In the carbon fiber yarn bedding process, quality problems such as short and long filaments, broken filaments, pilling, and splices may occur. In order to avoid these quality problems and consequences, it is necessary to strictly control the operating procedures, equipment maintenance and environmental conditions in the carbon fiber yarn bedding process to ensure the high quality and uniformity of the yarn.
The raw yarn quality inspection system adopts a machine vision + deep learning solution. It uses an 8K high-speed line scan camera to acquire yarn images in real time, transmits them to a computer via gigabit Ethernet, and uses dedicated software and algorithms to perform algorithmic analysis on the images to detect yarn defects in real time.