Real-time polymer flow state monitoring during fused filament fabrication based on acoustic emission

Abstract

Fused filament fabrication (FFF) has been widely used in various manufacturing situations and has shown that it is flexible and convenient in complex customize manufacturing scenarios. It is worth noting that the flow state of polymer melt inside the nozzle of the FFF process is closely related to the quality of the printed product, however, it is an unsolved problem to obtain effective information about this critical flow state by an in-situ and nondestructive method. In this work we proposed a strategy for identifying the polymer melt flow state inside the nozzle in the FFF process. Firstly, we summarized three types of printing products of different quality which was drawn by three different flow state, specifically an acoustic emission sensor was adopted to collect the information of flow state due to its real-time and nondestructive capability, the experiment of collecting acoustic emission (AE) signals under three different flow conditions was established, then an deep learning based pattern recognize model named AETCN was adopted to identify difference flow state without complex data preprocessing, and the accuracy rate is more than 98%. Results of our works can provide a novel perspective for FFF quality monitoring and benefit to the development of an FFF advanced monitoring system.