A study on factors influencing dynamic response of thin-walled box-girder railway bridge subjected to high speed train load through regression modelling

Abstract The issue of evaluating the dynamic characteristics of a bridge due to the presence of rapidly moving vehicles has considerable importance. The objective of this study is to conduct a comprehensive study on the variables that influence the dynamic behavior of a thin-walled box-girder bridge exposed to high-speed train loads using regression analysis. The high-speed train is mathematically represented by a system with 38 degrees of freedom, while the sub-track system utilizes China's Railway Track System slab track. The numerical modeling of the bridge is accomplished using computationally efficient finite elements that represent thin-walled box-beams. The rail's imperfections are also accounted for, and they are represented using a power spectral density function. The dynamic response of the bridge is calculated using the Newmark Beta technique, considering several degrees of freedom and stress resultants. A thorough parametric analysis on the factors affecting dynamic response of the bridge is conducted and a regression model has been proposed. The regression equation yields an excellent fit for shear force, distortional moment, and distortional bi-moment, with an R² value near to one. It has been alsp been observed that the range of the coefficient R² in case of bending moment, torsion, torsional bi-moment, and vertical deflection typically falls between 0.82 and 0.9. R² value near to 1 indicates that it is quite accurate in forecasting the dynamic influence of high speed trains on bridge’s response.