A Simplified Current Feature Extraction and Deployment Method for DC Series Arc Fault Detection

DC series arc fault is one of the main causes of low-voltage dc distribution system fire accident. Traditional methods use time-frequency domain features to make judgments directly or use machine learning to further classify them. In this article, a method is proposed to simplify the feature extraction process while ensuring accuracy. First, a fully automated process is used to establish dc arc datasets. A composite bandpass filter is designed to extract the typical frequency segment of dc arc. Besides, an arc detection neural network based on temporal convolution network is proposed to extract current waveform features. Principal component analysis is used to process these features to reduce correlation. Finally, a single hidden layer neural network is used as classifier. The database is collected from different scenarios and working conditions. By measuring the dc raw current, the arc fault detector can achieve a test set accuracy of 99.88% at a sampling rate of 250 kHz. The model is also deployed on Jetson Nano with an average real-time detection of 0.15 s/sample.