Detection of gas turbine engine blade abnormalities based on light reflections

1. A method of inspecting blades of a gas turbine engine for abnormalities, comprising:

projecting light from a light source into an illumination area, wherein said projecting light includes projecting the light through at least one cylindrical lens disposed between the light source and the illumination area, and utilizing the at least one cylindrical lens to arrange the projected light into a generally planar light sheet in the illumination area which forms a line or curve on a surface of the blade;

utilizing a sensor to record data of at least one reflection of the projected light from a blade that is part of a gas turbine engine and is disposed in the illumination area;

determining, based on the recorded data, whether the blade is abnormal, wherein the determining comprises comparing a value describing an aspect of a brightness of the at least one reflection to an expected value; and

based on the determining indicating that the blade is abnormal, providing a blade abnormality notification.

2. The method of claim 1 , wherein:

the blade is one of a plurality of blades that extend radially outwards from a hub;

said projecting is performed while the plurality of blades rotate about a longitudinal axis during operation of the gas turbine engine, such that the light source projects light onto each of the plurality of blades as they pass through the illumination area; and

said utilizing a sensor and said determining are performed for reflections of the projected light from each of the plurality of blades.

3. The method of claim 2 , wherein said operation of the gas turbine engine corresponds to a flight, and said projecting, utilizing, determining, and providing are performed during the flight.

4. The method of claim 2 , wherein based on a rotational speed of the hub, the notification is provided in a manner that indicates at least one of:

a quantity of blades that are determined to be abnormal; or

which one or more particular ones of the plurality of blades are abnormal.

5. The method of claim 1 , wherein:

the light source includes a laser;

the at least one cylindrical lens includes a first cylindrical lens and a second cylindrical lens;

the first cylindrical lens is concave, is disposed between the light source and illumination area, and provides a decollimating feature that causes the projected light to diverge into a first light sheet as the projected light approaches the second cylindrical lens; and

the second cylindrical lens is convex, is disposed between the first cylindrical lens and the illumination area, and provides a collimating feature that causes the diverged projected light to become a second light sheet that is more collimated than the first light sheet as the projected light approaches the illumination areas.

6. The method of claim 5 , wherein:

the first cylindrical lens extends along a first longitudinal axis;

the at least one cylindrical lens includes a third cylindrical lens that is concave, is disposed between the second cylindrical lens and the illumination area, and extends along a second longitudinal axis; and

the second longitudinal axis is rotated approximately 90° with respect to the first longitudinal axis.

7. The method of claim 1 , wherein:

the light source includes a plurality of light-emitting diodes; and

the at least one cylindrical lens includes a convex cylindrical lens that causes the projected light to converge as it approaches the illumination area.

8. The method of claim 1 , wherein said utilizing a sensor to record data of at least one reflection of the projected light comprises utilizing at least one photodiode to record the data.

9. The method of claim 1 , wherein:

said utilizing the sensor to record data includes recording a time trace of sensor data; and

said determining, based on the recorded data, whether the blade is abnormal includes utilizing a neural network to analyze the time trace and determine whether the blade is abnormal, wherein the neural network is trained with historical data of reflections of projected light from blades of one or more gas turbine engines.

10. A gas turbine engine, comprising:

a light source configured to project light into an illumination area;

a hub and a plurality of blades that extend radially outward from the hub and are configured to rotate about a longitudinal axis through the illumination area;

at least one cylindrical lens disposed between the light source and the illumination area, wherein the light source is configured to project the light into the illumination area through the at least one cylindrical lens, and wherein the at least one cylindrical lens is configured to arrange the projected light into in a generally planar light sheet in the illumination area that forms a line or curve on a surface of said one of the plurality of blades;

a sensor configured to record data of at least one reflection of the projected light from one of the plurality of blades disposed in the illumination area; and

processing circuitry configured to:

determine, based on the recorded data, whether the blade is abnormal, wherein the determination includes a comparison of a value describing an aspect of a brightness of the at least one reflection to an expected value; and

based on the determination indicating that the blade is abnormal, provide a blade abnormality notification.

11. The gas turbine engine of claim 10 , wherein the blades are fan blades in a fan section of the gas turbine engine.

12. The gas turbine engine of claim 10 , wherein:

the sensor is configured to measure reflections of the projected light as the blades rotate through the illumination area; and

the processing circuitry is configured to:

based on a rotational speed of the hub, determine at least one of a quantity of blades that are abnormal or which one or more particular ones of the plurality of blades are abnormal; and

provide the notification in a manner that indicates said at least one of the quantity of blades that are abnormal or which one or more particular ones of the plurality of blades are abnormal.

13. The gas turbine engine of claim 10 , wherein:

the light source includes a laser;

the at least one cylindrical lens includes a first cylindrical lens and a second cylindrical lens;

the first cylindrical lens is concave, is disposed between the light source and illumination area, and provides a decollimating feature that causes the projected light to diverge into a first light sheet as the projected light approaches the second cylindrical lens; and

the second cylindrical lens is convex, is disposed between the first cylindrical lens and the illumination area, and provides a collimating feature that causes the projected light to become a second light sheet that is more collimated than the first light sheet as the projected light approaches the illumination area.

