HAL Université de Toulouse, et Toulouse INP
Not a member yet
159308 research outputs found
Sort by
Has “Ecological Economics” betrayed its roots? Revealing its state, internal tensions and evolution through a multi-level and multi-scale bibliometric assessment
No full textInternational audienceThis paper uses a mixed methods approach to investigate the development of Ecological Economics (EE), combining a critical literature review with quantitative bibliometric analysis. It maps the intellectual trajectory of the field, its thematic shifts and its authorship networks from inception to the present time. Three phases are identified: (i) 1983-1997: the founding period; (ii) 1998-2011: the establishment of ecosystem services' valuation; and (iii) 2012-2024: the contemporary EE, characterized by internal tensions and increasing integration into mainstream environmental initiatives. Our findings reveal a global surge in interest in the field, accompanied by an expansion in the intellectual scope and research themes, as well as a broader geographic distribution of authors. This is associated with the emergence of numerous new journals, with Ecological Economics maintaining a central, albeit declining, position. Our research shows that while EE began as a critique of neoclassical economics, during its establishment phase it embraced market-based solutions and monetary valuation. However, contemporary trends indicate a critical reassessment of these approaches, with a shift towards exploring issues such as climate justice, degrowth and social metabolism. This dynamic evolution highlights the ongoing importance of EE as a platform for transdisciplinary research that challenges conventional economic thinking
TRACT : A Transformer and Statistical Framework for Anomaly Detection in Multivariate Non-stationary Time Series
No full textInternational audiencePhysiological and inertial signals analysis supports numerous healthcare applications, including disease detection, rehabilitation, and treatment. Advancements in signal processing enable the representation of most stationary and non-stationary phenomena using mathematical expressions. These representations provide valuable insights and help in identifying distinctive patterns of interest. In this paper, we propose TRACT, a deep learning and statistical framework designed to detect anomalies in non-stationary environments. It comprises two main components, a transformer-based reconstruction model that captures signal patterns through multi-resolution attention, extending the standard attention mechanism in transformer architecture. During inference, reconstruction errors are computed by comparing observed signals with their reconstructed versions. Statistical modeling is applied to these errors, with parameters estimated directly from the data. TRACT adapts to varying data rates across datasets without imposing strict distribution assumptions, resulting in enhanced robustness and accuracy in anomaly detection for multivariate non-stationary time series. We evaluate TRACT on 12 real-world multivariate time series datasets from diverse domains, demonstrating its performance in anomaly detection tasks with various constraints and its ability to provide early warnings for anomalous events
Positive to negative photoconductance switching in plasmonic gold nanoparticle networks
No full textInternational audienceThe elaboration of versatile materials in which electrical conduction is tuned by light irradiation is of paramount relevance to such diverse applications as photodetectors, photodiodes, solar cells or light sensors. Although positive photoconductance is increasingly common, systems with negative photoconductance are scarcely reported. In this work, a switching from positive to negative photoconductance is observed upon simple annealing of well-organized networks of gold nanoparticles stabilized by dodecanethiols. The photoconductance properties are investigated experimentally using impedance spectroscopy. The measured Nyquist plots and resonance curves of the impedance are analyzed in terms of equivalent electrical circuits consisting in parallel resistance, capacitance and photoconductance. The positive and negative photo-current conversion efficiencies of the nanoparticles network are estimated k PPC = 389 ± 40 nS.W -1 .cm 2 and k NPC = -241 ± 40 nS.W -1 .cm 2 , respectively. With the aid of density functional theory calculations, the origin of the photoconductance is discussed, at the nanoscale level, in terms of changes of molecular conformation. Such molecular effects assist charge carrier tunneling between first neighbor nanoparticles, and favor the formation of traps introduced by the annealing of the sample. The present work contributes to the understanding of plasmo-electronic properties of hybrid molecule-nanoparticle self-assembled nano-structures
Influence of tow-preg composition on microstructure and mechanical behaviour of oxide/oxide CMCs
No full textInternational audienceCeramic matrix composites could soon be used in an ever-increasing number of high-temperature applications. This study explores the potential of a new tow-preg process for the manufacturing of oxide-based ceramic matrix composites. With these techniques, the choice of impregnation slurries is crucial for the quality of the materials. This study focuses on the development of several grades of alumina matrices CMCs. The impact of the slurry composition on the CMCs microstructure and mechanical behaviour is evaluated. The benefits of a minimum amount of boehmite, a gel-forming agent which is a precursor of alumina, on inter-ply cohesion have been demonstrated
Howard T. Odum’s contribution to Marxian ecological thought
