Photo: United Engine Corporation
United Engine Corporation has introduced a new fluorescent penetrant inspection method for quality control of aircraft engine components using computer vision and neural network technology. The innovation of Rybinsk-based UEC-Saturn has also allowed to computerize the quality control process for power unit blades and includes the use of advanced image processing algorithms.
The new computer-aided fluorescent penetrant inspection method provides imaging of all component surfaces, defect finding, geometrical calculation, classification and product suitability assessment. This test method is used in a production process to increase the accuracy and validity of results.
Formation of a “digital footprint” for the products to be manufactured is an additional effect of this method that facilitates retrospective analysis of the production process for optimization purpose.
“We are persistently working on various computer vision applications specifically for product quality control. Image processing algorithms and neural network technologies were tested and finalized using a test bench. The computer vision technique used for our manufacturing process increases exponentially the quality control capacity and reduces the personnel requirements,” pointed out Evgeny Alekseev, Director of Information Technology, UEC-Saturn.
UEC-Saturn was awarded a gold medal of the Archimedes 2023 International Salon of Inventions and Innovative Technologies.
