Low voltage DC series arc fault

Citation Author(s):
Yufei
Mao
Siemens AG
Submitted by:
Yufei Mao
Last updated:
Fri, 03/07/2025 - 10:06
DOI:
10.21227/kzr1-1w26
Data Format:
Research Article Link:
License:
0
0 ratings - Please login to submit your rating.

Abstract 

This dataset is collected to investigate detection algorithms for DC series arc faults on resource constrained devices. The data is current signal collected from test-bench build to simulate arc fault under various conditions according to the standard UL1699B. Different condition includes the power mode regulated by the elctronic load in the system for the simulation of different system dynamics under topology series and parallel.

The code attached is used in the comparative study for data evaluation with visulisation of the current signal and their representative FFT signal. Examples of detection algorithms, including a conventional method with threshold, SVM and CNN, are listed for cross-dataset validation and algorithm robustness evaluation. 

Instructions: 

The file includes four datasets arranged for detection algorithm, including parallel basic, parallel load, series basic and series load.

Funding Agency: 
European Commission
Grant Number: 
957197
Data Descriptor Article DOI: