System and method for identifying and recovering from a temporary sensor failure

• US 10,101,194 B2
• Filed: 12/31/2015
• Issued: 10/16/2018
• Est. Priority Date: 12/31/2015
• Status: Active Grant

• Current Assignee: General Electric Company
• Original Assignee: General Electric Company
• Inventors: Gonzaga, Carlos, Parolini, Luca
• Primary Examiner(s): McCall, Eric S
• Assistant Examiner(s): Keramet-Amircola, Mohammed E
• Application Number:
  US14/985,816
• Publication Number:
  US 20170191862A1
• Time in Patent Office: 1,020 Days
• CPC Class Codes: E21B 47/001 for underwater installation
• CPC Class Codes: E21B 47/06 Measuring temperature or pressure
• CPC Class Codes: F01D 15/00 Adaptations of machines or engines for special use; Combinations of engines with devices driven thereby
• CPC Class Codes: F05D 2270/00 Control
• CPC Class Codes: G01F 25/10 of flowmeters

Abstract

A sensor system for identifying a transient sensor failure in an industrial system and for recovering from an erroneous estimation of an expected mass flow rate resulting from the transient sensor failure. The sensor system includes one or more sensors for measuring at least one fluid property of the industrial system. The sensor system includes an enhanced flow soft sensing (EFSS) computing device configured to determine an estimated mass flow rate. The EFSS computing device is also configured to generate expected measurements to be received from one or more sensors. If an error value is not within predetermined parameters, the transient sensor failure is detected. The EFSS computing device is further configured to identify a resurgence of the sensor from the transient sensor failure. An erroneous expected mass flow rate then converges toward a correct expected mass flow rate.

Claims

1. A sensor system for identifying a transient sensor failure in an industrial system and recovering from an erroneous estimation of an expected mass flow rate resulting from the transient sensor failure, the sensor system including one or more sensors for measuring at least one fluid property of the industrial system, the sensor system comprising:

an enhanced flow soft sensing (EFSS) computing device configured to determine an estimated mass flow rate, the EFSS computing device comprising a processor and a memory coupled to the processor, the EFSS computing device in communication with the one or more sensors, the EFSS computing device configured to;

generate, based on at least a historical estimated mass flow rate and a correction term, an expected mass flow rate;

generate, based on at least the expected mass flow rate, expected measurements to be received from the one or more sensors;

receive measurements from the one or more sensors;

compare the expected measurements with the measurements received from the one or more sensors to determine an error value;

compare the error value to predetermined parameters, wherein if the error value is not within the predetermined parameters, the transient sensor failure of a sensor is detected, thereby generating the erroneous estimation of the expected mass flow rate; and

identify a resurgence of the sensor from the transient sensor failure,

wherein an erroneous expected mass flow rate converges toward a correct expected mass flow rate.

2. The sensor system in accordance with claim 1, wherein the measurements received from the one or more sensors include at least a vector of pressure and temperature measurements.

3. The sensor system in accordance with claim 1, wherein the convergence of the erroneous expected mass flow rate toward the correct expected mass flow rate is based on at least the correction term.

4. The sensor system in accordance with claim 1, wherein the correction term is based upon at least measurement errors and an internal model.

5. The sensor system in accordance with claim 1, wherein the expected measurements are based upon the expected mass flow rate and an internal model.

6. The sensor system in accordance with claim 1, wherein the predetermined parameters are obtained through a probability distribution of a difference between the expected measurements and the measurements received from the one or more sensors.

7. The sensor system in accordance with claim 1, wherein the EFSS computing device is coupled to a user interface, the EFSS computing device further configured to transmit a notification to the user interface that the transient sensor failure was detected.

8. A method for identifying transient sensor failure in an industrial system and recovering from an erroneous estimation of an expected mass flow rate resulting from the transient sensor failure, the method implemented using an enhanced flow soft sensing (EFSS) computing device configured to determine an estimated mass flow rate, the method comprising:

generating, based on at least a historical estimated mass flow rate and a correction term, an expected mass flow rate;

generating, based on at least the expected mass flow rate, expected measurements to be received from one or more sensors;

receiving measurements from the one or more sensors;

comparing the expected measurements with the measurements received from the one or more sensors to determine an error value;

comparing the error value to predetermined parameters, wherein if the error value is not within the predetermined parameters, the transient sensor failure of a sensor is detected, thereby generating an erroneous estimation of the expected mass flow rate; and

identifying a resurgence of the sensor from the transient sensor failure, wherein an erroneous expected mass flow rate converges toward a correct expected mass flow rate.

9. The method in accordance with claim 8, wherein the measurements received from the one or more sensors include at least a vector of pressure and temperature measurements.

10. The method in accordance with claim 8, wherein the convergence of the erroneous expected mass flow rate toward the correct expected mass flow rate is based on at least the correction term.

11. The method in accordance with claim 8, wherein the correction term is based upon at least measurement errors and an internal model.

12. The method in accordance with claim 8, wherein the expected measurements are based upon the expected mass flow rate and an internal model.

13. The method in accordance with claim 8, wherein the predetermined parameters are obtained by means of a probability distribution of a difference between the expected measurements and the measurements received from the one or more sensors.

14. The method in accordance with claim 8, wherein the EFSS computing device is coupled to a user interface, the method further comprising transmitting a notification to the user interface that the transient sensor failure was detected.

15. An enhanced flow soft sensing (EFSS) computing device configured to detect a transient sensor failure in an industrial system and recover from an erroneous estimation of an expected mass flow rate resulting from the transient sensor failure, the EFSS computing device further configured to determine an estimated mass flow rate, the EFSS computing device comprising:

a processor in communication with a memory and one or more sensors, the processor configured to;

generate, based on at least a historical estimated mass flow rate and a correction term, an expected mass flow rate;

generate, based on at least the expected mass flow rate, expected measurements to be received from the one or more sensors;

receive measurements from the one or more sensors;

compare the expected measurements with the measurements received from the one or more sensors to determine an error value;

compare the error value to predetermined parameters, wherein if the error value is not within the predetermined parameters, the transient sensor failure of a sensor is detected, thereby generating an erroneous estimation of the expected mass flow rate; and

identify a resurgence of the sensor from the transient sensor failure, wherein an erroneous expected mass flow rate converges toward a correct expected mass flow rate.

16. The EFSS computing device in accordance with claim 15, wherein the measurements received from the one or more sensors include at least a vector of pressure and temperature measurements.

17. The EFSS computing device in accordance with claim 15, wherein the convergence of the erroneous expected mass flow rate toward the correct expected mass flow rate is based on at least the correction term.

18. The EFSS computing device in accordance with claim 15, wherein the correction term is based upon at least measurement errors and an internal model.

19. The EFSS computing device in accordance with claim 15, wherein the expected measurements are based upon the expected mass flow rate and an internal model.

20. The EFSS computing device in accordance with claim 15, wherein the predetermined parameters are obtained by means of a probability distribution of a difference between the expected measurements and the measurements received from the one or more sensors.

21. The EFSS computing device in accordance with claim 15, wherein the EFSS computing device is coupled to a user interface, the EFSS computing device further configured to transmit a notification to the user interface that the transient sensor failure was detected.