MS 29

Optimal Design of Uncertain Dynamical Systems

Organizers

Marcos Valdebenito,
TU Dortmund University, Germany

Héctor Jensen,
Santa Maria University, Chile

Seymour M.J. Spence,
University of Michigan, USA

André Beck,
University of Sao Paulo, Brazil

Matthias Faes,
TU Dortmund University, Germany

Zhan Kang,
Dalian University of Technology, China

Abstract

Optimization strategies offer the means to explore different design configurations of systems and structures subject to dynamic loading, thus allowing to make efficient use of available resources. However, this task can become quite involved in practice due to inherent uncertainties associated with dynamic loading, boundary conditions and material properties, for example. In such situations, probability theory and other uncertainty models offer the means for characterizing uncertainty using the tools of stochastic dynamics. Nonetheless, this typically leads to what is commonly denominated a double loop problem, which involves performing optimization at the outer loop and uncertainty quantification (reliability analysis) at the inner loop. In general, the solution to such a class of problems is quite demanding, particularly in view of phenomena encountered when dealing with dynamic systems, such as resonances, mode veering, and nonlinear behaviours, to name a few. In this context, the aim of this MS is to bring together some of the latest developments on methods for optimal design of dynamical systems subject to uncertainties, including (but not limited to):

• Approaches for reliability-based optimization in dynamics, including single-loop or sequential approaches.
• Robust optimization of uncertain dynamic systems and structures.
• Development and application of specialized algorithms for uncertainty quantification and sensitivity analysis in dynamics.
• Topology optimization involving uncertainty in dynamics.
• Life-cycle and performance-based optimal design of dynamical systems, including optimal inspection and maintenance scheduling.
• Supercomputing and artificial intelligence (AI) for analysis in dynamics.
• Multi-Objective design optimization of uncertain dynamical systems.
• Practical applications of optimal design for large-scale systems subject to dynamic loading.