摘要
在五桥臂逆变器双永磁电机系统中,传统预测控制策略是一个控制周期内每台电机上等效于只作用一个幅值、方向均固定的矢量,矢量调节范围有限,系统运行时电流波动较大。为减小双电机系统中电机电流波动,采用一种三矢量预测控制策略,该策略在每个控制周期内作用两个有效矢量和一个零矢量。首先依据两个有效矢量所对应公共桥臂的开关状态将其分组,然后通过两台电机各自独立的价值函数筛选出每组中最优的矢量组合,最后比较不同组的矢量组合,从中进一步选择系统最优矢量组合,并按规则合成五桥臂电压矢量作用于双电机系统。对矢量进行分组后,该策略每个控制周期只需要计算30次价值函数,有效减小了多矢量预测控制的计算次数。实验结果表明,与传统预测控制策略相比,该策略在保持动态响应快等优点的同时,有效减小了电流波动,并具备一定的抗参数扰动性,提高了系统的稳态性能。
In the dual permanent magnet synchronous motors system fed by the five-leg inverter,only one voltage vector with fixed amplitude and direction is applied on each motor in a control cycle in the traditional predictive control strategy.The vector regulation range is limited and the motor currents fluctuate greatly.To reduce the motor current fluctuation in the dual-motor system,a three-vector-based predictive control strategy was adopted.In this method,two effective voltage vectors and one zero voltage vector were applied in each control period.Firstly,the two effective vectors were grouped according to their switch states.And then the optimal vector set of each group was selected by the cost function of each motor.Finally,the selected vector sets of different groups were compared,and the optimal vector set of the system was further selected from them.Based the optimal vector set,the five-leg voltage vector was synthesized and applied on the dual motors system.After grouping vectors,the strategy only needs to calculate the value of cost function 30 times per control cycle,which effectively reduces the number of calculations of multi-vector predictive control.Compared with the traditional predictive control strategy,the experimental results show that the strategy can effectively reduce the current fluctuations while maintaining the advantages of fast dynamic response,and has a certain resistance to parameter disturbance,which improves the steady-state performance of the system.
作者
耿强
王亮
周湛清
李新旻
王志强
Geng Qiang;Wang Liang;Zhou Zhanqing;Li Xinmin;Wang Zhiqiang(School of Electrical Engineering and Automation Tiangong University,Tianjin,300387,China;School of Artificial Intelligence Tiangong University,Tianjin,300387,China)
出处
《电工技术学报》
EI
CSCD
北大核心
2021年第1期87-95,共9页
Transactions of China Electrotechnical Society
基金
国家自然科学基金(51907142)
天津市自然科学基金(18JCQNJC74400)
天津市教委科研计划重点项目(2019ZD09)
天津市教委科研计划项目(2018KJ208)资助。
关键词
永磁同步电机
三矢量
预测控制
五桥臂逆变器
Permanent magnet synchronous motor
three-vector
predictive control
five-leg inverter