Portail HAL IMT Mines Albi
Not a member yet
5440 research outputs found
Sort by
Inspection of mechanical assemblies based on 3D Deep Learning approaches
No full textInternational audienceOur research work is being carried out within the framework of the joint research laboratory ”Inspection 4.0” between IMT Mines Albi/ICA and the company DIOTA specialized in the development of numerical tools for Industry 4.0. In this work, we are focused on conformity control of complex aeronautical mechanical assemblies, typically an aircraft engine at the end or in the middle of the assembly process. A 3D scanner carried by a robot arm provides acquisitions of 3D point clouds which are further processed by deep classification networks. Computer Aided Design (CAD) model of the mechanical assembly to be inspected is available, which is an important asset of our approach. Our deep learning models are trained on synthetic and simulated data, generated from the CAD models. Several networks are trained and evaluated and results on real clouds are presented
Risk-aware business process management using multi-view modeling: method and tool
No full textCorrection to: Risk‑aware business process management using multi‑view modeling: method and tool : https://doi.org/10.1007/s00766-021-00356-2International audienceRisk-aware Business Process Management (R-BPM) has been addressed in research since more than a decade. However, the integration of the two independent research streams is still ongoing with a lack of research focusing on the conceptual modeling perspective. Such an integration results in an increased meta-model complexity and a higher entry barrier for modelers in creating conceptual models and for addressees of the models in comprehending them. Multi-view modeling can reduce this complexity by providing multiple interdependent viewpoints that, all together, represent a complex system. Each viewpoint only covers those concepts that are necessary to separate the different concerns of stakeholders. However, adopting multi-view modeling discloses a number of challenges particularly related to managing consistency which is threatened by semantic and syntactic overlaps between the viewpoints. Moreover, usability and efficiency of multi-view modeling have never been systematically evaluated. This paper reports on the conceptualization, implementation, and empirical evaluation of e-BPRIM, a multi-view modeling extension of the Business Process-Risk Management-Integrated Method (BPRIM). The findings of our research contribute to theory by showing, that multi-view modeling outperforms diagram-oriented modeling by means of usability and efficiency of modeling, and quality of models. Moreover, the developed modeling tool is openly available, allowing its adoption and use in R-BPM practice. Eventually, the detailed presentation of the conceptualization serves as a blueprint for other researchers aiming to harness multi-view modeling
Review on the catalytic tri-reforming of methane - Part I: Impact of operating conditions, catalyst deactivation and regeneration
No full textInternational audienceTri-reforming of methane (TRM) allows the production of syngas with a low environmental impact, an optimal energetic consumption, and a modular H2/CO molar ratio. However, despite a large number of publications devoted to TRM reaction, this process is still in its infancy and faces technical issues due to the catalyst deactivation by the formation of solid carbon, thermal sintering, vapor-solid reactions, and poisoning. Moreover, TRM reaction is also highly dependent on the operational conditions. This article provides a critical analysis of the last achievements on the TRM reaction. First, the thermodynamic, kinetic and mechanism aspects are presented and commented. Then, the impact of the operational conditions is analyzed. Finally, the main reasons of catalyst deactivation and the associated methods for catalyst regeneration are discussed. In parallel, catalytic efficiency is tentatively linked to physico-chemical properties of the catalyst, and recommendations are proposed for the future work on TRM process
Impact of Atmosphere on Recovered Carbon Fibers From Poly Ether Ether Ketone (PEEK) Based Composites During Thermoconversion
No full textInternational audienceCarbon Fibers Reinforced Polymer composites (CFRP) are high added value materials used in many manufactured products. Especially in aeronautics, thermosetting resins tend to be replaced by heat-resistant thermoplastic polymers. The aim of the work is to evaluate suitable operating conditions of thermoconversion for the recovery of carbon fibers from Poly Ether Ether Ketone (PEEK)/carbon fibers composites. Micro and pilot scale tests have been performed in nitrogen, wet nitrogen, air, and wet air. Thermogravimetric analysis of PEEK/carbon fiber composites showed a moderate decomposition onset temperature of the composite at 515 °C and 510 °C in dry nitrogen and dry air respectively. The oxidative atmosphere did not significantly impact this temperature since the first mass loss was not atmosphere dependent. However, after the first PEEK degradation reaction, the nature of the atmosphere appeared as a great issue. Total mass loss was significantly improved with temperature (full oxidation of polymer and carbon fibers up to 800 °C) and with reaction time using air. Indeed, at pilot scale, the matrix was fully degraded in air atmosphere at 550 °C for 1 h while only 42% and 46% was reached in nitrogen and steam/nitrogen respectively. Comparison of thermogravimetric data between wet and dry atmospheres revealed that steam plays a thermal retardant role leading to some differences on the matrix degradation and on the surface morphology of the carbon fiber at pilot scale. Air treatments at pilot scale induced a reduction of fiber diameters (< − 3.7%) but the tensile strengths of recovered carbon fibers were preserved with an average retention of mechanical properties of 81%-85%. It has been concluded that recycling of PEEK/carbon composite required an oxidant to split up carbon fibers from the PEEK matrix
