HAL Portal UTC Université de Technologie de Compiègne
Not a member yet
11652 research outputs found
Sort by
Enhancing the texture and modulating digestive behavior of gluten-free quinoa sponge cakes via microwave-assisted alkaline amino acid treatment
No full textInternational audienceThis study investigated the effects of microwave treatment combined with lysine and arginine on gluten-free quinoa sponge cakes. The results indicated that the addition of these amino acids during microwave treatment significantly increased the cakes' specific volume by 49 %. X-ray diffraction analysis revealed that cake crystallinity reached 56.23 % and 49.17 % when lysine and arginine were used, respectively. Fourier Transform Infrared spectroscopy showed confirmed the absence of new functional groups or chemical bonds under different treatments, but the absorbance ratio at 1047 cm−1 to 1022 cm−1 was higher in microwave-treated cakes with lysine compared to traditionally steamed cakes. Simulated digestion experiments demonstrated that microwave-treated cakes, especially with added amino acids, exhibited higher protein digestibility but lower starch digestibility. Confocal Laser Scanning Microscopy observations further showed that proteins formed a denser network structure around starch granules during in vitro digestion, suggesting improved protein functionality and structure in microwave-treated quinoa sponge cakes
Lidar Pole Detection Training using Vector Maps for Localization
No full textInternational audienceAutonomous navigation requires accurate and reliable localization. In urban environments, infrastructure such as buildings and bridges disrupts Global Navigation Satellite Systems (GNSS), which requires the implementation of robust perception systems combined with inertial navigation. Roadside poles like traffic signs or light poles can serve as stable landmarks for map-based localization. When georeferenced in high-definition vector maps, these features enable reliable localization through detection pipelines and data association methods. While lidar captures their 3D geometry, distinguishing mapped poles in raw point clouds remains challenging. To train pole detectors tailored to the specific map used, we propose an automatic annotation framework that integrates lidar data, a vector map, and offline semantic segmentation to generate precise labeled data. By combining annotated pole clusters from the map with semantic segmentation, annotation errors can be minimized. This enables the training of a map-specific classifier optimized to detect mapped poles while filtering out irrelevant structures. It eliminates the need for manual labeling and ensures adaptability to the map used for online localization. Thanks to real data acquired in real-world urban scenarios, we show that this approach enhances significantly localization accuracy
Current Transformer-Based Power Supply for Pulse Encoders on Motor Power Lines in Adjustable Speed Drive Systems
No full textInternational audienceThis paper proposes a novel power supply system for an encoder in motor drive systems. The proposed system eliminates the need for a dedicated power supply cable for the encoder by utilizing a current transformer connected to the load motor. A design method for the current transformer is presented, considering its operation under non-linear loads and variable frequencies. The design process is simplified by modeling the diode rectifier as an equivalent voltage source. The validity of the proposed design is verified through experiments, demonstrating that the output voltage closely matches the design values. Additionally, a DC-DC converter is implemented at the output side of the diode rectifier to stabilize the output voltage. Experimental results confirm voltage stabilization across the entire operating range on the N-T map
Information System Architecture for Supervision and Control of Railway Smart Grids
No full textInternational audienceThis paper presents an information system architecture for controlling a storage system connected to a DC railway substation with local photovoltaic production (PV). The proposed architecture collects various information and data measurements necessary for the proper operation of the storage system enabling both day-ahead and real-time control
Laboratory Measurements for Machine Learning-based Modelling of LV Harmonic Sources
No full textInternational audienceAs renewable energy grows, power electronic converters increase harmonic disturbances, affecting power quality. This study introduces a novel modelling methodology based on machine learning and laboratory databases for harmonic current prediction. A model using electric vehicle charger data demonstrates promising performance. Integrating these models into low-voltage network improves frequency-domain simulation and enhances grid analysis
Transition Control from Dual to Single Motor to Prevent Speed Reduction during an Open-Circuit Fault in a DC Power System
