Degradation Classification of 3D Printing Thermoplastics Using Fourier Transform Infrared Spectroscopy and Artificial Neural Networks

Abstract

Fused deposition modeling (FDM) is the most popular technology among 3D printing technologies because of inexpensive and flexible extrusion systems with thermoplastic materials. However, thermal degradation phenomena of the 3D-printed thermoplastics is an inevitable problem for long-term reliability. In the current study, thermal degradation of 3D-printed thermoplastics of ABS and PLA was studied. A classification methodology using deep learning strategy was developed so that thermal degradation of the thermoplastics could be classified using FTIR and Artificial Neural Networks (ANNs). Under given data and predefined rules for ANNs, ANN models with nine hidden layers showed the best results in terms of accuracy. To extend this methodology, other thermoplastics, several new datasets for ANNs, and control parameters of ANNs could be further investigated.