Gas turbine sensor failure detection utilizing a sparse coding methodology
First Claim
1. A method for detecting gas turbine sensor failure based upon collected readings from at least one sensor, comprising:
- creating a dictionary of basis vectors defining values associated with known sensor readings for normal operating conditions;
applying a sparse coding process to a set of sensor reading data, using the created dictionary, to identify a set of L-1 norm residual data;
evaluating the L-1 norm residual data to categorize a predetermined subset of the largest-valued L-1 norm residual data as abnormal sensor readings;
comparing the abnormal sensor readings to a plurality of prior sensor readings and defining the designated abnormal sensor readings as associated with a sensor failure if a predefined number of the prior sensor readings are also designated as abnormal sensor readings;
transmitting a sensor failure signal to gas turbine personnel, identifying the particular sensor; and
removing the failed sensor from service, and repairing or replacing the failed sensor.
2 Assignments
0 Petitions
Accused Products
Abstract
A method and system for recognizing (and/or predicting) failures of sensors used in monitoring gas turbines applies a sparse coding process to collected sensor readings and defines the L-1 norm residuals from the sparse coding process as indicative of a potential sensor problem. Further evaluation of the group of residual sensor readings is perform to categorize the group and determine if there are significant outliers (“abnormal data”), which would be considered as more likely associated with a faulty sensor than noisy data. A time component is introduced into the evaluation that compares a current abnormal result with a set of prior results and making the faulty sensor determination if a significant number of prior readings also have an abnormal value. By taking the time component into consideration, the number of false positives is reduced.
8 Citations
20 Claims
-
1. A method for detecting gas turbine sensor failure based upon collected readings from at least one sensor, comprising:
-
creating a dictionary of basis vectors defining values associated with known sensor readings for normal operating conditions; applying a sparse coding process to a set of sensor reading data, using the created dictionary, to identify a set of L-1 norm residual data; evaluating the L-1 norm residual data to categorize a predetermined subset of the largest-valued L-1 norm residual data as abnormal sensor readings; comparing the abnormal sensor readings to a plurality of prior sensor readings and defining the designated abnormal sensor readings as associated with a sensor failure if a predefined number of the prior sensor readings are also designated as abnormal sensor readings; transmitting a sensor failure signal to gas turbine personnel, identifying the particular sensor; and removing the failed sensor from service, and repairing or replacing the failed sensor. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10)
-
-
11. A system for detecting failure of a gas turbine sensor comprising
a database of sensor readings; -
a sensor monitoring system component in communication with the database of sensor readings, the sensor monitoring system component comprising a program storage device and a processor, the program storage device embodying in a fixed tangible medium a set of program instructions executable by the processor to perform the method steps of; creating a dictionary of basis vectors defining values associated with known sensor readings for normal operating conditions; applying a sparse coding process to a set of sensor reading data, using the created dictionary, to identify a set of L-1 norm residual data; evaluating the L-1 norm residual data to categorize a predetermined subset of the largest-valued L-1 norm residual data as abnormal sensor readings; comparing the abnormal sensor readings to a plurality of prior sensor readings and defining the designated abnormal sensor readings as associated with a sensor failure if a predefined number of the prior sensor readings are also designated as abnormal sensor readings; and transmitting a sensor failure signal to gas turbine personnel, including information identifying the failed sensor, so as to be repaired or replaced; a monitoring system database for storing sensor readings classified as abnormal. - View Dependent Claims (12, 13, 14, 15, 16, 17, 18)
-
-
19. A method for predicting failure of a gas turbine sensor based upon collected readings from at least one sensor, comprising:
-
creating a dictionary of basis vectors defining values associated with known sensor readings for normal operating conditions; applying a sparse coding process to a set of sensor reading data, using the created dictionary, to identify a set of L-1 norm residual data; evaluating the L-1 norm residual data to categorize a predetermined subset of the largest-valued L-1 norm residual data as abnormal sensor readings; comparing the abnormal sensor readings to a plurality of prior sensor readings and defining the designated abnormal sensor readings as associated with a predicted sensor failure if a predefined number of the prior sensor readings are also designated as abnormal sensor readings; and transmitting a message to gas turbine personnel identifying a sensor for potential repair or replacement. - View Dependent Claims (20)
-
Specification