Development of automation and artificial intelligence technology for welding and inspection process in aircraft industry

Abstract

This paper will first give a brief overview of advanced production system. It will describe the intelligent manufacturing system developed based on this system concept through the development of automation and AI (artificial intelligence) technologies for welding and inspection processes for aeroengine parts. In the process of establishing new welding conditions for the aeroengine parts to be TIG welded, it digitized welding operations, welded part conditions, and welding equipment conditions to standardize and quantify man, material, and machine and to understand production conditions on a time axis. AI program was created that enables the robot to always perform welding under optimal conditions. As a result, it established the robot welding system and automated the skilled welding operator technique. In the process of establishing inspection program for the welded aeroengine parts, the camera image of the welded position to be inspected was digitized to improve the data accuracy. Subsequently, it was developed and applied a technology that uses a machine learning method based on a multilayer neural network which is currently attracting the most attention among machine learning methods, to judge whether the obtained and conversed image data is pass or fail. As a result, it established the automatic imaging and judgment system and automated the skilled inspection operator technique. Finally, it will summarize the stage of the engineering strategy and future automation and AI technologies for the welding and inspection process of the Digital Smart Factory in the aircraft industry.