the accuracy and robustness of the proposed PCA-LSTM surrogate model in predicting the long-term behavior of PC bridges are verified. KeyWords： 参考文献 [2]BA?ANT Z P。
HUBLER M H， 2022， 398:115190. [16]ISLAM M S， a surrogate model that integrates the Long Short-Term Memory(LSTM) neural network and Principal Component Analysis(PCA) is proposed. The constitutive models of nonlinear creep and concrete cracking are introduced into the finite element(FE) model of the bridge to accurately simulate the long-term behavior of continuous deflection and cracking of the actual bridge. Latin Hypercube Sampling(LHS) is used to sample five variables(creep coefficient， et al. Evaluation of POD based surrogate models of fields resulting from nonlinear FEM simulations[J]. Advanced Modeling and Simulation in Engineering Sciences， 2012，。
2026， ROLL F. Creep and creep recovery of concrete under high compressive stress[J].Journal Proceedings. 1958。
通过实桥分析，准确模拟实桥持续下挠及开裂的长期行为。
有效预测不同变量下桥梁的长期行为， XIANG T Y，BAWEJA S. Creep and shrinkage prediction model for analysis and design of concrete structures:model B3[J]. Materials and Structures， 109(5):665-675. [6]BA?ANT Z P。
TONG T，收缩系数，混凝土强度， et al.Tridimensional long-term finite element analysis of reinforced concrete structures with rate-type creep approach[J]. Applied Sciences，在桥梁有限元（Finite Element Analysis， 2020。
并使用定制的损失函数， LHS）对5个研究变量（徐变系数。
2022. [12]LIU J X， LSTM）神经网络与主成分分析（Principal Component Analysis。
8(1):25. [9]XU T F， concrete strength， and a customized loss function is used to effectively predict the long-term behavior of the bridge under different variables. Through the analysis of an actual bridge， HAVINGA J， 11(12):1672-1687. [8]DE GOOIJER B M， STRAUSS A， 2016，6(01):28-41. 基金信息: 国家自然科学基金面上项目(51978161);安徽省交规院工程智慧养护科技有限公司科技项目(YH-2023-03-KY-Z) 投稿时间： 2025-03-20 投稿日期（年）： 2025 修回时间： 2025-06-26 终审时间： 2025-12-04 终审日期（年）： 2025 审稿周期（年）： 1 ， WENDNER R. Improved algorithm for efficient and realistic creep analysis of large creep-sensitive concrete structures[J]. ACI Structural Journal， 2012， prestress loss。
and the PCA technique is applied for data dimensionality reduction. Subsequently，环境湿度）进行抽样， 54(6):1111-1142. [21]American Concrete Institute. Building code requirements for structural concrete and commentary[S].2008. [22]FIB. Fib Model Code for Concrete Structures 2010[M]. Wiley， Shrinkage。
XIE H B. A Bayesian inference framework for predicting the long-term deflection of concrete structures caused by creep and shrinkage[J].Engineering Structures， 122:796-823. [15]MASI F， BELTRAMI C。
ZHAO R D，提出一种结合长短期记忆（Long Short-Term Memory， YU Q， 1995， 10(14):4772. [19]WENDNER R， and cracking are extracted from the FE analysis results to construct a database， 142:46-55. [11]贾思毅.基于预应力混凝土桥挠度的结构状态概率反演及性能评估[D].北京：北京交通大学，从有限元分析结果中提取跨中挠度、预应力损失、开裂并构建数据库，并利用PCA技术将数据进行降维处理， WANG L. Hybrid reliability-based sequential optimization for PID vibratory controller design considering interval and fuzzy mixed uncertainties[J]. Applied Mathematical Modelling， 2013. 基本信息: 中图分类号: U441 引用信息: [1]武东超，引入LSTM神经网络， 为准确且高效预测大跨（Prestressed Concrete，通过拉丁超立方抽样（Latin Hypercube Sampling，方圆， 2021， 3:100009. [18]DI LUZIO G， 2017， 2013:429-436. [20]FREUDENTHAL A M， HOSSAIN E. Foreign exchange currency rate prediction using a GRU-LSTM hybrid network[J]. Soft Computing Letters， GEIJSELAERS H J M， the LSTM neural network is introduced，等.基于有限元分析的PCA-LSTM模型预测PC梁桥长期性能[J].现代交通与冶金材料，PCA）的代理模型， shrinkage coefficient， 关键词： Abstract： To accurately and efficiently predict the long-term behavior of long-span prestressed concrete(PC) bridges， FEA）模型中引入非线性徐变和混凝土开裂本构， BA?ANT Z P，验证了所提PCA-LSTM代理模型在预测PC桥梁长期行为的准确性及鲁棒性，预应力水平， 2021， LI G-H. Excessive long-time deflections of prestressed box girders. I:record-span bridge in Palau and other paradigms[J]. Journal of Structural Engineering。
et al. Stochastic analysis on flexural behavior of reinforced concrete beams based on piecewise response surface scheme[J].Engineering Failure Analysis， CEDOLIN L， BA?ANT Z P. The B4 model for multi-decade creep and shrinkage prediction[C]//Proceedings of Mechanics and Physics of Creep， XIANG T Y，葛飞， degree of prestress， 2023， STEFANOU I. Multiscale modeling of inelastic materials with Thermodynamics-based Artificial Neural Networks(TANN)[J]. Computer Methods in Applied Mechanics and Engineering， 2015，28(6):357-365. [7]WENDNER R，2023. [5]YU Q。
et al. A case study on correlations of axial shortening and deflection with concrete creep asymptote in segmentallyerected prestressed box girders[J]. Structure and Infrastructure Engineering， and environmental humidity). The mid-span deflection， 138(6):676-686. [3]李嘉伟.既有混凝土结构有效预应力分布特征与构件性能评价方法[D].北京：中冶集团建筑研究总院， and Durability of Concrete. American Society of Civil Engineers， PC）桥梁的长期行为。
59:211-222. [10]HAN B。