Device in a process system for detecting events

• US 6,017,143 A
• Filed: 03/28/1996
• Issued: 01/25/2000
• Est. Priority Date: 03/28/1996
• Status: Expired due to Term

• Current Assignee: Rosemount Incorporated (Emerson Electric Company)
• Original Assignee: Rosemount Incorporated (Emerson Electric Company)
• Inventors: Warrior, Jogesh, Eryurek, Evren
• Primary Examiner(s): Grant, William
• Assistant Examiner(s): Rao, Sheela S
• Application Number:
  US08/623,569
• Time in Patent Office: 1,398 Days
• Field of Search: 364/148, 364/164, 364/571, 364/140.04, 395/3, 395/183.02, 702/34, 702/45, 702/57, 702/116, 702/183, 702/185, 706/10, 706/23, 706/46, 706/52
• US Class Current: 700/51
• CPC Class Codes: G05B 13/0275 using fuzzy logic only
• CPC Class Codes: G05B 21/02 electric
• CPC Class Codes: G05B 23/0221 Preprocessing measurements, e.g. data collection rate adjustment; Standardization of measurements; Time series or signal analysis, e.g. frequency analysis or wavelets; Trustworthiness of measurements; Indexes therefor; Measurements using easily measured parameters to estimate parameters difficult to measure; Virtual sensor creation; De-noising; Sensor fusion; Unconventional preprocessing inherently present in specific fault detection methods like PCA-based methods
• CPC Class Codes: G05B 23/0278 Qualitative, e.g. if-then rules; Fuzzy logic; Lookup tables; Symptomatic search; FMEA

Abstract

A process device couples to a process control loop. The process device receives process signals. A memory in the process device contains a nominal parameter value and a rule. Computing circuitry calculates a statistical parameter of the process signal and operates on the statistical parameter and the stored nominal value based upon the stored rule and responsively provides an event output based upon the operation. Output circuitry provides an output in response to the event output.

Claims

1. A process device coupled to a process control loop, comprising:

a process signal input providing a process signal related to a process;

memory containing a trained parameter representative of statistical function of an observed process signal and a rule;

computing circuitry calculating a current parameter representative of a statistical function of the process signal and operating on the current parameter and the trained parameter based upon the rule to thereby recognize a pattern in the process signal associated with a failure event and responsively providing a failure event output in response to an event in the process detected based upon the operation; and

output circuitry outputting the failure event output.

2. The apparatus of claim 1 wherein the memory further contains a sensitivity parameter and the failure event output is formed or based upon the sensitivity parameter.

3. The apparatus of claim 1 wherein the process control loop is selected from the group consisting of two wire process control loops, three wire process control loops and four wire process control loops.

4. The apparatus of claim 1 wherein the device is completely powered with power received from the process control loop.

5. The apparatus of claim 1 wherein the statistical function is selected from the group consisting of standard deviation, mean, sample variance, root-mean-square (RMS), range, and rate of change.

6. The apparatus of claim 1 wherein the computing circuitry provides the failure event output based upon a plurality of statistical functions.

7. The apparatus of claim 1 wherein the rule is selected from the group consisting of spike, drift, bias, noise and stuck.

8. The apparatus of claim 1 wherein the computing circuitry comprises fuzzy logic and the rule employs a membership function.

9. The apparatus of claim 8 including a fuzzy membership function stored in the memory and wherein computation circuitry applies the membership function to the trained parameter prior to operation of the rule.

10. The apparatus of claim 1 wherein the process signal input comprises a sensor input channel and the process signal comprises a process variable.

11. The apparatus of claim 10 wherein the sensor input channel includes:

a sensor providing a sensor output related to the process; and

an analog to digital converter which converts the sensor output into a digitized process variable.

12. The apparatus of claim 11 wherein the sensor is selected from the group consisting of pressure, temperature, pH, flow, turbidity, level sensors, position, conductivity, motor current, motor back emf and vibration.

13. The apparatus of claim 1 wherein the process signal input comprises a control channel and the process signal comprises a control signal.

14. The apparatus of claim 13 wherein the control channel includes:

a control element for controlling the process; and

an actuator for actuating the control element.

