风电机组RBF神经网络PID独立变桨控制研究  

Wind Turbine Individual Pitch Control Based on RBF Neural Network PID

作　　者：田猛[1] 周腊吾[1] 陈浩 韩兵[1] 

机构地区：[1]湖南大学电气与信息工程学院,湖南长沙410082 [2]湖南世优电气股份有限公司,湖南湘潭411100

出　　处：《电力电子技术》2015年第12期70-72,共3页Power Electronics

基　　金：国家国际科技合作专项(2011DFA62890)~~

摘　　要：针对额定风速以上,风切变、风剪切和塔影效应引起的不平衡载荷进行优化控制,提出一种基于径向基函数(RBF)神经网络比例积分微分(PID)变桨控制方案。通过Park坐标变换与解耦的方法,将风轮不平衡载荷分解成俯仰力矩与偏航力矩,利用经典控制理论实现了风电机组独立变桨控制设计:通过RBF网络在线学习优化:PID控制器的参数,从而对塔架和桨叶上的不平衡载荷进行优化控制。在GH.Blade和Matlab中搭建3 MW风电机组联合仿真模型进行仿真。最后在3 MW风电机组变桨测试平台上进行现场实验,实验结果表明,提出的变桨控制方案能有效减小塔架与叶片等关键部件的载荷,而且提高了变桨控制系统的鲁棒性。For more than the rated wind speed,wind shear,wind shear and unbalanced load caused by the tower shadow effect optimal control,a pitch control scheme is proposed based on radial basis function（RBF） neural network proportional integral distortion（PID）.By park coordinate transformation and decoupling method, the wind wheel unbal- anced load down into pitching moment and yawing moment, classical control theory independent wind turbine pitch control design is used.RBF network through online learning to optimize PID controller parameters ,enabling the unbal- anced loads on the tower and the blades to optimize control.GH-Blade and Matlab 3 MW wind turbine combined simulation model is built, the experimental field is excuted in a 3 MW wind turbine test pitch platform.The test re- suits show that the pitch control scheme proposed in this paper can effectively reduce the load of the tower and blades and other key components, but also improve the pitch control system robustness.

关 键 词：风电机组 独立变桨控制 径向基函数 

分 类 号：TM315[电气工程—电机]