organised by Prof. K. Marti
Federal Armed Forces University, Munich, Germany
Prof. GI. Schuëller
Leopold-Franzens University, Innsbruck, Austria
and Prof.Dr.-Ing. T. Vietor
Technical University of Braunschweig, Germany
In optimization of concrete mechanical structures and (dynamic) systems only partial information is known about the model parameters. This information must be taken into account in order to reach optimal designs/decisions/controls which are least sensitive with respect to parameter variations in the known range of the model parameters. In order to
stabilize mechanical structures under dynamic stochastic applied loadings, active control strategies are taken into account. in order to counteract heavy applied loads. Hence, the original optimization/optimal control problem with
unknown parameters must be replaced by an appropriate deterministic substitute problem incorporating the given information about the unknown (stochastic) parameters. Using decision theoretical principles, substitute problems may be obtained - depending on the available information about the type of uncertainty - by using usually minimax or expectation operators. Typical substitute problems demand then, either, the minimization of the maximum or the expected total costs (initial plus recourse costs) subject to the remaining (mostly simple) design/control constraints, or the minimization of a (simple) expected initial cost function subject to certain reliability constraints involving the probability and/or the expected costs of structural/system failure. Hence, efficient methods have to be determined for coping with several types of uncertainty in order to get robust, i.e., most parameter-insensitive optimal designs/controls.
On the other hand, since material, model and design parameters have a stochastic character, in many application areas the importance of the stochastic modeling has been accepted more and more. Hence, increased demand for robust optimization arises in many technical/industrial areas like aerospace development, passenger vehicle engineering, railway engineering, construction of ships. Thus, a further aim of this special session is to support the exchange between different areas with theoretical and practical focus. This will contribute to an extended view in application areas which includes stochastic modeling and optimization. Furthermore, the demands of application areas are formulated and communicated to areas with more theoretical background.
Please note that papers that are found to fall outside the scope of
this session may be considered for other sessions.