A Proposed Scheme for Fault Discovery and Extraction Using ANFIS: Application to Train Braking System

Abstract

This paper showcases the use of model oriented techniques for real time fault discovery and extraction on train track unit. An analytical system model is constructed and simulated in Mathlab to showcase the fair and unfair status of the system. The discovery and extraction phases are centered on a hybrid adaptive neuro-fuzzy inference feature extraction and segregated module. Output module interprites zero (0) as a good status of the traintrack unit and one (1) as an unpleasant status. Final results showcase the robustness and ability to discover and extract multitude of unpleasant scenarios that hinder the smooth operations of train track units due to its high selectivity and sensitivity quality.