Hierarchical wind farm layout optimization considering different constraint conditions

Low prediction accuracy of wake model for wind farm layout and the impact of different constraint regions on the optimization results hinder the accurate prediction of the wind turbine number and minimize the energy cost. Based on an improved 3D wake model, a 2 km × 2 km wind farm layout is optimized using a hierarchical optimization strategy by taking different constraint conditions into account. The coordinate method is used to perform the second-order optimization on the first-order optimization results obtained by the grid method. It is shown that compared to the Jensen wake model, an improved 3D wake model is more reasonable to optimize the wind farm layout. Considering the convergence speed and the power cost, when the constraint interval is l_\textimit = 150\;\mathrmand\; 200 \;\mathrmm , the second-order algorithm has a faster iterative speed, and the power cost is 0.8498/W, which is 3.33% lower than the first-order optimization. The optimized layout shield light on the wind turbines installation.