Performance monitoring of advanced process control systems

• US 8,712,560 B2
• Filed: 12/14/2010
• Issued: 04/29/2014
• Est. Priority Date: 12/08/2010
• Status: Active Grant

• Current Assignee: L'Air Liquide Societe Anonyme Pour L'Etude et L'Exploitation Des Procedes Georges Claude (L'Air Liquide SA)
• Original Assignee: L'Air Liquide Societe Anonyme Pour L'Etude et L'Exploitation Des Procedes Georges Claude (L'Air Liquide SA)
• Inventors: Prabhu, Amogh Vishwanath, Carlos, Geraldine
• Primary Examiner(s): Padmanabhan, Kavita
• Assistant Examiner(s): CUMMINS, PATRICK D
• Application Number:
  US12/967,742
• Publication Number:
  US 20120150335A1
• Time in Patent Office: 1,232 Days
• Field of Search: 700/110
• US Class Current: 700/79
• 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: G05B 23/0294 Optimizing process, e.g. process efficiency, product quality

Abstract

Techniques are described for monitoring the performance of advanced process control (APC) systems used to control complex industrial processes. By correlating an alarm status of the APC system with an alarm status associated with the service factor (or service factor thresholds), a plant operator can understand how well the APC system performs over time, as well as identify when operator intervention may be required (or alternatively, may be the source of poor plant performance). Accordingly, embodiments of the invention can significantly impact both the reliability and efficiency of complex industrial process, e.g., the production of industrial gases using air separation units (ASUs) and steam methane reformer (SMR) furnaces.

Claims

1. A computer-implemented method for monitoring the performance of an advanced process control (APC) system having a plurality of system variables used to control an industrial process, the method comprising:

determining a count of performance metrics associated with the industrial process in an alarm state;

determining, based on the count, whether the APC system is in an alarm state;

determining a service factor for the APC system, wherein the service factor is determined as a ratio of a number of the system variables then currently being manipulated to control the performance metrics to a total number of the system variables of the industrial process available to be manipulated; and

correlating the alarm state of the APC system and an alarm state of the service factor in order to generate a recommendation for controlling the industrial process using the APC system.

2. The computer-implemented method of claim 1, wherein the service factor has an associated threshold for determining whether the service factor is in an alarm state.

3. The computer-implemented method of claim 2, wherein the threshold is determined relative to a collection of historical values of the service factor.

4. The computer-implemented method of claim 1, wherein at least a first performance metric has an associated threshold for determining whether the first performance metric in an alarm state.

5. The computer-implemented method of claim 4, wherein the threshold is determined relative to a collection of historical values of the performance metric.

6. The computer-implemented method of claim 1, wherein correlating the alarm state of the APC system and the service factor alarm state comprises, determining that the APC system is in an alarm state and that service factor is not in an alarm state, and wherein the generated recommendation is to lower the service factor.

7. The computer-implemented method of claim 1, wherein correlating the alarm state of the APC system and the service factor alarm state comprises, determining that the APC system is not in an alarm state and that service factor is in an alarm state.

8. The computer-implemented method of claim 1, wherein correlating the alarm state of the APC system and the service factor alarm state comprises, determining that neither the APC system nor the service factor is in an alarm state.

9. The computer-implemented method of claim 1, wherein correlating the alarm state of the APC system and the service factor alarm state comprises, determining that both the APC system and the service factor are in an alarm state, and wherein the generated recommendation is for operator intervention to restore the performance metrics to a desired set of operating ranges.

10. A non-transitory computer-readable storage medium containing a program, which, when executed on a processor, performs an operation for monitoring the performance of an advanced process control (APC) system having a plurality of system variables used to control an industrial process, the operation comprising:

determining a count of performance metrics associated with the industrial process in an alarm state;

determining, based on the count, whether the APC system is in an alarm state;

determining a service factor for the APC system, wherein the service factor is determined as a ratio of a number of the system variables then currently being manipulated to control the performance metrics to a total number of the system variables of the industrial process available to be manipulated; and

correlating the alarm state of the APC system and an alarm state of the service factor in order to generate a recommendation for controlling the industrial process using the APC system.

11. The computer-readable storage medium of claim 10, wherein the service factor has an associated threshold for determining whether the service factor is in an alarm state.

12. The computer-readable storage medium of claim 11, wherein the threshold is determined relative to a collection of historical values of the service factor.

13. The computer-readable storage medium of claim 10, wherein at least a first performance metric has an associated threshold for determining whether the first performance metric in an alarm state.

14. The computer-readable storage medium of claim 13, wherein the threshold is determined relative to a collection of historical values of the performance metric.

15. A system, comprising:

a processor; and

a memory storing a monitoring application, which, when executed by the processor, performs an operation for monitoring the performance of an advanced process control (APC) system having a plurality of system variables used to control an industrial process, the operation comprising;

determining a count of performance metrics associated with the industrial process in an alarm state,

determining, based on the count, whether the APC system is in an alarm state,

determining a service factor for the APC system, wherein the service factor is determined as a ratio of a number of the system variables then currently being manipulated to control the performance metrics to a total number of the system variables of the industrial process available to be manipulated, and

correlating the alarm state of the APC system and an alarm state of the service factor in order to generate a recommendation for controlling the industrial process using the APC system.

16. The system of claim 15, wherein the service factor has an associated threshold for determining whether the service factor is in an alarm state.

17. The system of claim 16, wherein the threshold is determined relative to a collection of historical values of the service factor.

18. The system of claim 15, wherein at least a first performance metric has an associated threshold for determining whether the first performance metric in an alarm state.

19. The system of claim 18, wherein the threshold is determined relative to a collection of historical values of the performance metric.