基于神经滑模控制的风机变桨系统控制与研究
李 洁，孟德洲
(内蒙古科技大学 信息工程学院,内蒙古 包头 014010)
关键词：独立变桨控制，RBF神经网络，滑模控制
中图分类号：TP29 文献标志码：A
摘要：以风力发电系统为背景，由于风速波动及风机叶片扫掠面积上风资源的不均匀分布，风机叶片的变桨需要根据自身风况单独控制，即实现独立变桨。提出一种基于RBF(径向基函数)神经网络的滑模控制策略，优化了风机变桨距的控制方法，提高了风力发电系统的稳定性。将算法植入10kW风机缩比模型实验台，控制伺服变桨电机，实验台模拟运行结果表明，RBF神经网络滑模控制策略能够改善变桨控制的效果，提高系统的鲁棒性。
Wind Turbine Pitch System Control and Research Based on Neural-Sliding Mode Control
Li Jie, Meng Dezhou
(Information Engineering School, Inner Mongolia University of Science and Technology, Baotou 014000, China)
Keyword:independence pitch control, RBF neural network, sliding mode control
Abstract: For the wind power generation system, wind speed is fluctuant and wind resource distribution is uneven in the wind turbine blade sweep area, so the wind turbine blade needs to be controlled separately according to its own wind condition to achieve independence pitch control. A method was presented based on RBF (Radial Basis Function) neural network sliding mode control strategy, which optimized the turbine pitch control methods and improved the stability of wind power generation system. The algorithm is implanted in the experimental platform of 10kW wind turbine scale model to control the servo motor. The simulation run results show that, the RBF neural network sliding mode control strategy can improve the effect of the variable pitch control and improve the system robustness.