Automated turbine blade to shroud gap measurement
Data is received which can characterize a two-dimensional image and a set of three-dimensional surface points. The two-dimensional image can characterize at least a tip of a blade and a portion of a shroud, and the three-dimensional surface points can characterize the tip of the blade and the portion of the shroud. A first location of an edge of the blade and a second location of a surface of the shroud can be determined based on the data. A distance between the first location of the edge of the blade and the second location of the surface of the shroud can be determined based on the data. The distance can be provided. Related apparatus, systems, techniques and articles are also described.
1. A method comprising:
receiving, by at least one processor of a computing system, data characterizing a plurality of two-dimensional images of a blade, including a blade edge and a blade surface, and a portion of a shroud, wherein each two-dimensional image of the plurality of two-dimensional images is captured at different brightness level;
combining, by the processor, the plurality of two-dimensional images to form a high-dynamic range (HDR) image including the blade, the portion of the shroud and a set of three-dimensional surface points characterizing the blade and the portion of the shroud, wherein each point in the set of three-dimensional surface points is associated with a pixel in each of the plurality of two-dimensional images;
determining, by the processor, a plurality of blade edge pixels at the blade edge;
determining, by the processor, a fitted shroud surface based on the three-dimensional surface points;
determining, automatically by the processor, a three-dimensional fitting of the blade surface in the vicinity of each blade edge pixel of the of plurality of blade edge pixels;
determining, by the processor, a plurality of three-dimensional viewing rays associated with the plurality of blade edge pixels;
determining, by the processor, a plurality of three-dimensional blade edge points at which the plurality of three-dimensional viewing rays intersect the three-dimensional fitting of the blade surface;
determining, by the processor, one or more distances between the plurality of three-dimensional blade edge points and the fitted shroud surface; and
providing the one or more distances.
2. The method of claim 1 , further comprising:
acquiring, by a borescope inserted into a turbine, one or more structured light images of the blade and the portion of the shroud;
acquiring, by the borescope, the plurality of two-dimensional images, wherein each of the plurality of two-dimensional images is a white-light image; and
determining, based on the acquired structured light images, the set of three-dimensional surface points.
3. The method of claim 2 wherein determining the plurality of blade edge pixels includes performing edge detection on the HDR image and the one or more structured light images.
4. The method of claim 1 , further comprising:
determining a line characterizing the blade edge, based on the plurality of blade edge pixels.
5. The method of claim 1 , further comprising:
clustering the plurality of blade edge pixels to form segments;
determining a final blade edge segment from the segments and based on a number of pixels within each segment, the determining the final blade edge segment further based on pixel intensity values;
filtering the final blade edge segment; and
determining, for each point in the filtered final blade edge segment, a three-dimensional edge point by at least projecting a position of a point in the filtered final blade edge segment onto a fitted three-dimensional blade surface.
6. The method of claim 1 , wherein determining the one or more distances comprises determining a minimum distance, a maximum distance and an average distance.
7. The method of claim 1 , wherein the providing includes:
displaying, within a first graphical user interface display space, one or more of plurality of two-dimensional images and first graphical icons characterizing a first location of a first blade edge pixel of the plurality of blade edge pixels within the one or more of plurality of two-dimensional images and a second location of the fitted shroud surface within the one or more of plurality of two-dimensional images; and
displaying, within a second graphical user interface display space, a rendered three-dimensional image including the three-dimensional surface points and second graphical icons characterizing the first location and the second location.
8. The method of claim 1 , wherein the receiving is performed by a borescope inserted into a port of a turbine that includes the blade and the shroud.
9. The method of claim 8 , further comprising:
receiving, from a turning tool and by a borescope, an instruction indicating that the blade has rotated within a field of view of a camera of the borescope; and
performing the receiving of the data characterizing the plurality of two-dimensional images, the determining of the plurality of blade edge pixels and the fitted shroud surface, and the determining of the distance in response to receiving the instruction.
10. The method of claim 9 , further comprising:
transmitting an instruction to a turning tool to rotate another blade into a field of view of a camera of the borescope; and
repeating, for the another blade, the receiving, the determining the plurality of blade edge pixels and the fitted shroud surface, and the determining the one or more distances.
11. The method of claim 9 , wherein the turning tool includes an electronically controlled motor and wherein the turning tool is mechanically coupled to the turbine and configured to rotate blades of the turbine through the field of view of the camera of the borescope.
12. The method of claim 1 wherein the providing includes displaying within a graphical user interface display space, a rendered two-dimensional image representing a three-dimensional space that includes the surface points and graphical icons representing a first location of a first blade edge pixel of the plurality of blade edge pixels within the three-dimensional space.
13. The method of claim 1 , wherein the providing includes displaying, within a graphical user interface display, a rendered two-dimensional image representing a three-dimensional space that includes the three-dimensional surface points and graphical icons representing the three-dimensional blade edge points within the three-dimensional space.
14. The method of claim 13 wherein at least one of the three-dimensional blade edge points does not match the three-dimensional coordinates of any of the three-dimensional surface points such that the graphical icons are displaced from the three-dimensional surface points in the rendered two-dimensional image.
15. The method of claim 1 wherein the providing includes displaying, within a graphical user interface display space, a rendered two-dimensional image representing a three-dimensional space that includes the three-dimensional surface points and graphical icons characterizing a first location of a first blade edge pixel of the plurality of blade edge pixels and wherein the three-dimensional surface points characterizing the portion of the shroud are automatically displayed in a consistent orientation relative to the three-dimensional surface points characterizing the blade regardless of the relative orientations of the blade and shroud in the rendered two-dimensional image.
16. A borescope system comprising an image sensor, a display, and a processor, wherein the at least one processor is configured to:
receive a plurality of two-dimensional images generated from data provided by the image sensor, the plurality of two-dimensional images including a blade, including a blade edge and a blade surface, and a portion of a shroud, wherein each two-dimensional image of the plurality of two-dimensional images is captured at different brightness level;
combining, by the processor, the plurality of two-dimensional images to form a high-dynamic range (HDR) image including the blade, the portion of the shroud;
compute, automatically, a plurality of three-dimensional surface points characterizing the blade and the portion of the shroud, wherein each point in the set of three-dimensional surface points is associated with a pixel in each of the plurality of two-dimensional images;
determine, automatically, a plurality of blade edge pixels at the blade edge;
determine, automatically, a fitted shroud surface based on the three-dimensional surface points;
determine, automatically, a three-dimensional fitting of the blade surface in the vicinity of each blade edge pixel of the plurality of blade edge pixels;
determine, automatically, a plurality of three-dimensional viewing rays associated with the plurality of blade edge pixels;
determine, automatically, a plurality of three-dimensional blade edge points at which the plurality of three-dimensional viewing rays intersect the three-dimensional fitting of the blade surface;
determine, automatically, one or more distances between the plurality of three-dimensional blade edge points and the fitted shroud surface;
provide the one or more distances to a display communicatively coupled to the borescope system.
17. The borescope system of claim 16 , wherein the at least one processor is further configured to display, on the display, an image generated from data provided by the image sensor and graphical icons indicating a first location of a first blade edge pixel of the plurality of blade edge pixels.