Randomness and time-varying characteristics of chloride ion transport in existing harbor concrete structures.

This paper presents a study on the durability testing of a concrete facility located within the North Bay Port region, which has been in service for 200 months. Based on the results of 20 sets of concrete samples, an improved Kolmogorov-Smirnov (K-S) test method and gross error discrimination method are proposed to analyze the random distribution functions of the apparent chloride diffusion coefficient (D app) and the apparent surface chloride concentration (C s), and further explore the time-varying characteristics of chloride ion transmission in concrete. The research shows that the improved K-S method presented in this paper can effectively reduce the probability of type II errors by selecting a rational sample size, which can also improve detection efficiency. When the sample size does not exceed 10, the 2 σ criterion proposed in this paper can effectively identify and eliminate gross errors. In general, D app and C s follow the log-normal distribution, while the time factor (m) follows a normal distribution parameter. The value of C s basically stabilized after exposure to 80 months, whereas D app tended to stabilize after exposure to 100 months. The value of m obtained through fitting short-term experimental data is typically overestimated, and its value decreases as the exposure time prolongs. Therefore, m should also be considered a time-varying parameter that decreases with time. The tested RC structure will begin to experience rebar corrosion after 48.5–60 years of exposure to the marine environment based on the probabilistic assessment. • The 2 σ criterion was introduced to identify and rectify significant errors. • The improved K-S method was developed to minimize the likelihood of type II errors. • Chloride profiles were extracted from 200 months aged harbor concrete structures. • Both D app and C s demonstrated a good fit with the log-normal distribution. [ABSTRACT FROM AUTHOR]