Failure of steel sandwich structures in compression: Modelling and testing

点击翻译按钮获取中文摘要

The transportation industry is a significant contributor to global green-house gas emissions. The shipping industry in specific currently accounts for approximately 3% of CO2 emissions, and this number is predicted to rise to 18% by 2050. If we can reduce the weight of vehicles, we will reduce the fuel consumption. In some types of ships, the ratio of structural steel weight to total weight is very large, for example 40-60% in container ships, ferries and cruise ships. It is possible to decrease some of that weight by using lightweight structures. Steel sandwich panels are lightweight, relatively cheap, recyclable and easy to implement in large structures like ships, but also bridges and buildings. A type which is easy to produce on a large scale is one where unidirectional stiffeners in the core are laser-welded to the face plates. A highly optimized version of the core shape consists of vertical stiffeners, the so-called webs, forming a web-core sandwich panel. In order to be able to use these panels as load-carrying members safely, their mechanical behaviour needs to be better understood. This proposal focuses on their failure under in-plane compressive loading, both uni-axial and bi-axial. The behaviour is complex, non-linear and influenced by numerous factors. These panels are especially sensitive to loading perpendicular to the webs, since the voids in the core can be empty, which results in very low transverse shear stiffness and causes unique structural features. The behaviour will be studied using numerical and experimental methods. Furthermore, this proposal aims to develop an effective strength prediction method that can model the progress of the failure and the load redistribution to the surrounding structure in a large assembly, e.g. a ship. The method will be based on equivalent single-layer theory (ESL) and nonlinear homogenized stiffnesses. The method will be implemented via user-defined subroutine in commercial finite element method (FEM) software, thus the design companies do not need to purchase additional software packages. The time savings are enormous in comparison with alternative FEM analyses where the topology of the panels is explicitly modelled. Thick-face effect and modified couple stress theory will be potentially used within the ESL method in the case of very low shear stiffness of the panels and for cases where continuum assumption is not strictly valid. In addition, methods for ultimate strength will be developed, which will allow the determination of the reserve strength of the panels, i.e. the difference between first and final failure. The developed methods will be validated with experiments. For the first time the tests will be carried out on the full-scale panels. The research program will train engineers who will be specialized in numerical and experimental methods in solid mechanics and will possess a broad understanding of structural analysis.