Automatic signal processing-based learning in a process plant

• US 10,282,676 B2
• Filed: 10/06/2014
• Issued: 05/07/2019
• Est. Priority Date: 10/06/2014
• Status: Active Grant

• Current Assignee: Fisher-Rosemount Systems Incorporated (Emerson Electric Company)
• Original Assignee: Fisher-Rosemount Systems Incorporated (Emerson Electric Company)
• Inventors: Nixon, Mark J., Zornio, Peter, Wojsznis, Wilhelm K., Lucas, J. Michael, Muston, Paul R., Rotvold, Eric D., Blevins, Terrence L.
• Primary Examiner(s): Rutten, James D.
• Application Number:
  US14/507,252
• Publication Number:
  US 20160098647A1
• Time in Patent Office: 1,674 Days
• Field of Search: 700 47
• CPC Class Codes: G05B 13/0265 the criterion being a learning criterion
• CPC Class Codes: G05B 13/028 using expert systems only
• CPC Class Codes: G06N 20/00 Machine learning

Abstract

Techniques for automatically or autonomously performing signal processing-based learning in a process plant are disclosed. Generally, said techniques automatically or autonomously perform signal processing on a real-time signal that is generated based on the process plant controlling a process. Typically, the signal corresponds to a parameter value that varies over time, and the signal is processed as it is generated in real-time during on-line plant operations. Results of the signal processing may indicate characteristics of the signal, and one or more analytics functions may determine the sources of the characteristics, which may include a process element or device, a piece of equipment, and/or an asset of the process plant that is upstream, within the process, of the source of the signal. An autonomous signal processor may be integrated with or included in a process control device and/or a big data node of the process plant.

Claims

1. A system for providing big data based learning in a process plant controlling a process, the system comprising:

a process control device that is one of a controller, a field device performing a physical function to control at least the portion of the process, or an input/output (I/O) device communicatively connecting the controller and the field device;

a signal processing module that is integral with the process control device, the integrated process control device and signal processing module including at least one memory and at least one processor, and the signal processing module including;

an input to receive a signal generated by a signal source associated with the process plant as the signal source generates the signal in real-time, the signal indicative of parameter value that varies over time based on the process plant controlling the process by using at least one field device that performs a respective physical function, and the process control device being the signal source;

an output to provide, to one or more respective recipient applications, an indication of at least one characteristic of the signal; and

computer-executable instructions that are stored in the at least one memory and that, when executed by the at least one processor, cause the signal processing module to perform real-time signal processing on the signal as the process control device generates the signal in real-time to determine the at least one characteristic of the signal, the real-time signal processing including at least one of;

magnitude or amplitude analysis, power analysis, intensity analysis, phase analysis, frequency analysis, spectrum or spectral analysis, correlation analysis, convolution, smoothing, Hilbert transformation, level detection, linear signal processing, or non-linear signal processing; and

an analytics module including;

an input to receive the indication of the at least one characteristic of the signal from the signal processing module; and

an output to provide an indication of a source of the at least one characteristic of the signal to the one or more respective recipient applications, the one or more respective recipient applications including at least one of a user interface application or another application; and

wherein;

the analytics module is configured to determine that at least one member of a set of upstream elements is the source of the at least one characteristic of the signal, the set of upstream elements being elements that are associated with the process plant and that are upstream, within the process, of the signal, and the determination that the at least one member is the source of the at least one characteristic of the signal is based on respective strengths of impact of members of the set of upstream elements on the at least one characteristic of the signal; and

the source of the at least one characteristic of the signal includes at least one of a process control device, a process variable, a piece of equipment, or an asset of the process plant that is upstream of the signal source.

2. The system of claim 1, wherein:

the process control device is a first signal source, the signal generated by the first signal source is a first signal, and the signal processing module is a first signal processing module; and

the system further comprises a second signal processing module integral with a second signal source generating a second signal, the second signal processing module performing real-time signal processing on the second signal as the second signal is generated in real-time by the second signal source, the second signal source being one of;

a second process control device controlling a respective at least a portion of the process, a big data node included in a process control big data network of the process plant, a piece of equipment within the process plant, or an asset of the process plant.

3. The system of claim 2, wherein the second signal source is the big data node included in the process control big data network of the process plant, and wherein the big data node is one of:

a local big data node, a regional big data node, a centralized big data node, or another big data node.

4. The system of claim 2, wherein the system further comprises a third signal source and a third signal processing module performing real-time signal processing on a third signal generated by the third signal source as the third signal is being generated in real-time by the third signal source, the third signal source included in a first device, and the third signal processing module included in a second device coupled to the first device in which the third signal source is included.

5. The system of claim 4, wherein:

the third signal source is a third process control device controlling a respective at least a portion of the process, the third process control device being one of a third controller, a third field, or a third input/output (I/O).

