Search

Chapeau. Le 9 janvier 2018, autour de 7h15, une avalanche d'ampleur significative descend le couloir du Bourgeat. L'écoulement de neige poudreuse, également appelé aérosol, détruit la forêt sur une largeur très importante et franchit le paravalanche du Bourgeat. Une quarantaine de chalets en aval de la digue terminale du paravalanche sont recouverts de neige par l'aérosol, et de nombreux arbres sont détruits et transportés par l'aérosol, ce qui cause les principaux dégâts qui restent fort heureusement uniquement matériels. Lors de sa propagation dans le couloir, l'écoulement dense, de neige froide au départ, se transforme en un écoulement plus lent, de neige humide, et interagit de manière complexe avec le paravalanche qui, in fine, contient bien la masse de neige lourde. Cet article propose une brève analyse qualitative des différentes hypothèses concernant les conditions de déclenchement, de propagation de l'aérosol et des écoulements denses et de leur interaction avec le paravalanche. Il dresse également une liste de questions qui peuvent se poser suite à cet événement.

The current code approaches for engineered systems are generally conceived on the basis of component designed philosophy, where explicit requirements are mostly specified at component level or at individual members. There is very little guidance on the overall safety and the approach for design to prevent progressive collapse as a result of abnormal loading is not standard. Several reference papers on the subject treat the problem qualitatively. Following a recent scheme in the literature, in which structural integrity is captured in the structural stiffness matrix, in this presentation we will treat the problem through a quantitative approach extending the analysis from the linear elastic case to the plastic deformation range in terms of frames removal or plastic hinges insertion. With this aim we discuss system integrity measures, and we suggest that the inverse of the stiffness matrix condition number (the so called metric distance) gives a better representation of structural integrity. We also discuss the compartmentalization problem and introduce for the first time quantitative measures of it. We finally apply such new measures to simple examples and show an optimization problem related to the structural integrity.

International audience; Within the Operational programme 'Italy - France (Alps - ALCOTRA) 'Project DynAval - Dynamique des avalanches: départ et interactions écoulement/obstacles three test sites were built or developed in France and Italy. The goal of the paper is to present the experimental devices and typical pressure generated by avalanches in these 3 sites differing mainly by their scale. (a) Seehore test site is located in Aosta Valley on Monte Rosa Massif. The slope, with an altitude difference of about 300 m (from 2300 to 2570 m a.s.l.), has a mean dip of about 38°. The site is instrumented with a steel obstacle on which load cells and other devices are mounted in order to measure the effects of impact of the avalanche. Surveys are made before and after each artificial event: snow density, front velocity, erosion and deposition are measured and photogrammetric and laser-scanner views are taken. (b) Col du Lautaret test site is located near the Lautaret pass (2058 m a.s.l) between Cerces and Ecrins range. Different avalanche paths are located on the south-east slope of Chaillol Mountain (2600 m a.s.l.). Small to medium avalanches occur at a sufficient frequency (up to 3 or 4 each winter). Avalanche flows are generally dense, wet or dry, with sometimes a small but fast powder cloud (or saltation layer). The dense part is usually less than one meter thick. The run-out distance is 500 to 800 m with an average gradient of 36°. Typical released volume is about 5000 m3 and front velocity can reach 30 m/s. Instrumentation includes a 3 m-high mast recording pressure and velocity each 20 cm, and a one square meter plate integrating the pressure all over the flow height. A high speed photogrammetric system is able to measure the avalanche front velocity. (c) The Taconnaz avalanche path is located in the Arves valley, close to Mont Blanc in France. The Taconnaz path is 7 km long, has a mean slope of 25° and a mean width of 300400 m. A defence structure system made of breaking mounds and dams was designed in 2009 based on a 100-year return period event of 1.6 Mm3 volume. In 2010, velocity and pressure sensors were set up on the breaking mounds in order to improve our knowledge of the interaction between avalanches and breaking dams.

The damage acting on a structure can lead to disproportionate consequences, i.e., the global collapse. This extreme situation has to be avoided and, thus, structural monitoring is requested in those structures where human losses are possible and large economic consequences are expected. Static measurement devices are the most economic instrumental set-ups able to highlight the presence of progressive damages. However, this monitoring system suffers from the structural behaviour under the external loads. In many situations, alternate load paths shown the non-effectiveness of the measurement system since the instrumentation are installed on elements not relevant for the response under the given loads. The same problems occur when the structure is compartmentalized, i.e. the structural responses of the single parts dependent on the loads acting almost only on the same single component. In order to measure the degree of compartmentalization, different novel metrics based on stiffness matrix properties are proposed and their effectiveness discussed. The new idea of this paper is to connect compartmentalization of structures with a sort of distance of the stiffness matrix from the set of diagonal matrices. Few examples are illustrated

Different procedures for assessing the robustness of a reinforced concrete (RC) frame under progressive damage are proposed and compared. The removal of a column in a RC frame structure is modeled with a commercial nonlinear finite element software according to three alternative strategies: (i) reduction of mechanical properties of the damaged column, (ii) incre- mental loading of the structure after total removal of the damaged column, and (iii) incremental unloading of internal forces on the damaged column. Nonlinear analysis is performed under a prescribed load combination on three concrete frames designed with three Italian building codes in force in different periods. Despite the differences in the strategies for damage mod- eling, similarities between structural response predictions are highlighted. In addition, it is shown that seismic design provisions for RC building structures increase the ductility of the structure but do not guarantee the robustness against the removal of a column.