The hardening soil model with small-strain stiffness (HSS model) is widely applied in deep foundation pit engineering in coastal soft-soil areas, yet it is characterized by a multitude of parameters that are relatively cumbersome to acquire. In this study, we incorporate a genetic algorithm and a back-propagation neural network (BPNN) model into an inversion analysis for HSS model parameters, with the objective of facilitating a more streamlined and accurate determination of these parameters in practical engineering. Utilizing horizontal displacement monitoring data from retaining structures, combined with local engineering, both a BPNN model and a BPNN optimized by a genetic algorithm (GA-BPNN) model were established to invert the stiffness modulus parameters of the HSS model for typical strata. Subsequently, numerical simulations were conducted based on the inverted parameters to analyze the deformation characteristics of the retaining structures. The performances of the BPNN and GA-BPNN models were evaluated using statistical metrics, including R2, MAE, MSE, WI, VAF, RAE, RRSE, and MAPE. The results demonstrate that the GA-BPNN model achieves significantly lower prediction errors, higher fitting accuracy, and predictive performance compared to the BPNN model. Based on the parameters inverted by the GA-BPNN model, the average compression modulus Es1-2, the reference tangent stiffness modulus Eoedref, the reference secant stiffness modulus E50ref, and the reference unloading-reloading stiffness modulus Eurref for gravelly cohesive soil were determined as Eoedref=0.83Es1-2 and Eurref=8.14E50ref; for fully weathered granite, Eoedref=1.54Es1-2 and Eurref=5.51E50ref. Numerical simulations conducted with these stiffness modulus parameters show excellent agreement with monitoring data, effectively describing the deformation characteristics of the retaining structures. In situations where relevant mechanical tests are unavailable, the application of the GA-BPNN model for the inversion analysis of HSS model parameters is both rational and effective, offering a reference for similar engineering projects.

An integrated model that considers multiphysics is necessary to accurately analyze the time-dependent response of hydraulic structures on soft foundations. This study develops an integrated superstructure-foundation-backfills model and investigates the time-dependent displacement and stress of a lock head project on a soft foundation during the construction period. Finite element analyses are conducted, incorporating a transient thermal creep model for concrete and an elasto-plastic consolidation model for the soil. The modified Cam-clay model is employed to describe the elasto-plastic behavior of the soil. Subsequently, global sensitivity analyses are conducted to determine the relative importance of the model parameters on the system's response, using Garson's and partial derivative algorithms based on the backpropagation (BP) neural network. The results indicate that the integrated system exhibits pronounced time-dependent displacement and stress, with dangerous values appearing during specific periods. These values are easily neglected, highlighting the importance of integrated time-dependent analysis. Construction activities, particularly the backfilling process, could cause a sudden change in stress and significantly impact the stress redistribution of the superstructure. Additionally, the mechanical properties of concrete have a significant impact on the stress on the superstructure, while the mechanical properties of the soil control the settlement of the integrated system.