A characterization model for time‐dependent displacement of concrete dams

Time-dependent displacement caused by creep and shrinkage can significantly affect the structural performance and even shorten the service life of a concrete structure, which is designed for decades or longer, resulting in a safety risk. Unlike in long-spanning structures such as bridges, where the time-dependent displacement is readily observable, time-dependent displacement in dams is typically entangled with hydrostatic and thermal displacement, requiring specialized models and methods to isolate the time-dependent displacement from field monitoring data. Data-driven methods can decompose field data to estimate time-dependent displacement, but the computation and results are subject to uncertainty due to the “black box” property. Physics-based approaches utilize mechanical laws to simulate creep and shrinkage, providing a clear physical meaning; however, the limited availability of information and data in dam engineering hinders the deployment of physics-based methods. For this reason, a poromechanical modeling concept that couples creep and shrinkage is employed in this study. Then, a creep–shrinkage coupled modeling tool is developed based on the modeling concept, enabling the numerical simulation of time-dependent deformation due to creep and shrinkage. This model is further integrated with monitoring data and an intelligent optimization algorithm to enable the real-time characterization of time-dependent dam displacements. Specifically, environment data dynamically updates loads and boundary conditions on the dam, while effect data serve as a benchmark to calibrate model parameters through multiobjective optimization. From material to structure, two concrete experiments and one dam project are employed to demonstrate the model's capability in characterizing time-dependent deformation. Comparisons with experiment and engineering data confirm that the model effectively captures the deformation behavior of the concrete material and structure, providing a physically consistent and data-informed framework for characterizing time-dependent displacement in concrete dams.