电池储能型模块化多电平变换器的混合模型预测控制方法

将模块化多电平变换器(MMC)作为电池储能系统(BESS)的并网变换器,可在实现高压并网的同时兼具控制的灵活性。针对电池储能型模块化多电平变换(B-MMC)系统,提出一种可有效减小计算量的混合型模型预测控制(H-MPC)方法。该H-MPC方法由PI控制和MPC组成。其中,PI控制部分用于求取满足交流电流输出和环流控制要求的子模块接入个数;MPC则负责共模电压(CMV)抑制,对子模块接入个数进行适当调整。结合子模块接入个数与电池组荷电状态(SOC)的排序结果,即可产生具体开关信号。针对不同应用场合,PI控制部分和MPC的控制目标选取要更为灵活。以环流控制为例,对其包含于MPC部分的情况进行简要分析。最后通过Matlab/Simulink仿真和实验,验证了该方法的正确性与有效性。

A Hybrid Model Predictive Control Method for Modular Multilevel Converter of Battery Energy Storage System

Modular multilevel converter (MMC) is deemed to be an attractive topology to realize grid-connection of battery energy storage system (BESS), due to its high output voltage level and great control flexibility. In this paper, a novel hybrid model predictive control method (H-MPC) is employed to the battery integrated modular multilevel converter (B-MMC), which can attenuate the computation burden of processors effectively. PI control part and MPC control part were included in the proposed H-MPC. Firstly, the inserted numbers in each arm were solved in PI control part, based on ac-side currents and circulating currents equations. MPC control part was responsible for common mode voltage (CMV) elimination, which would further modify the inserted numbers properly. Then, the switching signals were generated based on the inserted number and sorted state of charge (SOC) of battery packs. In fact, it is flexible to select different control objects for both PI and MPC parts. A simple analysis, where circulating currents control was included in MPC control part, was also proposed. Finally, Matlab/Simulink and experimental results verified the performance of the proposed method.