Intelligent ranging method for radiation network fault by utilizing S transformation spectrum characteristics

The invention discloses an intelligent ranging method for a radiation network fault by utilizing S transformation spectrum characteristics. The method comprises the following steps of: when a multi-branch radiation network is in fault, immediately initiating an initiating element, and solving zero-sequence current according to three-phase current of each line which is measured on a protection andinstallation part by utilizing Clark transformation; selecting a faulty line according to a principle that the polarity of the zero-sequence current in the faulty line is reverse to that of a sound line, and selecting a faulty line zero-sequence current signal with a sampling sequence length of 200 points for S transformation, wherein the transformation result is a 101*200 time frequency complex matrix; solving modes of various elements in the complex matrix; and extracting energy in each frequency range and taking the energy as an input vector of a hierarchical distributed neural network according to the obtained mod