A Glowworm Swarm Optimization and Variable-step Perturbation & Observation Algorithm for Photovoltaic Array under Partial Shadow

Under local shading conditions outdoors, the output P-V curve of photovoltaic arrays exhibits multiple peaks, which makes the traditional maximum power point algorithms fail to track its global peak. Even though the published swarm intelligence maximum power point tracking algorithms can get the glo...

Full description

Saved in:
Bibliographic Details
Published in2024 9th Asia Conference on Power and Electrical Engineering (ACPEE) pp. 1460 - 1464
Main Authors Yinuo, Shi, Baohong, Liu, Wei, Rao, Renxian, Zeng, Li, Lv
Format Conference Proceeding
LanguageEnglish
Published IEEE 11.04.2024
Subjects
Electrical engineering
glowworm swarm optimization
Maximum power point trackers
maximum power point tracking
partial shadow
Perturbation methods
photovoltaic array
Photovoltaic systems
Proposals
Simulation
Steady-state
variable step perturbation& observation
Online AccessGet full text

Cover

More Information
Summary:Under local shading conditions outdoors, the output P-V curve of photovoltaic arrays exhibits multiple peaks, which makes the traditional maximum power point algorithms fail to track its global peak. Even though the published swarm intelligence maximum power point tracking algorithms can get the global peak, which removes the drawbacks of the traditional algorithms, the shortages of long optimization time, easily local peak capture, as well as large steady oscillation, restrict their application. For this reason, an MPPT algorithm named as glowworm swarm optimization and variable-step perturbation & observation(GSO-VP&O) is proposed. The proposed algorithm is realized with two stages of searching known as global searching and local searching. In the global searching, the adaptive adjustment ability of glowworm swarm optimization(GSO) and the excellent global convergence of the multimodal function can mitigate the risk of local optima. In the local searching, the variable-step perturbation & observation is employed to quickly arrive at the global maximum power point and reduce the ripple of steady output. To validate the proposal, the platforms for the algorithms are constructed and simulated. Simulation results show that compared with the other swarm intelligent algorithms, the proposed GSO-VP&O can get a shorter optimization time and faster tracking speed as well as the minimum ripple in a steady state.
DOI:10.1109/ACPEE60788.2024.10532661