摘要
随着风电机组基础结构的不断增大,风电机组的控制方法面临新的机遇和挑战,而遥感测量技术的发展给传统风电机组控制策略提供一个新的研究领域。该文提出了基于激光雷达(light detection and ranging,LIDAR)辅助风电机组模型预测控制方法来实现控制系统对风速扰动的前馈补偿控制。首先根据叶素动量理论分析风电机组的载荷情况和LIDAR预测风轮迎风面的有效风速,利用扩展卡尔曼滤波重建噪声状态的非线性风电机组模型的未知状态,对预测时域状态值的进行预测实时处理,以求解最小目标函数获取系统当前时刻的最优化控制,使得系统参考轨迹和未来输出值之间差值实现最小化。最后,通过进行风电机组传统控制方法与LIDAR辅助线性模型预测控制、非线性模型预测控制的对比实验,证明LIDAR与模型预测控制相结合的控制方式能在一定程度上提高大型风电机组的风能利用系数,缓解风电机组的疲劳载荷。
With increasing size of large wind turbine, control methods faces new opportunities and challenges, the development of remote sensing technology provides a new research field of the traditional control strategy. This paper focused on the design of light detection and ranging(LIDAR) assisted model predictive control(MPC) of wind turbine, it achieved wind speed disturbance feedforward compensation control. First, the blade element momentum(BEM) theory have analyzed the wind turbine loads and LIDAR forecast wind speed of the rotor windward side, used of extended Kalman filter reconstruct unknown nonlinear wind turbine model for prediction horizon state values real-time processing, it solved the minimum objective function to get the current system time of the optimal control strategy, to minimize the reference trajectory and the output value. Finally, the experiment of traditional control method comparative with LIDAR assisted LMPC and NMPC, the results show that combination of LIDAR and MPC can improve power coefficient of large wind turbines and mitigate the fatigue load of wind turbine.
出处
《中国电机工程学报》
EI
CSCD
北大核心
2016年第18期5062-5069,5131,共8页
Proceedings of the CSEE
基金
科技部国际合作项目(2011DFA62890)~~
关键词
风电机组
模型预测控制
二次规划
扩展卡尔曼滤波
激光雷达
等效疲劳载荷
wind turbine
model predictive control(MPC)
quadratic programming
extended Kalman filter
light detection and ranging(LIDAR)
equivalent damage loads(DELs)