Anticipatory high frequency noise compensation in a distributed process control system

• US 20040148038A1
• Filed: 01/28/2003
• Published: 07/29/2004
• Est. Priority Date: 01/28/2003
• Status: Active Grant

• Current Assignee: Rosemount Analytical, Inc. (Emerson Electric Company)
• Original Assignee: Rosemount Analytical, Inc. (Emerson Electric Company)
• Inventors: Keyes, Marion A.
• Granted Patent: US 7,103,426 B2
• Time in Patent Office: Days
• US Class Current: 700/73
• CPC Class Codes: G05B 17/02 electric
• CPC Class Codes: G05B 21/02 electric

Abstract

A control system includes a measurement device capable of gathering high frequency process parameter data, a modeling routine which uses the high frequency parameter data to develop a model of the high frequency noise and a compensation routine that employs the developed model to adjust a control signal to thereby compensate for the high frequency noise present within the process parameter. The measurement device may measure a process variable at a relatively high frequency and send a subset of the measured data (e.g., the low frequency data) to a standard controller that generates a control signal to control the measured process parameter in any known manner. The modeling routine analyzes the high frequency data and develops a mathematical model of the high frequency noise within the process variable. The resulting mathematical model may then be used by the compensation routine to alter the standard control signal produced by the controller before that control signal is delivered to or used by the device being controlled.

Claims

1. A control system for use in a distributed process control network, the control system comprising:

a controller that generates a controls signal based on measurements of at least one process variable, the process variable measurements being made at a first frequency;

a controlled device communicatively connected to the controller;

a measurement device communicatively connected to the controller, the measurement device adapted to collect measurements of the process variable at a second frequency greater than the first frequency and to provide measurements of the process variable at the first frequency to the controller;

a modeling routine stored in a memory and adapted to be executed on a processor, the modeling routine adapted to use the measurements of the process variables made at the second frequency to derive a noise model of the high frequency noise present within the process variable; and

a signal combiner communicatively connected to the controller and to the modeling routine that generates an adjusted control signal from the control signal and the noise model, the signal combiner being further communicatively connected to the controlled device to provide the adjusted control signal to the controlled device to reduce the high frequency noise present within the process variable.

2. The control system of claim 1, wherein the controlled device is a smart actuator.

3. The control system of claim 1, wherein the modeling routine is adapted to derive a noise model that includes one or more of a time variant linear component, a time variant cyclical component and a time variant white noise component.

4. The control system of claim 1, wherein the modeling routine is adapted to use a Fourier analysis to determine an estimate of the high frequency noise.

5. The control system of claim 1, wherein the memory storing the noise modeling routine is in a device other than the controller.

6. The control system of claim 1, wherein the memory storing the noise modeling routine is located within the controlled device.

7. The control system of claim 1, wherein the memory storing the noise modeling routine is located within the measurement device.

8. The control system of claim 1, wherein the noise modeling routine includes a transfer function routine for determining a transfer function of the high frequency noise within the process variable, and an inverter routine adapted to invert the transfer function.

9. The control system of claim 8, wherein the inverter routine is located within the controlled device.

10. The control system of claim 8, wherein the inverter routine is located within a device other than the controller.

11. The control system of claim 8, wherein the inverter routine is integrated within the controller.

12. The control system of claim 11, wherein the inverter routine inverts the noise model transfer function and the signal combiner multiples the resulting inverted noise transfer function by the control signal.

13. The control system of claim 1, wherein the controller is communicatively connected to a network and is adapted to transmit the control signal to a controlled device communicatively connected to the network through a subnet.

14. The control system of claim 13, wherein the measuring device is communicatively connected to the subnet.

15. The control system of claim 14, wherein the memory storing the noise modeling routine is integrated into the measurement device, and the measurement device is communicatively connected to the controlled device through the subnet.

16. The control system of claim 15, wherein the signal combiner is integrated into the controlled device.

17. The control system of claim 1, wherein at least one of the controller, the controlled device and the measurement device includes a pseudo-random sequence generator adapted to generate pseudo-random sequences and wherein the signal combiner uses the pseudo-random sequences to generate the adjusted control signal.

18. The control system of claim 1, further including a data compression routine adapted to compress data.

19. The control system of claim 1, wherein the signal combiner continuously determines the adjusted control signal.

20. The control system of claim 1, wherein the signal combiner is adapted to generate the adjusted control signal so as to minimize residual cyclic error within a process variable.

21. The control system of claim 1, further including a source of a synchronized time source located in at least two of the controller, the controlled device and the measurement device.

22. A method of reducing high frequency process variations within a process control system, the method comprising the steps of:

generating a control signal in response to at least one process variable received at a first rate in a first device;

collecting measurement, away from the first device, of at least one process variable at a second rate greater than the first rate;

determining, away from the first device, a noise model based on the measurements of the at least one process variable captured at the second rate;

generating an adjusted control signal based on the control signal and the noise model, and using the adjusted control signal to reduce the high frequency noise within the process variable.

23. The method of claim 22, wherein the step of determining the noise model further includes the step of calculating a transfer function representing the high frequency noise present within the process variable.

24. The method of claim 23, wherein the step of generating the adjusted control signal further comprises the step of calculating the inverse of the noise model transfer function.

25. The method of claim 24, wherein the step of generating the adjusted control signal further comprises the step of multiplying the control by the inverse of the noise model transfer function.

26. The method of claim 22, further including the steps of generating pseudo-random sequences and using the pseudo-random sequences to generate the adjusted control signal.

27. The method of claim 22, further including the step of compressing data prior to sending the data between two or more devices within the process control system.

28. The method of claim 22, wherein the step of generating the adjusted control signal includes the step of continuously determining the adjusted control signal and further including the step of continuously applying the adjusted control signal to a controlled device.

29. The method of claim 22, wherein the step of generating the adjusted control signal includes the step of determining the adjusted control signal so as to minimize residual cyclic error within a process variable.

30. The method of claim 22, further including the step of time synchronizing data within the process control system using a time source located in at least two devices within the process control system.

31. The method of claim 22, wherein the step of determining includes the step of using a Fourier analysis to determine the noise model.