Exploring the Impact of Internal Lattice Angles on the Bending Behavior and Ultimate Bearing Capacity of Concrete‐Filled Steel Tubes Under Different Loading Conditions

ABSTRACT Concrete‐filled steel tubes (CFSTs) are a composite structural solution that enhances load‐bearing capacity and structural performance by fusing the stiffness and compressive strength of concrete with the high strength and ductility of steel. The integration of latticed angles may require precise manufacturing methods, leading to higher production costs and longer construction timelines. Additionally, as the added strength increases the bending capacity, it complicates design calculations and necessitates more sophisticated analysis techniques. This manuscript proposes the impact of internal latticed angles on the bending characteristics and load‐bearing capacity of concrete‐filled steel tubes (ILA‐BB‐UBC‐CFST). The primary objective of the proposed technique is to increase the bending capacity of CFSTs with internal steel angles and enhance their thickness and yield strength. The complex‐value spatiotemporal graph convolutional neural networks (CVSTGCNN) is employed to forecast the CFST's ultimate bearing capacity. The proposed model is implemented on the MATLAB platform. According to the results, the steel tube has the greatest impact on bending behavior, with yield strength and thickness having the most significant effects on bending capacity. The proposed ILA‐BB‐UBC‐CFST method is compared with existing techniques like the artificial neural network (ANN), radial basis function neural network (RBFNN), and back propagation neural network (BPNN). The proposed approach achieves the highest, coefficient of determination ( R 2 ) of 0.95, correlation coefficient ( R ) of 0.85, and a standard deviation (SD) of 0.22, all of which are better than those of existing methods.