Research On Single-phase To Ground Fault Isolation Method Of Distribution Network

Connection type of distribution network in our country generally adopt neutral point non-effectively grounded system or neutral point earthed via arc-suppression coil system. Most faults in distribution network are single-phase to earth faults. Nowadays,the processing method of single-phase to earth fault is mainly using faulted feeder selection device which will find faulty lines and tripping off the whole line when fault happens, this leads to cutting non-fault region, reducing power supply reliability, extending recovery time of power supply,and the selection accuracy of faulty line is not satisfactory. It is easy to find that distributed fault isolation method gives protection function to FTU after comparing centralized control method and distributed control method. It can succeed to realize fault isolation through tripping off the switch at the two end of fault line by using the fault information between adjacent switch node. This method does not affect non-faulty region and is rapid in fault processing, so it can realize precise fault location and improve reliability and quality of power supply.This paper puts forward a kind of distributed fault isolation scheme based on negative sequence current and zero sequence voltage by analyzing the fault characteristic of single-phase to earth fault in distribution network. Firstly, we divide the topology of distribution network into several groups. Secondly, we begin to determine the number of adjacent node and switch type of every node in each group. When single-phase to earth fault happens, the FTU of each node detects zero sequence voltage and starts protection, by comparing the amplitude of negative sequence current of adjacent node in feeder line, then it uses distributed algorithm to find fault region and finally realize fault isolation.The paper does simulation analysis for negative sequence current and zero sequence voltage before and after the fault point, verified why it can action as criteria. At last, we elaborate the whole process in detail that how to use distributed algorithm to realize fault location in a typical distribution network.