14. The gas turbine engine of claim 13 , wherein:

the first cylindrical lens extends along a first longitudinal axis;

the at least one cylindrical lens includes a third cylindrical lens that is concave, is disposed between the second cylindrical lens and the illumination area, and extends along a second longitudinal axis; and

the second longitudinal axis is rotated approximately 90° with respect to the first longitudinal axis.

15. The gas turbine engine of claim 10 , wherein:

the light source includes a plurality of light-emitting diodes;

the at least one cylindrical lens includes a convex cylindrical lens that causes the projected light to converge as it approaches the illumination area.

16. The gas turbine engine of claim 10 , wherein the sensor includes at least one photodiode.

17. A method of inspecting blades of a gas turbine engine for abnormalities, comprising:

projecting light from a light source into an illumination area, wherein said projecting light includes projecting the light through at least one cylindrical lens disposed between the light source and the illumination area, and utilizing the at least one cylindrical lens to arrange the projected light into a generally planar light sheet in the illumination area which forms a line or curve on a surface of the blade;

utilizing a sensor to record data of at least one reflection of the projected light from a blade that is part of a gas turbine engine and is disposed in the illumination area;

determining, based on detecting glare in the recorded data, that the blade is abnormal; and

based on the determining that the blade is abnormal, providing a blade abnormality notification.

18. The method of claim 1 , wherein said determining comprises:

extracting, from the sensor data, a rise time of a brightness of the at least one reflection from a first value to a second value;

comparing the extracted rise time to an expected rise time; and

determining that the blade is abnormal based on the extracted rise time differing from the expected rise time by more than a predefined threshold.

19. The method of claim 1 , wherein said determining comprises:

extracting, from the sensor data, a fall time of a brightness value of the at least one reflection from a first value to a second value;

comparing the extracted fall time to an expected fall time; and

determining that the blade is abnormal based on the extracted fall time differing from the expected fall time by more than a predefined threshold.

20. The method of claim 1 , wherein said determining comprises:

extracting, from the sensor data, a peak brightness value of the at least one reflection;

comparing the extracted peak brightness value to an expected peak brightness value; and

determining that the blade is abnormal based on the extracted peak brightness value differing from the expected peak brightness value by more than a predefined threshold.

21. The method of claim 1 , wherein said determining comprises:

extracting, from the sensor data, a sharpness of a peak of a peak brightness value of the at least one reflection;

comparing the extracted sharpness to an expected sharpness; and

determining that the blade is abnormal based on the extracted sharpness differing from the expected sharpness by more than a predefined threshold.

22. The method of claim 1 , wherein said determining comprises:

extracting, from the sensor data, an average brightness of a peak of a peak brightness value of the at least one reflection;

comparing the extracted average brightness to an expected average sharpness; and

determining that the blade is abnormal based on the extracted average brightness differing from the expected average brightness by more than a predefined threshold.

23. A gas turbine engine, comprising:

a light source configured to project light into an illumination area;

a hub and a plurality of blades that extend radially outward from the hub and are configured to rotate about a longitudinal axis through the illumination area;

at least one cylindrical lens disposed between the light source and the illumination area, wherein the light source is configured to project the light into the illumination area through the at least one cylindrical lens, and wherein the at least one cylindrical lens is configured to arrange the projected light into in a generally planar light sheet in the illumination area that forms a line or curve on a surface of said one of the plurality of blades;

a sensor configured to record data of at least one reflection of the projected light from one of the plurality of blades disposed in the illumination area; and

processing circuitry configured to:

determine, based on detecting glare in the recorded data, that the blade is abnormal; and

based on the determination that the blade is abnormal, provide a blade abnormality notification.

24. The gas turbine engine of claim 10 , wherein to determine whether the blade is abnormal, the processing circuitry is configured to:

extract, from the sensor data, a rise time of a brightness of the at least one reflection from a first value to a second value;

compare the extracted rise time to an expected rise time; and

determine that the blade is abnormal based on the extracted rise time differing from the expected rise time by more than a predefined threshold.

25. The gas turbine engine of claim 10 , wherein to determine whether the blade is abnormal, the processing circuitry is configured to:

extract, from the sensor data, a fall time of a brightness value of the at least one reflection from a first value to a second value;

compare the extracted fall time to an expected fall time; and

determine that the blade is abnormal based on the extracted fall time differing from the expected fall time by more than a predefined threshold.

26. The gas turbine engine of claim 10 , wherein to determine whether the blade is abnormal, the processing circuitry is configured to:

extract, from the sensor data, a peak brightness value of the at least one reflection;

compare the extracted peak brightness value to an expected peak brightness value; and

determine that the blade is abnormal based on the extracted peak brightness value differing from the expected peak brightness value by more than a predefined threshold.

27. The gas turbine engine of claim 10 , wherein to determine whether the blade is abnormal, the processing circuitry is configured to:

extract, from the sensor data, a sharpness of a peak of a peak brightness value of the at least one reflection;

compare the extracted sharpness to an expected sharpness; and

determine that the blade is abnormal based on the extracted sharpness differing from the expected sharpness by more than a predefined threshold.

28. The gas turbine engine of claim 10 , wherein said determining comprises:

extract, from the sensor data, an average brightness of a peak of a peak brightness value of the at least one reflection;

compare the extracted average brightness to an expected average sharpness; and

determine that the blade is abnormal based on the extracted average brightness differing from the expected average brightness by more than a predefined threshold.