation associated with the first contribution value;

calculating a second unexplained variation and a second explained variation associated with the second contribution value;

displaying the first unexplained variation and the first explained variation in a bar chart within the second graphic by at least one of superimposing the first unexplained variation on the first explained variation or showing together the first unexplained variation with the first explained variation; and

displaying the second unexplained variation and the second explained variation in the bar chart by at least one of superimposing the second unexplained variation on the second explained variation or showing the second unexplained variation together with the second explained variation.

6. A method as defined in claim 5, further comprising:

receiving a selection of at least one of the first measured variable or the second measured variable from a portion of the second graphic;

displaying graphically the value associated with the selected measured variable and corresponding past values over a time period of the process in a fourth graphic via the user interface;

displaying an average of past values associated with the selected measured variable in the fourth graphic over the time period of the process; and

displaying a standard deviation of past values associated with the selected measured variable in the fourth graphic over the time period of the process.

7. A method as defined in claim 2, further comprising:

displaying the predicted process quality in relation to at least one of a confidence range of the predicted process quality or a threshold of the predicted process quality;

displaying the predicted process quality over a time period of the process; and

applying the at least one corrective action to the process.

8. A method as defined in claim 2, further comprising:

calculating a third value associated with a calculated quality variable that is based on at least one of the first measured variable or the second measured variable;

calculating a third contribution value based on a contribution of the calculated quality variable to the variation; and

determining the at least one corrective action based on the third contribution value.

9. A method as defined in claim 8, further comprising:

determining if the first value exceeds a first threshold;

determining if the second value exceeds a second threshold;

determining if the third value exceeds a third threshold;

if at least one of the first value, the second value, or the third value exceeds the respective one of the thresholds, indicating in a chart via the user interface at least one of a process fault, a deviation of the first measured variable, a deviation of the second measured variable, a deviation of the quality variable, a process deviation, a duration of the fault, a cause of the process fault, or the variation of the process; and

displaying the first graphic upon receiving a selection of a portion of the chart.

10. A method as defined in claim 9, further comprising:

displaying the first value and the second value in a fourth graphic via the user interface that is selectable from at least one of a portion of the first graphic, a portion of the second graphic or a portion of the chart; and

displaying within the fourth graphic at least one of a timeline corresponding to a time period of the process, a bar chart indicating the first value and the second value normalized in relation to a mean and a standard deviation, or spark lines corresponding to the first variable and the second variable showing past process values for the first variable and the second variable.

11. A method as defined in claim 10, wherein the timeline includes at least one of an indication of a fault associated with the variation of the process, a current progress of the process in relation to the time period of the process, a selectable arrow that changes at least one of the values displayed in the bar chart to values corresponding to a time position of the arrow on the timeline, or a progress of the spark lines corresponding to a time position of the arrow on the timeline.

12. A method as defined in claim 10, wherein the spark lines include an indication of a deviation if past values associated with the at least one first measured variable or the second measured variable exceed a second threshold.

13. A method as defined in claim 2, further comprising:

receiving process control information relating to the process at a second time including a third value associated with the first measured variable and a fourth value associated with the second measured variable;

determining if a second variation associated with the process based on the received process control information at the second time exceeds the threshold and graphically displaying the second variation in the first graphic;

if the second variation exceeds the threshold, calculating a third contribution value based on a contribution of the first measured variable to the second variation at the second time and a fourth contribution value based on a contribution of the second measured variable to the second variation at the second time;

displaying graphically the third contribution value and the fourth contribution value in the second graphic that is selectable from a portion of the first graphic;

determining at least one corrective action based on the third contribution value or the fourth contribution value and displaying the at least one corrective action in the second graphic; and

calculating the predicted process quality based on the at least one corrective action at a time after the second time and graphically displaying the predicted process quality in the third graphic that is selectable from a portion of the second graphic.

14. A method as defined in claim 1, wherein the first measured variable and the second measured variable comprise at least one of a measured process variable or a measured quality variable.

15. A method as defined in claim 1, wherein the variation is based on at least one of a multivariate analysis, an algebraic analysis, an optimization analysis, a past batch process analysis, a statistical analysis, a regression analysis, a correlation analysis, a repeatability analysis, a reproduction analysis, or a time series analysis of at least one of the first measured variable or the second measured variable.

16. A method as defined in claim 1, wherein calculating the first contribution value and the second contribution value includes determining a relationship between the first and second contribution values and the variation based on at least one of a multivariate analysis, an algebraic analysis, an optimization analysis, a past batch process analysis, a statistical analysis, a regression analysis, a correlation analysis, a repeatability analysis, a reproduction analysis, or a time series analysis.

17. A method as defined in claim 1, wherein calculating the predicted process quality includes:

determining a relationship of an overall quality variable associated with the predicted process quality with at least one of the first measured variable or the second measured variable based on at least one of a multivariate analysis, an algebraic analysis, an optimization analysis, a past batch process analysis, a statistical analysis, a regression analysis, a correlation analysis, a repeatability analysis, a reproduction analysis, or a time series analysis;

applying the at least one corrective action to the relationship; and

calculating the predicted process quality from the relationship.

