Series Arc Fault Detection Based on Multimodal Feature Fusion.

In low-voltage distribution systems, the load types are complex, so traditional detection methods cannot effectively identify series arc faults. To address this problem, this paper proposes an arc fault detection method based on multimodal feature fusion. Firstly, the different mode features of the current signal are extracted by mathematical statistics, Fourier transform, wavelet packet transform, and continuous wavelet transform. The different modal features include one-dimensional features, such as time-domain features, frequency-domain features, and wavelet packet energy features, and two-dimensional features of time-spectrum images. Secondly, the extracted features are preprocessed and prioritized for importance based on different machine learning algorithms to improve the feature data quality. The features of higher importance are input into an arc fault detection model. Finally, an arc fault detection model is constructed based on a one-dimensional convolutional network and a deep residual shrinkage network to achieve high accuracy. The proposed detection method has higher detection accuracy and better performance compared with the arc fault detection method based on single-mode features.

Series arc fault detection based on continuous wavelet transform and DRSN-CW with limited source data.

When a series arc fault occurs in an indoor power distribution system, the temperature of arc combustion can be as high as thousands of degrees, which can lead to an electrical fire. Deep learning has developed rapidly in recent years and is widely used in fault diagnosis. The problem is that the sourced data is challenging to obtain, and few public data sources affect the application of deep learning models in arc fault diagnosis. In order to solve this problem, an arc fault detection method based on continuous wavelet transform and deep residual shrinkage network with the channel-wise threshold (DRSN-CW) is proposed. First, the grayscale images of source data features are obtained by continuous wavelet transform. Then, the feature images are data enhanced to construct the dataset. Finally, the DRSN-CW model is constructed and used to detect arc fault. The results show that the highest accuracy of arc fault detection is 98.92%, and the average accuracy is 97.72%. This method has excellent performance, which provides a new idea for arc fault detection.