No description available.

Unmanned Aerial Systems (UAS) have become increasingly popular recently for surveying and mapping because of their efficiency in acquiring remotely sensed data in a short amount of time and the low cost associated with them. They are used to generate digital elevation models (DEM) derived from aerial photography for various purposes such as the documentation of cultural heritage sites, archaeological surveying or earthwork volume calculations. This thesis investigates the possible effects different file formats may have on the quality of elevation models. In this thesis, an UAS survey was simulated using a digital camera to produce six DEMs based on JPEG, TIFF and RAW format in Agisoft Photoscan by taking two sets of images of a city model, in different light conditions. Furthermore, a reference DEM was produced in Geomagic Studio using data from a Leica Nova MS50 Multistation. The DEMs were then compared in Geomagic Control. The results from the 3D comparison in Geomagic Control show that the standard deviation of all elevation models is 4 mm with the exception of the elevation model derived from raw-edited images taken with lighting, which has a standard deviation of nearly 6 mm. Also, all of the models have an average deviation of 0.4 mm or less. The significant deviations in all DEMs occur in areas where the multistation lacked vision of certain objects of the city model such as walls, or on the edges of the analysed area. Additionally, the georeferencing results from Photoscan show that the DEMs based on normal light condition images have slightly lower georeferencing errors than the DEMs with lighting. It has been concluded that it is difficult to say whether file formats have any noticeably effect on the uncertainty of digital elevation models.

The main study objective was to develop robust processing and analysis techniques to facilitate the use of small-footprint lidar data for estimating forest biophysical parameters measuring individual trees identifiable on the three-dimensional lidar surface. This study derived the digital terrain model from lidar data using an iterative slope-based algorithm and developed processing methods for directly measuring tree height, crown diameter, and stand density. The lidar system used for this study recorded up to four returns per pulse, with an average footprint of 0.65 m and an average distance between laser shots of 0.7 m. The lidar data set was acquired over deciduous, coniferous, and mixed stands of varying age classes and settings typical of the southeastern United States (37° 25' N, 78° 41' W). Lidar processing techniques for identifying and measuring individual trees included data fusion with multispectral optical data and local filtering with both square and circular windows of variable size. The window size was based on canopy height and forest type. The crown diameter was calculated as the average of two values measured along two perpendicular directions from the location of each tree top, by fitting a four-degree polynomial on both profiles. The ground-truth plot design followed the U.S. National Forest Inventory and Analysis (FIA) field data layout. The lidar-derived tree measurements were used with regression models and cross-validation to estimate plot level field inventory data, including volume, basal area, and biomass. FIA subplots of 0.017 ha each were pooled together in two categories, deciduous trees and pines. For the pine plots, lidar measurements explained 97% of the variance associated with the mean height of dominant trees. For deciduous plots, regression models explained 79% of the mean height variance for dominant trees. Results for estimating crown diameter were similar for both pines and deciduous trees, with R2 values of 0.62-0.63 for the dominant trees. R2 values for estimating biomass were 0.82 for pines (RMSE 29 Mg/ha) and 0.32 for deciduous (RMSE 44 Mg/ha). Overall, plot level tree height and crown diameter calculated from individual tree lidar measurements were particularly important in contributing to model fit and prediction of forest volume and biomass. / Ph. D.

This study focuses on the Triangulated Irregular Network (TIN) model for representing terrain. A TIN is a vector representation of terrain composed of vertices, lines, and triangular polygons. The elevation values are stored in the vertices. TINs are used in Geographic Information Systems (GIS) for terrain analyses. In particular 1 compare TIN extraction methods. A TIN extraction method is a method in which points are selected from a raster Digital Elevation Model (DEM) in order to construct a TIN. Evaluations of TIN extraction methods have been performed in the past. However, none have compared the TIN extraction methods with terrain texture. This study compared and analyzed the four most popular TIN extraction methods: VIP, Hierarchy, Drop Heuristic, and LATTICETIN. Comparison of the TIN extraction methods was based on the 90th percentile of the absolute error, the absolute error being the absolute difference between TIN elevation and DEM elevation at each point. The results indicated that LATTICETIN yielded the lowest 90th percentile errors for all types of terrain textures and Drop Heuristic was second best in most instances. The Hierarchy method was superior to VIP when less than 12% of the original DEM points were selected. When 12% of the original DEM points were selected, VIP outperformed the Hierarchy method. A correlation was found among TIN accuracy and terrain texture. There was a direct relationship between the two variables. / Master of Science

