基于扩散路径映射的混凝土碳化细观模拟方法

Mesoscopic Concrete Carbonation Simulation Method Based on Diffusion Path Dependent Mapping

摘要:

对碳化作用细观求解方法进行梳理改进，利用传输和反应过程状态变化的路径依赖特点，规避了传统方法大规模矩阵求解，提出了基于扩散路径映射的碳化细观模拟方法。方法基于Dijkstra路径算法对界面层影响进行修正，建立水泥浆体扩散路径算法，得到环境表面到水泥浆体的最短路径网络；通过水泥浆体代理模型，模拟水化反应、碳化反应以及材料孔隙率变化的耦合碳化过程，得到关键指标（氢氧化钙浓度）沿扩散路径的时变分布；最终依据相同的扩散路径长度进行结果映射，实现碳化作用中关键物质浓度随时间空间分布的高效精准预测。通过材料试验对比校验，扩散路径映射方法模拟结果能够准确反应混凝土碳化深度范围；与传统偏微分求解方法对比，扩散路径映射方法能够在细观尺度精准模拟关键物质空间分布情况，且求解效率提升20倍以上，存储空间需求降低为偏微分求解的2%以内。

Abstract:

The solution foundation of mesoscopic carbonation is summarized and improved. According to the path dependent feature of transport and reaction， the solution of huge-size matrix in traditional methods is avoided， and a diffusion path dependence mapping （DPDM） method is proposed. Based on the Dijkstra shortest path calculation algorithm， the paths from environmental boundary to any nodes in cement mortar are calculated with the modification of the influence of the interfacial transition zone. With the agent model of cement mortar， the complex coupling process of the hydration reaction， the carbonation reaction， and the change of material properties（concentration of Ca（OH）₂）can be solved. Finally， the solution results of the agent model are mapped on the mesoscopic model with the same length of diffusion path， and the temporal-spatial distribution of critical substance concentration can be accurately and efficiently predicted. The comparison and verification with material test indicate that the simulation results of diffusion path dependence mapping method can accurately simulate the carbonation depth in concrete. Compared with the traditional solution based on partial difference equation， the proposed method can precisely reflect the mesoscopic distribution of critical components， where the solution efficiency can be improved by more than 20 times， and the storage requirement can be reduced to less than 2%.