Failure analysis of complex bonded structures: Experimental tests and efficient finite element modelling by tied mesh method

This paper aims at assessing the accuracy and applicability of an efficient finite element (FE) computational method for the prediction of the post-elastic response of large and complex bonded structures. In order to overcome the limitations found in the technical literature, such as the use of user defined elements, the present work assesses the applicability of a reduced computational technique, named Tied Mesh (TM) method, previously presented by the authors. The method is based on standard modelling tools, describes the adherends by semi-structural elements (plates or shells), and the adhesive by means of a single layer of cohesive elements. The benchmarks for the TM method are the force–displacement curves obtained by experimental tests on a complex, industrial-like structure. A square thin-walled beam is considered, which made of two different portions butt joined by overlapping thin plates on each side. Two different geometries of the beam are loaded by a three point bending fixture up to failure, thus originating a complex stress field on the bonded region. The comparison with the experimental data shows a good accuracy of the proposed TM method in terms of structural stiffness, maximum load (error below 10%) and post-elastic behaviour up to the collapse of the structure. The numerical precision and the computational speed make the TM method very useful for the efficient analysis of complex bonded structure, both in research and industrial world.