基于K-SVD字典学习与ResNet的混凝土结构损伤识别

打开文本图片集

中图分类号：TU375 文献标识码：A DOI： 10. 7535/hbgykj. 2025yx05005

Damage identification of concrete structures based on K-SVD dictionary learning and ResNet

LI Hongxin¹，CHEN Zonggang²，HAN Song²

（1.Maerdang Branch of Qinghai Electric Power Company Limitedof National Energy Group，Maqin，Qinghai 814099，China; 2.Power China Northwest Engineering Corporation Limited， Xi′an ，Shaanxi 710100，China）

Abstract：Toaddress theissuesoflowsignal-to-noiseratio，highbackgroundnoise，and nonstationarityindetectingconcrete structure source signals based on piezoelectric wave method，a piezoelectric signalfiltering method based on K-singular value decomposition（K-SVD）toupdate the dictionary was proposed，andthedamageofconcrete structures were identified.Firstly， piezoelectric signals from concrete structures in both crackedand intact states were collcted and clasified. Secondly，the acquired piezoelectric signals werefiltered，and thetheresultsusing theK-SVDdictionarylearning method werecomparedand analyzedwiththeunfilteredresults toevaluatetheapplicablityoftheK-SVDdictionarylearningfiltering method.Finally，the filteredpiezoelectric signals using ResNet wereclassifiedandrecognized.Theresults showthatthemethod basedonK-SVD dictionarylearningandResNetcanstably identifythepiezoelectricsignalsof internal damageinconcretestructures.The accuracy of damage signal recognition in training set and test set is 93.25% and 92.38% ，respectively. The recognition accuracy of lossless signal is 95.41% and 94.67% ，respectively，which is more than 1O percentage points higher than that of unfilteredsignal.Theefectivedamage identification inconcretebridge structures throughtheintegrationof K-SVDdictionary learningandResNet hasachieved thelocalizationof internal damageareas intheconcretestructures.Theresearch findings present a novel approach to data processing in the health monitoring of concrete bridge structures.

Keywords：concrete and reinforced concrete structures；piezoelectric wave method;K-SVD；ResNet；damage identificatio1

近年来，中国交通强国战略推动公路、桥梁等基础设施投资持续增长[1-2]。（剩余14472字）