📖 ABSTRACT/OVERVIEW
Additive manufacturing technologies are increasingly being evaluated for producing replacement components and custom mechatronic assemblies for oil and gas field equipment operated in the Niger Delta and Deep Offshore environments, yet the fatigue behaviour of additively manufactured metals is governed by process-induced microstructural features including residual stresses, porosity distributions, anisotropic grain textures, and surface topography that differ fundamentally from those in wrought equivalents and are inadequately represented by existing classical fatigue life prediction theories. This dissertation develops a computational theory for microstructure-informed fatigue life prediction of additively manufactured steel and titanium components, integrating crystal plasticity finite element modelling, machine learning-assisted microstructure characterization from X-ray computed tomography data, and probabilistic damage mechanics. A key original contribution is the derivation of a microstructure-aware fatigue notch factor formulation that translates AM process parameters and resultant microstructure statistics into probabilistic fatigue life distributions compatible with reliability-based design frameworks. The theory is validated against a programme of 240 fatigue coupon tests spanning selective laser melted 316L stainless steel and Ti-6Al-4V specimens manufactured under varied process parameter combinations. Model predictions of fatigue life at the median and 10th percentile survival probabilities agree with experimental data within a scatter factor of 1.8 across all tested material-process combinations, compared to a scatter factor of 4.6 for standard S-N curve prediction. Applications to downhole tool and subsea valve actuator component design are analysed within an oil field equipment qualification framework. Keywords: additive manufacturing, fatigue life prediction, microstructure, crystal plasticity, oil and gas components
Need Complete Chapters of the Above Topic?
Get high-quality, Zero-AI research materials with current citations.
Request via WhatsApp 💬