Method and system for position control based on automated defect detection feedback

A method of performing position control on defective blades in an aircraft engine, the method comprising:

• providing a storage medium that stores data and programs used in processing video images and a processing unit that processes the video images;
• receiving from an image capture device video images of a plurality of the blades in motion inside of the engine;
• the processing unit detecting a potential defect in the first blade of the plurality of blades by performing Robust Principal Component Analysis to simultaneously decompose the video images into a low rank matrix representing the first blade and a sparse matrix representing the potential defect;
• the processing unit implementing a classifier to confirm the potential defect is one of a defect or a non-defect;
• the processing unit selecting a reference blade from the plurality of blades inside of the engine;
• indexing a position of each of the plurality of the blades inside of the engine relative to the reference blade;
• using the indexed positions of each of the plurality of the blades to locate the first blade in a current position;
• and
• after the defect is detected and confirmed in the first blade and the first blade is located, providing instructions to move the first blade having the defect from the current position to an inspection position within the engine.

A computer program product, comprising a non-transitory computer readable medium having a computer readable program code stored on the non-transitory computer readable medium, the computer readable program code adapted to be executed to implement a method for performing position control on defective blades in an aircraft engine, the method comprising:

• receiving from an image capture device video images of a plurality of the blades of a stage in motion inside of the engine;
• detecting a potential defect in the first blade of the plurality of blades by performing Robust Principal Component Analysis to simultaneously decompose the video images into a low rank matrix representing the first blade and a sparse matrix representing the potential defect;
• implementing a classifier to confirm the potential defect is one of a defect or a non-defect;
• selecting a reference blade from the plurality of blades inside of the engine;
• indexing a position of each of the plurality of the blades inside the engine relative to the reference blade;
• using the indexed positions of each of the plurality of the blades to locate the first blade in a current position;
• after the defect is detected and confirmed in the first blade and the first blade is located, transmitting instructions to a turning tool rotate the first blade having the defect from the current position to an inspection position;
• and
• rotating the first blade from the current position to the inspection position.