Investigation of thermal conductivity and related parameters of early-age cement paste
Investigation of thermal conductivity and related parameters of early-age cement paste
复制标题
早龄期水泥浆体导热系数及相关参数研究
DOI：
10.1016/j.ijheatmasstransfer.2020.119798
复制
发表时间：
2020
影响因子：
5.2
通讯作者：
Huang Donghui
中科院分区：
文献类型：
--
作者：
Zhao Haitao;Fan Guangchao;Wei Zhenzhong;Wang Jian;Huang Donghui
水泥基材料的导热系数是进行早期温度场分析和裂缝控制的重要参数。本研究采用平行热线技术对早龄期水泥浆体的相变温度进行了实验测定。试验结果表明，早龄期水泥净浆的TC随龄期和水灰比(w/c)变化明显。不同w/C水泥浆的TC曲线相似，可分为三个阶段，即曲线先快速下降，然后急剧上升，最后趋于稳定。40小时后，水泥净浆的TC几乎与w/c呈线性下降。同时，基于广义自洽方法和CEMHYD3D相结合，提出了考虑热阻的早期水泥净浆TC多尺度预测模型。同时开发了基于粒子群算法的改进反演程序。利用该程序结合实验数据求出了水泥净浆中各相的相转变温度和热阻参数。反演结果表明，在不同W/C的条件下，各相TC和热阻参数的偏差最小，所提出的反演程序算法是稳健的。TC多尺度模型预测曲线与试验数据吻合较好，验证了TC多尺度模型对水泥净浆的预测能力。
Thermal conductivity (TC) of cement-based materials is a key parameter for temperature field analysis and crack control at early age. In this study, the TC of early-age cement paste was experimentally determined by the parallel hot wire technique. The test results showed that the TC of early-age cement paste changed obviously with age and water to cement ratio (w/c). The TC curves of cement pastes with differentw/cwere similar and could be divided into three stages, i.e., the curves first rapidly decreased, then sharply increased, and finally tended to stabilize. The TC of the cement paste almost linearly decreased withw/cafter 40 hours. Meanwhile, based on a combined generalized self-consistent method with CEMHYD3D, an effective TC multiscale predictive model for early-age cement paste was proposed considering the thermal resistance. An improved inversion program was simultaneously developed based on the particle swarm optimization. The TC of each phase in cement paste and the parameters of thermal resistance were obtained using the inversion program with experimental data. The inversion results showed that the deviations of the TC of each phase and the parameters of thermal resistance with differentw/cwere minimal and that the algorithm of the inversion program proposed was robust. The TC multiscale model predictive curves were in good agreement with the test data and the predictive capability of the TC multiscale model for cement paste was successfully validated.