双馈异步风力机BP神经网络空载并网控制  

Double-fed Asynchronous Wind Turbine BP Neural Network No-load Grid-connected Control

作　　者：王嘉良 WANG Jialiang(College of Energy and Electrical Engineering,Hohai University,Nanjing 211100,China)

机构地区：[1]河海大学能源与电气学院,江苏南京211100

出　　处：《机械制造与自动化》2021年第2期213-216,221,共5页Machine Building & Automation

摘　　要：为减小双馈异步风力发电机组并网时对电网产生的冲击电流,降低电压误差,实现柔性并网,在双馈异步发电机数学模型的基础上,对基于定子磁链定向矢量控制的空载并网原理进行分析,利用BP神经网络对PID控制器的参数进行最优调节,建立BP神经网络PID空载并网控制策略;在Matlab/Simulink软件中建立双馈异步风力发电机BP神经网络PID空载并网的仿真模型。通过与普通PID空载并网控制仿真结果的比较表明,该控制策略的响应速度更快,对电网电压的跟踪能力更强,精度更高,是一种良好的并网控制策略。In order to reduce the impulse current and voltage error and to realize pflexible grid connection of DFIG,the principle of no-load grid connection based on stator flux oriented vector control was analyzed through the mathematical model of DFIG.The parameters of PID controller were optimally adjusted by BP neural network.The strategy of BP neural network PID no-load grid connected control was proposed,and the simulation model of the doubly-ed asynchronous wind turbine BP neural network PID no-load grid connection was established by Matlab/Simulink software.By comparison with the ordinary PID no-load grid connected control simulation results,the new grid connected control strategy proves to be faster at response speed of the control strategy,stronger in ability to track the grid voltage and higher in degree of accuracy.

关 键 词：双馈异步发电机 风力机 空载并网 PID控制 BP神经网络 

分 类 号：TM315[电气工程—电机]