Research database

The Finite Element Method (FEM) is one of the most successful outcome of computational mechanics with high impact in various engineering fields, including automotive, civil, bio, aerospace and naval sectors. In the current version of FEM – referred to as FEM 1.0 - beam, plate, and shell elements are built using assumptions dating back to Leonardo da Vinci beam-like models, i.e., the deformation concerns reference lines or surfaces, and sections or normal remain unstretched. Over the last years, a vast body of publications has shown that these assumptions have various shortcomings and preclude the proper modeling of the physical behavior of a deformable body. An example is locking mechanisms limiting the analysis of local phenomena, laminates, and multifield problems. FEM users - strongly relying upon highly effective, efficient, and user-friendly commercial software - are not fully aware of such shortcomings. This proposal aims to overcome the assumptions of the structural theories of FEM 1.0 and unleash a new framework: FEM 2.0. The first objective is to improve the accuracy of FEM 1.0; then, to provide solutions to problems that, currently, cannot be tackled. The methodology is based on the compact notation of the Carrera Unified Formulations, CUF, allowing arbitrary expansion functions to model the unknown fields. Academic and non-academic international institutions will work with PI and disseminate the new tools. Partnerships with commercial software-house will be explored, and plug-in interfaces created. Open-source codes will be made available to the computational mechanics' community. FEM 2.0 aims to provide a computational tool for solving multiphysics and multiscale problems for large structures, create a direct link between CAD and CAM software with no need for post-processing, overcome the limitations of FEM 1.0 in solving thermal stress and layered structure problems, and create a platform for the nonlinear analysis of structures including multibody capabilities. FEM2.0 would represents a kind of ‘revenge’ of the decades of work done by eminent contributors to mechanics of structures vs an unprecednet application of their findings in a Second Generation Finite Element implementions.