The status of whitebark pine (Pinus albicaulis), a foundation and keystone species and a pioneer establisher at alpine treeline, is threatened by the invasive and exotic fungal pathogen (Cronartium ribicola) that causes white pine blister rust in five-needled pines. Originally thought to be limited to moderate environments, the disease is now found extensively throughout colder and dryer regions east of the Continental Divide, including alpine treeline. My research objective was to determine how blister rust infection of treeline whitebark pine varies across Glacier National Park. I present findings from field sampling conducted in July 2008 in Glacier National Park, Montana. Thirty plots were randomly placed at 6 different treeline study sites on the eastern slopes of the Continental Divide. Vegetative and geomorphic characteristics, along with presence/absence and level of blister rust intensity, were detailed within each plot. Vegetation measurements included conifer composition, tree island dimensions and windward growth patterns, evidence and intensity of blister rust, as well as shelter type. Field-measured topographic characteristics included elevation, aspect, and slope. In addition, high resolution GPS-derived DEMs were created at each plot in order to model the land surface and calculate detailed environmental variables in a GIS. Environmental and blister rust intensity variables were used to determine spatial correlates of blister rust infection at treeline. The resulting blister rust prediction model (P < 0.001, F(4,25) = 6.79, R2 = 0.52, Adjusted R2 = 0.44) suggests that areas exhibiting increased wind speed, northwest facing slopes, high flow accumulation rates, and close proximity to perennial streams have a higher likelihood of blister rust intensity, specifically total canker density. Results of this research may contribute to the understanding of the dynamics of this disease, and prove useful in whitebark ecosystem management and conservation. / Master of Science

For food industry processes packaging machines with high throughput are required and one way to improve the overall machine efficiency is to increase its working speed. However, testing of prototypes is time and cost expensive. Therefore, simulation is used to evaluate the process and adapt it. Optimization can help to find better machine designs by using simulations to evaluate one solution. This work uses the Discrete Element Method to model a vertical tubular bag machine for packaging basmati rice. The Covariance Matrix Adaption Evolution Strategy optimizes the simulation model and results in a significant machine speedup. This work is a guidance to adapt this method for similar problems.

