A two-stage form-parameter decoupled optimization framework for mooring system design of floating offshore wind turbines

Mooring systems significantly influence the dynamic performance, station-keeping capability, and economic viability of floating offshore wind turbines (FOWTs). Existing optimization studies primarily focus on parameter tuning within predefined mooring configurations, potentially overlooking superior solutions at the configuration level. To address this limitation, a two-stage form–parameter decoupled optimization framework is proposed for FOWT mooring system design. In Stage I, representative mooring forms with different restoring mechanisms are evaluated and screened considering dynamic performance, mooring-component material cost, safety, and installation feasibility under operational and extreme environmental conditions. A hybrid configuration incorporating elastic ropes and clump weights is identified as the preferred scheme, achieving a 15.5% mooring-component material cost reduction compared with a conventional all-chain baseline. In Stage II, a feasibility-aware deep neural network (DNN) surrogate model is coupled with the NSGA-III algorithm to perform multi-objective optimization of key mooring parameters. A steel–concrete hybrid semi-submersible platform integrated with the NREL 5 MW reference turbine is adopted as the case study. Results show that the optimized compromise solution reduces mooring-component material cost, maximum surge, pitch motion, and peak mooring tension by 56%, 29%, 24%, and 13%, respectively, while maintaining adequate safety margins. The proposed framework provides an efficient numerical methodology for preliminary design of FOWT mooring systems.