摘要
传统有限状态-模型预测控制(FCS-MPC)由于控制方式直接、控制性能优越而广泛应用于功率变换系统,但其在指标函数寻优时需考虑所有开关状态所对应的电压矢量,计算量较大。针对该问题,提出一种改进的MPC算法,与传统FCS-MPC方法相比,该方法无需循环寻优,预测耗时少,易于数字化实现。在两电平脉宽调制(PWM)整流器样机上的实验结果表明了所提方法的有效性和优越性。
Conventional finite control set model predictive control（FCS-MPC） is emerging as a popular control scheme in power converter,because it takes the superior performance into account and offers a direct control strategy.However, FCS-MPC requires to evaluate the cost function for each discrete switching state which poses high computation bur- den.Thus,an improved MPC algorithm is proposed.Compared with the conventional FCS-MPC,the proposed method does not require loop optimization and the predictive time is short, which makes it easy to obtain digital realization. Experimental tests are carried out in the two-level pulse width modulation（PWM） rectifier prototype.The results verify the effectiveness of the proposed algorithm.
作者
杨兴武
牛梦娇
陈磊
YANG Xing-wu;NIU Meng-jiao;CHEN Lei(Shanghai University of Electric Power,Shanghai Green Energy Grid Connection Technology Research Center, Shanghai 200090, Chin)
出处
《电力电子技术》
CSCD
北大核心
2018年第2期101-103,共3页
Power Electronics
基金
国家自然科学基金(51207086)
上海市教育委员会科研创新项目(14YZ126)
关键词
整流器
模型预测控制
寻优
rectifier
model predictive control
optimization
