Abstract
One of the materials used in the design of aircraft structures is fiber-reinforced composites due to their good tensile strength and resistance to compression. During the manufacturing process, these structures are thoroughly inspected for flaws and defects to ensure structural integrity during commercial use. This non-destructive inspection (NDI) process can be done using ultrasonic testing (UT) due to its effectiveness at detecting flaws embedded within the material surface. Currently, the NDI process is done manually and can be a significant bottleneck in the manufacturing workflow. Artificial intelligence (AI) has shown success in numerous fields including computer vision, natural language processing, and recommendation systems. In this paper, we develop an AI-based assistance tool to drastically reduce inspection time. Typical AI workflows require large amounts of labelled data but defects rarely occur resulting in strong class imbalance. To overcome this, our method utilizes self-supervised learning (SSL) and only requires non-defect data during training. We verify our method using fuselage data generated in a production environment. We show that our method can effectively identify three common types of defects: delamination, foreign object debris (FOD), and porosity.
