Fatigue damage characterisation in additively manufactured AlSi10Mg using integrated entropy generation and acoustic emission analysis

This paper discusses an approach on non-destructive method to classify and predict fatigue damage in additively manufactured AlSi10Mg alloy by integrating acoustic emission (AE) and entropy analysis. Fatigue crack growth was monitored using AE signals, while thermal data captured temperature variations associated with energy dissipation at the crack tip. These observations reflect irreversible energy losses during fatigue, offering insight into material degradation. AE and entropy data were analysed to identify different fatigue stages. The entropy generation values recorded for three block loading conditions were 3.446 MJ/m 3 K, 3.114 MJ/m 3 K, and 3.795 MJ/m 3 K. The proposed model achieved over 90% accuracy in fatigue stage classification using confidence interval boundaries, confirming a strong correlation between entropy, AE characteristics, and crack propagation rates. This integrated AE–entropy approach forms a foundation for developing intelligent, non-invasive predictive models for fatigue assessment in aluminium alloy components produced via additive manufacturing.