Intelligent Warping Detection for Fused Filament Fabrication of a Metal-Polymer Composite Filament

Fused Filament Fabrication (FFF) for a metal-polymer composite filament is receiving attention as an effective means to manufacture complex metal parts. However, warping deformation is one of the main problems that reduce part quality in the FFF of a metal-polymer composite filament. In this regard, an appropriate quality monitoring system for the warping deformation of FFF is essential to effectively manage the quality of metal-polymer composite 3D printing outputs. This study presents an intelligent warping detection process for FFF using a metal-polymer composite filament through vision monitoring and automatic object detection. First, five FFF samples are manufactured using the Ultrafuse-316L metal-polymer composite filament. Each fabrication process is recorded by a camera to construct an image dataset for warping deformation. Then, the YOLO v5 algorithm is applied to a training set (80%) of the image dataset to learn labeled FFF output and warping deformation areas. The constructed object detection model for warping deformation is tested using a validation set (20%) of the original image dataset. This study provides a basis to develop an intelligent quality management system for the FFF of a metal-polymer composite filament.