REVIEW OF THE UTILIZATION OF WASTE BIOMATERIAL RESOURCES OF SELECTED FOOD CROPS: RICE, CASSAVA, COCOA, SORGHUM AND OIL PALM
No full textPoor waste management is one of the major causes of environmental degradation. In a developing nation like Nigeria with huge agro potential, it is not unlikely that the bulk of our wastes are from biomaterial sources on farms. These supposed wastes, however, can be of immense Economic and Environmental benefits if re-used. This article therefore, presents a review of simple processing techniques to convert these biomaterials (wastes) into value added products (materials) and sustainable energy sources.The Federal Ministry of Agriculture and Rural Development (Nigeria) in its Minna 2011 Staff Retreat, unveiled an Agricultural Transformational Agenda in which rice, cassava, cocoa, oil palm, and sorghum were identified as target commodity value chain with which to drive the nation's economy. One of the transformational policies in view of this development is the financing of value chains for the processing of these target crops. However, along the value chain, Rice husks, Cassava peels, Cocoa pods, Oil Palm Empty Fruit Bunches (EFB), and Sorghum bagasse, each in millions of metric tons per annum are either inefficiently used, underutilized, or totally wasted. In this paper therefore, the author presents a review of simple processing techniques to convert these biomaterials (wastes) into value added products (materials) and sustainable energy sources. Furthermore, the economic cost and environmental benefits of the effective biomaterial utilization of these waste resources is drawn-out
Wear Behavior and Microstructure of Thermally Sprayed NiCrBSiFeC and Composite NiCrBSiFeC-WC(Co) Coatings
No full textInternational audienceIn this work, a study was carried out on the friction and wear behavior of flame thermal sprayed NiCrBSiFeC-WC(Co) composite and NiCrBSiFeCcoatings subjected to severe wear conditions. For this purpose, flame remelted samples were tested in reciprocating wear conditions based on a cylinder-on-flat configuration. The wear assessment of the coatings was achieved using scanning electron microscopy (SEM)and 3D optical profilometry. The microstructure and the mechanical properties of the coatings were investigated using SEM, EDS and XRDtechniquesas along withindentation tests. The tribological behavior of the substrate and the coatings was successfully studied thankstowear tests conductedon an adapted multi test apparatus. The results show that both NiCrBSiFeC and composite coatings induced a significant increase in the steel substrate hardness and wear resistance due to the formation of precipitates with high hardness well dispersed withinanultra-crystalline structure.Besides, adding WC(Co) to NiCrBSiFeC leads to a composite coating with hardnessandwear resistance further improved. In return, it increases the coefficient of friction (COF) and the coatings’ roughness. Furthermore, improvements in the surface hardness, the roughnessandthe coating-substrate adhesionwere attained after the remelting process for both NiCrBSiFeC and NiCrBSiFeC-WC(Co) coatings. Wear tracks investigations indicated that reciprocating dry sliding based on cylinder-on-flat test configuration promote several wear mechanisms that may occur simultaneously
Initial microstructure influence on Ti-Al-Mo-V alloy’s superplastic deformation behavior and deformation mechanisms
No full textInternational audienceSuperplastic forming is an effective way to manufacture complex-shaped parts of titanium-based alloys. This paper studies the influence of the initial microstructure and its strain-induced evolution on superplastic deformation behavior and the formability of a titanium-based alloy. Two types of Ti-Al-V-Mo alloy samples having a different fraction of recrystallized grains before the start of the superplastic deformation were studied. The deformation behavior, including strain hardening and strain rate sensitivity of the flow stress, was analyzed in a temperature range of 775°C to 900°C and a strain rate range of 10-5 to 10-2 s-1. Strain-induced changes of the microstructure within the bulk of the samples and on the surface of the pre-polished samples were studied during superplastic deformation with a constant strain rate. The dynamic recrystallization and dynamic grain growth in the volume of the samples and the multiple slip bands on the samples’ surface were revealed after superplastic deformation. The grain structure evolution and slip bands localization depended on the samples’ initial microstructure. The results showed that the samples with an increased fraction of recrystallized grains exhibited better superplasticity and higher quality of the formed parts with a more uniform thickness distribution across the section than the samples with a lower initial recrystallized fraction
