Abstract：An adaptive active arc suppression method based on S-transform correlation degree for distribution

network unidirectional grounding was proposed to effectively eliminate single-phase grounding fault arc in

distribution network and improve power system stability. The time series data of zero-sequence current of

distribution network lines were collected，the time-frequency information matrix of zero-sequence current through

S-transformation was obtained，the correlation matrix of time-frequency information matrix between each line was

calculated and obtained. Input the convolution neural network model，and output the single-phase grounding fault

line selection results；the double closed-loop adaptive control method based on fuzzy quasi-PR and PI control was

adopted in the fault line to inject zero-sequence current into the neutral point of the fault line of the distribution

network，effectively control the phase and amplitude of the injected current，force the voltage value of the phase of the fault line of the distribution network to be 0，and complete the adaptive active arc suppression. The experimental results show that this method can effectively identify the single-phase grounding fault lines in the distribution network，realize the single-phase grounding adaptive active arc suppression，and complete the single-phase grounding adaptive active arc suppression work as soon as possible to ensure the safe and stable operation of the distribution network.

      Key words：S-transformation；correlation degree；distribution network；single phase grounding fault；adaptive；active arc suppression