Unmanned aerial vehicle swarm placement optimization for offshore wind farms

Wind Turbines (WT) are large structures subject to various defects, requiring periodic maintenance. This research addresses the problem of finding the optimum number and placement of Unmanned Aerial Vehicles (UAVs) for Offshore Wind Farm (OWF) inspections, thereby improving placement efficiency and reducing the cost of automating wind farm inspection, which is preferable to human inspection as it is a hazardous task. This research formulates a new UAV placement optimization problem based on a newly defined objective function and algorithm called the Limited-Radius K-Nearest Coverage Algorithm (LRKNCA). Our approach is evaluated on two case studies - the 189 turbine Walney OWF in the UK and a simulated OWF represented by 47-randomy selected turbines from the Walney OWF. Results show a significant reduction in the number of UAVs required in both case studies compared to competing approaches in the literature, with an average placement efficiency of 79.36% and 84.12%, respectively, and a reduction in the average cost by 37.04% and 47.74%, respectively. Three optimization algorithms, Particle Swarm Optimization (PSO), Biogeography-Based Optimization (BBO), and Grey Wolf Optimizer (GWO), were deployed for evaluation, with GWO achieving the best overall performance.