| 引用本文: | 张珍珍,邹见效.基于神经网络的半直驱永磁同步风力发电机组有功功率控制[J].电力自动化设备,2013,33(9): |
| ZHANG Zhenzhen,ZOU Jianxiao.Active power control based on neural network for half-direct permanent-magnet wind power generator[J].Electric Power Automation Equipment,2013,33(9): |
|
| 摘要: |
| 针对半直驱风力发电机组，提出了一种基于神经网络的有功功率控制方法，分别对变桨控制器和转速控制器进行设计，实现机组在额定功率控制模式和非额定功率的恒功率控制模式的有功功率控制。着重研究了转速控制算法，采用线性反馈化方法对系统模型进行处理，然后根据动态面控制算法设计转速控制器，并利用RBF神经网络逼近特性避免由于对具有非线性特性风力发电机组线性化而导致的模型不确定性问题。最后基于MATLAB／Simulink平台，将该功率控制方法应用于2 MW半直驱永磁同步风力发电机组，验证了所提方法的有效性。 |
| 关键词: 半直驱永磁同步风力发电机组 有功功率 反馈线性化 动态面控制 神经网络 风电 功率控制 |
| DOI： |
| 分类号: |
| 基金项目:国家自然科学基金资助项目（６１２０１０１０）；中央高校基本科研业务费专项资金资助项目（ZYGX2010J081） |
|
| Active power control based on neural network for half-direct permanent-magnet wind power generator |
|
ZHANG Zhenzhen, ZOU Jianxiao
|
|
Institute of Automation，University of Electronic Science and Technology of China，Chengdu 611731，China
|
| Abstract: |
| A method of active power control based on neural network is proposed for half-direct permanent-magnet wind generator. The pitch controller and the speed controller are separately designed to realize the active power control in both the rated power mode and the non-rated constant power mode. The rotational speed control algorithm is emphasized. The feedback linearization is applied to system model processing and the dynamic surface control algorithm is used in rotational speed controller design，where RBF neural network is employed to approximate the unknown nonlinear functions and avoid the uncertainty of model. Based on MATLAB／Simulink platform，the proposed control method is applied to a 2 MW half-direct permanent-magnet wind generator，which verifies its effectiveness. |
| Key words: half-direct permanent-magnet wind generator active power feedback linearization dynamic surface control neural networks wind power power control |
