Dr Jincheng Zhang - University of Warwick

Research Fellowships 2025

Offshore renewable energy is crucial for the UK to achieve its energy and environmental targets, with the UK government aiming for up to 60 gigawatts of offshore wind by 2030, including up to 5 gigawatts of floating wind. By 2050, floating wind is expected to represent well over half of the UK's offshore wind generation. 

However, floating wind turbines currently face a high levelised cost of energy, primarily due to wave- and wind-induced platform motion impacting energy efficiency and structural durability. Advanced control strategies, such as model predictive control, have shown great potential in tackling this. However, existing work is limited by point estimation of wave and wind – they have no access to the spatiotemporal wave-wind flow fields. This lack of detailed wave-wind information greatly undermines control optimality and reliability, and poses greater challenges for future turbines producing over 20 megawatts, where the spatiotemporal wave-wind variability becomes more vital. Field estimation of waves and wind will thus be key to deliver the most promising control technology for future-generation floating wind turbines.

This Research Fellowship aims to significantly reduce the cost of floating wind energy, by developing the first control solution fully aware of its spatiotemporally varying wave-wind field. A physics-aware, data-driven machine learning framework will be developed to seamlessly integrate sparse sensor data, domain knowledge, and physical laws to predict and process spatiotemporal wave-wind flow fields, addressing the limitations of existing numerical, analytical, and data-centric approaches. The ultimate goal of the project is to deliver a cost-effective, reliable, and uncertainty-aware solution for current and future floating wind turbines, with spatiotemporal wave-wind-wake awareness. The outcomes will make a significant contribution to realising wind energy’s full potential in the transition to net zero. 

Personal website

LinkedIn