Reducing wind energy costs is a crucial goal to promote renewable energy and reduce dependencies on fossil fuels, both to fight climate change and to increase energy independence. Two main optimization problems arise during the design of offshore wind farms: the Wind Farm Layout Optimization (WFLO) problem and the Wind Farm Cable Routing (WFCR) problem. The goal of WFLO is to find the optimal turbine locations within a given area to maximize revenues from energy production minus construction costs. The goal of WFCR is to minimize the costs of subsea cables that transfer the electricity produced by turbines. Traditionally, the two NP-hard problems have been studied separately and are solved sequentially, despite being strongly related. The output of WFLO defines the input to WFCR, and their goals are opposing: while larger distances between turbines have a positive effect on WFLO due to reduced wake losses, they negatively impact the cable costs in WFCR. Hence, we study the combined optimization problem, unifying the two problems. We propose a novel heuristic that uses a combined local search to modify solutions of WFLO and WFCR simultaneously. While the combined approach is more computationally demanding than the sequential approach, the proposed method solves industry-scale problems and improves up to 12 million Euro the Net Present Value (NPV) of wind parks, particularly when the energy density is low. Furthermore, the combined approach reduces the up-front investment costs by up to 10%, significantly reducing investment risks, and enabling a faster expansion of wind energy.
Read full abstract