In the field of rock mass engineering, the precise simulation of the post-peak mechanical behavior of rocks is essential for ensuring engineering safety and for the prevention and management of disasters. To overcome the limitations associated with fixed parameter approaches in conventional Mohr-Coulomb models for simulating post-peak failure stages, this study introduces an innovative experimental data-driven multi-parameter dynamic collaborative correction method. Initially, by integrating Python with the FLAC3D platform, we developed a real-time backpropagation algorithm alongside a three-dimensional constitutive field dynamic iteration model. This framework facilitates the multi-threaded collaborative optimization of parameters, including cohesion, internal friction angle, and dilatancy angle, via embedded interfaces. Subsequently, utilizing strain gradient adaptive theory, we devised a real-time data assimilation engine capable of dynamically adjusting constitutive parameters through cyclic correction mechanisms. This approach effectively addresses the modeling challenges posed by the nonlinear coupling effects inherent in traditional static segmentation methods, which exhibit errors exceeding 15%.During the validation process, a numerical model for uniaxial compression, with dimensions of 50 mm×50 mm×100 mm and comprising 2 541 mesh elements, was utilized. A Python script was employed to dynamically invoke the s.stress()[2][2] function in FLAC3D, allowing for the extraction of stress fields. This process initiated multi-parameter collaborative corrections whenever the experimental data surpassed a deviation threshold of Δσ=0.01 MPa. The experimental findings demonstrate that the dynamically corrected model effectively captured the post-peak strain-softening behavior of the rock. The stress levels predicted by the original model consistently exceeded the actual values, whereas the stress simulations from the corrected model aligned closely with the experimental values, thereby completely mitigating the trend of overestimation. This study offers novel insights into optimizing the accuracy of rock parameter estimation and provides a scientific basis and guidance for geotechnical engineering design.
岩石是采矿、基坑和隧道等各类工程中的常见材料,对于各类工程的稳定性具有重要影响,因此研究其力学行为显得十分必要。岩石是一种非均匀脆性材料,在对其进行应力加载至峰值强度后,岩石材料的强度会迅速降低,但依旧会保有一定强度,这种由于变形引起岩石材料强度降低的现象称为岩石的应变软化(Wang et al,2020)。为准确模拟岩石的应变软化现象,数值软件通过特定的本构模型(刘志祥等,2023;Chen et al,2024)控制岩石力学参数,从而间接控制岩石的变形与破坏。然而,研究人员对岩体的本构关系和变形破坏机理的理解不够透彻,加之输入数值模型的数据参数不够准确,导致模拟结果的可靠性和准确性难以保证(刁心宏等,1999;唐春安,1999),因此构建精确岩石力学参数模型尤为关键。
在数值模拟中,构建真实的岩石应变软化模型是岩体工程计算分析的关键问题。数值模拟技术被广泛应用于揭示岩石峰后力学行为等方面,其突破主要体现在参数关联模型的构建和数值实现方法的创新上。通过在FLAC3D平台引入动态参数修正机制,实现了岩石本构行为与实验数据的精准映射。首先,建立了黏聚力和内摩擦角等关键参数与塑性应变梯度的非线性耦合方程,揭示了不同岩石在应变软化阶段的动态响应规律(张帆等,2008;王卫华等,2021);其次,提出了广义力学参数框架,通过状态参数重构屈服面模型,显著提升了软弱岩体在复杂围压下的模拟精度(陆银龙等,2010;孙闯等,2015)。同时,基于Voronoi多边形离散模型(裴书锋等,2024)和预制裂隙试件试验(蒲成志等,2010;Liu et al,2016),研究发现矿物接触参数、裂隙倾角与裂隙密度三者共同控制岩石强度劣化路径,在高围压条件下,闭合裂隙更易诱发剪切破坏模式(Sun et al,2012),而张开裂隙主要控制拉张破坏模式(Eftekhari et al,2016;王雄等,2025)。在此背景下,通过改进FLAC3D弹脆性本构与超细网格划分技术,双裂隙岩石破坏模拟精度提升至90%以上(付金伟等,2012)。三轴压缩试验进一步表明,围压增大能够有效抑制岩石脆性破坏,且残余强度和初始残余应变随围压呈指数增长(唐海燕,2016;刘冬桥等,2017),据此建立的损伤本构模型(Chen et al,2023)通过损伤变量量化了裂隙发展对强度退化的贡献。在理论建模方面,部分学者创新性地引入包含退化角等特征量的新本构方程,定量阐明了围压对峰后脆性退化的影响(于永江等,2012),以莫尔—库伦准则为框架,采用内摩擦角和黏聚力的应变相关函数构建峰后应力—应变关系(李文婷,2012),结合位移法解析软化段斜率(陈康等,2013),最终实现了从理论推导到有限元程序计算的闭环验证(王水林等,2014;Zhang et al,2022)。总体而言,关于岩石力学行为的研究范式已从单纯参数调整转向本构方程迭代优化,基于Fish语言的二次开发极大地增强了参数修正系统的自适应性(王嵩等,2017),使得不同尺度模拟结果与试验数据的匹配误差普遍控制在5%以内(王小平等,2018)。
ChenK, CudmaniR, OlarteA A P,2024.Validation of a damage constitutive model based on logistic model dedicated to the mechanical behavior of the rock[J].Geomechanics and Geophysics for Geo-Energy and Geo-Resources,10(1):15.
[2]
ChenY F, LinH, XieS J,et al,2023.Fracture closure empirical model and theoretical damage model of rock under compression[J].Materials,16(2):589.
