Method for analyzing and diagnosing large scale process automation control systems

• US 8,489,209 B2
• Filed: 01/24/2012
• Issued: 07/16/2013
• Est. Priority Date: 01/24/2011
• Status: Active Grant

• Current Assignee: ABB Incorporated (ABB Limited)
• Original Assignee: ABB Incorporated (ABB Limited)
• Inventors: Starr, Kevin Dale, Mast, Timothy Andrew, Garverick, Robert Trent
• Primary Examiner(s): HARTMAN JR, RONALD D
• Application Number:
  US13/357,413
• Publication Number:
  US 20130030555A1
• Time in Patent Office: 539 Days
• Field of Search: 700/17, 700/26, 700/28, 700/32, 700/52, 702/179, 702/183, 702/189
• US Class Current: 700/52
• CPC Class Codes: G05B 23/0216 Human interface functionality, e.g. monitoring system providing help to the user in the selection of tests or in its configuration

Abstract

A method is provided for analyzing and diagnosing a large scale process automation control system having a plurality of control loops. Assessments for pre-defined key performance indicators (KPIs) are automatically generated for control, process and signal sections of each control loop. The automatically generated assessments of the pre-defined KPIs may be displayed in a graphical user interface (GUI) of a computer. A user may change the automatically generated assessments of the pre-defined KPIs. Different data views may also be displayed in the GUI. The data views include time series trends for measured process variable, controller output, controller set point and error, as well as controller parameter clustering views in two-dimensional and three-dimensional plots.

Claims

1. A method performed by a computer for analyzing one or more of a plurality of control loops of a control system operable to control a process, the control loops each having a controller, a final control element and a measurement device for providing the controller with a measurement of a process variable being controlled by the controller, the method comprising:

receiving, from a user, a selection of one of the control loops for analysis;

generating an assessment for a pre-defined key performance indicator (KPI) for each of first, second and third sections of the selected control loop, the first section relating to the controller of the selected control loop, the second section relating to the process and the final control element of the selected control loop and the third section relating to the measurement of the process variable of the selected control loop;

displaying, in a single screen of a graphical user interface (GUI) of the computer, at least two different types of control data of the selected control loop, the types of control data being selected from the group consisting of the measurement of the process variable, output of the controller, set point of the controller, proportional tuning parameter of the controller, integral tuning parameter of the controller and error, which is the difference between the set point of the controller and the measurement of the process variable;

displaying, in the single screen of the GUI of the computer, the generated assessments of the pre-defined KPIs for the selected control loop;

receiving, from a user, a different assessment for one of the KPIs for the selected control loop; and

displaying, in the single screen of the GUI of the computer, the received different assessment of the one of the pre-defined KPIs for the selected control loop in lieu of the generated assessment for the one of the pre-defined KPIs for the selected control loop.

2. The method of claim 1, wherein the control data is displayed in the form of time series trends.

3. The method of claim 2, wherein a first one of the time series trends is for the set point of the controller and a second one of the time series trends is for the output of the controller.

4. The method of claim 2, wherein a third one of the time series trends is for the measurement of the process variable.

5. The method of claim 1, wherein each of the assessments of the pre-defined KPIs comprises a state indicating a problem or no problem.

6. The method of claim 5, wherein the assessment of each pre-defined KPI having a state indicating a problem further comprises a value indicating a severity of the problem.

7. The method of claim 5, wherein the pre-defined KPI for the first section is selected from the group consisting of over-control, under control, excessive manual control, oscillatory set point, significant deadband in the controller, non-zero control error, excessive movement of the final control element, slow execution of the controller and unknown control problem.

8. The method of claim 7, wherein the pre-defined KPI for the second section is selected from the group consisting of saturated output of the controller, operation of the final control element mostly in outer portions of its operating range, cyclic behavior of the measurement of the process variable and the output of the controller, a calibration error in the measurement device or a leakage in the final control element, a non-periodic disturbance in the process, externally triggered periodic variability in the measurement of the process variable and unknown process problem.

9. The method of claim 8, wherein the pre-defined KPI for the third section is selected from the group consisting of the measurement of the process variable being quantized, excessive noise in the measurement of the process variable, outliers in the measurement of the process variable and/or output of the controller, unnatural step changes in the measurement of the process variable, data compression, over-smoothness of the measurement of the process variable, inappropriate sampling rate for the measurement of the process variable, no measurement of the process variable, saturation in the measurement of the process variable and unknown measurement problem.

10. The method of claim 1, comprising generating assessments for a plurality of KPIs for each of first, second and third sections of the selected control loop.

11. The method of claim 1, wherein the control data is displayed in the form of a two dimensional plot of two different data types against each other.

12. The method of claim 11, wherein a first one of the data types is the measurement of the process variable and a second one of the data types is the output of the controller.

13. The method of claim 1, wherein the control data is displayed in the form of a three dimensional plot of three different data types against each other.

14. The method of claim 13, wherein a first one of the data types is the measurement of the process variable, a second one of the data types is the output of the controller and a third one of the data types is the set point of the controller.

15. The method of claim 1, further comprising performing numerical methods on one of the types of control data and displaying the results thereof in a table in the single screen of the GUI of the computer, the numerical methods comprising standard deviation, mean, average, maximum and minimum.