双馈感应电机最大功率跟踪鲁棒滑模控制设计

变速恒频风力发电技术已广泛用于实际风机中,其中一个经典问题即最大功率跟踪(Maximum Power Point Tracking,MPPT)。针对双馈感应电机(Doubly-Fed Induction Generator,DFIG)设计了一款基于扰动观测器的滑模控制(Perturbation Observer Based Sliding-Mode Control,POSMC)来实现MPPT。所提控制器将扰动观测器和滑模控制(Sliding-Mode Control,SMC)相结合,从而大幅提高了DFIG的鲁棒性。首先,应用扰动观测器对系统扰动(发电机非线性、参数不确定和随机风速)进行在线估计。随后,通过SMC对该扰动估计进行实时完全补偿,从而实现不同工况下的控制全局一致性以及各类不确定环境下的鲁棒控制。最后,POSMC无需精确的DFIG系统模型,仅需测量转子角速度和定子无功功率,易于实现。进行了三个算例研究,即阶跃风速、随机风速和发电机参数不确定性。仿真结果表明,与矢量控制(Vector Control,VC)、反馈线性控制(Feedback Linearization Control,FLC)和SMC相比,POSMC在各类工况下均可捕获最大的风能并具有最强的鲁棒性。

Optimal wind power extraction of doubly-fed induction generator via perturbation observer based sliding-mode controller

Variable-speed constant-frequency wind power generation technology has been widely used in actual wind turbines. A typical issue is the Maximum Power Point Tracking (MPPT). This paper designs a Perturbation Observer Based Sliding Mode Control (POSMC) scheme for MPPT of Doubly-Fed Induction Generator (DFIG). It combines the perturbation observer and Sliding-Mode Control (SMC) to significantly enhance the robustness. Firstly, the perturbation observer estimates the perturbation (generator nonlinearities, parameter uncertainties, and random wind speed) online. Then, the perturbation estimation is compensated in the real-time by SMC, such that a global control consistency and improved robustness can be achieved under different operation conditions. Finally, POSMC does not require an accurate DFIG model while only the rotor angular velocity and stator reactive power needs to be measured, thus it is easy to implement. Three case studies are undertaken, namely, step wind speed, random wind speed and uncertain generator parameters. Simulation results show that, compared with Vector Control (VC), Feedback Linearization Control (FLC) and SMC, POSMC can capture the maximum wind power under various operation conditions and can provide the strongest robustness. This work is supported by Science and Technology Project of China Southern Power Grid Company Ltd. (No. GDKJXM20180576).