电动汽车双向充电桩有限集模型预测控制研究

Predictive Control Study of Finite Set Model for Electric Vehicle Bi-Directional Charging Pile

为提高电动汽车双向充电桩并网电流质量和抗干扰能力,并抑制其电流谐波,结合模型预测控制特性分析,提出了电动汽车双向充电桩的有限控制集模型预测控制(FCS-MPC)方法。首先,在同步旋转坐标系下建立了变流器的离散数学模型。然后,在传统模型预测控制基础上设计了输出电流的(k+2)时刻预测值,再将预测值和采样时刻的电流误差以及直流(DC)母线电压的误差作为价值函数的控制因子,寻求价值函数的最小解,以达到最优控制效果。仿真结果表明,相比传统比例积分(PI)前馈控制策略,FCS-MPC有效提高了DC母线电压的稳定性和抗干扰性,并将网侧电流控制在更低的总谐波失真水平下。考虑到组成价值函数控制因子的电压、电流、开关频率等物理量直观且易于采集,FCS-MPC方法可推广应用于同类的电力电子变换器。

A finite control set-model predictive control(FCS-MPC)method for electric vehicle bi-directional charging pile is proposed to improve the grid-connected current quality and anti-interference capability and suppress its current harmonics,combined with model predictive control characteristics analysis.Firstly,a discrete mathematical model of the converter is established in the synchronous rotating coordinate system.Then,the predicted value of(k+2)moments of the output current is designed based on the traditional model predictive control,and then the predicted value and the error of the current at the sampling moment and the error of the direct current(DC)bus voltage are used as the control factors of the value function to seek the minimum solution of the value function to achieve the optimal control effect.Simulation results show that compared with the traditional proportional integral(PI)feedforward control strategy,FCS-MPC effectively improves the stability and immunity of the DC bus voltage,and the network-side current is controlled at a lower level of total harmonic distortion.Considering that the physical quantities such as voltage,current,and switching frequency that make up the value function control factor are intuitive and easy to collect,the FCS-MPC method can be extended to be applied to similar power electronic converters.