Hybrid hydraulic‐seasonal‐time model for predicting the deformation behaviour of high concrete dams during the operational period
Guang Yang, Hao Gu, Xudong Chen, Kupeng Zhao, Dong Qiao, Xingquan Chen
Abstract
Considering the limitations of the traditional hydraulic-seasonal-time (HST) model, this study proposes a hybrid modeling method for the deformation prediction of high concrete dams during the operational period. First, the elastic finite element (FE) method is applied to simulate the interactive effects of structural properties, topography, geology, and high hydrostatic load on the deformation behaviour of high concrete dams in operating conditions. The hybrid model of hydrostatic pressure deformation is established. The hybrid hydraulic-seasonal-time (HHST) model is proposed. Second, the self-adaptive stochastic inertia weight, dynamic learning factors, and velocity and position parameters are introduced to improve the particle swarm optimization (PSO) algorithm. The hybrid prediction approach is developed through the comprehensive application of the HHST model and the improved PSO algorithm. The proposed methodology is adopted for the Jinping I project, which is the highest concrete arch dam in the world. The analysis results indicate that the model accuracy is good and that the model performance is promoted.