Volume-query-based high-accuracy finite element modelling and capacity assessment of CFST columns with local random overlapping pitting corrosion

Existing finite element (FE) models for CFST columns with pitting corrosion are typically built upon the simplified assumption that corrosion pits do not overlap, which deviates from actual engineering scenarios. Furthermore, the pit-volume-superposition method used in these models overestimates the degree of corrosion, leading to potential errors in predicting structural behavior. Additionally, current capacity models are predominantly based on an equivalent-thickness approach, yet its rationale remains unvalidated. This paper developed a specimen-volume-query-based FE modelling method for the high-accuracy simulation of CFST columns with local random overlapping pitting corrosion. In addition, a user-friendly Graphical User Interface (GUI) module was developed. Then, the proposed FE model was adopted to conduct parameter studies on the compressive behavior of pitting-corroded CFST columns. Results suggested that the behavior of CFST columns with different spatial characteristics of pitting corroded zone under a same volume loss ratio( VLR ) may differ significantly, suggesting that the mechanical deterioration depends not only on the corrosion degree but also on the spatial characteristics of the corroded zone. Moreover, local random pitting corrosion degrades the capacity more severely than local uniform corrosion at an identical VLR , as pitting-induced concavities significantly amplify stress concentration and consequently increase the steel tube’s risk to premature local buckling. The capacities of CFST columns with uniform pitting corrosion generally fall between those of the aforementioned corrosion types. Finally, a load-bearing capacity model simultaneously considering the effect of spatial characteristics of pitting corrosion zone and the adverse effect of random pitting corrosion was proposed and its high accuracy was verified.