Improvement of model predictive control for single‐phase inverters by using sinusoidal signal observers. Issue 17 (19th January 2021)

• Record Type: Journal Article
• Authors: Li, Po; Li, Ruiyu; Feng, Haifeng; He, Ying; Zhang, Jingrui
• Is Part Of: IET power electronics. Volume 13:Issue 17(2020)
• Page Start: 3841
• Page End: 3850
• Publication Date: 2021-01-19
• Subjects: observers -- invertors -- field programmable gate arrays -- predictive control; distorted current tracking -- model predictive control -- single‐phase inverters -- sinusoidal signal observers -- finite control -- FCS MPC -- sinusoidal current tracking -- parameter uncertainties -- inverter system -- control performance -- unknown parameters -- accurate parameters -- current reference -- strong robustness -- unknown offset free MPC frame -- UOF MPC -- unknown sinusoidal signal -- linear observer -- instantaneous value observation -- time‐varying observer -- phase identification -- rapid control prototype -- maximum tracking current boundary; Power electronics -- Periodicals; 621.31705
• DOI: 10.1049/iet-pel.2020.0710
• Languages: English
• ISSNs: 1755-4535

Abstract

Abstract : Studies show that finite control set model predictive control (FCS MPC) is robust for single‐phase inverters in sinusoidal current tracking. That is, even if there are parameter uncertainties in inverter system, MPC can achieve accepted control performance. However, it is still necessary to reconstruct the unknown parameters. On the one hand, an accurate model is helpful to get better performance. On the other hand, accurate parameters can be used to determine the boundary of current reference. Relying on the strong robustness, under unknown offset free MPC (UOF MPC) frame, this study characterises the deviation of the parameters as an unknown sinusoidal signal, and designs two observers to reconstruct it. One is a linear observer for instantaneous value observation. The other is a time‐varying observer for the amplitude and phase identification. Its convergence is guaranteed by the Lyapunov function and the Laselle invariant set principle. All the proposed methods are validated by a rapid control prototype based on Labview FPGA platform. Moreover, experiments show that even if the reference exceeds the real maximum tracking current boundary, the proposed observers work well with accurate parameters identified and the reference can be adjusted online to avoid distorted current tracking.