Application of Component-Based Mechanical Models and Artificial Intelligence to Bolted Beam-to-Column Connections

Top and seat beam-to-column connections are commonly designed to transfer gravitational loads of simply supported steel beams. Nevertheless, the flexural resistance characteristics these type should be properly taken into account for design, when a reliable analysis semi-rigid structures is desired. In this research paper, different component-based mechanical models from Eurocode 3 (EC3) literature proposal (by Kong Kim, 2017) considered evaluate initial stiffness (Sj,ini) ultimate moment capacity (Mn) top-seat angle with double web angles (TSACWs). An optimized artificial neural network (ANN) model based on bee colony (ABC) algorithm proposed in paper acquire an informational available database experimental test measurements TSACWs. order expected effect each input parameter (such as thickness top flange cleat, bolt size, etc.) performance overall moment–rotation (M–θ) response selected connections, sensitivity presented. The collected comparative results prove potential ANN approach TSACWs, well its accuracy reliability prediction characteristic features similar joints. For most examined configurations, higher found estimates, compared 3- or et al.-based formulations.