System, device and method for frozen period detection in sensor datasets

1. A computer-implemented method of detecting at least one frozen period in at least one sensor dataset associated with at least one sensor in a technical system, the method comprising:

receiving, by at least one processor of a sensor monitoring device of the technical system, the at least one sensor dataset in time series;

computing, by the at least one processor, run-lengths for the at least one sensor dataset, wherein each of the run-lengths is a length of consecutive repetitions of a sensor value in the at least one sensor dataset;

clustering, by the at least one processor, the run-lengths into two clusters based on a run frequency such that a first portion of the run-lengths are clustered into a first cluster and a second portion of the run-lengths are clustered into a second cluster, wherein the run frequency is a number of times the run-lengths are repeated in the at least one sensor dataset;

identifying, by the at least one processor, a cluster from the two clusters with a lower run frequency;

detecting, by the at least one processor, the at least one frozen period in the at least one sensor dataset based on the identified cluster; and

displaying, by a display unit of the technical system, the detected at least one frozen period, therein providing an indication that a sensor of the at least one sensor is jammed and measurements of the sensor are anomalous,

wherein the technical system comprises a gas turbine, and

wherein the at least one sensor comprises a compressor sensor, an inlet pressure sensor, or a combustion temperature sensor.

2. The method of claim 1 , wherein the computing of the run-lengths for the at least one sensor dataset comprises:

identifying the consecutive repetitions of the sensor value in the at least one sensor dataset; and

computing the run-lengths by determining a number of the consecutive repetitions of the sensor value.

3. The method of claim 2 , wherein the computing of the run-lengths by determining the number of the consecutive repetitions of the sensor value comprises:

defining a minimum discount length for the at least one sensor dataset, wherein the minimum discount length is a number of allowed consecutive repetitions; and

computing the run-lengths for the at least one sensor dataset after subtracting the minimum discount length from each of the consecutive repetitions.

4. The method of claim 1 , further comprising:

defining a minimum run-length for the at least one frozen period; and

detecting the at least one frozen period when the consecutive repetitions in the at least one sensor dataset exceed the minimum run-length.

5. The method of claim 1 , wherein the clustering of the run-lengths into the two clusters based on the run frequency comprises:

determining the run frequency by computing the number of each of the run-lengths in the at least one sensor dataset; and

clustering the run-lengths into the two clusters based on the run frequency, wherein the two clusters comprise an un-anomalous cluster of the run-lengths with a higher run frequency and an anomalous cluster of the run-lengths with the lower run frequency.

6. The method of claim 1 , further comprising:

generating the at least one sensor dataset comprising sensor values corresponding to at least one operation parameter associated with the technical system.

7. The method of claim 1 , further comprising:

recording, by the at least one sensor in the technical system, the at least one sensor dataset comprising sensor values corresponding to at least one operation parameter associated with the at least one technical system.

8. A sensor monitoring device for detecting at least one frozen period in at least one sensor dataset associated with at least one sensor in a technical system, the sensor monitoring device comprising:

a receiver configured to receive the at least one sensor dataset in time series;

at least one processor;

a display unit; and

a memory communicatively coupled to the at least one processor, wherein the memory and the at least one processor are configured to:

compute run-lengths for the at least one sensor dataset, wherein each of the run-lengths is length of consecutive repetitions of a sensor value in the at least one sensor dataset;

cluster the run-lengths into two clusters based on a run frequency such that a first portion of the run-lengths are clustered into a first cluster and a second portion of the run-lengths are clustered into a second cluster, wherein the run frequency is a number of times the run-lengths are repeated in the at least one sensor dataset;

identify a cluster with lower run frequency from the two clusters; and

detect the at least one frozen period in the at least one sensor dataset based on the identified cluster,

wherein the display unit is configured to display the detected at least one frozen period, therein providing an indication that a sensor of the at least one sensor is jammed and measurements of the sensor are anomalous,

wherein the technical system comprises a gas turbine, and

wherein the at least one sensor comprises a compressor sensor, an inlet pressure sensor, or a combustion temperature sensor.

9. The sensor monitoring device of claim 8 , wherein the memory and the at least one processor are further configured to:

define a minimum discount length for the at least one sensor dataset, wherein the minimum discount length is a number of allowed consecutive repetitions; and

compute the run-lengths for the at least one sensor dataset after subtracting the minimum discount length from each of the consecutive repetitions.

10. The sensor monitoring device of claim 8 , wherein the memory and the at least one processor are further configured to:

define a minimum run-length for the at least one frozen period; and

detect the at least one frozen period when the consecutive repetitions in the at least one sensor dataset exceed the minimum run-length.

11. The sensor monitoring device of claim 8 , wherein the memory and the at least one processor are further configured to:

determine the run frequency by computing the number of each of the run-lengths in the at least one sensor dataset; and

cluster the run-lengths into the two clusters based on the run frequency,

wherein the two clusters comprise an un-anomalous cluster of the run-lengths with a higher run frequency and an anomalous cluster of the run-lengths with the lower run frequency.

12. A system for detecting at least one frozen period in at least one sensor dataset, the system comprising:

a server;

a network interface communicatively coupled to the server;

a display unit; and

at least one technical system communicatively coupled to the server via the network interface,

wherein the at least one technical system comprises a gas turbine,

wherein the server comprises a sensor monitoring device having:

a receiver configured to receive the at least one sensor dataset in time series;

at least one processor; and

a memory communicatively coupled to the at least one processor, the memory and the at least one processor configured to:

compute run-lengths for the at least one sensor dataset, wherein each of the run-lengths is length of consecutive repetitions of a sensor value in the at least one sensor dataset;

cluster the run-lengths into two clusters based on a run frequency such that a first portion of the run-lengths are clustered into a first cluster and a second portion of the run-lengths are clustered into a second cluster, wherein the run frequency is a number of times the run-lengths are repeated in the at least one sensor dataset, and identify a cluster with lower run frequency from the two clusters; and

detect the at least one frozen period in the at least one sensor dataset based on the identified cluster,

wherein the display unit is configured to display the detected at least one frozen period, therein providing an indication that a sensor is jammed and measurements of the sensor are anomalous, and

wherein the sensor comprises a compressor sensor, an inlet pressure sensor, or a combustion temperature sensor.

13. The system of claim 12 , wherein the at least one technical system comprises at least one sensor configured to generate the at least one sensor dataset comprising sensor values corresponding to at least one operation parameter associated with the at least one technical system.

14. The system of claim 13 , wherein the at least one sensor comprises one of a plurality of turbine engine sensors.