Afully funded PhD position to work on Bayesian methods for predicting failures in large offshore steel structures.
Project description: This study aims to develop a systematic framework that integrates physics-based modelling with SHM data to predict the true structural response of offshore steel structures with improved confidence. The improved predictions will then inform fatigue life analysis, including both damage accumulation and crack propagation, thereby providing a quantitative basis for risk assessment and maintenance planning. The starting point is a statistical FE method (statFEM) [Girolami et.al. 2021, Febrianto et.al. 2022] applied at known critical locations (hot spots). This model explicitly incorporates uncertainty in cyclic environmental loading and material strength variability, producing stress and strain predictions expressed as probability measures. In parallel, the SHM system provides strain-gauge data, which may be noisy or incomplete. These measurements are assimilated with FE predictions through Bayesian model calibration. With calibrated stresses in hand, fatigue analysis proceeds in two stages. First, fatigue damage accumulation is assessed using the stress cycles distributions using S–N models. Second, once crack initiation occurs, fracture-mechanics models (e.g., Paris–Erdogan law) are applied to predict crack propagation. The outcome of this study will be a robust framework for predicting crack initiation and growth in offshore steel structures which enables better-informed inspection planning, risk management, and life-extension decisions.
What is expected: Through this multidisciplinary project, you will gain broad expertise in areas such as Bayesian methods, structural health monitoring, and fatigue crack analysis of steel structures. Furthermore, you will have opportunities to work with industry partner on a data from real asset.
How to apply: To apply, please contact Dr Eky Febrianto (eky.febrianto@glasgow.ac.uk) or Dr Ji-Eun Byun (ji-eun.byun@glasgow.ac.uk) with: (1) A short statement of motivation outlining your interest, academic background, and suitability. (2) A Curriculum Vitae (CV) maximum two pages.
An interview will be conducted for shortlisted applicants.
Funding Notes
This fully-funded PhD position is open to any nationality. The funding will cover 3.5 years of tuition and stipend at standard UKRI rates.
Eligibility:
(1) A first class or upper second-class honours degree (or international equivalent) in Engineering, Mathematics, Statistics, or a related discipline.
(2) A relevant Master’s degree is desirable.
(3) Enthusiasm for working across disciplines at the interface of computational mechanics, Bayesian statistics, and structural engineering applications.
About the Author
