Behavior of exposed column base plate connection subjected to combined axial load and biaxial bending

Abstract

Column base plate (CBP) connections are one of the most crucial structural components of steel structures that act as a transfer medium for all the forces and moments from the entire building into the foundation. Importance of this type of connection becomes significant when the structure experiences dynamic loading, such as wind or earthquake, which incorporates dynamic effects in the structure that need to be transferred to the foundation. Considerable research efforts have been made over the past few decades on CBP connections, which led to the publication of AISC Design Guide 1 (2006) for CBP design. This design guide is still widely used in the industry. All the previous studies and design guidelines considered only the uniaxial (major axis) bending moment combined with axial load for CBP connection design. However, very often the base plate experiences a bidirectional bending moment from lateral loads during any dynamic loading event. Although, the column is designed and checked under combined axial load and bi-axial bending, when it comes to the base plate connection, only the axial load and major axis bending are considered. Therefore, the objective of this research is to investigate the behavior of CBP connections subjected to combined axial load and biaxial bending through an extensive numerical parametric study, using general purpose finite element software ABAQUS. For this numerical study, an accurate nonlinear finite element (FE) model is developed, considering both geometric and material nonlinearities and validated against experimental results that are available in the literature subjected to monotonic and uniaxial cyclic loading. Validation results show that the developed FE model can effectively simulate force transfer at major contact interfaces in the connection. Concurrently, a database of CBP connection subjected to axial load and uniaxial bending, is constructed from the literature to identify the influential parameters as well as different failure modes of the CBP connection, using Machine Learning (ML) approach. Among nine different ML models, the Decision tree based ML model provides an overall accuracy of 91% for identifying the failure mode whereas base plate thickness, embedment length, and anchor rod diameter are found to be the influential parameters that govern the failure mode of CBP connections. Therefore, a total of 20 different FE models that have different base plate thicknesses and yield strengths, anchor bolt sizes and quantity as well as embedment lengths, grout thicknesses and axial load ratios are developed. Furthermore, a bidirectional symmetric lateral loading protocol is developed and applied with constant axial compressive load in the developed models. The study reveals that the thickness of base plate and anchor rod diameter are the governing parameters for different base connection behavior such as moment rotation response, maximum bolt tensile force, and yield line pattern of the base plate. Moreover, the rigidity of the base plate connection is found to be in the semi-rigid region under biaxial bending condition. Finally, this study found that the available methods for uniaxial bending overpredicts the connection rotational stiffness compared to the stiffness obtained from numerical analysis considering biaxial bending.