15. The apparatus of claim 1 wherein the process signal input comprises input circuitry coupled to the control loop to receive the process signal from the process control loop.

16. The apparatus of claim 1 wherein the output circuitry couples to the process control loop and transmits the failure event output on the process control loop.

17. The apparatus of claim 1 wherein the output circuitry comprises user output circuitry for outputting the failure event output to a user.

18. The apparatus of claim 1 wherein process signal input comprises a diagnostic channel and the process signal comprises a diagnostic signal related to a control element used to control the process.

19. The apparatus of claim 18 wherein the diagnostic signal is selected from the group consisting of valve stem position, force, and pressure.

20. The apparatus of claim 14 wherein the actuator comprises a motor and the control element comprises a pump operated by the motor.

21. The apparatus of claim 1 including an inference engine coupled to the failure event output for performing diagnostics on the process control loop.

22. The apparatus of claim 2 wherein the computing circuitry monitors the current parameter during normal operation of the process control loop and thereby generates the trained parameter.

23. The process control device of claim 1 wherein the trained parameter is related to the process when the process has settled.

24. A transmitter in a process control loop, comprising:

a sensor sensing a process variable;

an analog to digital converter coupled to the sensor having a digitized process variable output;

a memory containing a trained parameter representative of statistical function of an observed process variable and a rule;

computation circuitry coupled to the memory and the analog to digital converter, the computation circuitry calculating a current parameter representative of a statistical parameter of the digitized processed variable, operating on the current parameter based upon the rule and the trained parameter to thereby recognize a pattern in the process signal associated with a failure event and responsively providing a process failure event output; and

output circuitry coupled to the process control loop outputting the process failure event output onto the loop.

25. The transmitter of claim 24 wherein the memory further contains a sensitivity parameter and the failure event output is further based upon the sensitivity parameter.

26. The transmitter of claim 24 wherein the device is completely powered with power received from the process control loop.

27. The transmitter of claim 24 wherein the rule is selected from the group consisting of signal spike, signal drift, signal bias, signal noise and signal stuck.

28. The transmitter of claim 24 wherein the sensor is selected from the group consisting of pressure, pH, flow, turbidity and level sensors.

29. The transmitter of claim 24 including compensation circuitry coupled to the analog to digital converter for compensating the digitized process variable.

30. The transmitter of claim 25 wherein the computing circuitry monitors the current parameter during normal operation of the process control loop and thereby generates the trained parameter.

31. The transmitter of claim 24 wherein the computation circuitry calculates standard deviation and range of the digitized process variable and responsively provides a noise detected output if a current standard deviation is greater than a trained standard deviation times a sensitivity parameter (β

) and if a current range is greater than a trained range.

32. A method performed in a process device for detecting a process failure event in a process control system, comprising:

obtaining a process signal related to a process variable;

retrieving a rule from a memory;

retrieving a trained parameter representative of statistical function of an observed process variable from the memory;

calculating a current parameter representative of a statistical function of the process signal;

comparing current parameter to the trained parameter in accordance with a relationship defined by the rule to thereby recognize a pattern in the process signal associated with a failure event; and

responsively providing a process failure event output based upon the step of comparing.

33. The method of claim 32 including retrieving a sensitivity parameter and where the step of comparing is further based upon the sensitivity parameter.

34. The method of claim 32 wherein the process signal comprises a process variable.

35. The method of claim 32 wherein the process signal comprises a control signal.

36. The method of claim 32 wherein the process signal comprises a diagnostic signal.

37. The method of claim 33 including calculating a current parameter of the process variable during normal operation and storing the current parameter in memory as the trained parameter.

38. The method of claim 32 wherein the rules are selected from the group consisting of spike, drift, bias, noise and stuck.

39. The method of claim 32 wherein the function parameter is selected from the group consisting of standard deviation, mean, sample variance, root-mean-square (RMS), range, and rate of change.

40. The method of claim 32 wherein the step of comparing includes performing a fuzzy logic operation.

41. The method of claim 32 including calculating standard deviation and range of the process signal and responsively providing a noise detected output if a current standard deviation is greater than a trained standard deviation times a noise sensitivity parameter (β

) and if a current range is greater than a trained range.

