Damage Assessment of Segmental Concrete Beams with Different Types of Joints and Prestressed External Tendons

Precast segmental beams with prestressed external tendons have become increasingly common in bridge constructions owing to the merits of high-quality control and fast onsite construction with minimum environmental impact. Due to the nonlinear characteristics of the dynamic behavior of these segmental beams associated to the joint opening, the traditional condition monitoring and damage assessment methods based on vibration parameters either in time or frequency domain using the linear theory are unsuitable. To overcome this challenge, this paper proposes a damage assessment method through analyzing nonlinear structural dynamic responses for monitoring the conditions of segmental beams with different joint types and prestressed external ten -dons. Two damage indices are proposed to identify the occurrence of damage and to qualitatively indicate the damage severity. In the proposed approach, the measured dynamic responses of the segmental concrete beams under hammer impact are first adap-tively decomposed into a finite number of mono-components by using variational mode decomposition (VMD) technique. Then, the instantaneous frequencies of the decom-posed mono-components from the dynamic responses are obtained by conducting the Hilbert transform. Two damage indicators are defined for the damage assessment and for the identification of the damage location, respectively. Three four-span segmental beams with different types of joints (i.e. dry joints or epoxy joints) and prestressing tendon types (i.e. steel and carbon fiber reinforced polymer tendons) were constructed and tested under five loading levels with different damage severities. The feasibility and accuracy of the proposed approach are verified by comparing the identification results with the secant and initial stiffness extracted from the load-deflection curves at different loading levels. The effectiveness of these two damage indicators is also validated with the structural conditions observed in the tests. The results show that the proposed approach can successfully identify damage in segmental concrete beams.