18. An apparatus to predict process quality in a process control system, comprising:

a batch data receiver to receive process control information relating to a process at a first time including a first value associated with a first measured variable and a second value associated with a second measured variable; and

a processor to;

determine if a variation based on the received process control information associated with the process exceeds a threshold;

if the variation exceeds the threshold, calculate a first contribution value based on a contribution of the first measured variable to the variation and a second contribution value based on a contribution of the second measured variable to the variation;

determine at least one corrective action based on the first contribution value, the second contribution value, the first value, or the second value; and

calculate a predicted process quality based on the at least one corrective action at a time after the first time.

19. An apparatus as defined in claim 18, further comprising a display manager to:

display graphically the variation in a first graphic via a user interface;

display graphically the first contribution value and the second contribution value in a second graphic via the user interface, wherein the second graphic is selectable from a portion of the first graphic;

display graphically the at least one corrective action via the user interface; and

display graphically the predicted process quality in a third graphic via the user interface, wherein the third graphic is selectable from a portion of the second graphic.

20. An apparatus as defined in claim 19, wherein the display manager is to:

display graphically the variation as an explained variation and an unexplained variation within the first graphic; and

display graphically the first contribution value and the second contribution value within the first graphic.

21. An apparatus as defined in claim 19, wherein the processor is to:

calculate a first unexplained variation and a first explained variation associated with the first contribution value; and

calculate a second unexplained variation and a second explained variation associated with the second contribution value.

22. An apparatus as defined in claim 21, wherein the display manager is to:

display the first unexplained variation and the first explained variation in a bar chart within the second graphic by at least one of superimposing the first unexplained variation on the first explained variation or showing together the first unexplained variation with the first explained variation; and

display the second unexplained variation and the second explained variation in the bar chart by at least one of superimposing the second unexplained variation on the second explained variation or showing together the second unexplained variation with the second explained variation.

23. An apparatus as defined in claim 21, wherein the display manager is to:

receive a selection of at least one of the first measured variable or the second measured variable from a portion of the second graphic;

display graphically the value associated with the selected measured variable and corresponding past values over a time period of the process in a fourth graphic via the user interface;

display an average of past values associated with the selected measured variable in the fourth graphic over the time period of the process; and

display a standard deviation of past values associated with the selected measured variable in the fourth graphic over the time period of the process.

24. An apparatus as defined in claim 19, wherein the display manager is to:

display graphically the predicted process quality in relation to at least one of a confidence range of the predicted process quality or a threshold of the predicted process quality; and

display graphically the predicted process quality over a time period of the process.

25. An apparatus as defined in claim 19, wherein the processor is to apply the at least one corrective action to the process.

26. An apparatus as defined in claim 19, wherein the processor is to:

calculate a third value associated with a calculated quality variable that is based on at least one of the first measured variable or the second measured variable;

calculate a third contribution value based on a contribution of the calculated quality variable to the variation;

determine the at least one corrective action based on the third contribution value; and

calculate the predicted process quality based on the at least one corrective action.

27. An apparatus as defined in claim 26, wherein the processor is to:

determine if the first value exceeds a first threshold;

determine if the second value exceeds a second threshold;

determine if the third value exceeds a third threshold; and

if at least one of the first value, the second value, or the third value exceeds the respective one of the thresholds, indicate in a chart via the user interface at least one of a process fault, a deviation of the first measured variable, a deviation of the second measured variable, a deviation of the quality variable, a process deviation, a duration of the fault, a cause of the process fault, or the variation of the process.

28. An apparatus as defined in claim 27, wherein the display manager is to:

display the first graphic upon receiving a selection of a portion of the chart;

display the first value and the second value in a fourth graphic via the user interface that is selectable from at least one of a portion of the first graphic, a portion of the second graphic, or a portion of the chart; and

display within the fourth graphic at least one of a timeline corresponding to a time period of the process, a bar chart indicating the first value and the second value normalized in relation to a mean and a standard deviation, or spark lines corresponding to the first variable and the second variable showing past process values for the first variable and the second variable.

29-33. -33. (canceled)

34. A machine-accessible medium having instructions stored thereon that, when executed, cause a machine to:

receive process control information relating to a process at a first time including a first value associated with a first measured variable and a second value associated with a second measured variable;

determine if a variation based on the received process control information associated with the process exceeds a threshold;

if the variation exceeds the threshold, calculate a first contribution value based on a contribution of the first measured variable to the variation and a second contribution value based on a contribution of the second measured variable to the variation;

determine at least one corrective action based on the first contribution value, the second contribution value, the first value, or the second value; and

calculate a predicted process quality based on the at least one corrective action at a time after the first time.

35. A machine-accessible medium as defined in claim 34, wherein the machine-accessible instructions, when executed, cause the machine to:

display graphically the variation in a first graphic via a user interface;

display graphically the first contribution value and the second contribution value in a second graphic via the user interface, wherein the second graphic is selectable from a portion of the first graphic;

display the at least one corrective action via the user interface; and

display graphically the predicted process quality in a third graphic via the user interface, wherein the third graphic is selectable from a portion of the second graphic.

36-47. -47. (canceled)