BROWSE
Detail
Conference Paper
시계열 2차원 변환을 통한 전기 아크 결함 탐지 성능 향상
Cited - time in 
Share 
Share
- Issue Date
- 2025-11
- Citation
- 대한전자공학회 학술 대회 (추계) 2025, pp.526-528
- Publisher
- 대한전자공학회
- Language
- Korean
- Type
- Conference Paper
- Abstract
- Arc faults are discharge phenomena caused by incomplete contact, contamination, or moisture, and are major sources of fires and equipment damage. With recent advancements in artificial intelligence, the need for deep learning-based arc fault detection models has increased. Detecting arc signals requires at least one full cycle of time-series data, and in some loads, a high sampling frequency is necessary, resulting in long input sequences. This makes it difficult for models to fully learn temporal dependencies. To address this issue, this study proposes an arc fault detection model that processes time-series data by transforming it into a two-dimensional representation. Experimental results on binary and multi-class classification tasks demonstrate that the proposed model outperforms existing approaches, confirming that the two-dimensional transformation of time-series data can enhance the accuracy and reliability of arc fault detection.