Influence of Reinforcing Steel Corrosion on Life Cycle Reliability Assessment of Existing R.C. Buildings

Abstract

Time-dependent reliability assessment is a crucial aspect of the decision process for rehabilitation of existing reinforced concrete structures. Since the assessment strongly depends on degradation of materials with time, the paper focuses on the influence of corrosion in reinforcing steel on time-reliability curves of relevant reinforced concrete (r.c.) structures, built in Italy in the 1960s, belonging to different building categories. To realistically represent the probability distribution functions (pdfs) of the relevant properties of reinforcing steel and concrete commonly adopted in the 1960s, stochastic models for steel yielding and concrete compressive strength have been derived, by means of a suitable cluster analysis, from secondary databases of test results gathered at that time in Italy on concrete and steel rebar specimens. This cluster analysis, based on Gaussian mixture models, provides a powerful tool to “objectively” extract material classes and associated probability density functions from databases of experimental test results. In the study, different degradation conditions and several reinforcing steel and concrete classes are considered, also aiming to scrutinize their influence on the time-dependent reliability curves. Finally, to stress the significance of the study, the time-dependent reliability curves so obtained are critically examined and discussed also in comparison with the target reliability levels currently adopted in the Eurocodes.

Bayesian Methodology for Probabilistic Description of Mechanical Parameters of Masonry Walls

“…A further analysis is then needed to identify homogenous statistical populations for masonry mechanical parameters. In particular, the general procedure already successfully applied by Croce et al (2018bCroce et al ( , 2020 to identify concrete and steel rebar classes can be used.…”

“…As already noted by Croce et al (2018bCroce et al ( , 2020, statistical parameters derived via cluster analysis are much more refined than those obtained analyzing the whole data set, and the estimate of the COV associated with each cluster is significantly improved. For example, in the case of k ¼ 3, the COV of compressive strength, which is 33% for the whole data set, reduces to 28% for the lower class, 14% for the intermediate class, and 11% for the upper class.…”

“…When all the distributions of the mixture belong to the normal family, the model is a GMM (Mclachlan and Peel 2000), but different distributions can be also adopted for the components of the mixture depending on the data. GMMs have been successfully used by Croce et al (2018bCroce et al ( , 2020 to identify coherent material resistance classes in a whole database of test results, even if the origin of individual data were unknown. GMMs will be also adopted in the following analyses for the investigated masonry parameters; in fact, statistical tests such as Kolmogorov-Smirnov carried out on the entire data set do not show a preference for distribution other than normal.…”

“…The proposed method relies on a more general original procedure for the identification of homogenous material classes in secondary databases of raw test results, illustrated by Croce et al (2018b). The rationale of the procedure, already successfully implemented to identify concrete classes (Croce et al 2018b) and reinforcing steel classes (Croce et al 2020) used in Italy during the 1960s, is to partition, by means of a cluster analysis based on Gaussian mixture models (GMMs) (Mclachlan and Peel 2000), a database of mechanical test results to recognize homogenous statistical populations of relevant mechanical parameters. In the following, the focus is on the main mechanical parameters of stone masonry: compressive strength, f m , elastic modulus, E, and shear modulus, G, but the method can be easily extended to other relevant parameters and different masonry typologies provided that adequate data sets of experimental results are available.…”

Life Cycle Prediction and Maintenance of Buildings

“…The influence of reinforcing steel corrosion on life cycle reliability assessment of existing reinforced concrete structures is analysed in [11]. This study evaluates the influence of different degradation conditions and several reinforcing steel and concrete classes on the time-dependent reliability curves proposed.…”