# III-1: Evaluation of coupled partial models for complex load-bearing structures with timedependent impact

## III-1: Assessment of coupled partial models of complex structures under time dependent actions

The prognosis of the behaviour of complex structural systems is often obtained from partial models of the whole system, however, such complex structures are actually a combination of different partial models. The coupling between these partial models and the level of abstraction used in modelling may strongly affect the prognosis of the whole. Moreover, when studying a single partial model, it is often tuned to serve a certain response quality or to fit an analysis method. In the context of the above, an appropriate quality assessment of the whole model is rarely performed but of need.

For a complex structural system, for example a bridge, a whole model is built using coupled partial models, starting from the foundation heading up to the substructure, superstructure and to the time dependent actions on the bridge due to traffic loading. By studying the sensitivity and the robustness of these coupled models with respect to the input parameters and abstraction levels, a conclusion is drawn and the quality of the whole model is quantified by its uncertainties.

The compatibility between the partial models is of major importance in order to couple them without significant loss of information. For example, applying that to the modelling of a bridge bearing as a coupling element between two partial models or the coupling of the time dependent actions with the structure. Besides the question of compatibility in the coupled partial models and the analysis methods, the response scenario is an active player in the quality assessment of the coupled models. The solution of the described problem is a theoretical systematic study applied to the partial models and their coupling for different response scenarios considering different abstract levels.

Tutors: K.-J. Witt; U. Freundt, C. Könke, Ch. Bucher

Kontakt: Ghada Karaki