三相异步电机电流多步预测控制方法

摘要

Model predictive control (MPC) has several important advantages such that concepts are very intuitive and the multivariable case with constraints can be easily considered. So, MPC has become a hot research topic in electric drives. A predictive current control method based on continuous control set MPC (CCS-MPC) is investigated in this paper for the vector control of three phase induction machine. Firstly, the augment model of three phase induction machine is established based on voltage equation in synchronous coordinate system with rotor flux orientation. After that, the offset-free predictive current control method is presented. Secondly, the determining method of horizon length and weighting factors is researched by using the closed-loop poles of the system. And then, parameter robustness of the predictive current control is discussed. This paper presents how to use the augment model to realize offset-free control which is the fundamental issue for the control method research. The paper focuses on the effect of horizon length and weighting factors on the closed-loop poles position and the determining method of these parameters. The experimental results show that the predictive current control method can completely replace PI regulator to realize asynchronous motor current control.%模型预测控制(MPC)具有概念简单、容易处理包含约束条件的多变量控制问题等优点,已成为电力传动领域研究的热点.针对三相异步电机磁场定向控制,研究了基于连续控制集模型预测控制的电流控制方法.首先,以同步坐标系(转子磁场定向)下电压方程为基础建立电机的增广模型,实现了电流无静差预测控制;其次,利用系统闭环极点研究了预测域长度和权重系数的确定方法,并分析了控制方法的参数鲁棒性.无静差是控制方法需要研究的基本问题,讨论了采用增广模型实现无静差预测控制的方法.确定预测域长度和权重系数是 MPC 设计的主要内容,重点分析了这些参数对系统闭环极点位置的影响,并讨论了这些参数的确定方法.实验结果表明,该方法能够完全替代PI调节器实现异步电机电流控制.