A novel neural network with simple learning algorithm for islanding phenomenon detection of photovoltaic systems

Authors:
Kuei-Hsiang Chao;Chia-Lung Chiu;Ching-Ju Li;Yu-Choung Chang
Affiliations:
Department of Electrical Engineering, National Chin-Yi University of Technology, Taiwan;Department of Electrical Engineering, National Chin-Yi University of Technology, Taiwan;Department of Electrical Engineering, National Chin-Yi University of Technology, Taiwan;Green Energy & Environment Research Laboratories, Industrial Technology Research Institute, Taiwan
Venue:
Expert Systems with Applications: An International Journal
Year:
2011

Citing 1
Cited 1

A Maximum Power Point Tracking Method Based on Extension Neural Network for PV Systems

ISNN '09 Proceedings of the 6th International Symposium on Neural Networks on Advances in Neural Networks

Characterization of Concentrating Photovoltaic modules by cooperative competitive Radial Basis Function Networks

Expert Systems with Applications: An International Journal

Quantified Score

Hi-index	12.05

Visualization

Abstract

This study aimed to propose an intelligent islanding phenomenon detection method for a photovoltaic power generation system. First, a PSIM software package was employed to establish a simulation environment of a grid-connected photovoltaic (PV) power generation system. A 516W PV array system formed by Kyocera KC40T photovoltaic modules was used to complete the simulation of the islanding phenomenon detection method. The proposed islanding phenomenon detection technology was based on an extension neural network (ENN), which combined the extension distance of extension theory, as well as the learning, recalling, generalization and parallel computing characteristics of a neural network (NN). The proposed extension neural network was used to distinguish whether the trouble signals at the grid power end were power quality interference or actual islanding operations, in order that the islanding phenomenon detection system could cut off the load correctly and promptly when a real islanding operation occurs. Finally, the feasibility of the proposed intelligent islanding detection technology was verified through simulation results.