Abstract
Due to the attack of harmful environmental substances, the reinforcements in concrete can be corroded. The corrosion of reinforcements induces the concrete crack and the reduction in structural capacity. The complex durability deterioration at the material scale is commonly oversimplified in the structural analysis, and the influence of reinforcement corrosion distribution can hardly be evaluated in traditional methods. Benefitted from the accurate material model and efficient solution technologies, a structural life-cycle probabilistic evaluation method is proposed based on the material uncertainty influence. The computational framework consists of three steps. First, the simulation of chloride ingress in structural cross-section considering mesoscopic variation; second, the finite element analysis of structural response with modifying reinforcement corrosion states; third, the structural life-cycle performances are evaluated probabilistically with the trained Artificial Neural Network model. Focussed on the life-cycle hysteretic performance of the reinforced concrete column, the whole procedure is operated for illustration, and the time-dependent decreasing structural performance are investigated. Compared with widely used macroscopic methods, the model precision can be refined over 90%. By discussion effects of different measures to improve structural performance are studied, including increasing concrete strength and concrete cover. The results show that the thicker concrete cover can postpone the initial deterioration and significantly increase structural life-cycle performance probabilistically, and with the same designed action effect, the structural service life can be extended over 85% to reach same reliability.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call