三相并网逆变器的鲁棒模型预测电流控制

Robust Model Predictive Current Control for Three-phase Grid-connected Inverter

模型预测控制因具有概念简单、动态响应快、跟踪性能良好等优点,在电力电子领域得到广泛应用。模型预测控制在实际应用中存在参数失配的情况,易产生预测电流误差,因此提出一种基于扩张状态观测器的鲁棒模型预测控制策略。首先,以采用LCL滤波器的并网电压源逆变器为研究对象,利用加权平均电流将LCL滤波系统的传递函数由三阶降为一阶,简化模型预测控制的预测模型;然后,以加权平均电流为控制目标间接控制并网电流。仿真结果表明,基于扩张状态观测器的模型预测电流控制具有良好的动态和稳态性能,同时在解决参数失配方面具有较好的系统鲁棒性。

Model predictive control has been widely used in the field of power electronics for its advantages of simple concept,fast dynamic response and excellent tracking performance.Model predictive control has parameter mismatch in practical application,which can easily cause predictive current errors.Therefore,a robust model predictive control strategy based on extended state observer is proposed.Firstly,the gridconnected voltage source inverter with LCL filter is taken as the research object.The weighted average current is used to reduce the transfer function of LCL filter system from the third order to first one,which simplifies the predictive model of model predictive control.Then,the weighted average current is used as the control target to indirectly control the grid-current.The simulation results show that the model predictive current control based on extended state observer has good dynamic and steady-state performance,and has good system robustness in solving parameter mismatch.