Over the next decades, there will be a huge expansion of offshore renewable energy facilities to add electricity to the grid and reduce greenhouse gas emissions around the world. Globally, an estimated 17% annual growth from 22 GW to 154 GW in total installed offshore wind power capacity will be seen by 2030. The UK's Offshore Wind Sector Deal (2019) also sets out a goal for the offshore wind sector output being 30 GW by 2030. To meet the ambition of offshore wind energy exploration, it is of great importance to design cost-efficient foundations which, due to the complexity of subsea soil behaviour, remains a major challenge.

Offshore foundation designs are well known to be conservative, which has led in part to the foundations accounting for 25-34% of the overall budget of offshore wind farms. The design of offshore foundations is particularly difficult for carbonate soils which cover roughly 35% of the ocean floor because (1) the complex mechanical behaviour of carbonate soils for which a reliable constitutive model is yet unavailable and (2) carbonate soils around foundations often experience large deformations, such as during foundation installation, leading to significant changes of their properties which are difficult to evaluate using traditional finite element techniques.

The research proposed in this project aims to develop advanced computer models capable of predicting the mechanical response of carbonate sands at offshore foundations from the installation stage to the operational stage. This will be achieved by developing a novel numerical approach called the particle finite element method (PFEM), for analysing large-deformation soil-water-foundation interactions, and a self-learning simulation framework based on advanced deep-learning techniques for training data-driven constitutive models for carbonate sands. The developed PFEM with the trained data-driven constitutive model for modelling the responses of carbonate sands at offshore structure foundations will be validated under both standard laboratory conditions and high gravity centrifuge testing conditions. The success of the proposed research will not only improve our understanding of the behaviour of carbonate sands surrounding offshore foundations but also provide engineers with a robust open-source computer tool to analyse interactions between submerged carbonate sands and foundations with large deformations and help achieve cost-effective foundation solutions for offshore renewable energy developments.