No full textInternational audienceThis study examined a tandem motor system for more electric aircraft (MEA). A method was proposed that, during a single motor's power loss, continuously outputs the necessary torque to maintain speed and suppress speed reduction compared with standard controls, achieved by directly modulating torque command distribution based on overmodulation detection
T-type Inverter as a Power-Pulsation Buffer for a PMSM with Large Load Variation
No full textInternational audienceThis paper presents a modulation strategy for a three-level T-Type inverter to utilize the neutral point as a power-pulsation buffer. The pulsating power of a swing compressor is compensated by the buffer, resulting in dc input power of the motor inverter, despite the varying load. Consequently, the overall drive system can be designed with significantly reduced dc-link capacitance, saving cost and size. Next to theoretical derivation, the proposed modulation strategy is verified in simulation and experiment
Modular Clamping Circuit for High Speed Solid State Circuit Breakers
No full textInternational audienceThe aviation industry is on the cusp of a revolutionary transformation with the advent of zero-emission aircraft. These innovative aircraft, powered by cutting-edge technologies such as electric propulsion and hydrogen fuel cells, promise to significantly reduce the environmental impact of air travel. As global awareness of climate change and sustainability grows, the development and adoption of zero-emission aircraft are becoming increasingly vital.However, the integration of these advanced technologies necessitates robust electrical network protection to ensure safety and reliability. This leads to the need for advanced protection circuits, such as Solid-State Circuit Breakers (SSCBs) and clamping circuits, which are essential for managing the complexities of modern electrical systems.This paper focuses on an advanced clamping circuit that enables efficient voltage clamping for high-speed SSCBs. The method effectively clamps overvoltage in failure cases, realizing almost no voltage overshoot and a minimum time to clear the fault current. Due to its complete analog nature it is very suitable for safety critical applications where active digital control circuits have to be avoided. The proposed method is very efficient, ensuring the stability and safety of electrical networks in zeroemission aircraft.</p
Prise en chargedes douleurs chroniquespar les dispositifs médicaux
No full textThe management of chronic pain represents a major public health challenge, both in France and globally. In 2023, it is estimated that around 12 million adults in France suffer from chronic pain, 70% of which still do not receive adequate care.Patient care is based on a multidisciplinary approach. It includes the diagnosis of chronic pain with functional imaging techniques, electromyography and in addition to rating scales. Management continues with treatment, whose therapeutic approaches for the relief of chronic pain include transcutaneous electrical stimulation (TENS), or photobiomodulation (PBM). Finally, as these conditions are long-term, the treatment logically includes an aspect of home follow-up.This essay means to showcase these medical devices used in the management of chronic pain, along with how they work, their pros and cons, for whom and how they are used. We aim at helping people interested and involved in the matter (medical, biomedical staff and patients) better understand the scope of these solutions.La gestion de la douleur chronique représente un défi majeur en matière de santé publique, tant en France qu'à l'échelle mondiale. En 2023, on estime qu'environ 12 millions d’adultes en France souffrent de douleurs chroniques, dont 70 % ne bénéficient toujours pas d’une prise en charge adéquate.La prise en charge du patient repose sur une approche multidisciplinaire. Elle inclut le diagnostic de la douleur chronique avec les techniques d’imagerie fonctionnelle, l'électromyographie et en complément des échelles d'évaluation. Les approches thérapeutiques pour le soulagement de la douleur chronique comportent notamment la stimulation électrique transcutanée (TENS), ou encore la photobiomodulation (PBM). Enfin, ces affections s’inscrivant dans la durée, la prise en charge inclut logiquement un aspect de suivi à domicile.Ce mémoire a pour but de mettre en lumière ces dispositifs médicaux utilisés dans la prise en charge de la douleur chronique, leurs principes de fonctionnement, leurs avantages, pour qui et comment ils s’appliquent. Nous espérons ainsi aider les personnes concernées (corps médical, biomédical, patients…) à avoir une meilleure compréhension du panorama de ces solutions
Minimization of the circulating currents of an interleaved converter for particle accelerator using inversion-based control
No full textInternational audienceAn interleaved DC/DC converter is used to supply a magnet for particle accelerator. A very high accuracy of 50 parts per million is required to control the particle beam. Any disturbance can reduce this accuracy. An inversion-based control is derived from the Energetic Macroscopic Representation of this power converter. This control is composed of several control loops and compensations. The control leads to the right accuracy despite some ripple due to current measurement