Abstract
Adaptive structures are equipped with hydraulic actuators for compensating external loads and damping vibrations, enabling lightweight construction and an extended operational lifetime. In the context of structural health monitoring, these actuators also provide the means for active diagnostic approaches, which can improve the diagnostic performance over conventional passive methods. This paper investigates a model-based active diagnosis method for structural faults resulting in a local loss of stiffness. Parametric uncertainties, which are often substantial in civil engineering and pose a major challenge in model-based structural health monitoring, are handled in a probabilistic framework. An upper bound on the probability of a diagnostic error is used as objective of the input optimization, with an alternative objective additionally taking the actuation energy into account. In a simulation study, the proposed method achieves notable improvements in the diagnosis performance and reduces either the required time by up to 75% or the actuation energy by up to 60%.
Users
Please
log in to take part in the discussion (add own reviews or comments).
