This paper presents work on a hybrid fuzzy control scheme to improve the performance of tilting trains using a nulling-based tilt strategy. Two multi-objective genetic algorithm tuning methods (MOGA and NSGAII) were employed to optimise both the fuzzy output membership functions and the controller parameters. The objective functions incorporated the tilt response and roll gyroscope signals for the deterministic (curved track) profile, and lateral acceleration for the stochastic (straight track) profile. Simulation results discuss the effectiveness of using the presented techniques for tuning the fuzzy control scheme via multiple objectives. The proposed scheme is compared with the conventional nulling-tilt approach and a manually tuned fuzzy controller.
No Reference information available - sign in for access.
No Citation information available - sign in for access.
No Supplementary Data.
multi objective genetic algorithm;
Document Type: Research Article
Department of Electronic and Electrical Engineering, Loughborough University, Leicestershire, UK,College of Science and Technology, Universiti Teknologi Malaysia, Kuala Lumpur, Malaysia
Department of Electronic and Electrical Engineering, Loughborough University, Leicestershire, UK
Publication date: 2008-09-01