Abstract: The topology of the distribution network is complex, and the passive detection method cannot reflect the fault location sensitively and reliably. Considering the variation rules of neutral grounding mode, the distributed autonomous mode is applied to detect the differences of fault amounts before and after the fault and the variation of neutral point, so as to realize the joint activepassive detection and isolation of radial feeder fault section. The passive detection criterion of the fault segment on the main line is constructed by the transient zero-sequence current direction parameters of the fault point upstream and downstream. According to the phase difference of the power-frequency zero-sequence current of the same monitoring point before and after the compensation variation of the arc suppression coil, the active criterion for the fault segment of the branch line are constructed, and the joint activepassive detection of the radial feeder fault segment is completed. After determining the fault segment, a small resistance is put at the neutral point, and the fault isolation is determined according to the zero-sequence current variation of the fault point upstream.The method utilizes the active variation of the neutral point operation mode and the distribution automation to complete the fault section location and isolation, which can enhance the section location reliability of the single-phase grounding fault of the distribution network, and reduce the burden of the main station in the distribution network.