[Oral Presentation]An Improved Model-Free Predictive Power Control for Three-phase Grid-connected Inverter Based on Hybrid SVM Under Unbalanced Grids

IMPORTANT DATES

Digest Submission Opens: Oct. 30, 2022
Digest Submission Deadline: Mar. 20, 2023
Acceptance Notification: Apr. 28, 2023
Full Paper Submission Deadline: May. 20, 2023

An Improved Model-Free Predictive Power Control for Three-phase Grid-connected Inverter Based on Hybrid SVM Under Unbalanced Grids

ID：47 View Protection：PRIVATE Updated Time：2023-06-13 17:12:54 Hits：503 Oral Presentation

Start Time：2023-06-18 16:00 （Asia/Shanghai）

Duration：20min

Session：[S] Oral Session » [S4] Oral Session 7 & Oral Session 10

Presentation File

Tips: The file permissions under this presentation are private. You have not logged in yet and cannot view it temporarily.

Abstract

Model based predictive power control based on space vector modulation (SVM-MBPPC) is a popular control method for three-phase grid-connected inverter with the advantage of quick response and excellent performance. However, the excellent control performance of the conventional SVM-MBPPC mainly depends on the accuracy of system parameters. To overcome the poor robustness of SVM-MBPPC, some scholars propose model-free predictive power control based on SVM (SVM-MFPPC). In spite of the strong robustness of the SVM-MFPPC, the steady-state performance of conventional SVM-MBPPC with accurate parameters is greater than the SVM-MFPPC. This paper proposes an improved MFPPC based on hybrid SVM (HSVM-MFPPC), which has equal robustness to conventional SVM-MFPPC and achieves better steady-state performance. The proposed HSVM-MFPPC is also extended to unbalanced grid condition by adding an compensation power value. The simulated results confifirm that the proposed HSVM-MFPPC inherits strong robustness and can reduce the current total harmonic distortion (THD) by 30.6% than conventional SVM-MFPPC when the average modulation index is high enough

Keywords

Three-phase grid-connected inverter, Power control, predictive control, model-free, hybrid SVM.

WELCOME TO PRECEDE 2023