Abstract: Air film holes are one of the important cooling structures in turbine blades, it has great significance to improve the cooling efficiency of aircraft engines, reduce the temperature of the combustion chamber and even increase the service life of the engine. However, in the actual course of working, it is easy to have internal defects such as unopened air film holes, deflection, and damage to the inner wall due to external factors. Traditional detection methods mainly use artificial film X-ray method to detect internal defects, but there are problems such as low detection efficiency, high cost and high dependence on the experience and skills of the inspectors. This paper proposes an automatic detection method for air film hole defects based on X-ray digital imaging technology. By studying the equivalence of DR digital images and film images, the law of the effect of the transillumination angle on defect recognition and the optimization of the detection process, key technologies based on deep learning neural network defect self-recognition, high-precision detection and intelligent recognition of blade air film hole processing defects are realized.

Key words: non-destructive detection, aircraft engine blades, film holes, intelligent identification, X-ray digital radiography