Portail HAL IMT Mines Albi
Not a member yet
5440 research outputs found
Sort by
Catalyst-free synthesis of 5-hydroxymethylfurfural from fructose by extractive reaction in supercritical CO2 – subcritical H2O two-phase system
No full textInternational audienceAn extractive reaction configuration using supercritical carbon dioxide (scCO2) as the extracting solvent was tested for the production of 5-hydroxymethyl furfural (HMF) from a 5 wt% fructose aqueous feed. In this configuration, extraction of HMF by scCO2 prevents HMF degradation in the aqueous phase. Because of water co-extraction by scCO2, the volume of the reactional mixture was maintained by continuous injection of water. Reaction was operated in a 90 mL high pressure reactor, where an HMF maximum yield of 62.4% was achieved at 160 °C and 25 MPa, with a CO2 flow rate of 20 g.min-1 for 420 min. This is the first time that HMF is reportedly produced with such a yield by a catalyst-and organic solvent-free process. Besides, the separation efficiency reached 97.3% and the relative purity of HMF in the extract was 95.8 wt%. Therefore, this configuration avoids post reactional purification which is needed in conventional batch processes or in extractive reaction processes using organic solvents. Based on kinetic and thermodynamic studies, modeling of the extractive reaction process was developed to perform a sensitivity analysis for CO2 flow rate and extraction efficiency, upon the HMF yield. As an example, it was shown that for 800 min reaction duration, a CO2 flow rate of 100 g.min-1 or an extraction efficiency increase by a 10-fold factor could theoretically led to HMF yields of 73.0% and 73.7%, respectively
Co-optimization of a high temperature thermal storage as per its modeling accuracy
No full textInternational audienceCoupling energy networks becomes unavoidable in order to decarbonize human usages, increase global energy efficiency and ensure flexibility in so-called “multi-energy” network. In such a network, high temperature thermal energy storage (HTTES) can be a relevant solution when designed and managed in an optimal way. However, the precise modeling of its physical behavior requires complex models whose computational costs are not compatible with optimal control. A fortiori, a co-optimization approach requires to select a less precise but faster model. This article proposes to study the consequences of using a panel of such lighter models, in particular by discussing the modeling of losses. To do so, two business models will be discussed on a case study composed of a heat network linking a concentrated solar power (CSP) to thermal industrial load. When losses are not a consideration, the use of very simplistic models is sufficient to determine a good estimate of storage sizing. However, if losses are included, a proper co-optimization can only be achieved by using a metamodeling approach
Dynamical multi-parameter sizing of DDMRP buffers in finite capacity flow-shops
No full textInternational audienceThe DDMRP (Demand Driven Material Requirements Planning) methodology uses buffer stocks to (i) maintain a high level of service, (ii) stop the spread of uncertainty and (iii) adapt to market changes. According to theory, the size of these buffer stocks should be defined regularly. This sizing involves several parameters and policies to update them, but very little information is available on this subject. We aim to help practitioners choose sizing policies, while maximizing the performance of a given workshop. We have developed an experimental design to compare many combinations of flow-shops and bottleneck constraints, taken from industrial use-cases, and using discrete event simulation. The results show that (i) different degrees of dynamism are needed depending on the performance metric chosen by practitioners, (ii) completely dynamic control does not systematically lead to better performance, and (iii) contrary to what the existing literature on DDMRP suggests, varying buffer sizes may be less effective than fixed ones for an important part of use-cases
Understanding polymer nucleation by studying droplets crystallization in immiscible polymer blends
No full textInternational audienceThis Feature Article reviews our recent work on the nucleation and crystallization of finely dispersed semicrystalline polymeric droplets in immiscible matrices. Droplets dispersions can be used as a toolbox to investigate polymer nucleation. Studying the overall crystallization kinetics of the droplets can be a way to separate the contributions of nucleation and growth rates depending on the sample. To characterize the relative importance of nucleation versus growth, we have defined a dimensionless parameter named the “Turnbull number”. This number equals the ratio of the time needed for a crystal to grow inside the average droplet volume, divided by the required time for nucleation to occur in the same droplet. We show examples for Turnbull numbers close to unity, where the overall crystallization kinetics of the droplets is dominated by crystal growth, and a sigmoidal (second or third order) kinetics is obtained. On the other hand, when the overall crystallization of the droplets is controlled by nucleation, very low Turnbull numbers are obtained (e.g., 0.1–0.01) with concomitant first-order kinetics. We also show how the nucleation step in the droplets can be skipped through the strategic use of self-nucleation. In the case of double crystalline polymer blends, the self-nucleation of the matrix can be used to study the surface nucleation of the droplets at the interface. The role of interfacial roughness in promoting droplet nucleation is also addressed, together with the addition of heterogeneous nucleating agents. Considering all the body of information, we demonstrate that studying the overall kinetics of different droplet dispersions can contribute to the fundamental understanding of polymer nucleation
Errors in surface energy estimation of fibres for Liquid Composite Moulding processes and potential origins
No full textInternational audienceFor a better understanding of the behaviour of new eco-reinforcements, such as recycled carbon fibres, during Liquid Composite Moulding processes (LCM), fibres surface analysis and wetting properties are studied. However, this type of analysis, using the Owens and Wendt relation requires special procedures, specifically for estimation of the contact angle. Based on two tensiometric methods, carbon and basalt fibres with different sizing are characterised in first approach. The main contribution of this study is to evaluate the error in surface energy and its components determination associated to the measurement of an alleged equilibrium contact angle deriving from static or quasi-static data
