Arttu Polojärvi

Aalto University is seeking a Doctoral Researcher to work on physics-enhanced machine learning for floating offshore wind digital twins . The project sits at the intersection of marine hydrodynamics, offshore structural dynamics and data-driven modelling, with a focus on estimating and monitoring the behaviour of floating wind turbines in demanding sea conditions. Floating wind turbines experience complex motions in six degrees of freedom under irregular waves and turbulent wind. Their mooring lines and other critical components are difficult to sense directly, so the project aims to develop digital twins that combine hydrodynamic knowledge with measurements from a small number of sensors. The goal is to infer unmeasured motions and loads, track structural condition, detect early degradation, and update predictions as new data arrive. The position is based in the Marine and Arctic Technology research group at Aalto University’s Department of Energy and Mechanical Engineering. The group studies ships and offshore structures in waves, wind and ice, combining machine learning and data-driven methods with hydrodynamic theory, numerical simulation and model-scale experiments. The successful candidate will work closely with this team and with international academic collaborators. The research themes include real-time estimation of wave-induced motions and mooring loads, physically consistent learning across changing sea states, and uncertainty-aware monitoring using probabilistic methods. The project specifically highlights model ideas such as Koopman operator theory and dynamic mode decomposition, and also refers to hydrodynamic simulations, potential-flow methods, CFD and coupled simulation tools such as OpenFAST. This is a research-intensive doctoral opening in a multidisciplinary environment at Aalto’s Otaniemi campus in Espoo, where the work is framed around safer and more cost-effective floating wind farms and the indirect monitoring of hard-to-access components.