Open AccessFault Diagnosis in Distributed Power-Generation Systems Using Wavelet Based Artificial Neural Network
Author(s) -
Jiahui Chen,
Jason Gao,
Yiwei Jin,
Peng Zhu,
Qinzhen Zhang
Publication year - 2021
Publication title -
european journal of electrical engineering
Language(s) - English
Resource type - Journals
eISSN - 2116-7109
pISSN - 2103-3641
DOI - 10.18280/ejee.230107
Subject(s) - fault (geology) , artificial neural network , electric power system , computer science , wavelet transform , wavelet , discrete wavelet transform , power (physics) , artificial intelligence , data mining , pattern recognition (psychology) , real time computing , physics , quantum mechanics , seismology , geology
In recent years, research on fault diagnosis of grids is becoming increasingly important, because it ensures the stable operation of power systems, and meets high demands on the power quality by power customers. In this paper, an intelligent approach for fault diagnosis of distributed power generation systems is proposed based on maximum overlap discrete wavelet transform and artificial neural network. In the proposed scheme, the fault data are first collected. Then, maximum overlap discrete wavelet transform is applied to detect faults and extract features. Finally, artificial neural network is constructed to classify the fault types. Results show that the method can identify faults precisely, classify fault types accurately, and is not affected by the change of electrical parameters. In addition, compared with several existing intelligent diagnosis techniques, the proposed approach can provide better fault classification accuracy. To evaluate the performance, the algorithm is verified by the case of the modified simulation model of IEEE-13 bus standard system.