Intelligent controllers for vechicle suspension system using magnetorheological damper

Abstract

Semi-active suspension control with magnetorheological (MR) damper is one of the most fascinating systems being studied in improving the vehicle ride comfort. This study aims to investigate the development of intelligent controllers for vehicle suspension system using MR damper, namely, the proportional-integral-derivative (PID) and fuzzy logic (FL) controllers optimized using particle swarm optimization (PSO), firefly algorithm (FA) and advanced firefly algorithm (AFA). Since the conventional optimization method always has a problem in identifying the optimum values and it is time consuming, the evolutionary algorithm is the best approach in replacing the conventional method as it is very efficient and consistent in exploring the values for every single space. The PSO and FA are among of the evolutionary algorithms which have been studied in this research. Nevertheless, the weakness of FA such as getting trapped into several local minima is an attractive area that has been focused more as a possible improvement during the evolutionary process. Thus, a new algorithm based on the improvement of the original FA was introduced to improve the solution quality of the FA. This algorithm is called advanced firefly algorithm. A parametric modelling technique known as Spencer model was proposed and employed to compute the dynamic behaviour of the MR damper system. The Spencer model was experimentally validated and conducted to capture the behaviour of the Lord RD-1005-3 MR damper with the same excitation input. A simulation of a semi-active suspension system was developed within MATLAB Simulink environment. The effectiveness of all control schemes were investigated in two major issues, namely the ability of the controller to reject the unwanted motion of the vehicle and to overcome the damping constraints. The result indicates that, the PID-AFA control scheme is more superior as compared to the PID-PSO, PID-FA, FL-PSO, FL-FA, FL-AFA and passive system with up to 27.1% and 19.1% reduction for sprung mass acceleration and sprung mass displacement, respectively. Finally, the performance of the proposed intelligent control schemes which are implemented experimentally on the developed quarter vehicle suspension test rig shows a good agreement with the results of the simulation study. The proposed control scheme of PID-AFA has reduced the sprung mass acceleration and sprung mass displacement over the FL-AFA and passive system up to 28.21% and 16.9%, respectively.