基于递归量化分析和主成分分析的桥梁结构损伤识别方法
张维锋 , 刘威 , 刘鹏飞 , 王爱华 , 段元锋 , 朴星月
结构工程师 ›› 2026, Vol. 42 ›› Issue (3) : 64 -69.
基于递归量化分析和主成分分析的桥梁结构损伤识别方法
Damage Identification of Bridges Using Recurrence Quantification Analysis and Principal Component Analysis
在实际桥梁运营过程中,许多年久失修的桥梁由于缺乏完整的有限元模型和结构参数,难以通过数值模拟生成大量有标签的样本数据,限制了有监督学习方法的应用。针对这一问题,本文提出了一种基于无监督学习的桥梁损伤识别方法。该方法基于单个加速度传感器采集的数据,首先通过自然激励技术处理原始加速度信号以构建递归图,随后对递归图进行递归量化分析,提取特征指标,并利用主成分分析模型进行训练与异常检测,从而判断传感器所监测区域对应的位置是否存在结构损伤。通过对桥梁的数值模拟进行损伤识别,验证了所提方法的有效性与鲁棒性。
In real-world bridge operations, many aging bridges lack complete finite element models and structural parameters, making it difficult to generate sufficient labeled sample data through numerical simulations. Consequently, the application of supervised learning methods is limited. To address this issue, this paper proposes an unsupervised learning-based bridge damage identification method. First, raw acceleration signals from a single accelerometer are processed using the natural excitation technique to construct recurrence plots. Subsequently, recurrence quantification analysis is performed on these plots to extract characteristic indicators. A principal component analysis model is then trained for anomaly detection, enabling the determination of structural damage within the monitored region. The effectiveness and robustness of the proposed method are validated through damage identification tests conducted on a cable-stayed bridge numerical model.
| [1] |
蔚文杰,王楠,赵正松.工程失败与工程科学:以塔科马海峡大桥事故为例[J].工程研究—跨学科视野中的工程,2020,12(5):488-498. |
| [2] |
WEI Wenjie,WANG Nan,ZHAO Zhengsong.Engineering failures and engineering science:the case of the Tacoma Narrows Bridge collapse[J].Journal of Engineering Studies,2020,12(5):488-498.(in Chinese) |
| [3] |
LIFSHITZ J M,ROTEM A.Determination of reinforcement un-bonding of composites by a vibration technique[J].Journal of Composite Materials,1969,3(3):412-423. |
| [4] |
MESSINA A,WILLIAMS E J.Structural damage detection by a sensitivity and statistical-based method[J].Journal of Sound and Vibration,1998,216(5):791-808. |
| [5] |
FARRAR C R,BAKER W E,BELL T M,et al.Dynamic characterization and damage detection in the I-40 bridge over the rio grande:Los Alamos National Laboratory Report LA-12767-MS[R].Los Alamos National Laboratory,1994. |
| [6] |
FARRAR C R,DOEBLING S W,NIX D A.Vibration-based structural damage identification[J].Philosophical Transactions of the Royal Society A:Mathematical,Physical and Engineering Sciences,2001,359(1778):131-149. |
| [7] |
郏亚坤.基于小波总能量相对变化的桥梁结构损伤识别研究[D].杭州:浙江大学,2017. |
| [8] |
JIA Yakun.Study on damage identification of bridge structures based on relative change of wavelet total energy[D].Hangzhou:Zhejiang University,2017.(in Chinese) |
| [9] |
PANOFSKY H A.Atmospheric Turbulence[M].Albuquerque:Sandia Corporation,1959. |
| [10] |
丁幼亮,李爱群.基于小波包分析的Benchmark结构损伤预警试验研究[J].工程力学,2008(11):128-133. |
| [11] |
DING Youliang,LI Aiqun.Experimental study on damage early warning of benchmark structures based on wavelet packet analysis[J].Engineering Mechanics,2008(11):128-133.(in Chinese) |
| [12] |
BERMAN A.Mass matrix correction using an incomplete set of measured modes[J].AIAA Journal,1979,17(10):1147-1148. |
| [13] |
黄民水,吴玏,朱宏平.噪声影响下基于改进损伤识别因子和遗传算法的结构损伤识别[J].振动与冲击,2012,31(21):168-174. |
| [14] |
HUANG Minshui,WU Le,ZHU Hongping.Structural damage identification under noise influence based on improved damage identification factor and genetic algorithm[J].Journal of Vibration and Shock,2012,31(21):168-174.(in Chinese) |
| [15] |
KHODABANDEHLOU H,PEKCAN G,FADALI M S.Vibration‐based structural condition assessment using convolution neural networks[J].Structural Control and Health Monitoring,2019,26(2):2308. |
| [16] |
YU Y,WANG C,GU X,et al.A novel deep learning-based method for damage identification of smart building structures[J].Structural Health Monitoring,2018,18(1):143-163. |
| [17] |
ECKMANN J P,KAMPHORST S O,RUELLE D.Recurrence plots of dynamical systems[J].Europhys Lett,1987,4(9):973-977. |
国家自然科学基金(52361165658)
/
| 〈 |
|
〉 |