提高双有源桥变换器动态性能的预测电流模式控制
Predictive Current Mode Control of Dual-active-bridge Converter to Improve Its Dynamic Performance
俞宏洋 1李辉1
作者信息
- 1. 上海电力大学自动化工程学院,上海 200090
- 折叠
摘要
提出一种预测电流模式控制算法,以提高双有源桥变换器在负载切换时的动态性能.首先,分析了双有源桥变换器开关周期内的电压、电流波形,推导出预测方程,在下一控制周期立即根据预测方程更新控制量,使得电流响应带宽接近开关频率,因此提高输出电压动态响应速度.然后,推导了输出电流瞬时值和电感电流瞬时值的代数关系,使用输出电流代替受开关振荡影响的电感电流以提高抗噪声能力.其次,建立了双有源桥的降阶模型,推导了传统双极点双零点2P2Z(two-pole tow-zero)电压模式控制器并和所提算法进行比较.最后,通过仿真和多种工况的实验结果表明,相比2P2Z电压模式控制,所提算法降低了变换器闭环输出阻抗,使得电感电流能够在一个控制周期内快速响应,极大提高了输出电压在负载切换时的动态性能.
Abstract
A predictive current mode control algorithm is proposed in this paper to improve the dynamic performance of a dual-active-bridge ( DAB ) converter during load switching.First,the voltage and current waveforms of the DAB converter during a switching cycle are analyzed,and the prediction equation is derived.In the next control cycle,the control value is updated immediately according to the prediction equation,so that the current response bandwidth is close to the switching frequency,thereby improving the output voltage dynamic response speed.Then the algebraic relationship between the instantaneous value of output current and that of inductance current is also deduced,and the output current is used to replace the inductance current affected by the switching oscillation to improve the noise immunity.Second,a reduced-order model of DAB is established,and the traditional two-pole two-zero( 2P2Z ) voltage mode controller is deduced,which is further compared with the proposed algorithm.Finally,simulation results and experimental results under various operating conditions show that compared with the 2P2Z voltage mode control,the proposed algorithm reduces the closed-loop output impedance of the converter,enables the inductance current to respond quickly in a control cycle,and greatly improves the dynamic performance of output voltage during load switching.
关键词
双有源桥变换器/预测电流模式控制/输出阻抗/动态性能Key words
Dual-active-bridge ( DAB ) converter/predictive current mode control/output impedance/dynamic performance引用本文复制引用
出版年
2024
国家科技期刊平台
NSTL
万方数据