Contrôle du paramètre Decoupled Lead Time (DLT) et des temps deréponse dans un système piloté en DDMRP
No full textJour de soutenance fictif car non disponible, mois et année corrects.The Demand Driven Material Requirements Planning (DDMRP) method is a recent production management method that integrates various industrial engineering principles such as push flow, pull flow, the use of inventory buffers, and the mechanism of reorder point methods. By combining these elements, the authors of DDMRP claim that their method is the solution to the limitations of traditional production management methods.One of the most important parameters of DDMRP is the Decoupled Lead Time (DLT). It is defined as the time that production orders are allowed to be completed. The time actually taken to produce an order is called flow time. Thus, if the flow time of an order is greater than the DLT, the order is completed late and there is a risk of stock-outs. Moreover, this parameter is used to size the stock buffers that generate the production orders. Consequently, the DLT is also linked to the company's inventory levels. It is therefore interesting to reduce this parameter as much as possible, to avoid unnecessary storage costs.This is how the compromise around the DLT dimensioning appears, which is the main issue of this thesis: on the one hand, we try to dimension the DLT large enough so that the flow times are as often as possible lower than or equal to the DLT, and on the other hand, we try to reduce as much as possible this parameter to reduce the costs related to the stocks. Our objective is therefore to propose methods allowing either to control the flow times so that they are lower than the DLT, or to adjust the DLT accordingly.After having surveyed the literature on the different topics and performed a critical review of the literature to define three different research objectives, we use an experimental protocol based on discrete event simulation to achieve these objectives, as it allows to model complex environments subject to high variability. First, we propose a control loop that dynamically adjusts the DLT parameter according to the real times observed in the workshop. We then use simulation to create decision support charts. These tools allow to choose the best possible capacity option for a given forecasted demand. The idea is to control the flow times by controlling the load rate of bottleneck resources. Finally, we couple DDMRP to the Constant Work-In-Process (ConWIP) method to create a more robust model for generating production orders than the classical DDMRP method. Our model takes into account the variability of demand and production to adapt the manufacturing batch sizes, thus smoothing theload and making the shop floor more fluid.The main results of our research are the demonstration that when production orders use the same bottleneck resource, their flow times are approximately the same. Therefore, the DLT parameter of the product buffers must be the same for all. Moreover, the dynamic adjustment of this parameter makes it possible to reduce the stocks, in particular at the time of a bad initial parameter setting. Secondly, it seems more sensible to try to control the flow times rather than adjusting the DLT.This control is achieved by controlling the load rate of bottleneck resources. This is why wepropose visual decision support tools to identify the best capacity adjustment options. These tools allow to correlate the resource load rate, the flow time distribution, the DLT parameter, and the workshop and customer service rates. Finally, the coupling of DDMRP and ConWIP methods allows for the automatic adjustment of batch sizes according to the state of the shop floor, making the shop floor more fluid. This coupling makes the system more robust and offers better performance than the traditional DDMRP method, in terms of WIP volume, flow time distribution, and service rate.La méthode Demand Driven Material Requirements Planning (DDMRP) est une méthode récente de gestion de la production, intégrant différents principes du génie industriel tels que le flux poussé, le flux tiré, l’utilisation de stocks tampons, ou encore le mécanisme des méthodes à point de commande. En combinant ces éléments, les auteurs du DDMRP affirment que leur méthode est la solution aux limites des méthodes traditionnelles de gestion de la production. Un des paramètres les plus importants du DDMRP est le Decoupled Lead Time (DLT). Il est défini comme le temps que l’on alloue aux ordres de fabrication pour être complétés. Le temps réellement mis pour fabriquer un ordre est appelé temps de réponse. Ainsi, si le temps de réponse d’un ordre est plus grand que le DLT, l’ordre est alors terminé en retard et l’on risque des ruptures de stocks.De plus, ce paramètre sert à dimensionner les stocks tampons qui génèrent les ordres de fabrication. Par conséquent, le DLT est également relié aux niveaux de stocks de l’entreprise. Il est donc intéressant de réduire au maximum ce paramètre, pour éviter des coûts de stockage inutiles. C’est ainsi qu’apparaît le compromis autour du dimensionnement du DLT, enjeu principal de cette thèse : d’un côté on cherche à dimensionner le DLT assez grand pour que les temps de réponse y soient le plus souvent possible inférieurs ou égaux, et de l’autre on cherche à diminuer au maximum ce paramètre pour réduire les coûts liés aux stocks. Le DLT est alors relié à deux