M. Corey, B. Eslami
Widener University,
United States
Keywords: FDM 3D printing, machine learning, PLA, Schlieren imaging
Summary:
Fused deposition modeling (FDM) suffers from unpredictable mechanical properties in nominally identical prints. Current quality assurance relies on destructive testing or expensive post-process inspection, while existing machine learning approaches focus primarily on printing parameters rather than real-time thermal environments. This proof-of-concept study introduces thermal fingerprinting: a novel non-destructive technique that combines background-oriented Schlieren (BOS) imaging with machine learning to predict tensile strength during printing. We captured thermal gradient fields surrounding PLA specimens (n=30) under six controlled cooling conditions using consumer-grade equipment (Nikon D750 camera, household hairdryers) to demonstrate low-cost implementation feasibility. BOS imaging was performed at nine critical layers during printing, generating thermal gradient data that was processed into engineered features for analysis. Our preliminary dual-model ensemble system successfully classified cooling conditions (100%) and showed promising correlations with tensile strength (initial 80/20 train-test validation: R² = 0.808, MAE = 0.279 MPa). However, more rigorous cross-validation revealed the need for larger datasets to achieve robust generalization (5-Fold Cross Validation R² = 0.301, MAE = 0.509 MPa), highlighting typical challenges in small-sample machine learning applications. This work represents the first successful application of Schlieren imaging to polymer additive manufacturing and establishes a methodological framework for real-time quality prediction. While our limited sample size constrains immediate predictive accuracy, the demonstrated correlation between thermal gradients and mechanical properties validates the underlying concept. Future work will focus on expanding the dataset and refining feature engineering to develop production-ready quality assurance systems. The study confirms that thermal environments encode mechanical property information, opening new avenues for non-destructive 3D printing quality control.