适用于提高T型三电平储能变流器功率响应特性的模型预测控制算法

摘要

Model predictive control ( MPC ) has been widely used in battery energy storage system. During each switching cycle of conventional neutral point clamped( NPC) three-level energy storage converter, its output voltage jumping can not exceed the half of the DC bus voltage, which will seriously affect the power response characteristics of the converter MPC. Therefor, this paper proposes the MPC algorithm based on T-type three-level converter for energy storage. Under the premise of ensuring the active power, reactive power and neutral point potential deviation of the converter, we adopt traversing method to obtain the output voltage vector with the minimum switching frequency of the converter, which gives full play to the advantages of no output voltage jump limitation of T-type three-level topology converter. Finally, we use geometric analysis method to compare the power control characteristics between T-type topology and NPC topology converters for energy storage. The analysis results show that the proposed MPC algorithm can greatly improve the power control characteristics of T-type three-level converter for energy storage.%模型预测控制(model predictive control,MPC)已经广泛应用于电池储能系统.在传统中点钳位式(neutral point clamped,NPC)三电平储能变流器的每个开关周期内,其输出电压跳变不能超过直流母线电压的一半,该特点将严重影响变流器模型预测控制的功率响应特性.为此,提出一种基于T型三电平储能变流器的模型预测控制算法.在确保变流器输出有功功率、无功功率及中点电位偏差一定的前提下,采用遍历法寻求使变流器开关频率最低时的输出电压矢量,充分发挥T型三电平拓扑输出电压跳变不受限制的优势.采用几何分析法对比T型拓扑与NPC拓扑储能变流器的功率控制特性,分析结果表明,所提模型预测控制算法极大地提高了T型三电平储能变流器的功率控制特性.