Implementation of Adaptive Neuro-Fuzzy Inference System in Fault Location Estimation

摘要

This paper proposed a new scheme using hybrid intelligent technique that combines artificial neural network and fuzzy inference system. This technique, known as Adaptive Neuro-Fuzzy Inference System (ANFIS) has associated with the advantage of wavelet transform as a pattern recognition method. The algorithm used to identify the type of fault either single line to ground, double line, double line to ground or three phase occur on a power transmission line. Other than that, this scheme is capable to analyze the fault location without the knowledge of line parameters. A power clustering algorithm called Gustafson Kessel is implemented for better performance. Alternative Transient Program/Electromagnetic Transient Program (ATP/EMTP) is used for simulation purposes and Matlab for further analysis. Outcomes indicated that the scheme is efficient and has a high percentage of accuracy.