Robust Predictive Current Control With Variable-Gain Adaptive Disturbance Observer for PMLSM

For high-velocity/high-precision linear motion systems, one of the most important factors that influence their dynamic performance is the characteristic of the inner current loop. The proportional-integral controller is the most practical strategy used for current control. However, its linear struct...

Full description

Bibliographic Details
Main Authors: Rui Yang, Ming-Yi Wang, Li-Yi Li, Cheng-Ming Zhang, Jia-Lin Jiang
Format: Article
Language:English
Published: IEEE 2018-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/8310031/
Description
Summary:For high-velocity/high-precision linear motion systems, one of the most important factors that influence their dynamic performance is the characteristic of the inner current loop. The proportional-integral controller is the most practical strategy used for current control. However, its linear structure and imperfect decoupling capability make it difficult to obtain satisfying transient response under multiple operation conditions. The predictive current control (PCC) is designed as the current controller contributes to its superior performance. The main drawback of the PCC lies in its sensitivity against the unavoidable disturbances due to the parameters mismatch and the unmodeled dynamics. In this paper, an online adaptation-gain update method that can extend the inductance robust limit is proposed. First, by analyzing the closed-loop transfer function of the PCC system in the discrete domain, the effect of disturbances is discussed. Then, to eliminate the static current errors and improve the transient response, an adaptive disturbance observer is introduced. However, the direct dependence of the equivalent integral gain of the observer on the inductance in the controller leads to the deteriorative current response as the larger inductance mismatch exists. Therefore, an improved variable-gain method utilizing the current estimation errors is developed to reduce the current overshoot and the oscillation. Meanwhile, the design consideration for the two additional parameters of the proposed method is made by full-digital simulation. Finally, the effectiveness of the proposed method is uniformly verified with both simulation and experimental results.
ISSN:2169-3536