Numerical and Hybrid Ann-Hho-Based Evaluation of Single-Anchored Sheet Pile Walls in Medium-Depth Excavations

OpenAlex · Turkish

This study investigates the behavior of single-anchored steel sheet pile walls used as temporary retaining systems in medium-depth excavations ranging from 5 to 6 meters in depth. A series of parametric analyses were performed using PLAXIS 2D under static conditions to evaluate the effects of excavation depth, soil internal friction angle, unit weight, and vertical anchor position on wall performance. Consequently, the optimum vertical distance of the anchor from the ground surface was determined to be between 0.25 and 0.3H in terms of displacement. The results of the numerical simulations provided the basis for developing a predictive model for maximum horizontal wall displacement. To obtain accurate, generalized predictions, a hybrid Artificial Neural Network (ANN) model optimized using the Harris Hawks Optimization (HHO) algorithm was developed. In this approach, HHO was used to optimize the ANN hyperparameters, including transfer functions and the number of hidden-layer neurons. The findings highlight the potential of combining finite element analysis with hybrid soft-computing approaches to enhance the design and performance evaluation of anchored retaining systems in temporary excavations.