The invention discloses a series fault arc detection method and
system, equipment and a storage medium. The detection method is characterized by, firstly obtaining a sampling sequence of each
power frequency period, carrying out
standardization processing, calculating a plurality of characteristic quantities of the sequence data of each
power frequency period, and describing the
AC current waveform by employing the plurality of characteristic quantities, so that the characteristics of the current data are well reflected from multiple dimensions, and the detection accuracy is improved; and then, setting a sliding detection window, wherein the length of the sliding detection window is N
power frequency periods, making the sliding detection window slide on a characteristic quantity sequence formed by the multiple power
frequency periods, and judging the series fault arc based on a distance calculation result between two sets of characteristic value data before and after window sliding. The overall detection and
recognition algorithm is realized based on the distance calculation result of the two feature matrixes, and feature quantity comparison after current
data decomposition is not needed, so that the
algorithm is relatively simple, and the requirement for the performance of an
embedded system is not too high.