Optimization of Wind Farm Layout Using Flow-Sensing Genetic Algorithm Considering Wake Detection and Local Adjustments

Wind power, as a clean and renewable energy source, has garnered substantial research interest. The optimization of wind farm layouts plays a critical role in enhancing energy conversion efficiency through the strategic placement of wind turbines. This paper proposes a novel Flow-Sensing Genetic Algorithm (FSGA) for wind farm layout optimization. The FSGA integrates a local wake effect assessment mechanism to direct turbine repositioning. Specifically, before the crossover and mutation operations in each generation, the algorithm performs an initial adjustment of the lowest-performing turbine in a given layout using a surrogate model. A flow-sensing operator is then employed to quantitatively evaluate the wake effects induced by this initially adjusted position and its neighboring candidate positions on a downstream local region. By comparing these assessment results, the position yielding the minimal downstream wake effect is selected for secondary fine-tuning, thereby further improving energy conversion efficiency. Extensive numerical experiments under diverse simulated and real-world wind conditions demonstrate that the FSGA outperforms existing state-of-the-art algorithms.