基于模型预测控制算法的NPC三电平逆变器

NPC Three-Level Inverter Based on Model Predictive Control Algorithm

应用于二极管箝位型(neutral point clamped,NPC)三电平逆变器的空间矢量脉宽调制(space Vector pulse width modulation,SVPWM)方法,由于零矢量参与调制或滚动优化,无法有效降低共模电压(commonmode voltage,CMV),且运算时间长。针对上述问题,提出一种综合考虑中点电位、开关频率和共模电压的模型预测控制算法。首先,建立NPC三电平逆变器的负载电流、中点电位离散化数学预测模型。然后,构建中点电位、开关频率、CMV的多约束目标函数,滚动筛选最优矢量作用NPC三电平逆变器,降低共模电压且减少运行时间。最后,在MATLAB/SIMULINK平台搭建仿真模型,并对算法的运行性能进行验证。

The space vector pulse width modulation(SVPWM) method applied to the neutral point clamped(NPC)three-level inverter cannot effectively reduce the common mode voltage(CMV) due to the zero vector participating in the modulation or rolling optimization, and the operation time is long. To solve the above problems, a model predictive control algorithm considering neutral point potential, switching frequency and common mode voltage is proposed in this paper. Firstly, the discrete mathematical prediction model of load current and neutral point potential of NPC three-level inverter is established. Then, the multi constraint objective functions of neutral point potential, switching frequency and CMV are constructed, and the optimal vector is selected to act on NPC three-level inverter, reducing common mode voltage and running time. Finally, the simulation model is built on matlab/simulink platform, and the performance of the algorithm is verified.