三电平双向DC-DC变换器的模型预测控制方法

Model Predictive Control Method for Three-level Bi-directional DC-DC Converter

针对传统双闭环控制动态响应慢、PI控制器控制参数选择困难的缺点,提出采用模型预测控制(MPC)方法对三电平双向DC-DC变换器进行控制。该方法通过预测模型的建立和目标函数的最小化两个环节实现对控制目标性能的优化。对传统PI控制和模型预测控制搭建仿真模型,仿真结果显示模型预测控制方法可以使直流母线电压超调量控制在0.25%以内,使蓄电池充放电电流波动范围控制在0.1%以内,由此证明模型预测控制方法可以更好地实现直流母线电压的稳定,优化蓄电池的充放电过程。同时,采用模型预测方法控制三电平双向DC-DC变换器可以实现中点电压的自平衡。

The traditional double closed loop control method for three-level bi-directional DC-DC converterintroduces the drawbacks like slow dynamic response,complex controller parameters selection. In order to solve theseproblems,a method based on model predictive control(MPC)was proposed. In this method,the prediction modelswere established and the cost function was minimized to optimize the performance of control target. The simulations fortraditional PI control method and MPC method were carried out,the results shows that DC bus voltage overshoot isreduced to less than 0.25% and the battery charging and discharging current ripple is reduced to less than 0.1% byadopting the proposed control method,which prove that the stability of DC bus voltage and the optimization of chargingand discharging process of battery have been well realized through the model predictive control. Meanwhile,thebalance of the midpoint voltage can be maintained automatically.