Structural Damage Identification Method Based on Flexural Difference Curvature Matrix

Wen Jing Gao *

School of Civil Engineering and Transportation, North China University of Water Resources and Electric Power, Zhengzhou 450045, China.

*Author to whom correspondence should be addressed.


Abstract

Engineering structures continuously accumulate damage during their service life due to various factors, making effective damage identification and health monitoring a vital necessity. Addressing the inaccurate localization issue of traditional flexibility-based damage identification methods under slight or multiple structural damages, this paper proposes a novel structural damage identification method based on the flexibility difference curvature matrix. Utilizing the modal information of the structure before and after damage, the flexibility difference matrix is first calculated, and then differentiated once by column and by row to obtain the flexibility difference curvature matrix. The main diagonal elements are extracted to form a column vector (δ FCMD) as the damage detection index. Taking an engineering beam as an example, numerical simulations are conducted on simply supported and fixed-ended beams under single and multiple element damage conditions. Comparative analysis with existing flexibility indices, such as flexibility difference (δ F), rate of flexibility diagonal (RFD), and change in uniform load surface curvature (ULSC), shows that the δ FCMD index requires only the first few low-order modes of the structure. It can not only accurately locate single and multiple damages but also effectively evaluate the severity of the damage based on the magnitude of the curve mutation. Its comprehensive performance is significantly superior to other comparative indices.

Keywords: Engineering structure, damage identification, flexibility difference curvature ma-trix, numerical simulation


How to Cite

Gao, Wen Jing. 2026. “Structural Damage Identification Method Based on Flexural Difference Curvature Matrix”. Advances in Research 27 (2):163-74. https://doi.org/10.9734/air/2026/v27i21610.

Downloads

Download data is not yet available.