III-2: Assessment of multiscale partialmodels of structural elements and their interaction

Continuously increasing computational power and sophisticated softwares tempt to raise the complexity of structural modeling to gain more accurate results. With increasing complexity more effects are considered by the model at the same time. Such models can be universally applicable. On the other hand, complex models mostly demand the knowledge of more input parameters in order to provide better results. The concrete determination of these parameters is naturally subjected to uncertainties, which inevitably influence the simulation results. Although more complex models can give more accurate results, the uncertainties accompanying the structural response values increase. From an engineer point of view this unavoidable contiguity between complexity and uncertainty leads to the question, which model should be chosen for a given situation? More precise, what to select from a space of possible partial-models that would represent a balanced compromise between accuracy in the result and the effort spend in modeling?

The aim of this project is to develop methods that enable to determine and assess the model prognosis ability in a quantitative manner. The focus is on models for primary and secondary structural elements of multi-storey RC-frame buildings. Examples for these elements can be columns, beams or frame-joint as primary members and infills as secondary members. Particular attention is paid on the interaction between the primary and secondary members studying their bearing behavior. Such interactions may be observed during extreme loading, such as earthquakes. Beside the development of assessing methods an additional part of the work is to define an appropriate scale for the model assessment. The identification of appropriate damage indicators (e.g. deformations and/or consideration of energy balance) is the basis of the assessment scale to capture the damage of frame and infill quantitatively and qualitatively. Furthermore, an attempt to deploy the considered models in the form of numerical simulations will be performed, in order to describe the consequence of the models choice of the structural elements in the context of a global response of the entire building. The challenge is to perform reliable simulations on element level and building level as well.

Tutors: J. Schwarz, F. Werner, P. Gülkan, M. Bikçe

Contact: Mathias Leipold