基于RBF神经网络整定PID的风力发电变桨距控制

Variable Pitch Control of Wind Power Generation Based on RBF Neural Network Tuning PID Control

为了改善变速恒频风力发电系统在恒功率输出运行区域内的动态性能,在分析系统变桨距控制研究现状的基础上,基于RBF神经网络(RBFNN)整定PID控制理论设计风力发电系统变桨距控制器,建立了风力机及变桨距机构模型,以发电机转速测量值与额定转速相比后误差为输入设计控制器。在随机风作用下对设计的RBFNN整定PID控制器进行仿真,结果表明基于RBFNN整定PID控制理论的变桨距控制器具有良好的动态性能及对风速扰动的鲁捧性,能够有效改善风力发电系统变桨距控制效果。

In order to improve dynamic performances of variable speed constant frequency （VSCF） in constant power output running area, research situation of variable pitch control is studied, based on RBF neural network （RBFNN） tuning PID control, wind turbine and variable pitch control model is built, generator rotate speed measurements and speed error as input contoller. Simulation results at random wind speeds show the controller has good dynamic performance, good robustness of anti-wind-interference, and it can can effectively improve the performance of variable pitch controL.