Computational & Technology Resources
an online resource for computational,
engineering & technology publications
Civil-Comp Proceedings
ISSN 1759-3433
CCP: 88
PROCEEDINGS OF THE NINTH INTERNATIONAL CONFERENCE ON COMPUTATIONAL STRUCTURES TECHNOLOGY
Edited by: B.H.V. Topping and M. Papadrakakis
Paper 78

Optimization of H_infinity Controller with Preview for Semi-active Magnetorheological Suspension Systems

R.S. Prabakar, S. Narayanan and C. Sujatha

Machine Design Section, Department of Mechanical Engineering, Indian Institute of Technology Madras, Chennai, India

Full Bibliographic Reference for this paper
R.S. Prabakar, S. Narayanan, C. Sujatha, "Optimization of H_infinity Controller with Preview for Semi-active Magnetorheological Suspension Systems", in B.H.V. Topping, M. Papadrakakis, (Editors), "Proceedings of the Ninth International Conference on Computational Structures Technology", Civil-Comp Press, Stirlingshire, UK, Paper 78, 2008. doi:10.4203/ccp.88.78
Keywords: H_infinity active control with preview, semi-active MR damper with preview, equivalent linearization, modified Bouc-Wen model, NSGA II.

Summary
Of late electrorheological (ER) and magnetorheological (MR) fluids, which can change their viscosity significantly on the application of suitable electrical and magnetic fields respectively, have been used in vehicle suspensions as a semi-active suspension system by a number of researchers. These authors have used different control policies like skyhook, hybrid control, linear quadratic regulator (LQR), H_infinity to control the input current supplied to the MR damper. The weighting functions required in the controller design have been chosen arbitrarily. Moreover, the semi-active suspension with preview has not been considered in the literature so far.

In this paper, the response of a quarter car vehicle model with semi-active MR damper and with preview control is considered for the first time. The effect of preview control on the MR damper performance is investigated. The input currents to the MR damper are chosen in such a way that the control force generated by the MR damper is equivalent to the control force of an active control scheme like the H_infinity control. In this way an attempt is made to enhance the performance of the semi-active MR damper to levels of performance of H_infinity control.

The vehicle is assumed to be moving over a random rough road, which is modelled as the output of a first order linear filter to white noise excitation, with a constant velocity. The semi-active MR damper suspension with preview control is designed by minimizing a performance index which is a weighted function of performance measures such as sprung mass acceleration, suspension stroke, road holding and control force. The weighting constants with respect to the performance measures are optimally obtained by using a multi objective optimization technique non-dominated sorting genetic algorithm (NSGA II) by minimizing the three objectives namely the sprung mass acceleration, suspension stroke and road holding and with a constraint on the mean square control force to lie between the control force of the MR damper for zero and maximum input currents.

The MR damper used in this work is an actual MR damper fabricated in the laboratory which uses an indigenously developed MR fluid. The modified Bouc-Wen model is used to characterize the hysteretic behaviour of the MR damper and the same is used in a quarter car vehicle model. The model parameters of the MR damper are also evaluated using NSGA II, combined with a pareto optimal scheme minimizing the error between experimental and theoretical damper forces for two different input currents. The results show that the overall performance of the semi-active MR damper suspension with preview control improves with an increase in preview distance and saturates beyond a particular preview distance. The performance of the semi-active MR damper with preview control is much better than that of passive suspension for different preview distances and approaches H_infinity active suspension with preview.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description