1st International Conference on Modelling and Simulation on Engineering, Science and Technology (ICOMSEST) - Kuala Lumpur - Malaysia (2019-02-24)

A Study on Fault Location Technique for 154kV Substation using Weka Software
Recently, studies on the intelligent power facilities have been attempted to apply AI techniques according to the development of computer platform. In particular, faults in substations should be able to quickly identify possible faults and minimize power fault recovery time. A substation is composed of various components such as transmission line, transmission bus, transformer bank, distribution line, distribution bus, LS, DS and CB. So, it is difficult to find where the fault occurred. In this paper, based on the SOP (standard operation procedure) of KEPCO, we propose a fault location technique for domestic standard 154kV substation using Weka software. In order to identify the fault location of each component of the target 154 kV substation, such as line, bus and transformer, the training pattern matrix is based on the operating conditions of CB, DS, and IED. After learning the training pattern, the performance evaluation of the fault location of the substation through the proposed Neural Network by Weka software was tested under various conditions. Simulation results show that the proposed Neural Network has 100% fault locating performance, which shows that it has a perfect fault location for single fault situation.
Mr. Kyung-Min LEE, Mr. Jae-Young Hong, Mr. Chang-Gi Chae, Professor Tae-Won Kang, Ms. You-Jin Lee, Professor Chul-Won Park