Parametric structural optimization of conveyor idler rollers

Resumo

Conveyor rollers are widely used in the mining industry for ore transportation. However, due to the
severe operating conditions, these components may fail prematurely and thus cause high maintenance costs. The
failure of these rollers is usually directly related to the service life of their bearings. This service life is greatly
influenced by the angular deflection of the bearings. Within this context, this work presents an optimization
study of a metallic conveyor roller. The purpose is to find a roller with lower mass, however, maintaining the
angular deflection of the bearings in an acceptable range of operation, without significant changes in the roller's
shape to do not affect the current manufacturing processes. To define the optimization design problem, the study
is based on a Brazilian standard, where factors such as: type of roller, load, allowable stresses and angular
deflections are defined. The optimization algorithm is coupled to a radial basis functions (RBF) metamodel
which predicts the structural response of the roller. The RBF metamodel is built, and iteratively refined, through
finite element analyses performed in the commercial code Ansys. The optimization results indicate the
possibility of obtaining a roller design with lower mass and higher stiffness than those manufactured nowadays.