Robust model predictive power control for three-phase VSRs under unbalanced grid

• Published in: ISA transactions Vol. 133; pp. 450 - 462
• Main Authors: Guo, Xin, Ren, Hai-Peng
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01.02.2023
• Subjects: Model predictive power control; Parameters uncertainty; Three-phase VSRs; Unbalanced grid; Weights optimization; Model predictive power control; Three-phase VSRs; Unbalanced grid; Parameters uncertainty; Weights optimization
• ISSN: 0019-0578; 1879-2022
• DOI: 10.1016/j.isatra.2022.07.021

The unbalanced grid voltage and circuit parameter uncertainty are two main obstacles for three phase voltage source rectifiers (VSRs) to achieve high performance in the practical applications. According to the instantaneous power model of the three-phase VSRs, six power components have to be well-regulated using only four available current manipulated variables, which is a typical underactuated problem. The model predictive control (MPC) provides a unified framework to regulate six power components simultaneously. However, how to balance the six power components control efforts is a challenge task. Meanwhile, the predictive model maybe inaccurate because of circuit parameters uncertainty, which degrades the performance of the MPC as well. In this paper, a robust model predictive power control (RMPPC) method is proposed for the three-phase VSRs to overcome above twice obstacles. The contributions of the work are: (1) The proposed method achieves the balance six power components control of the three-phase VSRs under unbalanced grid by using the off-line optimized weights; (2) a soft robust item with time variant boundary is proposed to achieve robust predictive model to deal with parameter uncertainty. Comparing with the existing voltage oriented control (VOC), direct power control (DPC) and model predictive control (MPC) methods, the proposed method achieves the best power quality in the sense of highest power factor and the lowest power oscillation in experiment, which verify the superiority of the proposed method. •Robust power predictive control is achieved for VSRs under parameters uncertainty.•Optimal balance and simultaneous control for six power components is achieved under the MPC frame.•Satisfactory power quality of three-phase VSR system under unbalanced grid is achieved.