Plasticity inverse analysis for Zipingpu concrete-faced rockfill dam based on advanced cloud surrogate model via improved Jaya optimization algorithm

A notable prerequisite for state assessments of earth-rockfill dams is a reasonable and accurate description of the deformation. However, due to the complexity of rockfill, its mechanical properties are often difficult to fully describe using the forward finite element method (FEM). The result of FEM depends on the constitutive model and its parameters' selection usually obtained empirically or from indoor tests. In this paper, the improved Jaya optimization algorithm (iJaya) is combined with the advanced cloud surrogate model (ACSM) to perform inverse analysis, which has great advantages in computational speed and accuracy. The generalized plastic model (GPM) is chosen for simulation and its parameters are analyzed. Finally, the deformation field of the Zipingpu Dam is reconstructed. The results show that the ACSM can greatly replace the FEM even using GPM, achieving high prediction accuracy and reduced computation time. iJaya-ACSM can also identify the parameters of the GPM quickly and correctly. The deformation of ERDs can be well simulated based on the inverted parameters, which could help to know the state of the whole dam body. In conclusion, the proposed inverse method is promising in analyzing the deformation of ERDs.

Automated summary

A reasonable and accurate description of deformation is crucial for state assessments of earth-rockfill dams (ERDs). However, the mechanical properties of rockfill are difficult to fully describe using the finite element method (FEM) due to its complexity. This paper combines the improved Jaya optimization algorithm (iJaya) with the advanced cloud surrogate model (ACSM) for inverse analysis, achieving high prediction accuracy and reduced computation time. The proposed inverse method shows promise in analyzing the deformation of ERDs and simulating their state.