Damage Identification for Beam Structures Using the Laplace Transform-Based Spectral Element Method and Strain Statistical Moment

Abstract

The Laplace transform-based spectral element method (LTSEM) can avoid some limitations of the conventional Fourier transform-based spectral element method (FTSEM) in handling finite-length structures, such as the wrap effect. In this paper, a damage identification technique based on the LTSEM and strain statistical moment (SSM) is proposed, which is combined with a general genetic algorithm. The dynamic responses of beam structures were first analyzed using the LTSEM and were validated by finite-element analysis results. Then, the strain statistical moment was used as the damage indicator; the damage location and severity for beam structures were numerically identified on the basis of the proposed technique. Finally, an experimental investigation for damage detection on a simply supported steel beam was also implemented. The experimental results verified that the proposed damage identification technique is feasible and has robust antinoise performance.