基于系统模型在线估计的微网频率预测控制

Model Predictive Control-Based Frequency Control with Recursively Estimated System Model for Microgrids

频率控制是保证微网安全稳定运行的一项重要措施。在实际微网中,由于新能源出力的强波动性,微网系统参数和运行状态时刻发生变化,传统基于固定参数模型设计的频率控制策略自适应能力较差,无法实现系统频率的实时最优控制,为此提出了一种基于系统模型在线估计的微网频率预测控制方法。首先,基于实时量测数据,采用动态模式分解法,实时辨识微网系统参数,更新系统模型;其次,基于更新的系统模型,结合模型预测控制方法,设计微网频率控制策略,实现微网频率的实时最优调节;最后,在Matlab/Simulink上搭建微网系统对所提的方法进行仿真验证,结果表明所提方法能正确识别系统状态参数,并能对微网频率进行实时最优控制。

Frequency control is important to ensure safe and stable operation of microgrids.In an actual microgrid,owing to the variable generation of distributed energy resources,the microgrid system parameters and operating conditions change continuously.The traditional,frequency-control strategy designed based on the fixed-parameter model has poor self-adaptive capability and cannot realize real-time optimal control of the system frequency.In the new,real-time measurement data-based design,the dynamic mode-decomposition method is used to update the system model by identifying the system parameters in real time.The microgrid frequency-control strategy was designed to realize real-time,optimal,microgrid frequency regulation based on the updated system model and combined with the predictive-control method.Finally,the proposed method was simulated and verified through a detailed microgrid system using MATLAB/Simulink software.Simulation results indicate that the proposed method can correctly identify the system-state parameters and effectively control the microgrid frequency.