PSI - Issue 79

Available online at www.sciencedirect.com

ScienceDirect

Procedia Structural Integrity 79 (2026) 326–334

© 2025 The Authors. Published by ELSEVIER B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of IGF28 - MedFract3 organizers Keywords: Additive Manufacturing, Ti-6Al-4V, Fatigue Life Prediction Models, Optimisation Algorithms. Abstract The structural integrity of Ti-6Al-4V components produced via Selective Laser Melting (SLM) is critically dependent on understanding fatigue behaviour. However, accurately modelling the growth of small cracks originating from inherent manufacturing defects remains a significant challenge. This work introduces a novel methodology for accurately modelling small crack growth in Ti-6Al-4V alloy produced via Selective Laser Melting (SLM). Based on a modification of a S-N curve model proposal, to explore the use of nanotomography as a defect characterisation technique. This study proposes the use of optimisation algorithms, specifically Nelder-Mead and SLSQP, to determine an optimal predictive model. Evaluation against experimental data demonstrates the high predictive capability of the proposed methodology (R² = 99.28%), achieving a performance comparable to the traditional Basquin equation (R² = 99.29%). 28 th International Conference on Fracture and Structural Integrity - 3 rd Mediterranean Conference on Fracture and Structural Integrity A Novel Optimisation-Based Approach to Predict the S-N curves Behaviour based on Small Crack Growth in Additively Manufactured Ti-6Al-4V João Alves a,b *, Teresa Morgado a,b,c,d , Manuel Pereira e , António Pereira f , Rui F. Martins a,c a NOVA FCT/UNIDEMI, NOVA University of Lisbon, Almada, 2829-516, Portugal b UnIRE, Lisbon School of Engineering of Polytechnic University of Lisbon, Lisboa, 1959-007, Portugal c LASI, Intelligent Systems Associate Laboratory, Guimarães, 4800-058 Guimarães, Portugal d Navy Research Center, Marine Academy, Almada,2810-001, Portugal; tmorgado@dem.isel.ipl.pt e CERENA- University of Lisboa, Lisbpa 1000-268, Portugal f CDRsp & ESTG, Institute Polytechnic of Leiria, 2400-441, Portugal

* Corresponding author. Tel.: +351 212 948 300; E-mail address: jfg.alves@campus.fct.unl.pt

2452-3216 © 2025 The Authors. Published by ELSEVIER B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of IGF28 - MedFract3 organizers 10.1016/j.prostr.2025.12.341