On the monitoring-driven assessment of engineered systems

Abstract

The life-cycle management of structural systems operating under diverse loads involves the tasks of simulation (forward engineering), identification (inverse engineering) and maintenance/control actions. The efficient and successful implementation of these tasks is however non-trivial, due to the ever-changing nature of these systems, and the variability in their interactive environment. Two defining factors in understanding and interpreting such large-scale systems are nonlinear behavior and structural uncertainty. The former is related to the external dynamic loading that might shift the structural response from purely linear to nonlinear regimes, while the latter is related to erroneous modeling assumptions, imprecise sensory information, ageing effects, and lack of a priori knowledge of the system itself. This paper discusses implementation of methods and tools able to tackle the aforementioned challenges. Among other topics, the use of surrogate models and Bayesian-type filters for the reduced representation and identification of uncertain and nonlinear structural systems is discussed.