In response to the prominent problems of weak recognition ability and insufficient accuracy in the field of low-voltage fault arc recognition, this article aims to optimize and improve existing detection technologies through in-depth analysis of the typical characteristics of fault arcs, in order to achieve more accurate and efficient fault arc recognition and prevention. Firstly, this article introduces a high-precision and high-sensitivity dedicated detection chip, which can accurately capture the weak signals generated by fault arcs, providing a solid foundation for the accurate identification of fault arcs. Secondly, by adopting dual core current transformer technology, accurate measurement and comparison of current signals have been achieved, effectively improving the accuracy of fault arc identification. In addition, this article also combines cutting-edge technologies such as big data analysis and artificial intelligence to design a multi-channel high-speed processing circuit for fault arcs, deeply mining and analyzing fault arc data, and further improving the intelligence level of fault arc recognition. The research results of this article are of great significance for improving the accuracy and efficiency of low-voltage fault arc recognition, providing strong support for further research and application in related fields.