indicateurs de performances clés en gestion de la production : le taux de service client et le niveau des stocks.Notre objectif général est donc d’améliorer les performances des ateliers de production pilotés par la méthode DDMRP, en proposant des outils de maîtrise des temps de réponse ou d’ajustement du paramètre DLT.Après avoir fait le tour des publications sur les différents sujets abordés et réalisé une revue de la littérature permettant de définir trois objectifs spécifiques de recherche différents, nous utilisons un protocole expérimental basé sur la simulation à évènements discrets pour atteindre ces objectifs, car elle permet de modéliser des environnements complexes soumis à de fortes variabilités. Nous proposons dans un premier temps une boucle de régulation permettant d’ajusterdynamiquement le paramètre DLT en fonction des temps de réponse observés dans l’atelier. Nous utilisons par la suite la simulation pour créer des abaques d’aide à la décision. Ces outils permettent de choisir la meilleure option possible de capacité pour une demande prévisionnelle donnée. L’idée étant de maîtriser les temps de réponse en contrôlant le taux de charge des ressources goulots. Enfin, nous couplons le DDMRP à la méthode Constant Work-In-Process (ConWIP) pour créer un modèle de génération des ordres de fabrication plus robuste que la méthode DDMRP classique. Notre modèle prend en compte la variabilité de la demande et celle de la production pour adapter les tailles de lot de fabrication, et ainsi lisser la charge et fluidifier l’atelier.Les principaux résultats de nos travaux de recherche sont la démonstration que lorsque des ordres de fabrication utilisent la même ressource goulot, leurs temps de réponse sont sensiblement les mêmes. Par conséquent, le paramètre DLT des stocks tampons de produits doit être le même pour tous. De plus, l’ajustement dynamique de ce paramètre permet de réduire les niveaux de stocks, notamment lors d’un mauvais paramétrage initial. Ensuite, il paraît plus judicieux de chercher à maîtriser les temps de réponse plutôt que d’ajuster le DLT. Cette maîtrise passe par le contrôle du taux de charge des ressources goulots. C’est pourquoi nous proposons des outils visuels d’aide à la décision permettant d’identifier les meilleures options d’ajustement de la capacité. Ces outils permettent de corréler le taux de charge des ressources, la distribution des temps de réponse, leparamètre DLT, et les taux de service atelier et client. Enfin, le couplage des méthodes DDMRP et ConWIP permet d’ajuster automatiquement la taille des lots de fabrication en fonction de l’état de l’atelier, fluidifiant ce dernier. Ce couplage rend plus robuste le système et offre de meilleures performances que la méthode DDMRP classique, en termes de volume d’en-cours, de distribution des temps de réponse, et de taux de service
Engine Oil- Crankshaft Interaction Fem Modelling of an Air-Cooled Diesel Engine under Dynamic Severe Functioning Conditions
No full textInternational audienceTaking into account the interaction between the engine oil and the crankshaft to model crankshaft thermomechanical behavior under dynamic loading is very important. In particular, when the crankshaft is working in severe conditions. This paper deal with an air cooled direct injection-type engine crankshaft thermomechanical FEM modelling account for engine oil-cranks half interaction in severe working conditions. As case of application we consider the diesel engine Deutz F8L413. The model takes into account 2 forced convectives heat flux: engine oil and crankcase air. The severe mechanical and thermal characteristics of engine are experimentally measured on a bench test equipped with a hydraulic brake. The temperature distribution inside the crankshaft was computed using the measured temperature as boundary conditions. The most thermo-mechanical stressed zones of the crankshaft have been determined. The fatigue resistance of the crankshaft under thermo-mechanical conditions was examined using Dang-Van multi-axial fatigue criteria. To prove our model efficiency, we have compared crankshaft damage in service to the numerical simulation results. It was found the breakage occurred in an area where the numerical simulations give the highest stresses
A Latent Heat Storage System for Low-Temperature Applications: From Materials Selection to Prototype Performances
No full textInternational audienceThe industrial sector is increasingly obliged to reduce its energy consumption and greenhouse gases emissions to contribute to the world organizations’ targets in energy transition. An energy efficiency solution lies in the development of thermal energy storage systems, which are notably lacking in the low-temperature range (50–85 °C), for applications such as district heating or low-temperature waste heat recovery. This work aims to bring a latent heat storage solution from material selection to prototype evaluation. The first part of this paper is dedicated to the characterization and aging of a phase change material selected from a screening of the literature (fatty acid mixture mainly composed by stearic and palmitic acid). Then, this material is encapsulated and tested in a prototype whose performances are evaluated under various operating conditions. Finally, a numerical model validated by the experimental results is used to explore the influence of a wider range of operating conditions, dimensioning choices, and material conductivity improvements