MMC换流器虚拟子模块分层逆向模型预测控制

Hierarchical Reverse Model Predictive Control of MMC Converter Virtual Sub Module

传统有限集模型预测控制应用于MMC系统时存在权重因子难以精确计算,运算负荷过大等问题。文中提出一种分层逆向模型预测控制方法,通过逆向预测网侧电压以及桥臂不平衡压降计算最优导通子模块个数,实现交流电流控制和环流抑制。无需计算系统全部子模块投切状态省去权重因子设计过程降低循环预测次数,提高了控制器的动态性能。同时运用改进的虚拟子模块循环映射方法在均压环节引入换位裕度,降低电压排序次数减少器件开关频率和控制器计算负荷增加系统稳定性。在Simulink中搭建MMC系统进行控制策略的仿真,仿真结果显示在一个预测周期内控制器预测次数降低了16%,MMC子模块均压效果明显且系统动态性能良好,验证了所提方法的正确、有效性。

Aiming at the problems of poor robustness,complex calculation and low realizability when the traditional control theory is applied to MMC system,a hierarchical inverse model predictive control method is proposed in this paper. Without calculating the switching state of all sub modules in the system,the optimal number of conduction sub modules is calculated by inversely predicting the network side voltage and bridge arm unbalanced voltage drop,so as to reduce the number of cycle prediction and improve the dynamic performance of the controller. At the same time,the improved virtual sub module cyclic mapping method is used to introduce transposition margin in the voltage equalizing link,The switching frequency of power devices is reduced,the switching loss is reduced,and the system stability is increased. Finally,the MMC system is built in Simulink to simulate the control strategy. The simulation results show that the prediction times of the controller are reduced by16% in one prediction cycle,the voltage equalization effect of MMC sub module is obvious,and the system dynamic performance is good.