Abstract

In recent years, semi-active magnetorheological (MR) and electrorheological (ER) fluid technology based devices and systems have been developed and successfully utilized in many applications as valves, shock absorbers, dampers and clutch/brake systems. These promising devices have the adaptivity of the fully active systems to accommodate varying external excitations while maintaining the reliability and fail-safe features of the passive systems. Compared with ER based devices or systems, MR based devices have recently received special attention due to their high performance with minimal power requirements. Moreover MR fluids have significantly higher yield strength and are less sensitive to contaminants and temperature compared with the ER fluids.
The geometric optimal design of MR valves/dampers is an important issue to improve the damper performance, such as damping force, valve ratio and inductive time constant. Considering this, the primary purpose of this study is to establish a general design optimization methodology to optimally design single–coil annular MR valves constrained in a specific volume in MR damper. To accomplish this, first the damping force of MR damper has been modeled using Bingham plastic model. The magnetic circuit of MR damper has been analyzed using finite element method in ANSYS environment to obtain magnetic field intensity which can be subsequently used to obtain the yield stress of the MR fluid in the active volume where the magnetic flux crosses. Then the developed finite element model of the MR valve is effectively used to construct an approximate response function relating the magnetic field intensity to the identified design parameters in the selected design space using response surface method and design of experiment methodology.
Using the derived approximate relation for the magnetic field intensity in the MR damper model, the design optimization problem has been formulated using gradient based nonlinear mathematical programming technique based on the Sequential Quadratic Programming (SQP) technique and also stochastic optimization technique based on the Genetic Algorithm (GA) to find optimal geometrical parameters of the MR valve in order to maximize the damping performance under given constrained volume.
Finally a PID controller has been designed to evaluate the close-loop performance of the optimally designed MR damper in a quarter-car suspension model.