
科学技术与工程

中国技术经济学会
Early-age Creep Analysis of Reinforced Concrete Beams Incorporating Hydration-based Concrete Modeling Optimization
Sheng-qi MEI , Xiang-xu CHENG , Xing-ju WANG , Yun-xiao BU , Lei BIAN , Teng WU
Science Technology and Engineering ›› 2026, Vol. 26 ›› Issue (18) : 7879 -7890.
To more accurately analyze the early-age creep behavior of concrete, a creep model incorporating the degree of hydration and the effect of micro-prestress loss was developed and applied to the finite element analysis of beam elements. First, the predictive performance of four existing creep models (Delsaute, B4, B3, and MC2010) was evaluated against 640 sets of early-age concrete creep data from the NU-ITI(northwestern university-institute of infrastructure technology) Creep Database. The results show that the Delsaute model achieves the highest coefficient of determination (R2=0.65), with the overall ranking being Delsaute > B4 > B3 > MC2010, indicating a limited generalization capability of the existing models. Subsequently, based on the Delsaute model, viscoelastic theory, and microstructural analysis, the Kelvin-Voigt chain representing viscoelastic deformation is replaced by a viscous element correlated with the degree of hydration, enabling a more accurate representation of early-age viscoelastic characteristics. Meanwhile, an age-dependent viscosity coefficient is introduced to optimize the description of the effect of internal micro-prestress loss on creep behavior. Computational results of the modified model demonstrated a significant improvement in prediction accuracy, with the R2 value increasing to 0.86. Finally, to validate the applicability of the proposed model for the analysis of early-age creep in concrete beams, a constitutive UMAT subroutine is developed and implemented in ABAQUS for the finite element simulation of a concrete T-beam. The simulation results confirm that the proposed model effectively improved the computational accuracy of early-age concrete beam creep analysis. By incorporating the effect of micro-prestress loss, the prediction error is reduced by up to 53%, with the maximum error remaining below 14%.
early-age concrete / creep model / degree of hydration / stress loss / viscoelasticity
/
| 〈 |
|
〉 |