Enhancing deformation characteristics prediction of coarse-grained soils with time-series generative adversarial network-based data augmentation and pre-training  

作　　者：Ying ZHANG Meng JIA Xuedong ZHANG Liping CAO Ziying AN Hongchao WANG Jinyu WANG 

机构地区：[1]School of Control and Computer Engineering,North China Electric Power University,Beijing 102206,China [2]The State Key Laboratory of Simulation and Regulation of Water Cycle in River Basin,China Institute of Water Resources and Hydropower Research,Beijing 100038,China

出　　处：《Frontiers of Structural and Civil Engineering》2025年第3期396-410,共15页结构与土木工程前沿(英文版)

基　　金：supported by the National Natural Science Foundation of China(Grant No.52078212);the State Key Laboratory of Simulation and Regulation of Water Cycle in River Basin,and China Institute of Water Resources and Hydropower Research,China(No.IWHR-SKL-202003).

摘　　要：Coarse-grained soils are fundamental to major infrastructures like embankments,roads,and bridges.Understanding their deformation characteristics is essential for ensuring structural stability.Traditional methods,such as triaxial compression tests and numerical simulations,face challenges like high costs,time consumption,and limited generalizability across different soils and conditions.To address these limitations,this study employs deep learning to predict the volumetric strain of coarse-grained soils as axial strain changes,aiming to obtain the axial strain(ε_(a))-volumetric strain(ε_(v))curve,which helps derive key mechanical parameters like cohesion(c),and elastic modulus(E).However,the limited data from triaxial tests poses challenges for training deep learning models.We propose using a Time-series Generative Adversarial Network(TimeGAN)for data augmentation.Additionally,we apply feature importance analysis to assess the quality of the numerical augmented data,providing feedback for improving the TimeGAN model.To further enhance model performance,we introduce the pre-training strategy to reduce bias between augmented and real data.Experimental results demonstrate that our approach effectively predictscurve,with the mean absolute error(MAE)of 0.2219 and the R^(2) of 0.9155.The analysis aligns with established findings in soil mechanics,underscoring the potential of our method in engineering applications.

关 键 词：coarse-grained soils deformation characteristics TimeGAN data augmentation pre-training 

分 类 号：TU43[建筑科学—岩土工程]