6. The system of claim 2, wherein the signal generated by the second signal source is indicative of a parameter value that is indicative of a value of one of an output signal of the second process control device, a process variable, a measurement, a balance of energy, a balance of mass, or an output of an analytics module.

7. The system of claim 1, wherein:

the system further comprises a presentation module, the presentation module including;

an input to receive learned output from the analytics module; and

an output to provide the learned output to the one or more respective recipient applications; and

the learned output includes at least one of the indication of the source of the at least one characteristic of the signal, or other information learned by the analytics module.

8. The system of claim 7, wherein the presentation module is configured to determine the one or more respective recipient applications.

9. The system of claim 1, wherein the analytics module is a first analytics module, and the one or more respective recipient applications include the another application included in a second analytics module.

10. The system of claim 1, wherein the analytics module is a first analytics module, and the first analytics module determines the source of the at least one characteristic of the signal based on (i) the respective strengths of impact of the members of the set of upstream elements on the at least one characteristic of the signal, and (ii) information provided by a second analytics module.

11. The system of claim 10, wherein the second analytics module comprises a process element alignment module.

12. The system of claim 1, wherein the analytics module is integral with the process control device and the signal processing module, and the process control device is a single big data node that is communicatively connected to a process control big data network of the process plant.

13. The system of claim 1, wherein the integrated process control device and signal processing module is included in a first node corresponding to the process plant, the analytics module is included in a second node corresponding to the process plant, and the second node is communicatively connected to the first node.

14. The system of claim 13, the second node is a big data node included in a process control big data network of the process plant.

15. The system of claim 1, wherein the variations over time of the parameter value comprise one or more oscillations.

16. The system of claim 1, wherein the at least one characteristic of the signal comprises at least one of:

a dominant frequency, an n-th order frequency wherein n is an integer greater than one, a harmonic, a subharmonic, or a bifurcation of the signal.

17. The system of claim 1, wherein the signal processing module operates without supervision.

18. The system of claim 1, wherein the real-time signal processing further includes at least one of filtering or another signal processing technique.

19. A system for automatically performing big data based learning in a process plant controlling a process, the system comprising:

one or more processors;

one or more tangible, non-transitory memories;

a signal processing module comprising first computer-executable instructions that are stored on the one or more tangible, non-transitory memories and that, when executed by the one or more processors, cause the system to;

receive a signal generated by a signal source as the signal source generates the signal in real-time, the signal indicative of a value of a parameter that varies over time based on the process plant controlling the process by using at least one field device performing a respective physical function; and

perform real-time signal processing on the signal as the signal source generates the signal in real-time to determine one or more characteristics of the signal, the real-time signal processing including at least one of;

magnitude or amplitude analysis, power analysis, intensity analysis, phase analysis, frequency analysis, spectrum or spectral analysis, correlation analysis, convolution, smoothing, Hilbert transformation, level detection, linear signal processing, or non-linear signal processing; and

an analytics module comprising second computer-executable instructions that are stored on the one or more tangible, non-transitory memories and that, when executed by the one or more processors, cause the system to;

receive an indication of the one or more characteristics determined by the signal processing module;

determine one or more sources of the one or more characteristics of the signal, the one or more sources being one or more members of a set of elements that are upstream, within the process, of the signal, and each of the one or more members of the set of upstream elements having a respective impact on the one or more characteristics of the signal that is greater than a respective impact of at least one other member of the set of upstream elements that is not included in the one or more members; and

cause an indication of the one or more sources of the one or more characteristics of the signal to be provided to a recipient application, the recipient application being a user interface application or another application,

wherein the signal processing module is integral with a process control device that is one of a controller, a field device that performs a physical function to control at least a portion of the process, or an input/output (I/O) device communicatively connecting the controller and the field device, and

wherein the process control device is the signal source.

20. The system of claim 19, wherein:

the recipient application is a first recipient application;

the system further comprises a presentation module comprising third computer-executable instructions that are stored on the one or more tangible, non-transitory memories and that, when executed by the one or more processors, cause the system to receive an output of the analytics module and cause the output to be delivered to a determined set of recipient applications;

the first recipient application is included in the determined set of recipient applications; and

the indication of the one or more sources of the one or more characteristics of the signal is included in the output of the analytics module.

21. The system of claim 20, wherein the output of the analytics module further includes additional learned knowledge determined by the analytic module over various stages of learning.

22. The system of claim 19, wherein at least one of an indication of the signal, an indication of the signal source, or an indication of the one or more characteristics of the signal is provided to the recipient application.

23. The system of claim 19, wherein the integral process control device and signal processing module is a first node, and the analytics module is included in a second node communicatively coupled to the first node.

24. The system of claim 23, wherein:

the first node is a first big data provider node included in a process control big data network of the process plant; and

the second node is a second big data node included in the process control big data network.

25. The system of claim 19, wherein the analytics module is integral with the process control device and the signal processing module.

