Method for detecting and correcting sensor failure in oil and gas production system

• US 6,772,082 B2
• Filed: 04/26/2002
• Issued: 08/03/2004
• Est. Priority Date: 04/26/2001
• Status: Expired due to Term

• Current Assignee: Abb As (ABB Limited)
• Original Assignee: Abb As (ABB Limited)
• Inventors: Allen, John, Bringedal, Bjorn Oyvind, Morud, Svein Arne, van der Geest, Robert Anton Bernard
• Primary Examiner(s): NOLAND, THOMAS
• Application Number:
  US10/133,783
• Publication Number:
  US 20030005747A1
• Time in Patent Office: 830 Days
• Field of Search: 702/116, 702/100, 702/34, 702/104, 73/13.4
• US Class Current: 702/116
• CPC Class Codes: G01D 3/08 with provision for safeguarding the apparatus, e.g. against abnormal operation, against breakdown
• CPC Class Codes: G01F 1/74 Devices for measuring flow of a fluid or flow of a fluent solid material in suspension in another fluid
• CPC Class Codes: G01N 33/007 Arrangements to check the analyser calibrating G01N33/0006
• CPC Class Codes: G05B 9/02 electric

Abstract

Sensor failure in an oil and gas production system is monitored and detected by checking the consistency of the measurements of the sensor in question using a computer-based process model with the measurements of the other sensors in the system. The present invention also generates a back-up value for a failing sensor with the process model by identifying the value that is most consistent with the measurements of the other sensors in the system.

Claims

1. A method for detecting sensor failure in an oil and gas production system, comprising performing in a data processing system:

generating from a process model of the oil and gas production system using measurements from one or more sensors an expected value for a sensor measurement;

comparing the expected value with an actual measurement obtained from a first sensor in said oil and gas production system;

if the expected value is within acceptable specified limits of the actual measurement confirming the validity of the actual measurement obtained; and

if the expected value is not within acceptable specified limits of the actual measurement, indicating the failure of the sensor and replacing the value of the actual measurement with the expected value.

2. The method of claim 1, wherein said generating an expected value for a sensor measurement comprises performing in a data processing system obtaining a probability distribution of expected values for the sensor measurement.

3. The method of claim 2 wherein said obtaining of a probability distribution of expected values comprises performing in a data processing system multiple simulations of the oil and gas production system using sensors in the production system other than the first sensor over a range of possible sensor readings from the other sensors.

4. The method of claim 3 wherein said obtaining of a probability distribution of expected values comprises performing in a data processing system multiple simulations of the oil and gas production system using a range of model parameters for the production system.

5. The method of claim 1, wherein a probability distribution of measured sensor values is available and said comparing comprises performing in a data processing system obtaining a probability distribution of the difference between the expected value and the actual measurement obtained from the sensor.

6. The method of claim 5, wherein said comparing comprises performing in a data processing system obtaining the acceptable specified limits for the difference between the expected value and the actual measurement by means of said probability distribution of the difference between the expected value and the actual measurement.

7. A method for detecting sensor failure in an oil and gas production system, comprising the steps performed in a data processing system:

generating an expected value for a sensor measurement from a process model of the oil and gas production system, by means of one (or more) generated back-up value(s), the back-up value(s) primarily being generated by means of the process model;

comparing the expected value with an actual measurement obtained from a first sensor in said oil and gas production system;

if the expected value is within acceptable specified limits of the actual measurement confirming the validity of the actual measurement obtained; and

if the expected value is not within acceptable specified limits of the actual measurement, indicating the failure of the sensor and replacing the value of the actual measurement with the expected value.

8. The method of claim 7, wherein said generating an expected value for a sensor measurement comprises performing in a data processing system obtaining a probability distribution of expected values for the sensor measurement.

9. The method of claim 8 wherein said obtaining a probability distribution of expected values comprises performing in a data processing system multiple simulations of the oil and gas production system using sensors in the production system other than the first sensor over a range of possible sensor readings from the other sensors.

10. The method of claim 9 wherein said obtaining a probability distribution of expected values comprises performing in a data processing system multiple simulations of the oil and gas production system using a range of model parameters for the production system.

11. The method of claim 7, wherein a probability distribution of measured sensor values is available and said comparing comprises performing in a data processing system obtaining a probability distribution of the difference between the expected value and the actual measurement obtained from the sensor.

12. The method of claim 11, wherein said comparing comprises performing in a data processing system obtaining the acceptable specified limits for the difference between the expected value and the actual measurement by means of said probability distribution of the difference between the expected value and the actual measurement.

13. A computer program product for causing a data processing system to monitor for sensor failure in an oil and gas production system, the computer program product comprising:

a computer program storage device;

computer-readable instructions on the storage device for causing the data processing system to monitor for sensor failure in the oil and gas production system by performing;

generating from a process model of the oil and gas production system using measurements from one or more sensors an expected value for a sensor measurement;

comparing the expected value with an actual measurement obtained from a first sensor;

if the expected value is within acceptable specified limits of the actual measurement confirming the validity of the actual measurement obtained; and

if the expected value is not within acceptable specified limits of the actual measurement, indicating the failure of the sensor and replacing the value of the actual measurement with the expected value.

14. The computer program product of claim 13, wherein the computer-readable instructions further include instructions for causing the data processing system to perform obtaining a probability distribution of expected values for measurement from the sensor.

15. The computer program product of claim 13, wherein the computer-readable instructions further include instructions for causing the data processing system to perform multiple simulations of the production system using sensors in the production system other than the first sensor over a range of possible sensor readings from the other sensors.

16. The computer program product of claim 13, wherein the computer-readable instructions further include instructions for causing the data processing system to perform obtaining a probability distribution of the difference between the expected value and the actual measurement obtained from the sensor.

17. The computer program product of claim 16, wherein the computer-readable instructions further include instructions for causing the data processing system to perform obtaining the acceptable specified limits for the difference between the expected value and the actual measurement by means of said probability distribution of the difference between the expected value and the actual measurement.

18. A computer program product for causing a data processing system to monitor for sensor failure in an oil and gas production system, the computer program product comprising:

a computer program storage device;

computer-readable instructions on the storage device for causing the data processing system to monitor for sensor failure in the oil and gas production system by;

generating an expected value for a sensor measurement from a process model of the oil and gas production system, by means of one (or more) generated back-up value(s), the back-up value(s) primarily being generated by means of the process model;

comparing the expected value with an actual measurement obtained from a sensor in said oil and gas production system;

if the expected value is within acceptable specified limits of the actual measurement confirming the validity of the actual measurement obtained; and

if the expected value is not within acceptable specified limits of the actual measurement, indicating the failure of the sensor and replacing the value of the actual measurement with the expected value.

19. The computer program product of claim 18, wherein the computer-readable instructions further include instructions for causing the data processing system to perform obtaining a probability distribution of expected values for measurement from the sensor.

20. The computer program product of claim 18, wherein the computer-readable instructions further include instructions for causing the data processing system to perform multiple simulations of the production system using a range of model parameters for the production system.

21. The computer program product of claim 18, wherein the computer-readable instructions further include instructions for causing the data processing system to perform obtaining a probability distribution of the difference between the expected value and the actual measurement obtained from the sensor.

22. The computer program product of claim 21, wherein the computer-readable instructions further include instructions for causing the data processing system to perform obtaining the acceptable specified limits for the difference between the expected value and the actual measurement by means of said probability distribution of the difference between the expected value and the actual measurement.