Fault Line Identificaiton in the non-effectively grounded distribution networks Based on Feature Fusion of the Transient Signals

Reliable fault line identification is very important for keeping safety and reliability of the distribution network. However, wrong fault line identification results are commonly generated by using a single criterion due to the complex fault conditions and various unknown system parameters. In this paper, main transient fault features, barely affected by the fault condition and system parameters, are analyzed based on the equivalent sequence-circuit of the distribution network. A novel fault line criterion based on the information fusion algorithm is developed to realize reliable fault line identification. In the algorithm, the membership degree function for describing the contribution of each fault feature in fault line identification is computed to create the criterion based on fuzzy theory. The experiment of a distribution network is established by PSCAD/EMTDC. The experiment results demonstrate the effectiveness and reliability of the proposed method.