基于WLS-SVM的变速风力机有效风速估计

Effective Wind Speed Estimation for Variable Speed Wind Turbines Based on WLS-SVM

控制技术是大型变速恒频风力发电机组安全高效运行的关键。有效风速的精确估计可以提高风力发电机组的控制性能。由于风力机处于三维时变的风场环境中,风速在整个风力机旋转平面上分布不同,风速计测量得到的风速和整个风力机旋转平面所受到的有效风速有很大差别,因此有效风速是不能直接测量的。采用加权最小二乘支持向量机(WLS-SVM)对有效风速进行预估。仿真结果表明,该算法具有很高的估计精度,有效风速估计值能较好地跟踪有效风速的变化趋势。

Automatic control is crucial for the efficiency and reliability of a large-scale variable speed and variable pitch wind turbines. Accurate estimation of wind speed can improve control capabilities of wind turbines. Since the rotor is subject to a spatially distributed wind field which varies in time, and wind speed is different over wind turbine disc, there is considerably large discrepancy between the measured wind speed by an anemometer and effective wind speed experienced by wind turbine disc. The effective wind speed estimation is obtained based on weighted least square support vector machine. Simulation results show support vector machine is an effective method for effective wind speed estimation, which can exactly track the trend of wind speed and has high estimation precision.