26. The system of claim 25, wherein the process control device is a big data node included in a process control big data network of the process plant.

27. The system of claim 19, wherein the value of the parameter corresponds to one of a value of an output signal of the process control device, a value of a process variable, a measurement, a value indicative of a balance of energy, a value indicative of a balance of mass, or a value of an output generated by a big data analysis function.

28. The system of claim 19, wherein the one or more characteristics of the signal comprises at least one of:

a dominant frequency, an n-th order frequency wherein n is an integer greater than one, a harmonic, a subharmonic, or a bifurcation of the signal.

29. The system of claim 19, wherein the one or more sources of the one or more characteristics of the signal include at least one of the process control device, a process variable, a piece of equipment, or an asset corresponding to the process plant and upstream of the signal source.

30. The system of claim 19, wherein the analytics module is a first analytics module, the system further includes a second analytics module, and the recipient application is included in the second analytics module.

31. The system of claim 19, wherein the system operates autonomously in real-time without any user input.

32. The system of claim 19, wherein the real-time signal processing further includes at least one of filtering or another signal processing technique.

33. A method for providing big data based-learning in a process plant controlling a process, the method comprising:

obtaining, at a signal processing module included in a signal processing-based learning system, indications of variations over time of a value of a parameter, the variations over time of the parameter value being a signal generated by a signal source, the variations over time of the parameter value generated based on controlling the process in the process plant, the signal being obtained as the signal source generates the signal in real-time, the signal source being a process control device that is one of a controller, a field device that performs a physical function to control at least a portion of the process, or an input/output (I/O) device communicatively connecting the controller in the field device, and the signal processing module being integral with the process control device;

causing, by the signal processing module of the signal processing-based learning system, one or more signal processing functions to be applied in real-time to the signal as the signal source generates the signal in real-time, the one or more signal processing functions including at least one of;

magnitude or amplitude analysis, power analysis, intensity analysis, phase analysis, frequency analysis, spectrum or spectral analysis, correlation analysis, convolution, smoothing, Hilbert transformation, level detection, linear signal processing, or non-linear signal processing;

determining, by the signal processing-based learning system, at least one characteristic of the signal based on the application of the one or more signal processing functions;

determining, by the signal processing-based learning system, at least a subset of a set of elements that are upstream, within the process, of the signal to be one or more sources of the at least one characteristic of the signal, the determination of the at least the subset based on respective strengths of impact of the set of upstream elements on the at least one characteristic of the signal; and

causing, by the signal processing-based learning system, an indication of the one or more sources of the at least one characteristic of the signal to be provided to a recipient application, the recipient application being a user interface application or another application.

34. The method of claim 33, wherein obtaining the indications of the variations over time of the parameter value comprises obtaining indications of variations over time of a value of one of:

an output signal of the process control device, a process variable, a measurement, a parameter indicative of a balance of energy, a parameter indicative of a balance of mass, or an output of an analytics function performed on big data generated by the process plant.

35. The method of claim 33, wherein causing the signal processing to be performed on the signal to determine the at least one characteristic of the signal comprises causing the signal processing to be performed on the signal to determine at least one of:

a dominant frequency, an n-th order frequency wherein n is an integer greater than one, a harmonic, a subharmonic, or a bifurcation.

36. The method of claim 33, wherein determining the at least one characteristic of the signal based on the signal processing comprises determining the at least one characteristic of the signal based on the signal processing and based on at least one of:

measurements obtained while at least a portion of the process is on-line, measurements obtained while the at least the at least the portion of the process is off-line, or data obtained from a model of the at least the portion of the process.

37. The method of claim 33, wherein determining the at least the subset of the set of upstream elements to be the one or more sources of the at least one characteristic of the signal comprises determining at least one of the process control device, a piece of equipment, a process variable, or an asset to be the one or more sources of the at least one characteristic of the signal.

38. The method of claim 33, wherein at least one of:

the method is performed over a duration of time of one second or less, or

the method is an unsupervised method.

39. The method of claim 33, wherein at least one of:

obtaining the indications of the variations over time of the parameter value comprises detecting the variations over time of the parameter value;

or

obtaining the variations over time of the signal comprises obtaining an indication of one or more oscillations included in the signal.

40. The method of claim 33, wherein:

the signal processing-based learning system includes an analytics module coupled to the signal processing module; and

determining, by the signal processing-based learning system, the one or more sources of the at least one characteristic of the signal comprises determining, by the analytics module of the signal processing-based learning system, the one or more sources of the at least one characteristic of the signal.

41. The method of claim 40, wherein the analytics module is a first analytics module, and

causing the indication of the one or more sources of the at least one characteristic of the signal to be provided to the recipient application comprises causing the indication of the one or more sources of the at least one characteristic of the signal to be provided to a second analytics module.

42. The system of claim 33, wherein the one or more signal processing functions further includes at least one of filtering or another signal processing technique.