Ce travail de thèse concerne l’application d’une nouvelle technologie de préparation de catalyseurs propres Co/Al2O3 pour la synthèse de Fischer Tropsch (FT). Les matériaux sont préparés à l’aide d’un procédé de revêtement à sec dans un mélangeur à haut cisaillement le «Picomix». Les conditions opératoires, les supports en alumine, la taille des cristallites de cobalt et la teneur en charge de cobalt ont été optimisés pour augmenter l’activité catalytique du Co/Al2O3. De même, une modélisation DEM a été effectuée pour décrire le comportement des poudres dans le mélangeur. Les résultats expérimentaux montrent une très bonne adhésion des nanoparticules de Co3O4 à la surface des particules de Al2O3 pour une vitesse de rotation élevée pendant une courte période. Les particules γ-Al2O3 traitées thermiquement présentent une résistance mécanique améliorée des catalyseurs, mais une activité catalytique relativement faible. La taille des cristallites de Co3O4 a diminué de 64nm à 11nm après broyage dans un broyeur à boulets planétaire à 600 rpm/min pendant 40h. Enfin, le catalyseur optimal a été obtenu à 5000 rpm/min pendant 5min avec 5% en poids de cobalt. Le catalyseur obtenu présente des résultats catalytiques : conversion élevée en CO (37%) , sélectivité élevée en hydrocarbures en C5+ (75%) et faible sélectivité en CH4 (13%) à 250oC. La modélisation numérique par DEM a révélé que les paramètres liés aux conditions opératoires, la géométrie du dispositif et aux propriétés intrinsèques des particules avaient un impact sur le comportement des particules et la qualité de l’enrobage des matériaux / This thesis concerns the application of a novel preparation technology for Co/Al2O3 clean catalysts applied in Fischer Tropsch (FT) synthesis. The catalysts were prepared using a dry coating process in a high shear mixer “Picomix”. The operating conditions, alumina supports, cobalt crystallite size and cobalt loading content were optimized to increase the catalytic activity of Co/Al2O3 catalysts. Besides, DEM modeling was performed to describe the behavior or powders in the mixer. Experimental results showed a very good adhesion of the nano Co3O4 particles on the surface of Al2O3 particles after processed in the mixer under a high rotational speed and a short time. The heat-treated Al2O3γ particles presented enhanced mechanical strength of catalysts, however, exhibited relatively low catalytic activity. The Co3O4 crystallite size decreased from 64 nm to 11 nm after milling in planetary ball mill under 600 rpm for 40 h. Finally, the optimal catalyst was prepared by mixing milled-Co3O4 and Al2O3γ particles in “Picomix” under 5000 rpm for 5 min with 5 wt.% of Co. The obtained catalyst presented high CO conversion (37 %), high C5+ hydrocarbons selectivity (75 %) and low CH4 selectivity (13 %) in FT synthesis reaction at 250 oC. The DEM mumerical modeling revealed that the parameters related to operating condition, device geometry, and particle intrinsic properties had an impact on particle behavior and coating quality of materials

Les matériaux granulaires exhibe un spectre très large de propriétés constitutives, le long de chemins de chargement très divers. Développer des modèles constitutifs permettant de reproduire ces caractéristiques a demeuré un réel challenge scientifique au cours des dernières décennies. A cet égard, les approches multi-échelles constituent aujourd’hui une voie très prometteuse. Elles permettent de relier les propriétés macroscopiques à celles observées à l’échelle microscopique.Une investigation a été menée sur la base de simulations numériques discrètes (DEM)d’essais biaxiaux, afin d’identifier les caractéristiques micro-structurelles du matériau granulaire, la manière dont elles évoluent au cours d’un chemin de chargement, et le rôle qu’elles jouent dans l’émergence du comportement macroscopique. A l’échelle mésoscopique,le réseau de transmission de force (chaines de force) et les cellules définies parles vecteurs branches (meso-cycles) apparaissent jouer un rôle de première importance.Les meso-cycles, construits à partir du réseau de contact de l’assemblage, peuvent être caractérisés en fonction du nombre de cotés qu’ils contiennent (topologie). Leur influence sur le comportement volumique de l’échantillon est en effet étroitement liée à ce nombre de contact. En outre, leur interaction avec les chaines de force est également fortement dépendante de leur topologie. Ainsi, les cycles contenant 3 cotés (L3) participent activement à la stabilisation des chaines de force, alors que les cycles contenants au moins6 cotés (L6+) contribuent essentiellement au comportement dilatant de l’échantillon et à l’effondrement des chaines de force. Enfin, l’existence d’une méso-structure unique à l’état critique, au sein de la bande de cisaillement (rupture localisée) ou au sein de l’échantillon (rupture diffuse), est clairement démontrée.viii Sur la base de ces résultats, un modèle constitutif a été développé à partir du modèle H-directionnel (Nicot and Darve, 2011b). La structure du matériau granulaire est décrite par un assemblage d’hexagones (modélisant les cycles L6), orientés dans toutes les directions de l’espace. A partir d’opérations d’homogénéisation, les contraintes et les déformations incrémentales peuvent être reliées à l’échelle de l’assemblage, donnant lieu à un modèle de comportement dont la performance a pu être testée le long de chemins de chargements variés. / Granular materials exhibit a wide spectrum of constitutive features under various loading paths. Developing constitutive models which succeed to characterize these features has been challenging scientists for decades. A promising direction of achieving this can be the multi-scale approach. Through which the constitutive model is formulated in the way that relating material's macroscopic properties to their micro-scale essences, namely, corresponding micro-structure properties.To better characterize the micro-structure and formulate the relation between different scales, a comprehensive investigation have been carried out on the basis of numerical biaxial tests using 2D discrete element method (DEM), in order to ascertain the micro-structure characteristics of the granular material, the way they evolve versus loading and their corresponding rules in the macroscopic behaviors. In a meso-scale, intermediate between the single contact scale and the macro-scale, the force transmission network (force-chains) and area element enclosed by contacts branches (meso-loops) are highlighted in terms of their significant influences on material's macro-scale behavior. Meso-loops herein are tessellated from the whole area of the granular assembly by the contact branch network, and are subsequently categorized according to their side number.The development of meso-loops is observed to be intimately related to material's volumetric evolution, especially the plastic part. Then, the interaction between force-chains and meso-loops and its significance to the global volumetric behavior are revealed. Otherwise, in critical state, an identical meso-structure is found in the failure area of both localized and diffuse failure mode. Meso-loops with 3 sides (L3) appear to be indispensable for the force-chain stability, meanwhile, meso-loops with more than or equal to 6 sides (L6+) contribute much to the volume expansion and accelerate the force-chain buckling. The interplay between L3 and L6+ largely embody, or are representative of, the various mechanical performance of the granular material.A constitutive model has been developed by modifying the H-directional model. In this model, individual hexagons, representatives of L6+, construct the fabric as distributing along different directions in the space. Then multi-scale approach is then used to relate macroscopic properties from local ones, and eventually, to give the constitutive relation. This model is then validated in different loading paths, and eventually proved satisfying.

