Hydration plays a crucial role in cement composites, but the traditional methods for measuring hydration heat face several limitations. In this study, we propose a machine learning-based approach to predict hydration heat at specific time points for three types of cement composites: ordinary Portland cement pastes, fly ash cement pastes, and fly ash–metakaolin cement composites. By adjusting the model architecture and analyzing the datasets, we demonstrate that the optimized artificial neural network model not only performs well during the learning process but also accurately predicts hydration heat for various cement composites from an extra dataset. This approach offers a more efficient way to measure hydration heat for cement composites, reducing the need for labor- and time-intensive sample preparation and testing. Furthermore, it opens up possibilities for applying similar machine learning approaches to predict other properties of cement composites, contributing to efficient cement research and production.

This study proposes a multiscale prediction method that couples the DTW-CEMHYD3D hydration kinetics model with Mori–Tanaka homogenization theory to establish a quantitative relationship between hydration microstructure evolution and macroscopic heat and mass transport properties of cement-based materials. First, dynamic time warping (DTW) is introduced to correct the mapping relationship between the hydration cycle number in CEMHYD3D and the real hydration time. Then, cement-based materials are regarded as multiphase composites, and a unified calculation method for the effective diffusion coefficient and thermal conductivity is established. The results show that the DTW-CEMHYD3D model can markedly improve the prediction accuracy of early-age hydration heat. The average relative error between the predicted effective diffusion coefficient and the N-phase sphere model is 0.67%, while the coefficients of determination for thermal conductivity prediction and relative diffusion coefficient prediction during hydration reach 0.9812 and 0.987, respectively. Parametric analysis indicates that a higher water-to-binder ratio significantly increases the relative diffusion coefficient, and this effect exhibits nonlinear enhancement with hydration age. The influence of fly ash content is age-dependent. An increase in the degree of saturation reduces the chloride diffusion coefficient but only slightly increases the thermal conductivity. The proposed method provides a reference for durability analysis and parameter determination in multiphysics coupling models of cement-based materials.

The water-to-cement ratio and temperature are key factors governing ionic transport in cement-based materials. However, quantitative descriptions of their coupled effects on sulfate ion transport–reaction processes remain limited. In this study, a coupled transport–reaction model is developed to simulate sulfate ingress in cementitious materials subjected to external sulfate attack. The predictive capability of the model is validated through a series of well-designed experiments, and the distributions of sulfate ion concentration within the materials under different water-to-cement ratios and temperature conditions are systematically analyzed. The results demonstrate good agreement between model predictions and experimental measurements, indicating that increases in both water-to-cement ratio and temperature significantly accelerate sulfate diffusion and intensify chemical reactions between sulfate ions and cement hydration products, thereby promoting material degradation. These findings highlight that the coupled effects of water-to-cement ratio and temperature should be explicitly considered in durability assessment and numerical modeling of external sulfate attack in cement-based materials to improve predictive accuracy.