Dynamic Analysis of Magnetorheological (Mr) Fluid Based Semiactive Suspension System for Vehicular Application Using Nonparametric Approach

Abstract

The magnetorheological (MR) fluid dampers belong to a category of semi-active devices, in which damping force can be varied within a few milliseconds through the application of a magnetic field. The main aim of this project is to investigate the performance of MR damper used as a semi-active suspension system in vehicle models to improve the ride comfort and road holding quality of the vehicle, when subjected to average random road profile and road bump as inputs. Research work starts with design and development of MR damper, which includes optimization of MR damper to study the variation of magnetic flux density with variation of electromagnetic circuit parameters such as current magnitude, number of turns in the coil, coil core length, fluid flow gap and flange length. The optimization study shows that, the magnetic flux density induced in the fluid flow gap increases with increase of applied current, number of turns in the coil and coil core length. The magnetic flux density is seen to decrease with increase of fluid flow gap and flange length. The optimum fluid flow gap, which is obtained from the optimization technique has been considered for fabrication of MR damper. Experimental studies on a developed MR damper with different proportion of MR fluid have been conducted by using dynamic testing facilities at 1.5Hz and 2Hz operating frequencies. Based on the experimental results, the optimum level of parameters such as proportion of MR fluid and operating frequency are evaluated by using Taguchi design of experiments. Then, dynamic behaviour of MR damper with optimum level of parameters has been investigated. Developed damper shows the capability of improving both stiffness and damping properties with variation of electric current. Magnetostatic analysis of MR damper has been carried out, in order to find total magnetic flux density induced in the fluid flow gap. Total magnetic flux density induced in the fluid flow gap is divided into five categories by using statistical categorization technique. The average total magnetic flux density obtained from thestatistical categorization technique has been used to evaluate the damper force. Based on this, non-parametric model has been developed and polynomial function is used to relate the damper force as a function of current. Bouc-Wen model has been used to benchmark the developed non parametric model. The parameters of the Bouc-Wen model are evaluated by minimizing the error between the experimental and predicted force using non-dominated sorting genetic algorithm II (NSGA-II) optimization technique. The hysteresis behaviour of the MR damper is predicted by both models (non-parametric model and Bouc-Wen model) and validated with the experimental investigations. Both parametric and non-parametric models predict the behaviour, which is having good agreement with experimental results. Different mathematical models such as quarter car model (2 DOF), half car model (4 DOF) and full car model (7 DOF) of the vehicle with passive and semiactive suspension systems are formulated. Newly developed non-parametric model of MR damper is used in vehicle model as semi-active suspension system with suitable control strategy. Ride comfort and road holding performances of passive and semiactive suspension systems are found under average random road profile as input. In comparison, the vehicle with MR based suspension system provides better vibration isolation for a vehicle than passive suspension system.