结构可靠性评估机器学习方法研究进展

打开文本图片集

关键词：机器学习；结构可靠性评估；结构健康监测；失效模式搜索 中图分类号：TU31；TP181 文献标志码：A DOI：10.16385/j.cnki.issn.1004-4523.202506069

Abstract：Thougouttrng-tesic，eginegstructuresfacecmplexviomentalatiosndultiomponntinteactio effects，ladingtosinificantuncertaintyineirstateisdestraditionaleliabilityessentetodsicreasinglylt adresinglsliupleuremoes，iaritatefuciosxtrelyallfueprobbli，dgiial randomvariables.Inrecentyears，machinelearningmethodshavedemonstratedgreatpotential，offeringnewavenuestoovercomethe combinatorialexplosionandcurseofdimensionalityinreliabilityassessment，signiicantlyehancingcomputationaleficiency，and ultimatelymprovngevaationaccuracy.Tispaperprovdesasstematicevieofthlatestreseahadancs iaplyingmacheleaing toaddressealkylgsiraleblitsessnt.Ilsptseaplsofomaceamosdt structuralreliability evaluation basedonmonitoringdata，andconcludes witha discussionon future developmenttrends.

Keywords：machine learning;structuralreliabilityanalysis;structural health monitoring;failuremodes searching

工程结构在长期服役期内需承受多种环境与荷载作用以及地震等突发灾害事件等威胁，这些作用将导致工程结构不可避免地产生累积损伤，使结构安全性、适用性和耐久性下降[1]。（剩余25217字）