[3]
EftekhariM, BaghbananA, HashemolhosseiniH,2016.Crack propagation in rock specimen under compressive loading using extended finite element method[J].Arabian Journal of Geosciences,9(2):145.
[4]
LiuX W, LiuQ S, HuangS B,et al,2016.Fracture propagation characteristic and micromechanism of rock-like specimens under uniaxial and biaxial compression[J].Shock and Vibration,2016(1):6018291.
[5]
SunY Z, ZhaiX X, LiuJ M,2012.Analysis of failure mode and propagation for crack in uniaxial compression[J].Applied Mechanics and Materials,166/167/168/169:2929-2932.
[6]
WangZ H, WangJ C, YangS L,et al,2020.Failure behaviour and acoustic emission characteristics of different rocks under uniaxial compression[J].Journal of Geophysics and Engineering,17(1):76-88.
[7]
ZhangY T, LuoH W, ChenP S,et al,2022.Mechanical properties and binary-medium-based constitutive model for coral-reef limestone samples subjected to uniaxial loading[J].Sustainability,14(19):12193.
ChenKang, ZhaoGuangming, MengXiangrui,2013.Influence factors research on uniaxial compression strain softening in rock shear zone[J].Safety in Coal Mines,44(3):29-32.
FuJinwei, ZhuWeishen, WangXianggang,et al,2012.An improved elastic-brittle simulation method of crack propagation process in jointed rock mass and its application[J].Chinese Journal of Rock Mechanics and Engineering,31(10):2088-2095.
[14]
李文婷,2012.岩石峰后应变软化本构方程及数值模拟方法研究[D].济南:山东大学.Li Wenting,2012.The study on post-peak strain softening constitutuerit equation and numerical simulation of rock[D].Jinan:Shandong University.
LiuDongqiao, WangZhuo, ZhangXiaoyun,2017.Characteristics of strain softening of rocks and its damage constitutive model[J].Rock and Soil Mechanics,38(10):2901-2908.
LiuZhixiang, YanMengyang, ZhangShuangxia,et al,2023.Damage constitutive model considering the effect of rock microdefects[J].Gold Science and Technology,31(3):507-515.
LuYinlong, WangLianguo, YangFeng,et al,2010.Post-peak strain softening mechanical properties of weak rock[J].Chinese Journal of Rock Mechanics and Engineering,29(3):640-648.
[21]
裴书锋,郝文锋,王营利,等,2024.双江口水电站花岗岩单轴压缩微观破坏机制研究[J].西北水电,(4):62-68.Pei Shufeng,Hao Wenfeng,Wang Yingli,et al,2024.Study on micro failure mechanism of granite under uniaxial compression[J].Northwest Hydropower,(4):62-68.
PuChengzhi, CaoPing, ZhaoYanlin,et al,2010.Numerical analysis and strength experiment of rock-like materials with multi-fissures under uniaxial compression[J].Rock and Soil Mechanics,31(11):3661-3666.
PuCheng, ChenYunsheng, ZhouChong,et al,2018.Displacement loading rate of rock-like specimen under uniaxial compress test[J].Journal of Water Resources and Architectural Engineering,16(3):161-164.
SunChuang, ZhangShuguang, JiaBaoxin,et al,2015.Physical and numerical model tests on post-peak mechanical properties of granite[J].Chinese Journal of Geotechnical Engineering,37(5):847-852.
[28]
唐春安,1999.岩石破裂过程数值模拟方法发展的若干问题[C]//中国岩石力学与工程学会软岩工程专业委员会第二届学术大会论文集.北京:中国岩石力学与工程学会.Tang Chun’an,1999. Some problems in the development of numerical simulation methods for rock fracture process[C]//Proceedings of the Second Academic Conference of Soft Rock Engineering Committee of China Society of Rock Mechanics and Engineering.Beijing:China Society of Rock Mechanics and Engineering.
[29]
唐海燕,2016.深部灰岩力学性质及残余强度试验研究[J].金属矿山,47(8):43-47.
[30]
TangHaiyan,2016.Experimental studies on rock properties and residual strength of deep limestone[J].Metal Mine,47(8):43-47.
WangShuilin, ZhengHong, LiuQuansheng,et al,2014.Principle of analysis of strain-softening rock mass and its application[J].Rock and Soil Mechanics,35(3):609-622,630.
WangSong, ZuoShuangying, QuChuanqi,et al,2017.Correction of damaged mechanic parameter and numerical simulation on uniaxial compression test[J].Journal of Guizhou University(Natural Sciences),34(2):109-114.
WangWeihua, WangYongqiang, ZhangHenggen,2021.Construction and verification of post-peak strain softening model of rock[J].Chinese Journal of Underground Space and Engineering,17():546-551,608.
[37]
王小平,姜天洪,2018.基于FLAC3D的页岩单轴压缩数值模拟研究[J].山东交通科技,(6):99-101.Wang Xiaoping,Jiang Tianhong,2018.Numerical simulation of shale uniaxial compression based on FLAC3D[J].Shandong Jiaotong Keji,(6):99-101.
WangXiong, LiuXiaoyu, YuanChao,et al,2025.Study on the damage and rupture mechanism of red sandstone under the influence of fracture type and dip angle[J].Metal Mine,54(7):33-42.
YuYongjiang, ZhangChunhui, WangLaigui,2012.Post-peak strain softening model of rock based on degradation angle[J].Journal of China Coal Society,37(3):402-406.
ZhangFan, ShengQian, ZhuZeqi,et al,2008.Study on post-peak mechanical behaviour and strain-softening model of Three Gorges granite[J].Chinese Journal of Rock Mechanics and Engineering,27():2651-2655.