基于Python的再生混凝土细观性能模拟方法研究

(1.西安理工大学 土木建筑工程学院,陕西 西安 710048; 2.西北干旱地区生态水利国家重点实验室,陕西 西安 710048)

再生混凝土; 三维细观结构; 蒙特卡罗法; 随机骨料; 数值计算

Research on thesimulation method of mesoscopic performances of recycled concrete based on Python
YAO Zeliang1,2, CUI Tingting1,2, DUAN Dongxu1,2, DANG Faning1,2, WEN Shuo1,2

(1.School of Civil Engineering and Architecture, Xi'an Univ. of Technology, Xi'an 710048, China; 2.State Key Laboratory of Eco-hydraulics in Northwest Arid Region, Xi'an 710048, China)

recycled concrete; three-dimensional mesostructure; Monte Carlo method; random aggregate; numerical analysis

DOI: 10.15986/j.1006-7930.2021.02.003

备注

将再生混凝土看作由原生骨料、老砂浆基体、老界面区、新砂浆基体、新界面区组成的三维细观结构,运用具有动态语义特征的PYTHON语言编写基于蒙特卡罗法生成三维再生混凝土随机骨料模型的算法,通过PYTHON操纵ABAQUS软件内核,实现模型构建与数值分析的合理整合,采用该算法建立四种不同骨料含量的三维再生混凝土随机骨料模型,并对45%骨料含量的模型力学性能进行了数值计算分析. 结果表明:通过PYTHON建立的三维再生混凝土随机骨料模型与真实再生混凝土中骨料的随机性和投放率的切片细观结构相符合; 三维再生骨料模型的力学性能指标与试验结果吻合较好; 该方法较好地解决了构建再生混凝土三维随机骨料模型的技术难题,满足工程应用的需要.
A three-dimensional meso-structure model of recycled concrete is proposed which is composed of five-phase materials of newly hardened mortar, old hardened mortar, internal interface zone, external interface zone and natural aggregate. Using Python language with dynamic semantic features, an algorithm to generate three-dimensional recycled concrete random aggregate model is compiled based on Monte Carlo method. The kernel program of Abaqus software is manipulated by the algorithm to realize the reasonable integration of the model construction and numerical analysis. Three-dimensional recycled concrete random aggregate models are built with the above method based four different aggregate percent. The mechanical properties of the model with 45% aggregate percent are simulated and analyzed. The results show that the three-dimensional model established with Python is in good agreement with the real microstructure of recycled concrete. This method can meet the requirements of the aggregate generation randomness and the aggregate generation input rate. The mechanical properties of the model are in good agreement with the test results. This method is better able to solve the problem of constructing a three-dimensional random aggregate model of recycled concrete and can meet the needs of engineering applications.