Gain scaling tuning of fuzzy logic Sugeno controller type for ride comfort suspension system using firefly algorithm

• Main Authors: Ab. Talib, Mat Hussin, Mat Darus, Intan Zaurah, Mohd. Yatim, Hanim, Hadi, Muhamad Sukri, Mohd. Saufi, Mohd. Syahril Ramadhan, Ngadiman, Nor Hasrul Akhmal
• Format: Conference or Workshop Item
• Published: 2022
• Subjects: TJ Mechanical engineering and machinery
• Online Access: http://eprints.utm.my/id/eprint/99355/; http://dx.doi.org/10.1007/978-981-19-2095-0_29

A control system based on fuzzy logic (FL) is one of the effective controllers which operates using an inference mechanism rule base that requires a knowledge database. The system itself can remotely able to produce good linguistic variables depending types of output required. Nevertheless, the FL controller design still has a drawback that requires an improvement to give a very high capability in controlling a dynamic ride comfort of the vehicle suspension system. This study aims to improve the FL controller design by adding a gain scaling value for each input and output of the FL system. A metaheuristic-based firefly algorithm (FA) is used to optimize the value of each input and output of the FL system. Taking an acceleration of the suspension system response as an objective function, the FA strategy is an attempt to find and search for an optimum value of the gains that able to be as a sort of contact information for improving the targeted value obtained from the FL controller. In this work, an external disturbance in the form of sinusoidal waves is applied to the system to verify the sensitivity and durability of the proposed control schemes. Consequently, a comparative assessment between FL controller without having gain scaling and with the gain scaling tuned by FL strategy is investigated an analysis in the form of the amplitude reduction for both body displacement and acceleration responses. Simulation results indicated that the FL with gain scaling shows a good response compared to the FL without gain and its performance is improved by up to 52.1% compared to others.