HAL Portal UTC Université de Technologie de Compiègne
Not a member yet
11652 research outputs found
Sort by
Multi-Agent Reinforcement Learning Method for Disassembly Sequential Task Optimization Based on Human–Robot Collaborative Disassembly in Electric Vehicle Battery Recycling
No full textInternational audienceAbstract With the wide application of new Electric Vehicle (EV) batteries in various industrial fields, it is important to establish a systematic intelligent battery recycling system that can be used to find out the resource wastes and environmental impacts of the retired EV battery. By combining the uncertain and dynamic disassembly and echelon utilization of EV battery recycling in the remanufacturing fields, human–robot collaboration (HRC) disassembly method can be used to solve huge challenges about the efficiency of retired EV battery recycling. In order to find out the disassembly task planning based on HRC disassembly process for retired EV battery recycling, a dynamic disassembly sequential task optimization method algorithm is proposed by Multi-Agent Reinforcement Learning (MARL). Furthermore, it is necessary to disassemble the retired EV battery disassembly trajectory based on the HRC disassembly task in 2D planar, which can be used to acquire the optimal disassembly paths in the same disassembly planar by combining the Q-learning algorithm. The disassembly task sequence can be completed through standard trajectory matching. Finally, the feasibility of the proposed method is verified by disassembly operations for a specific battery module case
An Overview of Machine Learning-Enabled Network Softwarization for the Internet of Things
No full textInternational audienc
Internet of Things in healthcare: A survey on protocol standards, enabling technologies, WBAN architectures and open issues
No full textInternational audienc
FDEPCA: A Novel Adaptive Nonlinear Feature Extraction Method via Fruit Fly Olfactory Neural Network for IoMT Anomaly Detection
No full textInternational audienc
Synthesis of Fluorescent, Small, Stable and Non-Toxic Epitope-Imprinted Polymer Nanoparticles in Water
No full textInternational audienceMolecularly imprinted polymers (MIPs) are really interesting for nanomedicine. To be suitable for such application, they need to be small, stable in aqueous media and sometimes fluorescent for bioimaging. We report herein, the facile synthesis of fluorescent, small (below 200 nm), water-soluble and water-stable MIP capable of specific and selective recognition of their target epitope (small part of a protein). To synthesize these materials, we used dithiocarbamate-based photoiniferter polymerization in water. The use of a rhodamine-based monomer makes the resulting polymers fluorescent. Isothermal titration calorimetry (ITC) is used to determine the affinity as well as the selectivity of the MIP for its imprinted epitope, according to the significant differences observed when comparing the binding enthalpy of the original epitope with that of other peptides. The toxicity of the nanoparticles is also tested in two breast cancer cell lines to show the possible use of these particle for future in vivo applications. The materials demonstrated a high specificity and selectivity for the imprinted epitope, with a K d value comparable with the affinity values of antibodies. The synthesized MIP are not toxic, which makes them suitable for nanomedicine
#3828 KINETICS OF THE DE NOVO NAD/NADH PATHWAY FROM AKI TO CKD
No full textInternational audienceAbstract Background and Aims Mitochondrial dysfunction and energy metabolism deficiency are part of the acute kidney injury (AKI) pathophysiology. In particular, recent studies highlighted the role of nicotinamide adenine dinucleotide (NAD) production in the pathogenesis of AKI. NAD is an electron carrier for mitochondria and a cofactor for cytoplasmic redox reactions. Its production is decreased during AKI following a decreased of the expression of PGC1α (Peroxisome proliferator activated receptor gamma co-activator-1-α). PGC1α inactivation worsened and its overexpression prevented renal injury in a mouse model of renal ischemia. It was also shown that the consequences of PGC1α decreased expression included the alteration of the de novo NAD+ synthesis pathway. This was due to a reduction in the expression of the Quinolinate PhosphoRibosylTransferase (QPRT), which led to an accumulation of quinolinate. Consequently, the quinolinate/tryptophane (uQ/T) ratio increased in the urine. The modulation of this de novo has been studied mainly in the acute phase immediately following the injury. However its evolution with degree of severity and during the transition from AKI to CKD has not been described yet. In order to answer these questions, we designed the following studies: (1) A “severity” study in which we provoked ischemic AKI of increasing intensity in mice and quantified kidney PGC1alpha and QPRT mRNA expression. (2) An “AKI to CKD” study in which we followed kidney PGC1alpha and QPRT mRNA expression and uQ/T. Method Renal ischemia-reperfusion was performed in C57Bl6/J male mice using the new vascular occluder we recently developed (RIRI clamp), after nephrectomy of the contralateral kidney. For the ‘severity’ study, a 5–30 minutes ischemia was performed. Renal function (measurement of plasma creatinine and urea) and structure (Periodic Acid Schiff staining) were assessed 24h after ischemia. For the ‘AKI to CKD’ study, renal ischemia was performed during 10 minutes. Mice were sacrificed 1, 2, 3, 6 and 28 days after ischemia. For both studies, PGC1α and QPRT mRNA expression was quantified by qPCR in the kidneys. Urinary Q/T was determined by mass spectrometry. Results PGC1α and QPRT mRNA expression is inversely corrected to the AKI severity We induced AKI in mice by unilateral ischemia reperfusion injury of increasing time to induce several degrees of AKI severity. The measurement of plasma urea and creatinine concentration confirmed that the severity increased with the duration of ischemia. PGC1α and QPRT mRNA expression decreased progressively with ischemia severity until reaching a plateau at 15 minutes of ischemia for PGC1α (Fig. 1A), like plasma creatinine and urea, whereas QPRT (Fig. 1B) decreased is linear until 30 min of ischemia. PGC1α mRNA decrease is also inversely correlated to kidney dysfunction (p<0.001) (Figue 1C). PGC1α and QPRT expression recovery after AKI A 10-minute ischemia led to the development of Chronic Kidney Disease (CKD), as evidenced by an incomplete recuperation of the kidney function at 28 days. QPRT and PGC1α mRNA expression showed a progressive but incomplete recuperation during the transition from AKI to CKD compared to sham mice (Fig. 1D, E). uQ/T increases after ischemia and is restored during transition from AKI to CKD (Fig. 1F). Conclusion In addition to confirming the decrease of PGCα and QPRT mRNA expression during AKI, we show that it is correlated to the severity of the ischemic AKI. Furthermore, we describe the recovery during renal repair and transition to CKD
