Ajuste de un modelo directo de amortiguador MR mediante algoritmos genéticos

Abstract

This Bachelor Degree Project seeks to model a direct magneto-rheological damper aided by genetic algorithms. The main function of the suspension system of a vehicle is to ensure that all tires are in contact with the pavement, that all vibrations created by the contact of the tires are absorbed and that the suspensions assures safety and comfort to the vehicle’s passengers. There are three different set ups for the suspension system of a vehicle, based in its control system. Passive suspension is designed to function at specific fixed conditions. Active suspension adds actuators in the suspension system, instead of the conventional elements, in order to produce a force for any given situation. The possible set up is semi-active suspension that addresses the problem related with the running costs of active suspension and presents an intermediate design between active and passive suspension. The damper in a semi-active suspension is, according to the latest research lines, the magneto-rheological damper. These dampers have a magneto-rheological fluid inside, which contains a 40% of suspended ferromagnetic particles in a lubricant solution. These particles are able, when induced with an electric current, to polarize forming a magnetic structure that increases their hysteric resistance. With this magneto rheological phenomena the rigidity of the damper is able to vary at a reduce cost. There are two types of models. The direct model, which will be studied in this bachelor project, aims to predict the magneto rheological damper force by feeding a known current to the damper. The inverse model, on the other hand, estimates the current that needs to be fed to apply a determined force. This bachelor degree project implements a direct model to experimental magneto rheological data aided by Genetic Algorithms.