基于广义预测控制的风电场调频控制策略研究

为降低通信延迟对风电参与二次调频时频率稳定性的影响,提出一种基于系统辨识理论和广义预测控制算法控制风电机组参与电力系统二次调频的控制策略。首先,建立考虑通信延迟的单区域互联电力系统二次调频模型。其次,通过利用变遗忘因子递推最小二乘算法（FFRLS）进行在线辨识,建立单输入单输出被控系统的差分方程的受控自回归滑动平均模型（CARMA）,将系统频率偏差和风电场功率指令作为输出变量和控制变量设计广义预测控制器。最后,在不同延时参数以及参数摄动情况下对所提控制策略进行仿真验证。结果表明：所提控制策略在满足系统频率稳定性的同时,使得系统上升时间缩短,超调量大幅减小。

SECONDARY FREQUENCY REGULATION CONTROL OF WIND FARM BASED ON GENREALIZED PREDICTIVE CONTROL

In order to reduce the impact of communication delay on frequency stability when wind power participates in secondary frequency regulation,this paper proposes a control strategy based on system identification theory and generalized predictive control algorithm to control wind turbines participating in the secondary frequency regulation of the power system. First,the second frequency regulation model of single-area interconnected power system considering communication delay is established. Second,through the use of forgetting factor recursive least squares （FFRLS） algorithm for online identification,a controlled auto regressive moving average （CARMA） model of the difference equation of the controlled system with single input and single output is established. The system frequency deviation and wind farm power commands are used as output variables and control variables to design a generalized predictive controller. Finally,the predictive control strategy is simulated and verified under different delay parameters and parameter perturbations. The result shows that the control strategy not only satisfies the frequency stability of the system,but also shortens the rise time of the system and greatly reduces the amount of overshoot.