42. A transmitter coupled to a two wire process control loop which completely powers the transmitter, comprising:

a process signal input providing a process signal related to a process comprising a sensor providing a sensor output related to the process, and an analog to digital converter which converts the sensor output into a digitized process variable;

memory containing a trained parameter representative of statistical function of an observed process signal and a rule;

computing circuitry providing a control signal output as a function of the digitized process variable and calculating a current parameter representative of a statistical function of the digitized process variable and operating on the current parameter and the trained parameter based upon the rule to thereby recognize a pattern in the process signal associated with a failure event and providing a failure event output in response to a failure event in the process detected based upon the operation; and

output circuitry outputting the failure event output on the two wire loop.

43. The apparatus of claim 42 wherein the memory further contains a sensitivity parameter and a failure event output is further related to the sensitivity parameter.

44. The apparatus of claim 42 wherein the process control loop is selected from the group consisting of two wire process control loops, three wire process control loops and four wire process control loops.

45. The apparatus of claim 42 wherein the statistical function is selected from the group consisting of standard deviation, mean, sample variance, root-mean-square (RMS), range, and rate of change.

46. The apparatus of claim 42 wherein the computing circuitry provides the event output based upon a plurality of statistical functions.

47. The apparatus of claim 42 wherein the rule is selected from the group consisting of spike, drift, bias, noise and stuck.

48. The apparatus of claim 42 wherein the computing circuitry comprises fuzzy logic and the rule employs a membership function.

49. The apparatus of claim 48 including a fuzzy membership function stored in the memory and wherein computation circuitry applies the membership function to the trained parameter prior to operation of the rule.

50. The apparatus of claim 42 wherein the sensor is selected from the group consisting of pressure, temperature, pH, flow, turbidity, level sensors, position, conductivity, motor current, motor back emf and vibration.

51. The apparatus of claim 42 including an inference engine coupled to the failure event output for performing diagnostics on the process control loop.

52. The apparatus of claim 43 wherein the computing circuitry monitors the current parameter during normal operation of the process control loop and thereby generates the trained parameter.

53. A process device coupled to a process control loop, comprising:

a process signal input for providing a process signal related to a process, the process signal input comprising a control signal for controlling the process;

memory containing a trained parameter representative of statistical function of an observed process signal and a rule;

computing circuitry calculating a current parameter representative of a statistical function of the digitized process variable and operating on the current parameter and the trained parameter based upon the rule to thereby recognize a pattern in the process signal associated with a failure event and providing a failure event output in response to a failure event in the process detected based upon the operation; and

output circuitry outputting the failure event output.

54. The apparatus of claim 53 wherein the memory further contains a sensitivity parameter and the failure event output is further based upon the sensitivity parameter.

55. The apparatus of claim 53 wherein the process control loop is selected from the group consisting of two wire process control loops, three wire process control loops and four wire process control loops.

56. The apparatus of claim 53 wherein the device is completely powered with power received from the process control loop.

57. The apparatus of claim 53 wherein the statistical function is selected from the group consisting of standard deviation, mean, sample variance, root-mean-square (RMS), range, and rate of change.

58. The apparatus of claim 53 wherein the computing circuitry provides the failure event output based upon a plurality of statistical functions.

59. The apparatus of claim 53 wherein the rule is selected from the group consisting of spike, drift, bias, noise and stuck.

60. The apparatus of claim 53 wherein the computing circuitry comprises fuzzy logic and the rule employs a membership function.

61. The apparatus of claim 60 including a fuzzy membership function stored in the memory and wherein computation circuitry applies the membership function to the trained parameter prior to operation of the rule.

62. The apparatus of claim 53 including input circuitry coupled to the control loop to receive the process signal from the process control loop.

63. The apparatus of claim 53 wherein the control signal is provided to an actuator comprising a motor thereby actuating a pump.

64. The apparatus of claim 53 including an inference engine coupled to the event output for performing diagnostics on the process control loop.

65. The apparatus of claim 54 wherein the computing circuitry monitors the current parameter during normal operation of the process control loop and thereby generates the trained parameter.