Reliability in tension governed RC reservoirs considering autogenous self-sealing

Abstract

The autogenous self-sealing of cracks is widely acknowledged to reduce leakage in tension governed reinforced concrete reservoirs. The design of these reservoirs is typically governed by serviceability limit state (SLS) considerations, i.e., limiting crack widths to prevent unacceptable leakage. This research introduces a reliability analysis to determine the achieved level of SLS reliability in reservoirs, considering the effect of autogenous self-sealing. Using a reliability limit state defined as the difference between the allowable and predicted leakage, the analysis simulates the behaviour of a reservoir at water tightness test stage. It makes use of the fib Model Code 2010 crack prediction model and determines the achieved SLS reliability for reservoirs in the stabilized cracking phase using Monte Carlo simulations. A range of target crack widths and water tightness test stabilization periods are considered, along with the probabilistic effect of both crack width prediction and self-sealing. From the analysis of a test reservoir, greater stabilization periods and more stringent target crack widths were found to lead to significantly reduced leakage and thus increased reliability. For stabilization periods of 0, 3, 7 and 14 days, an adequate SLS reliability of 1.5≤βt was found to require target crack widths of 0.11, 0.14, 0.15 and 0.16 mm, respectively. The proposed reliability analysis allows for the determination of the achieved level of SLS reliability in tension governed RC reservoirs; a key step towards cost optimization.