Development of Hybrid Machine Learning Models for Predicting Permanent Transverse Displacement of Circular Hollow Section Steel Members under Impact Loads

The impact effect is a crucial issue in civil engineering and has received considerable attention for decades. For the first time, this study develops hybrid machine learning models that integrate novel Extreme Gradient Boosting (XGB) model with Particle Swam Optimization (PSO), Grey Wolf Optimizer (GWO), Moth Flame (MFO), Jaya (JA), Multi-Verse (MVO) algorithms predicting permanent transverse displacement of circular hollow section (CHS) steel members under loads. are developed using data collected from 357 tests CHS members. efficacy evaluated three performance metrics. results show GWO-XGB achieves high accuracy outperforms other models. values R2, RMSE, MAE obtained test set 0.981, 2.835 mm, 1.906 respectively. SHAP-based explanation shows initial velocity indenter, mass, ratio position to member length most sensitive parameters, followed by yield strength length; meanwhile, diameter thickness parameters least Finally, web application tool help rapidly estimate