Marine renewable energy (MRE) is one of the renewable energy sources (RES) with the greatest potential in the transition to a sustainable, green and clean energy mix. In particular, wave energy, with its global and almost untapped potential – around 16,000 TWh/year, close to half of global electricity consumption – can make a vital contribution to achieving the Sustainable Development Goals (SDGs) of the United Nations 2030 Agenda, namely Goals 7 (affordable and clean energy) and 13 (climate action), and the EU Green Deal (climate neutrality by 2050). In Portugal, not only is the resource abundant, but there are also projects underway, including CorPower and WaveRoller, which plan to test small farms of up to 2 MW in the coming years. In addition, an auction is underway for the installation of up to 2 GW of offshore renewable energy by 2030, supported by the Portuguese Government, with plans to expand to 10 GW thereafter. Wave energy has not yet been commercialised on a large scale. Among the challenges it faces in becoming a fully-fledged market is the need for a more comprehensive understanding of the design and optimisation of wave energy farms. Thus, optimising the layout of wave energy farms (ODPEO) is essential to maximise energy production, ensure cost competitiveness, minimise the impact on the marine environment, avoid conflicts over the use of marine space, and exploit potential synergies with other forms of MRE, such as offshore wind, and/or niche market applications, from green hydrogen to desalination. Despite its benefits, ODPEO presents a significant challenge in obtaining accurate estimates of both energy absorption and park effects (i.e., changes in local wave conditions caused by complex wave-converter interaction). This requires the use of computationally demanding numerical physics models that resolve these interactions. As such, previous research on ODPEO has been predominantly limited to cases with a small number of wave energy converters (WECs), sea states, and optimisation objectives. This has restricted its applicability to large-scale farms and prevented a comprehensive understanding of the relationship between the key variables that influence optimal configurations. The OPTIWEF.AI project aims to fill knowledge gaps by providing insight into the complex interaction of factors influencing the optimal design of CEO parks. To this end, a new optimisation algorithm will be developed. This will take into account key factors in the design of CEO parks, including energy absorption, cost-benefit ratio and potential synergies with other RES (e.g., offshore wind). The algorithm will integrate evolutionary computation (EC) and machine learning (ML) solutions with wave propagation models
(e.g., SNL-SWAN) to accurately estimate energy absorption and farm effects. The final trained and validated model is expected to deliver a significant increase in computational efficiency compared to conventional methods and the pioneering application of the algorithm to large-scale parks. In addition, the project contributes to a broader understanding of ODPEO by providing an open-source database with a wide range of case studies, including for Portugal. This tool will not only facilitate the reproducibility of the work, but also provide researchers and companies with a benchmark against which they can compare their algorithms, thus contributing significantly to the advancement of the field. Finally, the project will provide an initial framework of recommendations and guidelines, identifying the main variables that influence ODPEO and the ideal settings for the algorithm. In summary, the OPTIWEF.AI project aims to improve understanding of CEO park design by providing an innovative optimisation tool. This will increase the accuracy and computational efficiency of ODPEO, helping CEO park designers to improve their projects at a relatively low cost, making them more competitive. In general, the OPTIWEF.AI project has two main scientific contributions:
Expanding knowledge about the intricate interaction between key variables to achieve an optimal design of the CEO park layout;
Expanding knowledge about the application of AA to predict energy absorption and park effects.
These contributions respond to two fundamental strategic needs: To provide a versatile and robust tool for the design and optimisation of CEO parks, thereby increasing the competitiveness of the wave energy sector;
To facilitate the transition to a carbon-neutral and self-sufficient energy system, contributing to long-term environmental and social sustainability.