基于DE优化系统辨识的风力机叶片自校正PID振动控制

Vibration control of wind turbine blades based on the self-tuning PID control and differential evolution algorithm

智能叶片是风力机叶片智能控制的主要方向,它能根据叶片的振动情况,利用控制算法调整叶片尾端智能驱动器的驱动力,使叶片处于最小振动状态,关键问题是在复杂运行环境下叶片系统参数发生突变时,如何设计基于最优辨识的控制规则,从而施加最优控制驱动力。对此,以风力机叶片的振动位移为控制目标,利用差分进化算法(DE)优化传统递推最小二乘法辨识的遗忘因子,得到基于最优系统辨识的叶片最佳间接自校正PID控制规则,采用此控制方法对系统模型建立Matlab文本与Simulink相结合的联合仿真模型并进行控制。仿真结果表明,在经差分进化优化辨识后的间接自校正PID控制器可以很好的减小系统参数突变下风力机叶片的挥舞位移和扭转角,可进一步提高复杂运行环境下的叶片振动控制效果。

The smart blade is the main direction of the intelligent control of wind turbine blades.The vibration of a blade can be significantly suppressed by the smart actuator installed on the trailing of the blade and furnished with a control algorithm.The key problem is the design of control rule based on the optimal identification to deal with the parameters’change of the blade system in complicated operation conditions.Aiming at the control and suppression of blade vibration,the differential evolution(DE)algorithm was adopted to optimize the forgetting factor in the traditional recursive least-square identification and an optimal self-tuning PID controller was developed based on the DE optimized identification.The proposed controller was implemented to control the system model,established in Matlab/Simulink.The simulation results show that the DE optimized self-tuning PID controller performs well in suppressing the pitch angle and plunge displacement in the condition of parameter’s change.The overall vibration control effect might be further improved by the proposed controller in complicated operation conditions.