A framework for simultaneous design of wind turbines and cable layout in offshore wind

Abstract

Abstract. An optimization framework for simultaneous design of wind turbines (WTs) and cable layout for a collection system of offshore wind farms (OWFs) is presented in this paper. The typical approach used in both research and practical design is sequential, with an initial annual energy production (AEP) maximization, followed then by the collection system design. The sequential approach is robust and effective. However it fails to exploit the synergies between optimization blocks. Intuitively, one of the strongest trade-offs is between the WTs and cable layout, as they generally compete; i.e. spreading out WTs mitigates wake losses for larger AEP but also results in longer submarine cables in the collection system and higher costs. The proposed optimization framework implements a gradient-free optimization algorithm to smartly move the WTs within the project area subject to minimum distance constraint, while a fast heuristic algorithm is called in every function evaluation in order to calculate a cost estimation of the cable layout. In a final stage, a refined cable layout design is obtained by iteratively solving a mixed integer linear programme (MILP), modelling all typical engineering constraints of this particular problem. A comprehensive performance analysis of the cost estimation from the fast heuristic algorithm with respect to the exact model is carried out. The applicability of the method is illustrated through a large-scale real-world case study. Results shows that (i) the quality of the cable layout estimation is strongly dependent on the separation between WTs, where dense WT layouts present better performance parameters in terms of error, correlation, and computing time, and (ii) the proposed simultaneous design approach provides up to 6 % of improvement on the quality of fully feasible wind farm designs, and broadly, a statistically significant enhancement is ensured in spite of the stochasticity of the optimization algorithm.