The interest in photovoltaic systems installed on modular connected floating foundations has increased over the recent years due to limited space on land. For these floating installations, knowledge on the interaction dynamics of individual floats and also connected chains of floats and wind-induced water waves is lacking. It is still challenging to estimate the actual (design) forces and possible energy efficiency losses caused by the movement of the floating photovoltaic systems, throughout the stages of the design process. Therefore, in this study, a single float and an interconnected chain of floats of a floating photovoltaic system interacting with different wave conditions are investigated in a physical model study regarding measured forces and motion of the structure. The obtained data is used to compare the individual float behaviour and the non-linear behaviour of a modular, lightweight, interconnected floating structure of multiple floats. High connector forces were observed at the centre of the float chain for float chain length to wavelength ratios of approximately unity. Furthermore, the motion of the float chain revealed similar dependencies on investigated wave parameters as the individual float. However, the change of motion along the individual modules of the float chain was found to be highly non-linear, affecting the energy efficiency along the float chain. Energy efficiency losses were estimated to reach a level of up to nearly 20% and were found to increase in the float chain setup when compared with the single float. Further energy efficiency losses varied along the float chain, with a marginal tendency of decreased energy efficiency toward the centre of the float chain. • Investigation of both single FPV float and float chain of ten interconnected floats. • Float chain forces mainly depend on mass loading, not wave parameters. • Connector peak loads showed wavelength to float chain length dependency. • Energy efficiency was correlated based on pitch motion.

Accurate short-term significant wave height forecasting is critical for coastal navigation, offshore energy operations, and hazard preparedness. However, operational wave forecast skill is often limited by uncertainty in wind forecasts. This study develops attention-based Long Short-Term Memory (LSTM) networks that explicitly address wind forecast uncertainty through noise-augmented training strategy and probabilistic modeling. Using wind and wave observations from the Western Long Island Sound (WLIS) buoy and operational North American Mesoscale (NAM) wind forecasts at Sikorsky Station, we characterize wind forecast errors and incorporate their statistical properties in model design. Three model variants are evaluated: (1) a perfect-future model trained with ideal future winds, (2) a noise-augmented model trained with controlled wind perturbations to enhance robustness, and (3) a probabilistic model that outputs both mean forecasts and uncertainty estimates. Results show that noise-augmented training reduced RMSE and MAE by approximately 7% compared to the perfect-future model and enhanced robustness to a broader range of wind forecast uncertainties. Furthermore, it led to smoother and more consistent performance across lead times. The probabilistic model achieved comparable accuracy while producing well-calibrated uncertainty estimates. Both models outperform a baseline relying solely on past buoy observations at all forecast horizons. These findings demonstrate that explicitly incorporating wind forecast uncertainty significantly improves wave height forecasting accuracy and reliability. The proposed framework offers a practical and robust pathway for operational wave forecasting systems under imperfect atmospheric forcing, enabling better-informed risk management and coastal decision-making. • Attention-based LSTM models using future winds outperform past-only baselines. • Noise-augmented training improves robustness to large wind forecast errors. • Probabilistic modeling provides calibrated uncertainty for risk-aware decisions. • Under forecast winds, proposed models reduce RMSE and MAE by about 7%.

Small, non-interconnected island systems are at the forefront of the energy transition but their isolated condition makes them exposed to the natural variability of renewable resources. This work develops a static robust optimization framework to design multi-technology portfolios that explicitly account for the inter-annual variability of the wave climate. The methodology couples the EnergyPLAN simulation tool with a MATLAB-based NSGA-II algorithm. The design vector includes the installed capacities of onshore and offshore wind, photovoltaics, two different wave energy converters (Pelamis and CorPower) and a battery energy storage system (BESS). Robustness is assessed over a discrete uncertainty set composed of twenty years of hourly meteomarine data from Copernicus reanalysis for La Gomera (Canary Islands). Three system-level indicators are optimized in a worst-case sense: annual C O 2 emissions, a demand-generation mismatch metric ϕ , and the BESS exploitation index. A post-processing step selects all robust portfolios that satisfy a stringent emissions target, corresponding to roughly a 70% reduction with respect to the validated reference configuration. Within this low-carbon subset, C O 2 is treated as a saturated objective and the remaining trade-offs are explored in the two-dimensional ϕ - R B E S S plane, where a secondary Pareto front is identified. The resulting portfolios reveal a clear interaction between storage use and temporal balancing, with different BESS levels but a quite constant wave contribution. The framework is generic and transferable to other island systems and renewable technology combinations, providing a practical tool for integrating long-term resource variability and explicit decarbonization targets into energy system planning. • Static robust optimization of multi-RES portfolios. • Robustness evaluated over twenty years of hourly Copernicus wave climate data. • La Gomera island used as real-world testbed for energy system planning strategies. • Robust low-carbon portfolios consistently retain wave energy in the RES mix. • Results demonstrate the strategic value of robust wave energy inclusive portfolios.