Education, Science, Technology, Innovation and Life
Open Access
Sign In

Islanding Detection Based on S - Transform and Neural Network

Download as PDF

DOI: 10.23977/ieps.2017.1005

Author(s)

Man Weishi, Zhu Zongyao, Zhang Zhiyu, Ma Ruwei, Wang Jianhua

Corresponding Author

Man Weishi

ABSTRACT

The passive methods have a large nondetection zone (DNZ) and a long detecting time. In addition, the threshold of the passive method is difficult to set. However, the active method will affect the power quality. So, an novel islanding detection method based on S - transform(ST) and neural network was proposed. In this method, ST was adopted to extract feature vector from the voltage of the point of common coupling (PCC), and then the feature vector is sent to the neural network after training to determine whether islanding occured. The simulation results show that the islanding can be detected accurately and quickly even when the power is perfectly matched, and in the presence of disturbances such as grid voltage fluctuation, harmonic disturbance, load switching, grid fault, etc. Avoid the problem of false tripping caused by various disturbances.

KEYWORDS

Islanding Detection, S Transform, Neural Network, Nondetection Zone.

All published work is licensed under a Creative Commons Attribution 4.0 International License.

Copyright © 2016 - 2031 Clausius Scientific Press Inc. All Rights Reserved.