The safety assessment of casks for radioactive material at interim storage facilities or in final repositories includes the investigation of possible handling accidents if clearly defined test conditions are not available from the regulations. Specific handling accidents usually are the drop of a cask onto the transport vehicle or the floor as well as the collision with the wall of the storage building or another cask. For such load cases an experimental demonstration of cask safety would be difficult. Therefore, numerical analyses of the entire load scenario are preferred. The lessons learnt from dynamic finite element analyses of accident scenarios with thick-walled cubical containers or cylindrical casks are presented. The dependency of calculation results on initial and boundary conditions, material models, and contact conditions is discussed. Parameter sets used should be verified by numerical simulation of experimentally investigated similar test scenarios. On the other hand, decisions have to be made whether a parameter or property is modeled in a realistic or conservative manner. For example, a very small variation of the initial impact angle of a container can cause significantly different stresses and strains. In sophisticated cases an investigation of simpler limit load scenarios could be advantageous instead of analyzing a very complicated load scenario.
Metal gaskets in the lid system of transport and storage casks for radioactive material have to guarantee leak-tightness and safe enclosure of the radioactive inventory under normal and accident conditions during transport, in case of accidents and for the longterm interim storage. For safety assessments by nurnerical simulation of the thermomechanical behaviour of a lid system with metal gaskets, the finite element method offers three options to use: i) special gasket elements, ii) complex three-dimensional modeHing with solid elements, and iii) a simplified axisymmetric approach. Gasket elements can be adjusted by many parameters, but they give only a global representation of the experimentally observed gasket behaviour. For calculations of the entire cask including the lid system with gaskets, nonetheless this approach is recommended. To investigate the hardly measurable impact behaviour in detail or to extrapolate the long-term behaviour, a local modeHing of all parts of a gasket and their interactions with effects like elastic-plastic deformation, creep, relaxation, and friction is necessary. The three-dimensional model can describe the change of contact area between outer jacket of the gasket and flange dependent on the load conditions, what overcomes an essential limitation of special gasket elements. To simplify the problern of investigating the underlying physical effects, an axisymmetric lid system can be modelled with axisymmetric finite elements. Usually an approximation is only necessary for the helical spring of a metal gasket. This paper explains basic ideas for an adequate finite element simulation of cask lid systems with metal gaskets and their thermo-mechanical behaviour under specific load scenarios.
Influence of impact angle and real target properties on drop test results of cubic containers
(2017)
Drop test scenarios with cubic containers without impact limiters at interim storage sites or in a final repository have been investigated by numerical simulations. An ideally flat drop is impossible to conduct as a free fall of a container even under laboratory conditions. Dynamic stresses and strains inside the container structure are sensitive to the impact angle. Even very small impact angles cause remarkable changes in the experimental or numerical results when a flat bottom or wall of a container hits a flat target. For drop tests with transport packages the International Atomic Energy Agency (IAEA) regulations define an essentially unyielding target. In contrast, potential accident scenarios for storage containers are derived from site-specific safety analyses or acceptance criteria in Germany. Each interim storage site or repository has a yielding or so-called real target with individual structural and material properties. The real target acts as a kind of impact limiter. A more conservative container design is required if the impact limiting effect of the target is not considered.