基于滤波电感辨识及延时补偿的并网逆变器模型预测控制

Model Predictive Control for Grid-connected Inverter Based on Filter Inductance Identification and Delay Compensation

针对三相并网逆变器采用传统模型预测控制时,因模型失配造成输出电流畸变率大、谐波含量高、并网效率低的问题,文章提出了一种基于滤波电感辨识及延时补偿的模型预测控制方案,将在线辨识的滤波电感和延时补偿的电压电流应用到三相并网逆变器模型预测控制系统中,并在MATLAB/Simulink环境下进行了仿真,仿真结果表明:基于滤波电感在线辨识及延时补偿的模型预测控制具有很好的动、静态特性,输出电流正弦度好、谐波含量少,验证了方案的可行性和正确性.

The traditional model predictive control（MPC） for three-phase grid-connected inverter has its disadvantages, such as large distortion rate of output current, high harmonic content and low grid effi- ciency due to the model mismatch. In order to solve those problems, a MPC scheme based on filter induc- tor identification and delay compensation is proposed. The filter inductor with on-line identification togeth- er with the voltage and current with delay compensation are applied to the MPC system of the three-phase grid-connected inverter. The simulation is carried out at the MATLAB/Simulink environment. The results show that the proposed MPC scheme has good dynamic and static characteristics, and the harmonic content in the output current is very low, so the feasibility and correctness of the proposed MPC scheme are veri- fied