IMT Nord Europe

• Multiscale chain-based FE framework for double-network gel mechanics • Coupled damage and viscoelasticity drive path-dependent anisotropy • Predicts hysteresis, damage localization, and crack initiation • Energy-based formulation captures fracture resistance evolution • Physics-informed framework for designing tough soft materials Double-network (DN) gels exhibit an exceptional combination of stretchability, toughness, and energy dissipation, making them promising materials for soft robotics, biomedical devices, and flexible electronics. These outstanding properties arise from the interplay between a brittle, sacrificial primary network and a ductile, extensible secondary network. However, capturing the complex, anisotropic, and path-dependent mechanical response of DN gels - especially under multiaxial and non-uniform loading - remains a major modeling challenge. This work introduces a three-dimensional multiscale constitutive model that integrates progressive chain scission, anisotropic damage, viscoelastic relaxation, and inter-network interactions. The primary network is modeled as an entropic assembly of chains with stochastic lengths undergoing irreversible damage, upscaled via microsphere integration to capture directionality. The secondary network follows a generalized Maxwell formulation to reproduce rate-dependent dissipation. Energy coupling between the two networks enables the model to reflect microstructural rearrangement, directional softening, and residual anisotropy. The model is implemented as a UMAT in ABAQUS/Standard and validated against experimental data under both homogeneous and non-uniform loading conditions. It accurately captures nonlinear stress-strain behavior, hysteresis, and anisotropic softening resulting from prior deformation. In pre-notched specimens under pure shear, simulations reproduce force-displacement response, stress localization, and damage evolution up to crack initiation. An energy-based analysis highlights the roles of reversible and dissipated energy, with predicted critical energy release rates closely matching experiments across pre-stretch levels. The model offers a predictive, physically grounded framework for understanding and engineering tough, damage-resistant DN gels.

This study provides a detailed assessment of groundwater salinization in the Quaternary aquifer of the Samba Dia region, Senegal, using an integrated approach that combines hydrochemical, stable isotopic (δ2H, δ18O), and electrical resistivity tomography (ERT) techniques. Fourteen high-resolution ERT profiles, along with comprehensive chemical and isotopic analyses, were performed to identify the main causes of salinity and their spatial distribution. Results show that groundwater salinization in the area is primarily driven by three mechanisms: seawater intrusion, surface salt leaching, and ion exchange. Hydrochemical facies evolution diagrams, ionic ratios, and isotopic signatures helped differentiate marine-influenced zones from inland salinization areas. ERT imaging also mapped the three-dimensional extent and geometry of saline interfaces, confirming zone-specific mixing of seawater and freshwater. The findings indicate that salinization of the coastal aquifer has worsened over the past twenty years, mainly due to human activities and climate variability. This study recommends a sustainable monitoring strategy to support aquifer management, focusing on accurately identifying vulnerable zones and enabling adaptive resource planning in semi-arid Senegal.

International audience

International audience

Online continuous action recognition has emerged as a critical research area due to its practical implications in real-world applications, such as human-computer interaction, healthcare, and robotics. Among various modalities, skeleton-based approaches have gained significant popularity, demonstrating their effectiveness in capturing 3D temporal data while ensuring robustness to environmental variations. However, most existing works focus on segment-based recognition, making them unsuitable for real-time, continuous recognition scenarios. In this paper, we propose a novel online recognition system designed for real-time skeleton sequence streaming. Our approach leverages a hybrid architecture combining Spatial Graph Convolutional Networks (S-GCN) for spatial feature extraction and a Transformer-based Graph Encoder (TGE) for capturing temporal dependencies across frames. Additionally, we introduce a continual learning mechanism to enhance model adaptability to evolving data distributions, ensuring robust recognition in dynamic environments. We evaluate our method on the SHREC'21 benchmark dataset, demonstrating its superior performance in online hand gesture recognition. Our approach not only achieves state-of-the-art accuracy but also significantly reduces false positive rates, making it a compelling solution for real-time applications. The proposed system can be seamlessly integrated into various domains, including human-robot collaboration and assistive technologies, where natural and intuitive interaction is crucial.