海上风电场尾流协同优化方法研究

To address the energy loss caused by wake interference in offshore wind farms, a collaborative control strategy is proposed that integrates multi-dimensional parameter dynamic correction with a two-stage sequential optimization algorithm. Offshore wind farms are characterized by low turbulence intensity, small surface roughness, and significant changes in atmospheric stability, resulting in a slow wake recovery rate and a wider wake effect range, which differs substantially from onshore wind farms. Therefore, this paper proposes real-time correction of relative wind direction, turbulence intensity, and wind speed multiplier factors based on the characteristics of offshore wind farms, and designs a yaw control algorithm based on a two-stage sequential optimization strategy. Combined with a yaw anti-oscillation control strategy, it increases power generation while reducing fatigue damage to yaw bearings and blades, achieving a balance between power generation benefits and turbine longevity protection. Results show that in a scenario with 37 wind turbines, a single optimization takes approximately 0.83 minutes, power generation increases by 4.9%, and the fatigue damage increment to yaw bearings and blades is less than 1%. The system adopts a hybrid parallel architecture of OpenMP and Python multiprocessing, combined with real-time data correction technology, overcoming the shortcomings of traditional methods in terms of model accuracy, computational efficiency, and safety. The method presented in this paper provides an engineering solution for quality improvement and efficiency enhancement in large-scale offshore wind farms.