Contribution to the control of a reactive power compensation system powered by a photovoltaic generator

ZABIA, Djallal Eddine (2025) Contribution to the control of a reactive power compensation system powered by a photovoltaic generator. Doctoral thesis, Faculté des Sciences et de la technologie.

Text Finaldjallal.pdf Download (195MB)

Abstract

Photovoltaic (PV) systems under partial shading conditions (PSCs) exhibit complex power�voltage characteristics, reducing energy yield with conventional Maximum Power Point Tracking (MPPT) methods like Perturb & Observe. This thesis presents a comprehensive framework to optimize PV performance and grid integration through advanced control and emulation strategies. The Enhanced Eel-Grouper Optimization (EEGO) algorithm, a novel metaheuristic MPPT approach, achieves near-optimal tracking efficiency (>99%) by swiftly identifying the global maximum power point across diverse shading scenarios. A dynamic restart mechanism ensures reliable adaptation to sudden environmental changes, enhancing system stability. A cost-effective PV emulator, built around a DC programmable source (GW Instek APS-1102A, 1000 W), accurately replicates PSC-induced I-V curves in a low-scale laboratory grid setup (500 W), enabling robust algorithm validation. An Real-Time MPPT framework, implemented on a dSPACE 1104 platform, dynamically adjusts control parameters to minimize fluctuations and support grid stability. To enhance power quality, a Shunt Active Power Filter (SAPF) with Predictive Direct Power Control (PDPC) reduces current Total Harmonic Distortion to below 5%, achieving near-unity power factor and compliance with IEEE 519-2022 standards under variable loads. Experimental validation using the dSPACE 1104 and DC programmable source demonstrates superior energy efficiency and grid compatibility. This framework advances PV system optimization, offering a scalable solution for sustainable energy integration in smart grid applications.

Actions (login required)

View Item