Multi-objective optimisation of a seawater reverse osmosis desalination system driven by vertical axis wind turbines: technical, economic, and environmental perspectives

• Integrates multi-objective optimisation with life cycle assessment for desalination. • Machine learning surrogate models predict reverse osmosis performance. • Evaluates 118,800 configurations for off-grid sustainable water production. • Identified optimal trade-offs between cost, water output, and GWP by TOPSIS analysis. • Optimal scenarios achieved 1.39 US$/m³ LCOW and 0.70 kg CO₂eq/m³ GWP. Previous studies demonstrated the feasibility of vertical axis wind turbine (VAWT)-driven reverse osmosis (RO) units integrated with compressed air energy storage (CAES) to manage renewable energy intermittency. However, an integrated multi-objective strategy for optimal configuration was missing. This study addresses this gap by developing and applying a novel integrated framework coupling multi-objective optimisation with life cycle assessment (LCA) to evaluate technical, economic, and environmental performance. The system was modelled using six design variables, exploring 118,800 configurations to balance annual water production, levelised cost of water (LCOW), and global warming potential (GWP). Machine learning techniques were used to develop surrogate models to identify Pareto-optimal solutions, while a TOPSIS analysis selected four scenarios representing diverse stakeholder priorities. Results revealed that the sizing and number of VAWTs are the dominant factors influencing both LCOW and GWP. The optimisation yielded an LCOW-prioritised scenario that achieved a cost of 1.39 US$/m³, while the GWP-prioritised scenario reached a footprint of 0.70 kg CO₂eq/m³. This work provides a novel methodology for the simultaneous techno-economic and environmental design of off-grid water infrastructure, establishing optimal strategies for renewable desalination to address global water scarcity.