Structural optimization of a continuous reinforced concrete beam considering risks and progressive collapse

Resumo

Current normatives that guides the structural design against progressive collapse adopts a damage-
tolerant approach, where the system is design to withstand the loss of individual vertical elements due to abnormal

load conditions. However, the actual guidelines consider this individual element loss in a deterministic manner,
which can overestimate the damage occurrence and substantially increase the total expected costs. Aiming to
analyze how the optimal design of usual structural systems is affected by a column removal scenario, a Risk
Optimization is performed on a continuous beam of reinforced concrete subjected to the loss of the internal support,
which is considered by means of a latent probability of failure. In order to increase the efficiency of the
optimization process, an adapted system single loop approach to the risk optimization is employed herein, allowing
a very fast convergence to the optimal design. Considering the steel rebar areas as the optimal parameters, it is
found that the latent probability of failure substantially increases the steel rebar area directly affect by the internal
column loss when compared to the semiprobabilist design presented by the current normative. When the smallest
latent probability is considered, the optimal steel rebar area is identical to when this probability is null, however,
when the target reliability is over the reliability of the reference design, this area increases very fast. It is also
identified evidences of a threshold column loss probability, but only for the rebar area not affected by the column
loss removal, meaning that its design is indifferent to the objective consideration of the column loss.