Chloride diffusion modeling of concrete using tree-based forest models

摘要

Reinforced concrete structures can experience various harsh environmentsduring their service life, among which chloride ion exposure, especially inmarine environments, can cause the durability reduction and deterioration ofconcrete structures. Artificial intelligence (AI)-based modeling of the nonsteady-state apparent chloride diffusion coefficient (D_C) of concrete for a longexposure time using the experimental field results can assist in identifying theinfluential factors and better estimating the service life of a concrete structure.In this study, two novel extensions of ensemble AI algorithms, includinggenetic programming forest (GPF) and linear genetic programming forest(LGPF) algorithms, were proposed to model the D_C of concrete. The experimentalfield data were gathered from the literature. Different structures of theproposed ensemble methods were developed and examined, and the bestdevelopedmodel was selected for further analysis, including sensitivity analysisand parametric study. In addition, the random forest (RF) method was used as thecontrol ensemble technique to have a comparison. The results show that the bestLGPF model possesses superior performance than the best-developed GPF and RFmodels. In addition, the results show that silica fume-to-binder ratio, exposuretime, and exposure conditions have the most significant impacts on the D_C of concrete.This study contributes to the civil engineering practice by developing a newtool to model the D_C of concrete that facilitates the durability assessment of concretestructures.