ANN-based PV-integrated power conditioning system for MPPT optimization and power quality enhancement in hybrid microgrids
Document Type
Article
Publication Date
6-1-2026
Abstract
Harmonic distortions and voltage disturbances in hybrid renewable microgrids, driven by nonlinear converters and dynamic load variations, present major challenges to power quality (PQ) and operational stability. This paper proposes an Artificial Neural Network (ANN)-based dual control framework that combines: (i) an ANN-MPPT controller for photovoltaic (PV) systems to ensure fast, stable, and accurate maximum power point tracking under rapid irradiance fluctuations, and (ii) an ANN-driven intelligent power conditioning system for PQ improvement. The ANN-MPPT significantly enhances PV energy extraction efficiency by minimizing oscillations and reducing convergence time, while the ANN-controlled power conditioning system mitigates harmonics, compensates reactive power, regulates voltage under 20 % sag/swell conditions, suppresses unbalanced disturbances, and provides effective compensation during three-phase short circuits. Simulation results demonstrate that the proposed system maintains source current THD between 1.71 %–1.73 % and load voltage THD between 0.68 %–0.70 %, fully complying with IEEE-519 standards while sustaining unity power factor operation. The combined ANN-MPPT and ANN-based power conditioning approach improves both PV utilization and grid-side PQ, offering a robust and adaptive solution for hybrid PV–wind microgrids under dynamic and nonlinear conditions.
Keywords
Artificial neural networks, Harmonic mitigation, MPPT optimization, Photovoltaic systems, Power conditioning, Voltage vompensation
Publication Title
Energy Reports
ISSN
2352-4847
DOI
10.1016/j.egyr.2026.109129
Recommended Citation
Reguieg, Zakaria; Mehedi, Fayçal; Bouyakoub, Ismail; Kacemi, Walid Mohammed; Saidi, Fayssal; and Mekhilef, Saad, "ANN-based PV-integrated power conditioning system for MPPT optimization and power quality enhancement in hybrid microgrids" (2026). Research Publications (2026 to 2030). 584.
https://knova.um.edu.my/research_publications_2026_2030/584
Volume
15