基于时间辅助信息的感应电机预测电流控制  

Predictive Current Control of Induction Machines Based on Time Auxiliary Information

作　　者：齐昕[1] 吴文昊 吴琳 周珂[1] 马祥华[1] 周晓敏[1] QI Xin;WU Wenhao;WU Lin;ZHOU Ke;MA Xianghua;ZHOU Xiaomin(School of Mechanical Engineering, University of Science and Technology Beijing, Haidian District, Beijing 100083, China)

机构地区：[1]北京科技大学机械工程学院

出　　处：《中国电机工程学报》2019年第16期4927-4934,共8页Proceedings of the CSEE

基　　金：国家自然科学基金项目(51407007);中央高校基本科研业务费专项资金资助项目(FRF-TP-15-021A3);北京市重点学科建设项目(HK100080429)~~

摘　　要：模型预测控制具有滚动优化、反馈校正的优点,近年来在电力电子领域受到广泛关注,很多学者提出了诸如预测磁链控制、预测转矩控制、预测电流控制的电机控制方法,其中预测电流控制无需进行权重设计,具有直接输出电压矢量的特点,算法相对简洁,容易实现。预测控制在采取简单的事件触发机制时,电流轨迹超出边界后才会触发算法选择新的电压矢量;但由于不可避免的系统软硬件延迟,当新矢量作用于电机时,电流轨迹已经大幅超出边界。为此,提出一种电压矢量时间辅助切换机制,减小电流超出边界圆的几率,有效抑制电流畸变;并针对预测电流控制矢量轮询相互独立的特点,应用多核数字信号处理器(digital signal processor,DSP),将并行处理技术引入预测电流控制,提高系统的在线实时性。在两电平感应电机实验平台上进行测试,实验显示系统的处理速度提升了22%;同时,针对采样频率对开关频率的影响进行测试,发现提升系统的在线实时性有助于降低逆变器的开关频率。The model predictive control has the advantages of rolling optimization and feedback correction. In recent years,it has received extensive attention in the power electronics field.Many scholars have proposed motor control methods such as predictive flux control, predictive torque control, and predictive current control, in which the predictive current control don’t require weight design and has the characteristics of the direct output voltage vector. The algorithm is relatively simple and easy to implement. When a simple event trigging mechanism is applied in predictive control, algorithm starts to calculate a new voltage vector after the exceeding of current trajectory.However, due to the unavoidable delays in system hardware and software, current trajectory will exceed the boundary seriously, when a new voltage vector applies on the machine.Hence, a voltage vector time-assisted switching mechanism was proposed in this paper to reduce the probability that the current exceeds the boundary circle and the current ripple.Because predictive current control vector polling calculations were independent of each other, multi-core digital signal processor(DSP) was adopted to introduce parallel processing into predictive control, which improves the real-time performance of the system. On the two-level induction motor test platform, experiments show that the system processing speed increased by 22%. The influence of the sampling frequency on the switching frequency was tested at the same time. It is found that improving the on-line real-time performance of the system can help reduce the switching frequency of the inverter.

关 键 词：并行计算 模型预测控制 单边界圆预测算法 低开关频率 电压矢量时间辅助切换机制 

分 类 号：TM76[电气工程—电力系统及自动化]