Advancing service life estimation of reinforced concrete considering the coupling effects of multiple factors: Hybridized physical testing and machine learning approach
Chloride-induced corrosion is the primary factor that significantly impacts the service life of reinforced concrete structures. Therefore, it is crucial to investigate the transportation behavior of chloride ions within concrete structures to accurately estimate their service life. Previous studies...
Gespeichert in:
Veröffentlicht in: | Journal of Building Engineering 2024-05, Vol.84, p.108476, Article 108476 |
---|---|
Hauptverfasser: | , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | Chloride-induced corrosion is the primary factor that significantly impacts the service life of reinforced concrete structures. Therefore, it is crucial to investigate the transportation behavior of chloride ions within concrete structures to accurately estimate their service life. Previous studies have mostly examined the diffusion of chloride ions through physical tests on concrete specimens. Some data-driven models have also been developed to predict the chloride ion concentration at different depths in concrete structures. However, these prediction models often considered random factors in materials and the environment separately, neglecting their coupled effects. As a result, there can be a significant disparity between the predicted results and the actual test outcomes. To achieve a more precise evaluation of the service life, this study conducted a physical test to investigate the diffusion behavior of chloride ions in plain concrete specimens. Factors such as water-cement ratio (W/C), coarse aggregate volume fraction (v) (ratios of the mass of coarse aggregates to the total mass of concrete specimen), ratios of environmental temperature to maintenance standard temperature (T/T0), ratios of environmental humidity to maintenance humidity (h/h0), and ratios of exposure time to maintenance time (t/t0) were considered in the test. Based on the test results, machine learning approaches were employed to determine the influence of each factor on the chloride diffusion coefficient (D) and surface chloride concentration (Cs). Subsequently, a prediction model was developed using the second Fick's law, with the chloride ion concentration at each depth as the output and the aforementioned factors as the inputs. Furthermore, by analyzing the above results, the study examined the impacts of various factors on the service life of reinforced concrete structures. The findings indicate that W/C and v are the most significant factors affecting D and Cs. As W/C and T/T0 increase, both D and Cs also increase. Conversely, the volume fraction of coarse aggregate has a hindering effect on the diffusion of chloride ions in concrete structures. Additionally, the study explored the effect of each input factor on the service life of reinforced concrete structures considering the effect of the above factors. Overall, these findings provide valuable insights into improving the accuracy of estimating the service life of reinforced concrete structures by considering the coupled ef |
---|---|
ISSN: | 2352-7102 2352-7102 |
DOI: | 10.1016/j.jobe.2024.108476 |