Lightweight AC Arc Fault Detection Method by Integration of Event-Based Load Classification

As the main cause of electrical fires, arc fault detection attracts lots of attention in recent years. However, with the growing complexity of electric loads, arc fault detection becomes more difficult. This article proposes a lightweight arc fault detection method that integrates load classification into fault detection. First, according to the turn-on patterns, we divide loads into resistive, inductive, and switchable loads. Load classification is developed using an event-based method. Then, arc detection method is developed for each load category, which is achieved through sequential forward floating selection. Its performance is validated by the experiment data and comparison with other reported methods. Meanwhile, the selection of classifiers, sampling rate, and sampling periods for arc fault detection are also discussed in detail. The results show that the proposed method not only achieves a high fault detection performance but also keeps a relatively low sampling rate and short sampling period. Thus, it is beneficial for practical arc fault interrupter development.