受众多外界因素的影响，混凝土拱坝结构损伤与模态信息之间表现出明显的非线性特征，这使得传统的模态分析很难精确识别结构的损伤程度。针对上述问题，提出一种基于平均曲率模态和最小二乘支持向量机的混凝土拱坝损伤识别方法。该方法在数值模拟的基础上，首先利用平均曲率模态对混凝土拱坝损伤位置进行识别，然后利用最小二乘支持向量机建立平均曲率模态和损伤程度间的非线性关系，对混凝土拱坝损伤程度进行识别。工程实例分析表明，该方法能有效地识别混凝土拱坝同时发生多处不同程度损伤的位置及损伤程度。
Abstract
Affected by many external factors, the relation between concrete arch dam damage and modal information is apparently nonlinear, which makes it difficult to accurately identify the degree of structural damage by traditional modal analysis. Aimed at these problems, a method of identifying the damage of concrete arch dam by using mean curvature mode and least squares support vector machine is proposed. On the basis of numerical simulation, the damage location is firstly identified using mean curvature mode, then the non-linear relationship between mean curvature mode and damage degree is established using least squares support vector machine to identify the damage degree. Engineering example shows that by using this method, the location and degree of various damages occurring simultaneously can be effectively identified.
关键词
平均曲率模态 /
最小二乘支持向量机 /
混凝土拱坝 /
损伤识别
Key words
mean curvature mode /
least squares support vector machine /
concrete arch dam /
damage identification
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 刘义伦,时圣鹏,廖 伟. 利用曲率模态识别桥梁损伤的研究[J]. 振动与冲击,2011,30(8):77-81.(LIU Yi-lun,SHI Sheng-peng,LIAO Wei. Bridge Damage Identification Using Curvature Mode Shapes[J]. Journal of Vibration and Shock,2011,30(8):77-81.(in Chinese))
[2] 张 旭,葛继平,李胡生,等. 基于曲率模态的箱梁损伤识别方法研究[J]. 上海应用技术学院学报(自然科学版),2012,12(3): 201-206.(ZHANG Xu,GE Ji-ping,LI Hu-sheng,et al. Damage Identification of Box Beam Based on Curvature Theory Method[J]. Journal of Shanghai Institute of Technology (Natural Science),2012,12(3): 201-206.(in Chinese))
[3] 李小荣,郭永刚. 基于遗传算法优化神经网络权值的大坝结构损伤识别[J]. 震灾防御技术,2008,3(2):189-196.(LI Xiao-rong,GUO Yong-gang. Dam Damage Identification on the Basis of Optimizing Neural Network Weight by Genetic Algorithm[J]. Technology for Earthquake Disaster Prevention,2008,3(2): 189-196.(in Chinese))
[4] 于 菲,刁延松,佟显能,等. 基于振型差值曲率与神经网络的海洋平台结构损伤识别研究[J]. 振动与冲击,2011,30(10):183-187.(YU Fei,DIAO Yan-song,TONG Xian-neng,et al. Damage Identification of an Offshore Platform Based on Curvature of Modal Shape Difference and BP Neural Network[J]. Journal of Vibration and Shock,2011,30(10): 183-187.(in Chinese))
[5] 练继建,李松辉. 基于支持向量机和模态参数识别的导墙结构损伤诊断研究[J]. 水利学报,2008,39(6): 652-658.(LIAN Ji-jian, LI Song-hui. Damage Diagnosis of Spillway Guide Wall Based on Support Vector Machine and Modal Parameter Identification[J]. Journal of Hydraulic Engineering,2008,39(6): 652-658.(in Chinese))
[6] 马立英,周 鋐,彭晓俊. 曲率模态及其在汽车后桥损伤识别中的应用[J]. 同济大学学报(自然科学版),2011,39(8): 1208-1211.(MA Li-ying,ZHOU Hong,PENG Xiao-jun. Application of Curvature Mode Shape to Damage Identification of Rear Axle[J]. Journal of Tongji University (Natural Science),2011,39(8): 1208-1211.(in Chinese))
[7] 张 静,静 行,袁海庆. 基于曲率模态变化率指标的结构损伤识别[J]. 华中科技大学学报(城市科学版),2010,27(2): 82-86.(ZHANG Jing,JING Hang,YUAN Hai-qing. Structural Damage Detection Based on Change Rate of Curvature Mode[J]. Journal of Huazhong University of Science and Technology (Urban Science),2010,27(2): 82-86.(in Chinese))
[8] 李 波,顾冲时,李智录,等. 基于偏最小二乘回归和最小二乘支持向量机的大坝渗流监控模型[J]. 水利学报,2008,39(12): 1390-1394.(LI Bo,GU Chong-shi,LI Zhi-lu,et al. Monitoring Model for Dam Seepage Based on Partial Least Squares Regression and Least Square Support Vector Machine[J]. Journal of Hydraulic Engineering,2008,39(12): 1390-1394.(in Chinese))
[9] 李 波,徐宝松,武金坤,等. 基于最小二乘支持向量机的大坝力学参数反演[J]. 岩土工程学报,2008,30(11):1722-1725.(LI Bo,XU Bao-song,WU Jin-kun,et al. Back Analysis of Dam Mechanical Parameters Based on Least Squares Support Vector Machine[J]. Chinese Journal of Geotechnical Engineering,2008,30(11): 1722-1725.(in Chinese))
[10]GU C S,LI B. Back Analysis of Mechanical Parameters of Roller Compacted Concrete Dam[J]. Science China: Technological Sciences,2010,53(3): 848-853.
基金
“十二五”国家科技支撑计划项目(2012BAK10B04)
PDF(1151 KB)
