岩石与混凝土蠕变行为研究进展：实验方法、数值模拟与工程适用性综述

Research Progress on Creep Behavior of Rocks and Concrete: Experimental Methods, Numerical Simulation, and Engineering Applicability

摘要: 岩石与混凝土蠕变作为典型的时间依赖性力学行为，对深部地下工程和土木结构的长期变形与安全具有重要影响。随着工程规模扩大和环境复杂化，传统基于短期强度的分析方法已难以满足长期安全评估需求，蠕变效应逐渐成为工程稳定性分析中的关键问题。本文系统综述了岩石与混凝土蠕变研究的主要进展，重点从实验研究、数值模拟方法、蠕变本构模型及其工程适用性等方面进行评述。总结了单轴、三轴及多因素耦合蠕变试验在蠕变阶段划分、长期强度判定和损伤演化机制揭示方面的研究成果；对有限元法、有限差分法和离散元法等数值模拟方法及典型蠕变模型进行了对比分析，评估其在长期变形预测中的适用条件与局限性；进一步探讨了基于损伤力学、微观机制及机器学习方法在复杂工况下蠕变预测中的应用潜力。研究表明，蠕变分析正由经验模型向多尺度、多场耦合和数据驱动方向发展，但模型的可解释性与工程适用性仍需进一步提升。本文可为岩体与结构工程的长期稳定性分析及蠕变模型选取提供参考。

Abstract: This Rock and concrete creep, as typical time-dependent mechanical behaviors, play a critical role in the long-term deformation control and safety of deep underground engineering and civil structures. With the increasing scale of engineering projects and the growing complexity of service environments, conventional analysis methods based primarily on short-term strength are no longer sufficient for long-term safety assessment, and creep effects have become a key issue in engineering stability analysis. This paper systematically reviews the major research advances in rock and concrete creep, with a focus on experimental investigations, numerical simulation approaches, creep constitutive models, and their engineering applicability. The achievements of uniaxial, triaxial, and multi-factor coupled creep tests in identifying creep stages, determining long-term strength, and elucidating damage evolution mechanisms are summarized. Typical numerical methods, including the finite element method, finite difference method, and discrete element method, as well as representative creep models, are comparatively analyzed, and their applicability and limitations in long-term deformation prediction are evaluated. Furthermore, the potential of creep prediction under complex conditions based on damage mechanics, micromechanical mechanisms, and machine-learning approaches is discussed. The review indicates that creep analysis is evolving from empirical modeling toward multi-scale, multi-field coupling and data-driven frameworks; however, the interpretability and engineering applicability of existing models still require further improvement. This work provides a useful reference for long-term stability analysis of rock masses and structures, as well as for the selection of appropriate creep models in engineering practice.