Stochastic improvement of structural design
The paper presents some cases referred to the probabilistic analysis and optimization of structural details of aeronautical as well as of automotive interest; the authors examine the case of a riveted sheet structure, followed by the study of an aeronautical panel, whose residual strength in presence of propagating cracks has to be increased and finally the case of an absorber of the type used in cars to reduce acceleration impacting passengers during an accident and whose design has to be improved. In all cases the final behaviour is influenced by design, manufacturing process and working conditions.
modeFRONTIER was used in the first case regarding a riveting operation. The residual compression load between the sheets (or the traction load in the stem of the rivet) was maximized while keeping the radial stress acting on the wall of the hole as low as possible. After defining all variables and parameters, the work can be set to be run by means of an user-defined script (AnsysLsDyna.bat ); the input values were embedded in an input file for Ansys preprocessor, which built a file to be processed by LS-Dyna to simulate the riveting operation.
The first task in modeFRONTIER was carried out in such a way as to perform a simple 3-level DOE analysis of the problem; the second task assigned to the software was to build the response surface of the problem to save time since each LS-Dyna trial was very time-consuming if compared to the use of RSM. Thus, the Pareto-set for the riveting problem was obtained and the last step was the choice of the most meaningful result, which was carried out by means of modeFRONTIER's MCDM tool.