Open Access

ARTICLE

Line Fault Detection of DC Distribution Networks Using the Artificial Neural Network

Xunyou Zhang^1,2,*, Chuanyang Liu^1,3, Zuo Sun¹

1 Country College of Mechanical and Electrical Engineering, Chizhou University, Chizhou, 247000, China
2 School of Electrical Engineering, Southeast University, Nanjing, 210096, China
3 College of Electronic and Information Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing, 211106, China

* Corresponding Author: Xunyou Zhang. Email:

Energy Engineering 2023, 120(7), 1667-1683. https://doi.org/10.32604/ee.2023.025186

Received 27 June 2022; Accepted 16 December 2022; Issue published 04 May 2023

Abstract

A DC distribution network is an effective solution for increasing renewable energy utilization with distinct benefits, such as high efficiency and easy control. However, a sudden increase in the current after the occurrence of faults in the network may adversely affect network stability. This study proposes an artificial neural network (ANN)-based fault detection and protection method for DC distribution networks. The ANN is applied to a classifier for different faults on the DC line. The backpropagation neural network is used to predict the line current, and the fault detection threshold is obtained on the basis of the difference between the predicted current and the actual current. The proposed method only uses local signals, with no requirement of a strict communication link. Simulation experiments are conducted for the proposed algorithm on a two-terminal DC distribution network modeled in the PSCAD/EMTDC and developed on the MATLAB platform. The results confirm that the proposed method can accurately detect and classify line faults within a few milliseconds and is not affected by fault locations, fault resistance, noise, and communication delay.

---

Keywords

---