ABSTRACT
In the context of the examination of an existing bridge, reliability with respect to the Ultimate Limit States of fracture (ULS-STR) and fatigue (ULS-FAT) has to be verified. Traffic action models, even updated ones, and structural models used for the design of new bridges often involve significant conservatism and are thus not realistic for existing bridges. The need for such models can be eliminated, or reduced, by using data obtained from Structural Response Monitoring (SRM), in order to develop, in a direct way, case-specific probabilistic action effect models suitable for reliability verifications (Treacy & Brühwiler 2015).
In this paper, the above presented concept is investigated through a practical application in the case of a short span RC railway bridge for which a monitoring campaign was conducted. The strain profile variations at the critical location of the RC slab of the bridge was monitored over several months by means of a specifically developed device. The recorded signals are processed in order to obtain stress histories in the reinforcement, which can be used for damage accumulation calculations, and for modeling of the extreme action effect probability distribution. These data can be used for structural safety verifications at the ultimate limit states of fracture and fatigue.
The monitored structure is depicted in Figure 1. It is a short single span railway underpass carrying two tracks and consisting of a reinforced concrete slab, of 0.30 m thickness, simply supported on masonry abutments. Instrumentation is placed at the bottom surface of the slab, at the middle of the span and under the axis of one of the two tracks (track 1). An innovative device has been developed for monitoring the variations of the axial strain profile in the principal and the secondary bending direction. The measuring principle is illustrated in Figure 1. (a) Measurement principle, (b) Longitudinal section and (c) cross-section of the underpass with the installed device. https://s3-euw1-ap-pe-df-pch-content-public-p.s3.eu-west-1.amazonaws.com/9781315207681/cd556cd4-4dcf-4efe-8e29-56fc67b8bfbd/content/fig154_1.tif"/>
The evolution of the damage is calculated by the microplasticity theory proposed by the authors (Grigoriou & Brühwiler 2015) for both monitoring-derived and artificial stress histories.
The probability that a crack will be formed in a steel bar before a reference time t is obtained by an approach according to Grigoriou (2015).
In Figure 2 the PDFs calculated taking into account the probability of CAFL exceedance are compared. The comparison shows that considering the probability of failure instead of the probability of initial crack formation allows a considerable extension of the safe service life. PDF of the time to formation of the initial crack (black lines) or conventional bar failure (red lines). https://s3-euw1-ap-pe-df-pch-content-public-p.s3.eu-west-1.amazonaws.com/9781315207681/cd556cd4-4dcf-4efe-8e29-56fc67b8bfbd/content/fig154_2.tif"/>
The major contributions of this work are: (However, this conclusion need to be further validated by field 342data or laboratory experiments on cracked concrete sections.)
– A new monitoring technique for monitoring strain profile variations in RC sections has been developed.
– Extreme values of action effect can be modeled by a shifted exponential distribution. (However, this conclusion need to be further validated by field data or laboratory experiments on cracked concrete sections.)
– It has been shown that the use of monitoring stress histories has a strong beneficial influence in the calculated safe service life duration mainly because it avoids the use of dynamic amplification factors.
