基于神经元PID的风力发电机组独立变桨控制  

Individual Variable Propeller Control of Wind Turbine Based on Neural Network PID

作　　者：郑宇[1] 

机构地区：[1]西华大学电气信息学院,四川成都610039

出　　处：《水电能源科学》2012年第2期151-154,163,共5页Water Resources and Power

摘　　要：针对传统统一变桨距控制策略存在的不足,基于风力机空气动力学原理,提出了基于神经元PID的独立变桨距控制策略,将测量的桨叶根部My方向载荷通过神经元PID控制算出d、q轴的桨距角,经反Park变换得到3个桨叶的附加桨距角,将其与统一变桨距角相加作为独立桨距角的设定值。并借助Fast软件平台以2MW变速变桨风力发电机组为例,仿真比较了独立变桨距控制策略与统一变桨距控制策略。结果表明,独立变桨距控制策略能有效保证在额定转速下机组输出功率稳定,且能有效降低风力发电机组各零部件的疲劳载荷。In view of shortcomings of traditional unified variable propeller control strategy,based on the aerodynamics principle of wind turbine,this paper proposes neural network PID control strategy to accomplish the individual variable propeller control of wind turbine.According to measure load of propeller shank in direction My,the propeller angle between d and q axis is calculated by using neural network PID control.And additional propeller pitch angles of three propellers are obtained with inverse park transformation.Then the setting value of individual propeller pitch angle is got by adding the unified variable propeller pitch angle.Taking 2 MW variable speed variable propeller wind turbine for an example,this control strategy is simulated on the Fast software platform.Compared with the traditional method of propeller control strategy,the results show that the proposed control strategy can maintain a better power regulation at the rated speed and reduce fatigue load of components for wind turbine effectively.

关 键 词：神经网络 风力发电机组 独立变桨控制 桨叶根部载荷 Fast软件 

分 类 号：TM315[电气工程—电机] TK83[动力工程及工程热物理—流体机械及工程]