Islanding classification mechanism for grid-connected photovoltaic systems
Document Type
Article
Publication Date
4-1-2021
Abstract
This article develops an islanding classification technique by adapting signal processing and machine learning techniques. The proposed method trains with all the possible islanding conditions, by extracting their features and classifying them. The performance of the proposed method was tested on a single-phase grid-connected photovoltaic system simulated using MATLAB/Simulink environment. The classifier achieved 98.1% training and 97.8% testing efficiency and can effectively detect islanding under 0.2 s with low misclassification. Further, the developed algorithm is tested with a 10-kW grid-connected photovoltaic system to monitor the changes in voltage and power mismatch at the point of common coupling (PCC) and classify the state of the system efficiently.
Keywords
Inverters, Islanding, Power harmonic filters, Reactive power, Voltage control, Phase locked loops, Harmonic analysis, Distributed generation (DG), islanding, neural networks, photovoltaic (PV) systems, wavelet transform (WT)
Divisions
fac_eng
Publication Title
IEEE Journal of Emerging and Selected Topics in Power Electronics
Volume
9
Issue
2
Publisher
IEEE-Inst Electrical Electronics Engineers Inc
Publisher Location
445 HOES LANE, PISCATAWAY, NJ 08855-4141 USA