No full textInternational audienceThis article explores Howard T. Odum's contributions to Marxian ecological thought through an analysis of his energy systems theory, its intersections with Marxian critiques of capitalism, and its relevance for contemporary debates on global ecological crisis and inequalities. We first describe how Odum's system ecology and environmental accounting can provide complementary insights to inform a biophysical critique of capitalism, by revealing the hidden energy flows that underpin economic processes. In particular, Odum's concept of emergy and his critique of conventional economics anticipated central debates within ecological economics, notably the need to ground economic analysis in thermodynamic realities and to account for nature's essential, yet often disregarded, contributions to human well-being. Then, we explore the similarities between Odum's emergy valuation of real wealth and Marx labour theory of value. Lastly, we connect Odum's analysis to Marxian ecological theories, particularly the metabolic rift thesis, illustrating how Odum's quantification of energy hierarchies complements Marx's historical-materialist critique of capital's exploitation of socio-economic systems. We argue that integrating Odum's biophysical perspective with Marx's critique of capitalism's exploitative logic provides critical insights into capitalism's systemic reliance on thermodynamic inefficiency and global resource extraction. This convergence between systems ecology and historical materialism can offer valuable tools to rethink sustainability within social and biophysical planetary boundaries
Structure and speed of hydrogen–air triple flames
No full textInternational audienceTriples flames play an important role in flame stabilization processes, especially for hydrogen flames. Being able to predict the speed sT of a triple flame is crucial in many hydrogen combustor designs as it controls flame stabilization and flashback. These speeds highly depend on the mixing field between fuel and air in front of the flame, usually measured by the thickness of the profile of mixture fraction z: Δ=1/max(∂z/∂X). In the limit of high Δ values, triple flames propagate at a speed sT given by theory. When Δ decreases, sTis modified and for conventional fuels, theory indicates that it decreases.The present work focuses on the structure and speed of non-propagating H2-air triple flames when Δis varied. It uses a DNS of hydrogen–air combustion with detailed chemistry where the inlet velocity is set via a control loop ensuring that a steady triple flame is stabilized in a stream at a given value of Δ so that a (sT vs Δ) curve can be built. Results confirm that, even for hydrogen, sT can be predicted for large Δ, by the standard expression sT=(ρu/ρb)1/2sL if expansion ratio ρu/ρb and laminar burning velocity sL are evaluated at their maximum value, respectively near stoichiometry and at equivalence ratio 1.6 for H2. When Δis decreased, the triple flame speed decreases significantly but extinction is not observed.Novelty and significance statement: Few detailed studies on hydrogen triple flames exist, even though this information is of practical use for combustor design. This paper uses high-fidelity DNS to show that the theoretical expression for triple flame speeds sT=(ρu/ρb)1/2sLcan work for hydrogen flames with simple and logical adjustments. It can then be used at any pressure by experimentalists and engineers to predict stabilization of hydrogen - air diffusion flames.Moreover, this work studies triple flames stabilized in a steady mixing field, contrary to previous works focusing on the ignition and propagation of hydrogen triple flames. In these conditions, it assesses quantitatively, for the first time, how the gradient of mixture fraction decreases triple H2/air flame speeds. However, quenching is not observed. A scaling is obtained and extended to engine-relevant pressures
Non-isolated DC/DC boost converters design, volume and efficiency estimation for Fuel Cell system in railway application
No full textInternational audienceThis paper presents an analytical method for designing a nonisolated DC/DC converter compatible with multiple fuel cells for railway applications. The converter must operate over a wide range while being optimized for both efficiency and volume. To maximize fuel cell lifetime, the current ripple must remain very low, which complicates the efficiencyvolume trade-off. Two main levers are explored to achieve significant improvements over previous designs: the use of new components with an improved frequency-loss trade-off, and the adoption of topologies with better inherent performance. This paper details the design of several nonisolated boost converter architectures that meet the operating requirements of different fuel cells while providing an optimal balance between ripple, efficiency, and volume.</div
Predictive validity of cognitive abilities for air traffic controllers selection among proficient scientific candidates
No full textInternational audienceMost air traffic controllers (ATCO) are selected after high school graduation, partly based on cognitive abilities. On the contrary, French ATCO are selected after an intensive scientific post-secondary program without undergoing cognitive assessment. This study examines the potential incremental validity of cognitive predictors in ATCO training for such proficient applicants. Initial and unit training results were available for 414 and 277 ATCO students, respectively. All students took cognitive ability tests, encompassing visuospatial processing, quantitative knowledge, and work samples. We compared a baseline regression model with academic performance as sole predictor to an extended model with the addition of the cognitive abilities composite score. The R2 of the extended model increased significantly, by 0.04 (p < .001) and 0.06 (p < .001), respectively, for predicting practical initial and unit training outcome. In conclusion, even for scientific proficient candidates, cognitive assessment can enhance the success rate of ATCO training
A Gaussian correlation inequality for plurisubharmonic functions
No full textInternational audienceA positive correlation inequality is established for circular-invariant plurisubharmonic functions, with respect to complex Gaussian measures. The main ingredients of the proofs are the Ornstein-Uhlenbeck semigroup, and another natural semigroup associated to the Gaussian ∂-Laplacian