Experimental validation of flexible multibody dynamics beam formulations

In this paper, the accuracies of the geometrically exact beam and absolute nodal coordinate formulations are studied by comparing their predictions against an experimental data set referred to as the "Princeton beam experiment." The experiment deals with a cantilevered beam experiencing coupled flap, lag, and twist deformations. In the absolute nodal coordinate formulation, two different beam elements are used. The first is based on a shear deformable approach in which the element kinematics is described using two nodes. The second is based on a recently proposed approach featuring three nodes. The numerical results for the geometrically exact beam formulation and the recently proposed three-node absolute nodal coordinate formulation agree well with the experimental data. The two-node beam element predictions are similar to those of linear beam theory. This study suggests that a careful and thorough evaluation of beam elements must be carried out to assess their ability to deal with the three-dimensional deformations typically found in flexible multibody systems.