A robust approach of single-phase-to-ground fault location for distribution grids with neutrals non-effectively grounded

Locating single-phase-to-ground faults (SGFs) in distribution grids is of great importance to avoid the worsening of the fault and guarantee a continuous and steady power supply. This paper presents an innovative approach of SGf location by using the fault information which is acquired by the fault location devices and sent to the master station of the Distribution Automation System (DAS). The criteria successfully used for locating phase-to-phase faults (PPFs) are modified to locate SGFs. The approaches of making the information binary and differential are described to meet the requirement of the proposed location criteria. Nevertheless, sometimes erroneous information is sent or received. To enhance the robustness of the proposed approach, the Bayes theorem is introduced to take full advantage of the redundancy and correlation of the information from fault location devices with various mechanisms. Two examples are given to prove that a SGF can be located correctly even under the conditions of missing or erroneous information. The proposed approach has been embedded into the Open 5200 Distribution Automation System which is in service in Xi'an. More than ten SGFs have been successfully located.