An overview of studies in structural mechanics

The present report gives an overview of the ongoing research programmes in structural mechanics at CEA/DEN. On the whole, these contributions are well representative of the research work performed, more oriented by engineering concerns than driven by pure academic goals. Fundamentally, the developed knowledge results in new methods and improved engineering and computational tools that can be used for CEA needs and transferred to industrial clients and partners. Basic research is carried out with the help of university laboratories, what allows CEA teams to identify the underlying problems and to address them in an adequate manner. Confrontation with other viewpoints and backgrounds takes place in international cooperative actions conducted with academic or industrial research centres, often giving rise to benchmarks. Due to the wide range of problems submitted to CEA/DEN, the R and D topics are numerous and the effort devoted to each of them is limited and sometimes not continuous. Basic research is of course more limited and needs thorough preparation in order to ensure that the key questions, which lock the progress, are really addressed.. Before to end, it is worth mentioning two original research actions which have begun: -) identification of medium state and representation of its variability by a probabilistic approach: this original approach couples inverse method an probability to obtain non directly measurable value from global effect on structures (for example deduce damage from the displacement of a loaded beam) and should be applied to non destructive identification of present state of nuclear reactor enclosures, -) a program of numerical simulations of fluid-elastic instability of a tube bundle submitted to cross flow has been initiated with an Arbitrary Lagrangian Eulerian -ALE- finite element method to obtain a better knowledge and understanding of the phenomenon. From these simulations, the evolutions of pressure and velocity fields close to fluid-elastic instability velocity are being obtained. Moreover, influence of non-dimensional parameters that are characteristic of the coupled problem (Reynolds number, Stokes number, reduced velocity) and their influence on the evolution of modal parameters (added mass, damping and thickness) is being studied through parametric studies. The research actions described here, and those in progress are in agreement with the general pattern of the 'Mechanics 2010' program, but nevertheless the R and D effort is less extensive than it was expected. So, this activity would also deserve more effort in transverse fields like probabilistic methods, innovative measuring techniques and numerical methods. (authors)