摘要
自然界风速的多变性与风机变桨系统的迟缓性会导致风机输出功率的不稳定。为了改善风机输出功率的稳定,首先基于RBF神经网络RBFNN(radial basis function neural network),以功率差作为信号来源,设计了RBF-PID自适应变桨控制器,建立了风力机及变桨距机构仿真模型。其次,建立了2种风况模型,较好地模拟了自然界基本风况。仿真表明:在不同风况下对比常规模糊控制与PID控制,RBF-PID参数自适应方法在风速波动较大的情况下能够更好地稳定输出功率,且减小了变桨的幅值与频率,增加了风机的寿命。
The variability of natural wind speed and the sluggishness of a wind turbine pitch-regulated system lead to the instability of wind turbine output power. To solve this problem,based on a radial basis function neural network(RBFNN),an RBF-PID adaptive pitch controller is designed with power differences as the signal source,and a simulation model of wind turbine and pitch-regulated mechanism is established. Then,two kinds of wind speed models are established,which can better simulate the basic wind conditions in nature. Simulation results show that compared with the conventional fuzzy control and PID control,the RBF-PID parameter adaptive method can better stabilize the output power and reduce the amplitude and frequency of pitch under larger fluctuations of wind speed,thereby improving the service life of the wind turbine.
作者
张真源
刘国荣
杨小亮
刘科正
邓争
ZHANG Zhenyuan;LIU Guorong;YANG Xiaoliang;LIU Kezheng;DENG Zheng(Wind Power Equipment and Power Conversion Collaborative Innovation Center,Hunan Institute of Engineering,Xiangtan 411104,China;College of Electrical and Information Engineering,Hunan University,Changsha 410082,China;College of Information Engineering,Xiangtan University,Xiangtan 411101,China)
出处
《电力系统及其自动化学报》
CSCD
北大核心
2020年第5期16-23,共8页
Proceedings of the CSU-EPSA
基金
国家自然科学基金资助项目(51177040)。
关键词
径向基神经网络
变桨距
参数自适应
功率稳定
radial basis function neural network(RBFNN)
variable pitch
parameter adaptive
power stability
