A methodology for identifying the damage state of concrete beam structures under cyclic dynamic load based on internal acceleration response

Abstract In transportation engineering, three‐point bending tests are a standard method to evaluate the flexural performance of highways, runways, and bridges, as these structures primarily undergo bending stresses under repeated traffic loads. Conventional monitoring responses of three‐point bending tests are strain or displacement, which have the disadvantage of poor overall structural perception. To overcome this, this study introduces an innovative methodology for identifying damage states in concrete beams under cyclic dynamic loading, utilizing internal acceleration responses. Dynamic three‐point bending tests were conducted with integrated Micro‐Electro‐Mechanical System (MEMS) acceleration sensors embedded inside the beams, in addition to traditional crack mouth opening displacement (CMOD) and deflection monitoring. The acceleration data were validated using digital image correlation (DIC). Experimental results indicate that the ratio of the amplitude at twice the excitation frequency to that at the excitation frequency increases as the damage state worsens, demonstrating that acceleration monitoring is more sensitive to varying damage levels and can provide early fracture warnings compared with CMOD and deflection monitoring. The concrete beam structure with cracks behaves as a nonlinear system. Signal denoising represents a critical preprocessing step for acceleration analysis, and this study proposes a novel effective methodology to address this challenge. Acceleration monitoring offers advantages such as comprehensive structural perception and ease of embedding, which avoids interference from traffic loads and is more suitable for long‐term dynamic monitoring of internal damage in transportation structures primarily subjected to bending effects.