HAL Portal UTC Université de Technologie de Compiègne
Not a member yet
11652 research outputs found
Sort by
Assessing the Decisional Capability for an ODD-Compliant Automatic Lane Change System via Sense-Think-Act Paradigm
No full textInternational audienceAs autonomous vehicles become more advanced, ensuring their safe and reliable operation is critical. Intelligent Vehicles (IVs) must make real-time decisions while adapting to dynamic environments, such as changing weather conditions, road obstacles, or varying traffic scenarios. However, for an optimal and safe vehicle navigation, it must remain within its Operational Design Domain (ODD), the set of conditions under which it is designed to operate safely. If a vehicle exceeds these limits, the risks of failure, unsafe behavior, or costly interventions increase. A key challenge is aligning the vehicle's capabilities with its ODD. This means that before executing any maneuver, the vehicle must assess whether it has the necessary capabilities to perform it safely. Current monitoring approaches focus on detecting violations after they occur rather than anticipating them. Furthermore, as vehicles gain autonomy, scalable evaluation methods are needed for ODD compliance. This paper proposes an approach based on the Sense-Think-Act paradigm, utilizing skill graphs to dynamically assess the vehicle's decisional capability for a lane change maneuver. This allows the IV to assess whether it can successfully perform the maneuver while staying within its ODD. The modular structure of the approach makes it compatible with long-term decision-making. Finally, simulation results prove the efficiency of the approach in preventing the vehicle from exiting its ODD
Moving towards the eco-design of flexible piezoelectric energy Harverster: some issues and prospects
No full textInternational audienc
Multiscale modeling of the skeletal muscle: an original approach of periodic representative volume elements
No full textInternational audienceSkeletal muscle is a hierarchical structure composed of specific constituents and mechanical properties at different scales, that can influence the overall organ function, health and mechanical response to stress. Finite element (FE) models are increasingly used to analyse the effect of these components (muscle fibers, fascicles, myofibrils, and the extracellular matrix - ECM) on the overall mechanical properties of the organ. One of the critical points in the development of the FE models lies in modeling the transition from one anatomical scale to the others. We developed an original approach based on 3D periodic RVE generation for the multiscale FE modeling of the skeletal muscle. By the inclusion of custom geometrical and material features, this model could be used for the multiscale and multicomponent analysis of different muscle phenotypes
Integrating context and criteria: a multi-head attention-based approach for multi-criteria group recommender systems
No full textInternational audienceGroup recommender systems (GRS) are essential for collective decision-making across various domains. However, traditional approaches often struggle to balance diverse group interests and tend to overlook crucial contextual factors and multi-criteria evaluations inherent in real-world choices. To address these limitations, we propose a Context-Aware Multi-Criteria Group Recommender System (CA-MCGRS) that employs a multi-head attention (MHA) mechanism. This architecture dynamically integrates individual user preferences, multiple evaluation criteria, and diverse contextual influences, enabling adaptive and nuanced recommendations. We conducted experiments on two distinct datasets: the ITM-Rec educational dataset, which includes multi-criteria ratings (e.g., application relevance, data quality, usability) and specific academic contexts (e.g., class, semester, lockdown status), and the widely-used MovieLens-100k dataset, adapted for group scenarios with simulated multi-criteria and contextual dimensions (e.g., time of day, season). Our results demonstrate that the MHA-based CA-MCGRS significantly improves predictive accuracy compared to various baseline models, particularly when leveraging rich multi-criteria information and adapting to specific contextual situations, such as the impactful lockdown status in the educational setting or temporal variations in the entertainment domain. Although the model exhibits longer training durations, it demonstrates superior generalization capabilities. These findings underscore the substantial benefits of integrating multi-criteria analysis and contextual awareness in GRS, offering valuable insights for developing more precise and relevant recommender systems
Silk Lipopeptide Surfactants: Mechanochemical Synthesis, Properties and Characterization
No full textInternational audienceNatural-based surfactants are widely used in cosmetics, detergents, and various industrial applications due to their biodegradability and environmentally friendly properties. The production scale of surfactants is immense, and this market continues to experience significant annual growth. A new challenge has emerged that requires the identification of natural resources that are sustainable, do not compete with the agro-food industry, and are suitable for applications in cosmetics and detergents. Silk as a protein biopolymer is an interesting material for biomedical applications due to its biocompatibility, mechanical strength, and unique structural properties. In this context, the development of lipopeptide surfactants from silk biomass, which is not in competition with the food sector, presents an innovative and sustainable alternative.The main objective of this work is to valorize silk proteins as a non-food biomass for the development of amphiphilic molecules. The first step involves the separation of silk protein fractions using green processes. These proteins were subsequently modified through enzymatic hydrolysis and an acylation reaction. The originality of this study is the development of a mechanochemical approach as an alternative to the conventional lipopeptide synthesis method. The surface adsorption properties of the resulting silk lipopeptide surfactants, featuring variable structures, were thoroughly studied, with their emulsifying and foaming capacities. The synthesized lipopeptides demonstrated a strong potential to replace petroleum-based surfactants in different applications, especially in cosmetics and household detergents.</p
