Volume 15 (2019): Edizione 2 (June 2019)

In this paper, a probabilistic study on durability concrete was carried out. In such a design, initiation time of corrosion must be expressed as a mathematical model using Fick’s second law and the statistical distributions properties of theirs parameters was included in this model. The scatter both in the environmental exposure conditions and structural properties was considered as random fields in the mathematical model with a probabilistic design. The main objective of this study is predicted initiation time of corrosion of concrete structures in chloride containing environments. This probabilistic study is developed using Monte Carlo simulation to determine the contribution of each input parameters and the statistical parameters of the random variables on the probability distribution functions of the initiation time of corrosion. Also, a comparison study was carried out to analyze the impact of the probability distribution on the response (the initiation time of corrosion).

Atmospheric Boundary layer wind tunnels (ABLWT) dedicated to building safety and comfort have been operated by CSTB in Nantes since 1971. Because ABLWT only deal with reduced scale models of real structures, the necessity of a larger wind tunnel, the Jules Verne Climatic wind tunnel (CWT), able to reproduce extreme wind loads on real scale structures arose in the years 80. Hence, it became a major European facility operating for improvement of the safety, quality and environmental impact of buildings and civil engineering works as well as products from industrial fields (transportation, energy…) with respect to strong winds and other climatic hazards. Both wind tunnel types, the ABLWT and the CWT are complementary and used for studying the effect of wind on the same structures at two different scales, when the effect of wind scaling is important. During the 2018 year, several modifications were made to the CWT facility. The atmospheric test section of the existing facility was elongated preserving the initial advantages, very large test section (approximately 120 m²) with wind velocity performance compatible with many applications (up to 90 km/h). This new test section makes it possible to simulate turbulent wind and driving rain testing. The sand winds capabilities have been maintained in the new design, despite the closed loop configuration, by fitting a filtering. The modifications of the wind tunnel geometry now offer a long test section upstream the turning vanes where a whole set of new tests can be carried out, as windmill field, natural ventilation of urban environments, slender structures (large bridges, pylons, cable transport systems,)

The cubic translation model is a well know tool in wind engineering, which provides a mathematical description of a non-Gaussian pressure as a cubic transformation of a Gaussian process. This simple model is widely used in practice since it offers a direct evaluation of the peak factors as a function of the statistics of the wind pressure data. This transformation is rather versatile but limited to processes which are said to be in the monotonic region. For processes falling outside this domain, this paper describes an alternative which is based on the physics of the wind flow. First, it is shown, with a classical example of a flow involving corner vortices on a flat roof, that the pressure data which does not meet the monotonic criterion is in fact associated with a bimodal distribution. Then, the proposed approach is to decompose this data into the two governing modes (slow background turbulence and fast corner vortices) and apply the usual translation model to each of them.