考虑风速和负荷变化场景的风电机组下垂控制设计

Wind turbines droop control design considering wind speed and load change scenarios

由于湍流风速与负荷变化的影响,风电机组参与电网一次调频过程存在自身动态稳定与支撑电网频率两者难以平衡的问题。为此,针对额定以下低风速场景,该文分析了不同风速、不同负荷扰动对一次调频下垂系数设定的影响,并结合不依赖模型的Q学习,设计了一种改进下垂控制,根据平均风速和初始频率变化率反映出的场景变化调整下垂系数。该策略通过Q学习离线寻优与模糊在线控制结合的设计方法,实现不同场景的变下垂控制。最后,利用电力系统动模实验平台,实现改进控制策略,并比较不同风速和负荷场景下的调频效果,验证该文方法的有效性。结果表明,该方法提高了风机在不同风速以及负荷变化场景下一次调频的适应能力。

Due to the influence of turbulent wind speeds and load changes,it’s difficult for wind turbines to balance its dynamic stability and supporting grid frequency when participating in primary frequency regulation process of power grid.Therefore,this paper analyzes influences of different wind speeds and load perturbations on setting of primary frequency regulation droop coefficient for low-wind speed scenarios below rated.In combination with Q learning independent of models,an improved droop control is designed to adjust droop coefficient according to scene changes reflected by average wind speed and initial frequency change rate.This strategy uses Q-learning offline optimization and fuzzy online control to realize droop control in different scenarios.Finally,an improved control strategy is realized by using dynamic model experiment platforms of the power system,and frequency regulation effect under different wind speeds and load scenarios is compared to verify the effectiveness of the proposed method.The results show that this method improves adaptive capacity of wind turbines at frequency regulation under different wind speeds and load change scenarios.