Control system of train speed based on fuzzy logic controller

Abstract

This study aims to implement fuzzy logic controller using Mamdani and defuzzification method of Centroid, as a controller for train speed. Research methodology employed consists of Modeling, Design, Testing, and Analysis. At Modeling phase, the system is modeled utilizing transfer function model; at Design phase, the system with fuzzy logic controller is designed; at Testing phase, system output is examined by utilizing four types of testing (testing of rise and fall, testing of pulse input, and testing with simulated scenarios); at Analysis phase, system output is analyzed based on rise time, fall time, settling time, and average steady-state error. The results showed that when tested with the reference speed of rise and fall, the system with fuzzy logic controller has an average value of rise time 2.3 seconds, fall time 2.44 seconds, and settling time 3.58 seconds; when tested with testing of pulse input, the system has an average value of rise time 2.359 seconds, fall time 4.238 seconds, and settling time 4.1125 seconds. Train simulation results with additional resistance showed that system using fuzzy logic controller has an an average value of error 2%.