An Ensemble Kalman Filter approach for service life prediction of reinforced concrete structures subject to chloride-induced corrosion

Abstract

Chloride induced reinforcement corrosion is one of the most common causes of deterioration of reinforced concrete structures. Condition assessment of such structures is typically based on comparing the chloride content at steel level to the critical threshold chloride concentration. This study aims at developing a robust non-destructive structural health monitoring (SHM) methodology for assessing the state of the structure and predicting its remaining corrosion free service life. The presented methodology relies on embedded sensors that measure the chloride concentration at certain locations, and a predictive model that governs the dynamics of chloride ingress. The system is propagated forward in time using a finite element/finite difference scheme, and the Ensemble Kalman Filter (EnKF) is employed to calibrate the predictive model parameters. The calibrated parameters are then used to assess the remaining corrosion free service life of the structure. The robustness of the presented methodology is demonstrated on 1-D and 2-D numerical examples yielding accurate estimation of the spatial and temporal variability of the chloride content and consequently enhancing maintenance efficiency and structural sustainability.