Creep Analysis of Thin-Walled Structures

In addition to the elastic behavior, engineering materials at elevated temperatures tend to creep or, with other words, a spontaneous and a time-dependent material response can be observed. Due to the increasing safety requirements for power plants, aircraft components, equipment for chemical processes, etc. the time-dependent material behavior should be taken into account in the design process. Since many structural elements working under creep conditions can be classified as thin-walled structures the analysis is connected with the following three items: the choice of a suitable material behavior model, the choice of an adequate structural analysis model, and the choice of a suitable numerical solution technique. All three items are interlinked. For example, the choice of the structural analysis model (beam, plate, shell, etc.) has a significant influence on the numerical effort. On the other hand, the accuracy of the calculations is influenced by the material behavior model taking into account more or less effects. Creep mechanics is a branch of solid mechanics with a history of more than 100 years. After a brief discussion of the historical development and the introduction of some important references two creep behavior models are presented. For the first one the material behavior is not influenced by the kind of stress state while in the second case significant differences in dependence on the kind of the stress state can be observed. A typical example of such behavior is the different behavior in tension and compression. This behavior can be observed, for example, for the tertiary creep characterized by the damage evolution under tensile conditions. If we establish compression conditions, the creep deformations are changing but the damage state is partly frozen (no nucleation of new voids, for instance). The last part is devoted to the structural models. For thin-walled structures all simplifications of the analysis equations are founded on the assumptions with respect to the thinness of the structure. In this case, for instance, one can introduce some hypotheses for the stresses, strains and/or displacements in the thickness direction, and with the help of these hypotheses governing equations, reduced with respect to the dimension, can be established (instead of a system of three-dimensional equations we have a system of two- or one-dimensional equations). On the correctness and accuracy of such an approach will be reported.