Damage detection of civil structures based on hybrid optimization algorithm and combined correlation function of heterogeneous responses

作     者：Zhang, Guangcai Wan, Chunfeng Yang, Zhiyuan Xie, Liyu Xue, Songtao 

作者机构：Southeast Univ Minist Educ Key Lab Concrete & Prestressed Concrete Struct Nanjing Peoples R China Tongji Univ Dept Disaster Mitigat Struct Shanghai 200092 Peoples R China Tohoku Inst Technol Dept Architecture Sendai Japan 

出 版 物：《MEASUREMENT》 (Meas J Int Meas Confed)

年 卷 期：2025年第246卷

核心收录：

学科分类：08[工学] 080401[工学-精密仪器及机械] 0804[工学-仪器科学与技术] 081102[工学-检测技术与自动化装置] 0811[工学-控制科学与工程] 

基　　金：National Key R & D Program of China [2021YFE0112200] Japan Society for Promotion of Science (Kakenhi) [18 K04438] Tohoku Institute of Technology research Grant Postgraduate Research & Practice Innovation Program of Jiangsu Province [KYCX23_0273] 

主　　题：Broyden-Fletcher-Goldfarb-Shanno Combined correlation function Damage detection Heterogeneous responses Hybrid optimization algorithm Output-only method 

摘      要：Accurate and robust damage detection of civil structures subjected to ambient excitations is crucial for ensuring safety and maintaining structural integrity. Heuristic algorithms and traditional reference point-defined correlation function methods have been employed in recent years. However, most single-method optimization strategies suffer from drawbacks such as slow convergence for global search optimizers and strong reliance on the initial guess for local search optimizers. Additionally, selecting the optimal reference point a priori is difficult, and identification accuracy would be severely compromised if the selected reference point exhibits poor response sensitivity to damage. To overcome these limitations, this paper proposes a novel output-only damage detection method for civil structures based on the hybrid optimization algorithm and combined correlation function of heterogeneous responses. The hybrid optimization algorithm integrates the local search operator of the BroydenFletcher-Goldfarb-Shanno (BFGS) algorithm into the global search optimizer of the hybrid Jaya and differential evolution algorithm (HJDEA), leveraging their comprehensive exploration and good convergence capacities. A new objective function is formulated based on the combined correlation function among acceleration and strain, eliminating the need for a reference point. To demonstrate the effectiveness of the proposed method, numerical studies on the Guangzhou new TV tower and a complex three-dimensional space truss structure are conducted. Furthermore, the effect of noise level, sampling frequency, sampling duration, and number of data points are investigated. The results show that the proposed method based on HJDEA-BFGS algorithm and combined correlation function with heterogeneous responses can achieve less than 0.4% mean error for varying damage scenarios and maintains robustness under noise levels up to 20%, demonstrating its reliability for real-world applications.