Degradation of herbicide atrazine in water by high voltage electrical discharge in comparison with Fenton oxidation and ultrasound treatments
No full textInternational audienceAtrazine, the most commonly used herbicide, has been reported to pollute the water environment and do harm to human health. It is thus urgent to find an efficient way to degrade atrazine. Although various advanced oxidation processes including high voltage electrical discharge (HVED) have been applied to degrade atrazine, the formation kinetics of its metabolites are still incomplete, and the detoxification of the degradation process remains to be clarified. Here, the degradation of atrazine by HVED was investigated, in comparison with traditional Fenton oxidation and ultrasound treatment. Nineteen metabolites of atrazine degradation were identified and quantified by high performance liquid chromatography coupled with high resolution mass spectrometry (HPLC-HRMS) techniques. Results show that HVED is more advantageous because of its high degradation rate for atrazine (89%), short processing time (1000 s, corresponding to 10 ms effective time), and the presence of the less toxic main metabolite hydroxyatrazine. Hydroxyl radicals (˙OH) play an important role in atrazine degradation. Adding ferrous ions (Fe2+) during HVED and ultrasound processes is beneficial for the degradation of atrazine, because of the ˙OH radicals released from hydrogen peroxide (H2O2). Based on the formation kinetics of atrazine degradation metabolites, detailed mechanisms of atrazine degradation pathways were proposed
Microwave-Assisted Synthesis of Pd Nanoparticles into Wood Block (Pd@wood) as Efficient Catalyst for 4-Nitrophenol and Cr(VI) Reduction
No full textInternational audiencePalladium (Pd) nanoparticle catalysis has attracted increasing attention due to its efficient catalytic activity and its wide application in environmental protection and chemical synthesis. In this work, Pd nanoparticles (about 71 nm) were synthesized in aqueous solution by microwave-assisted thermal synthesis and immobilized in beech wood blocks as Pd@wood catalysts. The wood blocks were first hydrothermally treated with 10% NaOH solution to improve the internal structure and increase their porosity, thereby providing favorable attachment sites for the formed Pd nanoparticles. The stable deposition of Pd nanoparticle clusters on the internal channels of the wood blocks can be clearly observed. In addition, the catalytic performance of the prepared Pd@wood was investigated through two model reactions: the reduction of 4-nitrophenol and Cr(VI). The Pd@wood catalyst showed 95.4 g −1 s −1 M −1 of normalized rate constant k norm and 2.03 min −1 of the TOF, respectively. Furthermore, Pd nanoparticles are integrated into the internal structure of wood blocks by microwaveassisted thermal synthesis, which is an effective method for wood functionalization. It benefits metal nanoparticle catalysis in the synthesis of fine chemicals as well as in industrial wastewater treatment
Minimum ignition energy of hydrogen-air mixtures at ambient and cryogenic temperatures
No full textInternational audienceThe ignition and combustion of hydrogen in air is considered more hazardous compared to other fuels due to the lower minimum ignition energy (MIE) and the wider flammability range. Spark discharge is the most common type of electrostatic ignition hazard. There is a need in validated safety engineering tools to accurately calculate MIE in a wide range of temperatures from atmospheric to cryogenic which are characteristic for hydrogen systems and infrastructure. Current MIE assessment methodologies rely on the availability of experimental data on quenching distance and/or laminar burning velocity and thus are mostly empirical correlations. This prevents their application beyond the limited number of experimental data, i.e. to arbitrary composition of the hydrogen-air mixture at arbitrary temperatures including cryogenic. This work aims at the development of a model able to accurately predict MIE for hydrogen-air mixtures with arbitrary initial composition and temperature. Cantera and Chemkin software are used to calculate the properties and unstretched laminar burning velocity of hydrogen-air mixtures. The flame thickness is found to well represent the critical flame kernel in the suggested model. The model is validated against experimental data on MIE for mixtures at ambient and cryogenic (down to 123 K) temperatures. Results show that the effect of flame stretch and preferential diffusion shall be considered to accurately predict MIE for lean hydrogen-air mixtures, which was not possible for previous models
Improving Semantic Similarity Measure Within a Recommender System Based-on RDF Graphs
No full textInternational audienceIn today's era of information explosion, more users are becoming more reliant upon recommender systems to have better advice, suggestions, or inspire them. The measure of the semantic relatedness or likeness between terms, words, or text data plays an important role in different applications dealing with textual data, as in a recommender system. Over the past few years, many ontologies have been developed and used as a form of structured representation of knowledge bases for information systems. The measure of semantic similarity from ontology has developed by several methods. In this paper, we propose and carry on an approach for the improvement of semantic similarity calculations within a recommender system based-on RDF graphs