永磁直线同步电机的变权重系数多步模型预测电流控制方法

Variable Weight Coefficient Multi-step Model Predictive Current Control Method for Permanent Magnet Linear Synchronous Motor

该文将多步连续控制集模型预测电流控制(model predictive current control,MPCC)应用于永磁直线同步电机(permanent magnet linear synchronous motor,PMLSM)的电流控制。低权重系数MPCC理论上可实现高带宽的电流调节,但数控时延会造成低权重系数MPCC电流跟踪失稳,同时低权重系数导致的高增益会引起严重的稳态电流纹波。针对此问题,该文提出基于增广状态观测器的变权重系数多步MPCC方法。利用闭环极点研究多步MPCC的数控时延、权重系数和电流闭环稳定性的关系,并采用离散增广观测器超前估计电流及其增量以消除数控时延造成的权重系数选取范围受限的问题,从而提高MPCC的带宽。在此基础上,通过给定电流增量及电流反馈误差自适应地切换控制权重系数,以解决高动态性能与低纹波不能共存的问题。通过实验验证所提出的多步MPCC可实现高带宽、高精度和强鲁棒性的PMLSM电流控制。

This paper applied the multi-step continuous control set model predictive current control(MPCC) to the current control of the permanent magnet linear synchronous motor(PMLSM). The MPCC with low weight coefficients could theoretically achieve current regulation with high bandwidth, but the control delay would cause the instability of current tracking, and the high gain caused by the low weight coefficient will cause serious steady-state current ripple. To solve this problem, this paper proposed a multi-step MPCC method with variable weight coefficients based on augmented state observer. The closed-loop poles were utilized to study the relationship between the control delay, weight coefficients and current closed-loop stability, and the discrete augmented observer was used to estimate the current and its increment in advance to eliminate the limited selection range of weight coefficients caused by the control delay, thereby increasing the bandwidth of MPCC. On this basis, the control weight coefficient was adaptively switched through a given current increment and current feedback error to solve the problem that high dynamic performance and low ripple cannot coexist. It is verified through experiments that the proposed multi-step MPCC can achieve PMLSM current control with high bandwidth, high accuracy and strong robustness.