Le formage de matériaux multiphasés comprend des mécanismes complexes en lien avec la rhéologie,la morphologie et la topologie des phases.Du point de vue numérique,la modélisation de ces phénomènes en résolvant les équations aux dérivées partielles (EDP) décrivant le comportement continu des phases n'est pas trivial.En effet,de nombreuses discontinuités associées aux phases se déplacent et peuvent interagir.Ces phénomènes peuvent être conceptuellement déclicats à intégrer au modèlecontinu et coûteux en termes de calcul.Dans cette thèse,la méthode des éléments discrets (DEM) est utilisée pour modéliser phénoménologiquement les déformations finies inélastiques dans les multi-matériaux.Les lois d'interactions attractive-répulsive sont imposées à des particules fictives,dont les ré-arrangements collectifs modélisent les déformations irréversibles de milieux continus.Le comportement numérique des empilements de particules est choisi pourreproduire des traits caractéristiques de la viscoplasticité parfaite etisochore:contrainte d'écoulement,sensibilité à la vitesse de déformation,conservation du volume.Les résultats d'essais de compression de bi-matériaux simples,simulés avec la DEM,sont comparés à la méthode des éléments finis (FEM) et sont en bon accord.Le modèle est entendu pour pouvoir supporter des sollicitations de traction.Une méthode de détection de contacts et d'auto-contacts d'objets physiques estproposée,basée sur l'approximation locale des surfaces libres.Les capacités de la méthodologie globale sont testées sur des mésostructurescomplexes,obtenues par tomographie aux rayons X.La compression à chaud d'un composite métallique dense est modélisée.La co-déformation peut être observées à l'échelle spatiale des phases.Deux cas de matériaux ``poreux'' sont considérés.Premièrement la simulation de la compression puis traction d'alliagesd'aluminium présentant des pores.Ces pores proviennent du coulage du matériau,leur fermeture et ré-ouverture mécanique est modélisée,y compris la coalescence à grande déformation.Deuxièmement la simulation de la compression de mousse métallique de faibledensité.Typiquement utilisée dans le but d'absorber de l'énergie mécanique,la compression jusqu'à densification provoque de nombreuses interactions entreles bras de matière. / Forming of multiphase materials involves complex mechanisms linked with therheology,morphology and topology of the phases.From a numerical point of view,modeling such phenomena by solving the partial differential equation (PDE) system accounting for thecontinuous behavior of the phases can be challenging.The description of the motion and the interaction of numerous discontinuities,associated with the phases,can be conceptually delicate and computationally costly.In this PhD,the discrete element method (DEM) is used to phenomenologically model finite inelastic strain inmulti-materials.This framework,natively suited for discrete phenomena,allows a flexible handling of morphological and topological changes.Ad hoc attractive-repulsive interaction laws are designed betweenfictitious particles,collectively rearranging to model irreversible strain in continuous media.The numerical behavior of a packing of particles can be tuned to mimic keyfeatures of isochoric perfect viscoplasticity:flow stress, strain rate sensitivity, volume conservation.The results for compression tests of simple bi-material configurations,simulated with the DEM,are compared to the finite element method (FEM) and show good agreement.The model is extended to cope with tensile loads.A method for the detection of contact and self-contact events of physicalobjects is proposed,based on a local approximation of the free surfaces.The potential of the general methodology is tested on complex mesostructuresobtained by X-ray tomography.The high temperature compression of a dense metallic composite is modeled.The co-deformation can be observed at the length scale of the phases.Two cases of ``porous'' material are considered.Firstly,the simulation of the compression and the tension of aluminum alloys with poresis investigated.These pores stem from the casting of the material,their closure and re-opening is modeled,including the potential coalescence occurring at large strain.Secondly,the compression of a metallic foam,with low relative density,is modeled.Typically used in energy absorption applications,the compression up to densification involves numerous interactions between thearms.