Evaluation of the performance of Finite Element digital image correlation to characterize textile reinforcements
No full textInternational audienc
A numerical method to simulate the intra-tows mechanical behaviour of oxide/oxide CMCs: coupling of a lattice with a FE continuum
No full textInternational audienceOxide/oxide Ceramic Matrix Composites (CMCs) are used in many technical applications, including the manufacture of aircraft engine tailpipes [1]. It is essential to well understand the impact of the manufacturing process on the properties of such a material. Essentially, it is necessary to be able to anticipate the mechanical behaviour of the porous matrix but also of its bonding with the fibres.For this purpose, a multi-scale numerical method was developed to mechanically model the microstructure of a porous ceramic elaborated from submicron alumina particles and nanometric silica ones, based on its characteristics in terms of porosity and composition. This microstructure modelling by a lattice makes it possible to determine the mechanical behaviour of the matrix alone. It can also be directly coupled with a finite element model of alumina fibres.This numerical method is mainly based on the assumption that a network of silica bridges is formed between the alumina particles due to the decrease in silica viscosity and the effect of surface tension during the sintering of the green body [2]Firstly, a large cluster of independent particles, placed in contact with each other, is numerically built and a selected area of this cluster is retained. This subcluster is then subjected to an homothecy to consider the desired porosity and composition. In parallel, a series of silica bridges of different sizes are modelled using a finite element method and subjected to four mechanical loading modes (tension, shear, torsion, bending). These results are then compiled in the form of stiffness response surfaces as a function of the distance between the particles and of the involved volume of silica. A Representative Elementary Volume (REV) of the ceramics microstructure is then formed as a network of connectors whose positions and stiffnesses are determined by the two previous simulations.Once the REV has been formed, it can be used to determine the mechanical properties of the matrix via the application of modal analysis. It can also be linked to a finite element model of the alumina fibre in order to reconstruct a REV of a CMC and analyse it mechanically either via virtual tensile tests or modal analysis.Thus, this method makes it possible to determine the mechanical properties of an alumina/silica sintered material from its composition and porosity. In order to investigate the effect of manufacturing processes on the properties, further developments will focus on relating microstructure to sintering. References: [1] Ceramic Matrix Composites. Fiber Reinforced Ceramics and their Applications, Walter Krenkel, Wiley-VCH Verlag GmbH&Co. KGaA. 2008.[2] Review: liquid phase sintering, R. M. German et al. Journal of Materials Science, vol. 44, no. 1, pp. 1–39, Jan. 2009
DIAG Approach: Introducing the Cognitive Process Mining by an Ontology-Driven Approach to Diagnose and Explain Concept Drifts
No full textInternational audienceThe remarkable growth of process mining applications in care pathway monitoring is undeniable. One of the sub-emerging case studies is the use of patients’ location data in process mining analyses. While the streamlining of published works is focused on introducing process discovery algorithms, there is a necessity to address challenges beyond that. Literature analysis indicates that explainability, reasoning, and characterizing the root causes of process drifts in healthcare processes constitute an important but overlooked challenge. In addition, incorporating domain-specific knowledge into process discovery could be a significant contribution to process mining literature. Therefore, we mitigate the issue by introducing cognitive process mining through the DIAG approach, which consists of a meta-model and an algorithm. This approach enables reasoning and diagnosing in process mining through an ontology-driven framework. With DIAG, we modeled the healthcare semantics in a process mining application and diagnosed the causes of drifts in patients’ pathways. We performed an experiment in a hospital living lab to examine the effectiveness of our approach
A recommendation system for personalized daily life services to promote frailty prevention
No full textInternational audienceFrailty is a clinical syndrome that commonly occurs in older adults and characterizes an intermediate state between robust health and the loss of autonomy. As such, it is crucial to identify and evaluatefrailty to preserve the abilities of older adults and reduce the loss of their functional capabilities. This paper proposes a personalized service recommendation framework designed, to assist senior citizens in selectingappropriate services, to prevent frailty and improve autonomy. This framework is based on a multidimensional evaluation of the elderly person, considering both user’s status information and service-related data. It defines a four steps recommendation process, including personal characteristics identification, needs identification, service types identifi- cation, and service identification. To support this, both knowledge-based and rule-based approaches are defined and employed to generate personalized recommendations that align with the individual’s specific needs. The effectiveness of the proposal is evaluated using different representative scenarios, of which an example is described in detail that demonstrates the developed recommendation process, algorithms and relevant rules
Towards a Flow-Oriented Reference Model for Educational Organizations
No full textPart 2: Emerging Technologies Towards Circular and Resilient NetworksInternational audienceMore than ever, education is a global issue for human societies in our volatile, uncertain, complex and ambiguous world. Unfortunately, education systems often have a proven performance deficit. To change this situation, we assume in this paper that they could benefit from some technical inputs originally developed for other types of industrial systems and in particular those that contribute to supply chain improvement. Indeed, there are already plenty of tools and methodologies to diagnose, assess and improve industrial processes. Using this analogy, we will explore which levers can be used to improve educational systems and we present a new framework to assess them: the Educational Supply Chain Operational Reference inspired by the Supply Chain Operational Reference mode