| description abstract | The random field theory is frequently employed to characterize the inherent spatial variability of material properties. In order to incorporate sampled data from site investigations or experiments into simulations, and to mitigate the variability and randomness inherent in random field simulations, this paper introduces conditional random fields (CRFs) for the compaction quality of rockfill material, taking the filling process into consideration through the utilization of the Kriging algorithm. Quantitative relationships between porosity and parameters of the Duncan-Chang E-B model were established through triaxial tests. A random field of the model parameters is then constructed, and a Monte Carlo finite-element simulation is conducted to analyze the deformation of the dam body. By comparing the deformation of the dam body obtained using deterministic parameters, the variability parameter without consideration of the filling process, and the variability parameter with consideration of the filling process, the importance of considering the filling process in establishing a realistic and rational random field for the rockfill material is elucidated. The study’s findings indicate that deterministic analyses may underestimate dam body deformation, while consideration of material variability may exaggerate it, resulting in larger deformations if the filling process is neglected. However, the risk of exaggerating material variability can be somewhat mitigated by considering dam deformation. In this study, a methodology that employs the Kriging algorithm to generate a random field of compaction quality for rockfill materials is proposed. By taking a 211-m-high dam under construction as a case study, the reasonableness of the random field simulation is improved by introducing sampling information and discussing the value of influence domain. This study investigates the possible quality differences that may occur during the filling process of the dam material disturbed by uncertainties, and it demonstrates that ignoring the spatial variability of the material may lead to an incorrect estimation of the dam deformation. This study illustrates the importance of quality control during the construction of rockfill dam to ensure that the project meets the design standards, and also reduces the error of the traditional numerical simulation without considering the spatial variability. This can effectively improve the accuracy of the deformation prediction of dams during the operation period and provide an important support to ensure the long-term safe operation of dam. | |
