Aiming at the problem of fault arc detection in aviation line system, based on the analysis of the causes and basic characteristics of fault arc, a fault arc detection algorithm based on wavelet analysis is designed. This paper analyzes the AC series arc fault circuit based on Schavemaker arc model, uses sym5 wavelet to decompose the fault arc current data, extracts the high frequency part of the current, and proposes a fault arc detection method based on the d3 standard deviation of the high frequency component, and simulates the algorithm based on Simulink. The simulation results show that the proposed method has a higher degree of discrimination than the traditional method of directly setting the threshold after wavelet decomposition, and can quickly and accurately determine the existence of fault arc, which provides a useful reference for engineering practice.