Referee : 
Stéphane Morel

Members : 
Stéphane Morel, Christine Delisée, Jean-Christophe Mindeguia

Our research activities are related to the analysis and to the modelling of the properties and the behavior of building materials. We emphasize on the physical processes involved in damage development. Our work focuses on the study of quasi-brittle failure mechanisms in building materials (wood, concrete, masonry…). The interpretation of rupture includes various approaches such as the estimate of classical failure criteria (I, II and combined modes), NLFM modelling, damage modelling, the combining of fracture features with the thermal and hygroscopic behavior of wood-based materials, and the study of scale effects. The numerical forward modelling is based on codes using either finite element methods (e.g., modelling of cohesion, damage modelling, thermal-hygro-mechanical couplings) or discrete element methods (e.g., masonry, joints). We are also interested in the effects of severe loading, that includes the behavior of wood composite materials and masonry under seismic loading or during fires as well as the study of tree vulnerability to high winds. The design of materials can be improved eventually from a better understanding of the interactions between the materials morphology, their mechanical properties and their evolution with time. The purpose of material improvements can whether be to reduce the environmental impact (e.g., using composite materials based on natural fibers, developing engineering wood products), or to increase the durability of building structures. Either way, we keep focusing on innovation and optimization following these objectives.

(Left) In-situ measurements of the anchorage resistance of a tree. (Center) Measurements of the mechanical properties of multi glued-in rods joint. (Right) Numerical modelling of the damages in a wall during an earthquake.

Referee : 
Alain Denis

Members : 
Alain Denis, Colette Sirieix, Roland Lastennet

This group focuses on the study of the physical, chemical and mechanical properties of natural systems, including soils and rocks, involved in the transfer processes (e.g., hydrological/aerial)occurring in the near-surface. This work covers both natural and anthropic forcing. We develop reliable tools and measurement technics, along with the continuous improvement of original theoretical, experimental and numerical approaches. A significant part of the work relies on instrumental deployment in-situ (geophysics, water quality measurements, geotechnical analysis), and on the monitoring of experimental sites at the field scale.

The scientific work relates to four main challenges:

• to understand the physical mechanisms: 1. driving groundwater flow in epikarstic and karstic reservoirs and its implications for water management; 2. controlling underground aerial transfers, especially inside painted caves; 3. analyzing CO2 exchanges between the surface and underground cavities;
• to develop original methods for geophysical data processing: 1. for time-lapse monitoring of the near-surface; 2. from the joint interpretation of geotechnical and geophysical data to reduce the uncertainties about the soils mechanical properties;
• to design new experimental setups: 1. to improve usual geotechnical tests for a better estimation of the behavior of watering/dewatering shales; 2. to perform physical and chemical analysis along with microscopic scale mapping (SEM, X-ray spectrometry) to infer mechanical properties of shaly soils;
• to perform numerical modelling of the structure and of the mechanical properties of a rock mass or of a cliff using 3D photogrammetry surveys (on land or airborne survey), combined with the time-lapse monitoring of displacements occurring at a smaller scale. 

The experimental approaches are developed both in the laboratory and at the field scale. This relies on a wide range of expertise in various domains including geophysics (electrical tomography, electromagnetics, seismic) and non-destructive testing, geomechanics (strain measurements, photogrammetry, fracture mapping), analysis of physical parameters (moisture, temperature, CO2), and chemistry (main elements, isotopes). 

(Left) 3D geostatistical integration of geophysical data from the site of Lascaux. (Center) Factorial analysis of the parameters driving an epikarstic water flow. (Right) Monitoring of water suction in a watering soil.

Referee : 
Sidi-Mohammed Elachachi

Members : 
Sidi-Mohammed Elachachi, Jean-François Lataste, Zoubir-Mehdi Sbartaï

The group develops original methods to assess the influence of uncertainty within complex systems in the field of civil engineering. These methods are applied to various topics including the eco-design of construction materials, the soil-structure interactions, the on site assessment of building materials, the asset management of infrastructures, the hazard management associated with construction projects. All these issues are multi-objective problems with many interactions and in which uncertainties appear at many levels, from the measurement to the final decision.

The main challenge of these approaches is the conception of reliable and practical tools that can identify the main sources of intrinsic and epistemic uncertainty in a complex system, as well as the propagation of uncertainties within that system. The methods developed in our group help to interpret and accurately estimate these uncertainties. The design and development of numerical simulators reproducing the main issues of the real complex system is a way to better understand and control the role of uncertainties. Such tools are used for simulating the management of public heritage (networks, civil engineering structures) or improving the on site non-destructuve assessment of buildings. Our work covers data analysis, knowledge modelling, multi-criteria estimation methods, as well as optimization methods for decision making.

(Left) Influence of the number of specimens on the accuracy of the estimate of concrete strength. (Center) Pareto front of optimal solution in a soil-structure interaction problem. (Right) Multi-criteria estimation and optimization