Humanités numériques : théorie et outils pour les éditions scientifiques
No full textNational audienc
Maximizing C-phycocyanin purification efficiency from Spirulina: A synergistic strategy combining CaCl2 precipitation and membrane diafiltration
No full textInternational audienceC-Phycocyanin (C-PC) is a bioactive compound known for its potent antioxidant and anti-inflammatory properties, making it widely used in health supplements. Ultrafiltration membranes are commonly utilized to isolate C-PC from Spirulina extract due to their operational simplicity. However, the presence of various proteins in the crude Spirulina extract complicates the separation of C-PC, resulting in significant membrane fouling. Although threshold flux operation can minimize foulant-foulant deposition, the strong foulant-membrane interactions significantly narrow the pore size, leading to poor separation selectivity of C-PC from other impurities. Drawing inspiration from the specific calcium-binding sites on photosystem proteins in the thylakoid membrane, we propose, for the first time, a calcium chloride (CaCl2) precipitation pretreatment method to remove these proteins from the extract, increasing C-PC purity from 0.56 to 1.02. Subsequently, we employed membrane diafiltration technology to eliminate salts and other organics, further enhancing C-PC purity to 1.23, with a concentration of 7.58 g/L. This work presents a practical and efficient method for C-PC extraction and purification, paving the way for future applications
Off-Grid Wind Energy Conversion System with Supercapacitor-Assisted Control for Optimal Green Hydrogen Production
No full textThis paper was presented at the EPE 2025 conference and was subsequently selected for publication in the special issue of Elsevier's PEDC journal. Therefore, only the abstract appears on the conference website.International audienceThis paper presents and validates a novel control strategy for directly integrating wind energy conversion systems (WECSs) with electrolyzers (ELZs) in off-grid systems maximizing electrical power and hydrogen production. The study builds on previous research that directly integrates WECSs with ELZs, eliminating the need for energy storage systems by using the ELZ as a controllable load, managed through maximum power point tracking (MPPT). Although this approach has shown effective results, it also led to significant current fluctuations in the ELZ. The new contribution of this study is the integration of supercapacitors (SCs) and modification of the control to smooth current fluctuations in the ELZ, reducing its degradation and improving operational efficiency. Simulations demonstrate the effectiveness of this control strategy, showing enhanced power management and reduced current peaks as SCs absorb power surges. This reduces ELZ cycling and degradation, extending its lifespan and offering a possible cost-effective solution for green hydrogen production in remote or offshore locations
An AC fault ride-through control of modular multilevel converters for HVDC systems
No full textInternational audienceThis paper proposes a modified threshold virtual impedance (MTVI) control to ensure an AC fault ride-through (FRT) for modular multilevel converters (MMCs). The proposal limits the AC current of the MMC by virtually increasing its output impedance. It prevents to trigger its protection systems when a fault occurs. Compared with conventional TVI control, the proposed method avoids not only MMC blocking but also increases its power quality by providing undistorted voltage and current waveforms in the case of asymmetrical faults. Simulations in MATLAB/Simulink are carried out to validate the effectiveness of the proposed MTVI control, which is compared with the conventional TVI control. The results show the ability of the MTVI control to limit voltage and current perturbations in the MMC and to provide better voltage and current profiles compared to the classical TVI during unbalanced faults
Estimation de la température interne de cellules Li-ion de grande capacité par impédancemétrie embarquée
No full textInternational audienceLa température interne des accumulateurs électrochimiques constitue un paramètre crucial, ayant un impactsignificatif sur leur vieillissement et leur sécurité. En particulier,le phénomène d’emballement thermique doit être détecté précocement en raison de son évolution rapide et de ses conséquencespotentiellement catastrophiques. Cependant, la mesure de latempérature pour chaque cellule au sein d’un pack batterie estcoûteuse et ne permet généralement d’accéder qu’à la températureexterne des éléments. Par ailleurs, l’augmentation de la taille descellules s’accompagne d’une hausse de leur capacité thermique,entraînant un décalage temporel significatif entre la températureinterne et la température de surface. Les techniques d’analysepar impédancemétrie offrent une méthode précise pour évaluerles paramètres et phénomènes électrochimiques, lesquels sontétroitement corrélés à la température interne des batteries. Cetarticle propose une approche innovante en utilisant une versionembarquée de la technique d’impédancemétrie pour estimerla température interne des accumulateurs électrochimiques degrande capacité énergétique