Admin CliCours
Energy supply by nuclear power plants constitutes, nowadays, one of the main technologies adopted in this industrial sector. The treatment and isolation of the radioactive waste generated by activities based on nuclear energy have become fundamental concerns. In principle, more intensive emissions of radionuclides from radioactive waste would require more isolated disposal facilities. In France, about the 90% of total wastes (Low- and Intermediate-Level, Short-Lived wastes, LIL-SL, and Very Low-Level wastes, VLL) are disposed in existing, sub-surficial locations. At the national level, the principal issue is the disposal management of the minor part of nuclear waste, constituting, nonetheless, more than 90% of the total radionuclides emissions: the High-Level and Intermediate-Level, Long-Lived wastes (HL and IL-LL) [5]. Among other countries, France has identified the deep geological repository (DGR) as a suitable and safe technique to conceive and implement deep disposal facilities for radioactive waste correspondent to HL and IL-LL types. DGR is based on the individuation of favourable geological formations characterized by structural, hydro-mechanical and geochemical properties to isolate as best the radionuclides dispersion and assure the most efficient construction and operation of disposal facilities. These type of formations are identified as host media, or host rocks, because of their geological features (e.g. claystones and shales). In France, the mission of the site identification and feasibility studies for a DGR concept is in charge to the French national radioactive waste management agency, Andra (Agence Nationale pour la gestion des déchets radioactifs). The agency has individuated the deep geological formation of Callovo-Oxfordian claystone as an appropriate host rock constituting the final barrier for the disposal of HL and IL-LL wastes. Characteristics such as very low permeability and reduced molecular diffusion constitutes favourable elements for this choice .
Medium-term reversibility, for about 100 years, and safety after the repository closure, for thousands of years, are key requirements related to the conception of a geological repository. Andra has conceived the project Cigéo (Centre Industriel de Stockage Géologique / Industrial center for geological disposal) to forward these demands ([3], [31]): it will be composed of horizontal cells connected to access galleries and containing the radioactive wastes . Once the feasibility study phase started, Andra began, in 2000, the construction of the Underground Research Laboratory (URL) at the agency Centre in the Meuse-Haute Marne department (CMHM). Excavated in the Callovo-Oxfordian formation, at about 500 m depth, the laboratory is a network of drifts to characterize the chemo-thermo-hydro-mechanical properties and behaviors of the confinement claystone, as well as test, demonstrate the feasibility, optimise the concept of the future DGR.
Drifts are intended as real-scale experiments to characterize the response of the rock according to the excavation method, structure geometry, supports system and orientations with respect to the principal stresses’ directions (e.g. [7], [8], [91]).In geological formations such as the CallovoOxfordian claystone, the excavation of deep tunnels creates a surrounding area characterized by cracking and diffused failure, whose shape and extension are the main issues for the repository safety. The terminology to define this zone has undergone several changes with progresses in geomechanics. Generally, researchers divide the overall zone in different concentric parts depending on the induced modifications on the hydraulic and transmission properties of the material ([71], [103]). In the proximity of the gallery perimeter, the formation of several interconnected macro-fractures leads to irreversible changes of these properties. Moving outward, the influence of the excavation decreases, together with the induced damage interconnections. Further, only temporary and reversible modifications of the material properties occur. Thus, in the area where failure occurs, variations of the excavation fractures’ entity and interconnections corresponds to different modifications of initial properties. These zones can be all referred as Excavation Damage Zones, EDZs, according to the classification review published by Perras and Diederichs [71]. Collecting different related works, the authors provide a scheme distinguishing four concentric levels of EDZs, from the excavation perimeter:
1. Construction Damage Zone (CDZ): it includes inevitable excavation consequences and additional damage effects induced by the construction method. These second effects may be adjusted according to the excavation / support techniques.
2. Highly Damaged Zone (HDZ): it includes inevitable damage causing geometry, structure, and/or induced stress changes, independent of excavation method. Highly interconnected macro-fractures are typically observed.
3. Excavation Damage Zone (EDZ): it corresponds to the transition from a connected damaged area to a partially connected or isolated damage area. In these areas, irreversible microdamage phenomena occur .
4. Excavation Influence Zone (EIZ): this is a stress-strain influence zone involving only reversible (elastic) changes. The outer limit, occurring at a large radial distance, is of minimal interest for a single excavation. On the contrary, in case of multiple, parallel excavations the interaction of adjacent EIZs should be considered.
In this work, for sake of simplicity in the notation, the ensemble of the three zones CDZ, HDZ and the inner EDZ is identified by the single acronym EDZ (D = Damaged). A unique notation identifying the area characterized by irreversible changes in the material were already proposed by Emsley et al. [35] and also adopted by Tsang et al. [103]. The transition zone, moving from the inner limit of the EDZ to the EIZ (irreversible to reversible changes) is identified with the acronym EdZ (d = disturbed). Many efforts are addressed by Andra to understand the phenomena leading to the formation of an EDZ-EdZ system around the drifts in the Callovo-Oxfordian claystone at the Underground Research Laboratory.
These efforts have undergone several scientific campaigns, at different levels, from laboratory micro-scale to in-situ experiments and monitoring, with the cooperation of different research groups (e.g. [91], [93]). This thesis project aims to provide a contribution on these efforts. Numerical analyses on computational models at the underground structure scale constitute the subject of this work and are presented and discussed in this manuscript. The approach adopted to model the EDZ-EdZ formation around deep galleries will account for a reduction of the host rock resistance when the reversible (elastic) conditions are overcome and material failure is attained. In general, the extension of this zone is estimated from a stress field calculated in elasticity or based on an elastic-plastic calculation ([3], [25]); if the first method does not take into account the redistribution of stresses due to irreversible phenomena, the conventional elastic-plastic modelling seems insufficient to explain the geometry of the failure zone encountered in some cases of deep structures in quasi-brittle rocks ([77], [78] and [102]). Observations suggest that damage mechanics phenomena, while the material resistance decreases (i.e. softening), are crucial to the development of these zones. Predictions of the material short-term failure will be performed by a purely mechanical mathematical and numerical modelling, based on damage mechanics. In the following, a description of the Andra URL, with relevant in-situ instrumentations and measurements is reported.
Chapter 1 : Introduction and Context of the Research |
Télécharger le document complet