La dégradation géométrique de la voie ballastée sous circulation commerciale nécessite des opérations de maintenance fréquentes et onéreuses. La caractérisation du comportement des pro-cédés de maintenance comme le bourrage, la stabilisation dynamique, est nécessaire pour proposer des améliorations en terme de méthode, paramétrage pour augmenter la pérennité des travaux. La simulation numérique d'une portion de voie soumise à un bourrage ou une stabilisation dynamique permet de comprendre les phénomènes physiques mis en jeu dans le ballast. Toutefois, la complexité numérique de ce problème concernant l'étude de systèmes à très grand nombre de grains et en temps de sollicitation long, demande donc une attention particulière pour une résolution à moindre coût. L'objectif de cette thèse est de développer un outil de calcul numérique performant qui permet de réaliser des calculs dédiés à ce grand problème granulaire moins consommateur en temps. La méthodologie utilisée ici se base sur l'approche Non Smooth Contact Dynamic s(NSCD) avec une discrétisation par Éléments Discrets (DEM). Dans ce cadre, une méthode de dé-composition de domaine (DDM) alliée à une parallélisation adaptée en environnement à mémoire partagée utilisant OpenMP sont appliquées pour améliorer l'efficacité de la simulation numérique. / The track deterioration rate is strongly influenced by the ballast behaviour under commercial traffic. In order to restore the initial track geometry, different maintenance processes are performed, like tamping, dynamic stabilisation. A better understanding of the ballast behaviour under these operations on a portion of railway track is a key to optimize the process, to limit degradationand to propose some concept for a better homogeneous compaction. The numerical simulation isdeveloped here to investigate the mechanical behaviour of ballast. However, the main difficultiesof this research action concerns the size of the granular system simulation increasing both in termof number of grains and of process duration. The purpose of this thesis is to develop an efficient numerical tool allows to realize faster computations devoted to large-scale granular samples. In this framework, the Non-Smooth Contact Dynamics (NSCD) of three-dimensional Discrete ElementMethod (DEM) simulations, improved by Domain Decomposition Method (DDM) and processedwith the Shared Memory parallel technique (using OpenMP) has been applied to study the ballast media mechanics.