京都大学 / 新制・課程博士 / 博士(エネルギー科学) / 甲第24717号 / エネ博第460号 / 新制||エネ||86(附属図書館) / 京都大学大学院エネルギー科学研究科エネルギー応用科学専攻 / (主査)教授 藤本 仁, 教授 白井 康之, 教授 濵 孝之 / 学位規則第4条第1項該当 / Doctor of Energy Science / Kyoto University / DFAM

Sweden’s real estate portfolio has a major renovation requirement. A large part of the renovations that need to be carried out are pipe replacement projects. These are major projects and not all housing associations are aware of the extent financial consequences it brings. Therefore, planning in the economy throughout the project are important. This thesis aims at seeing what affects a production estimate to differ from the actual final cost.ÅF infrastructure division thought that their calculations of the estimated cost for production too often differs to the actual final cost. At a first meeting with ÅF, it appeared that there has been discussion and thought for solutions to the problem for some time. The thesis is based on 3 issues that will provide answers on how to make a better production estimate.In order to answer the questions, project managers have been interviewed and literature has been studied. The assigned reference objects have been analyzed and compared in tables. This thesis has resulted in the insight of that the project managers at ÅF are relatively good at estimating the original cost of what the contractor produces for the pipe replacement projects. After the analysis, recommendations have been presented that can be developed and implemented in the ÅF approach to calculate their estimated cost of production even better. By calculating the costs that are not included in the original cost for the contractors in a more methodical manner, based on experience, the production cost can be optimized.The work has been going on for a limited time with delimitations and a selection of projects. The results presented in the report should be applied with caution and awareness about the differences in the projects.

No description available.

No description available.

With the evolution of GPU performance, the interest of using AI for all kinds of purposes has risen. Companies today put a great amount of resources to find new ways of using AI to increase the value of their products or automating processes. An area in the wood industry where AI is widely used and studied is in defect detection. In this thesis, the combination of using AI and hyperspectral images is studied and evaluated in the case of segmenting defects in hardwood with a U- Net network structure. The performance is compared to another known method usually used when dealing with high-dimensional data: PLS-DA. This thesis also compares the use of RGB image data in combination with AI, to further analyze the usefulness that the hyperspectral data provide. The results showed signs of improvement when using hyperspectral images com- pared to RGB images when detecting blue stain and red heartwood defects. De- tection of the defects rot and knots did however show no sign of improvements. Due to the annotations being more accurate in the RGB data, the results from the hyperspectral data-fed networks would suggest that blue stain and red heartwood could be of interest regarding further investigation. Computational performance is shown to vary across the different reduction meth- ods, and the results from this thesis provides some insight that might aid in the reasoning regarding how to choose an appropriate reduction method.

L’ascomicete Guignardia bidwellii, agente causale del marciume nero della vite, è un patogeno economicamente importante in alcuni areali viticoli. La conoscenza, disponibile sul marciume nero dell’uva, è stata recuperata dalla letteratura, analizzata e sintetizzata per sviluppare un modello meccanicistico del ciclo di vita del patogeno, guidata dalle variabili meteorologiche e dalla fenologia della vite, e basata sull'analisi dei sistemi. Il modello è stato poi valutato per la sua capacità di rappresentare il sistema reale e la sua utilità per la comprensione di epidemie di marciume nero su foglie e grappoli in un vigneto del Nord Italia, nel 2013 al 2015. Successivamente, le mancanze di conoscenza sono state analizzate, studiate e quindi colmate attraverso specifici esperimenti. In un primo passo, le dinamiche dell’inoculo primario e dei modelli di dispersione (di entrambi ascospore e conidi) da mummie svernate sono state studiate in un vigneto sperimentale per tre anni. In un secondo passo, l'effetto della temperatura e dell'umidità sulla formazione di picnidi di G. bidwellii e la successiva estrusione di cirri, nelle lesioni su foglia, la produzione e la germinazione dei conidi (inoculo secondario), e la lunghezza del periodo di latenza sono stati studiati sia in condizioni di campo che in ambiente controllato. In un terzo passo, sono stati condotti studi in ambiente controllato per studiare la produzione di conidi di G. bidwellii sulle lesioni di foglie, influenzata da lavaggi ripetuti e alternando periodi di secco ed umido. Il modello epidemiologico sviluppato in questa tesi può essere utilizzata da viticoltori come strumento predittivo per la pianificazione di trattamenti fungicidi nei vigneti. / The ascomycete Guignardia bidwellii, causal agent of black-rot on grapevines, is an economically important pathogen in some viticultural areas. The available knowledge on black-rot of grape was retrieved from literature, analyzed, and synthesized to develop a mechanistic model of the life cycle of the pathogen, driven by weather and vine phenology, and based on the systems analysis. The model was then evaluated for its ability to represent the real system and its usefulness for understanding black-rot epidemics on leaves and bunches in a vineyard of north Italy, in 2013 to 2015. Thereafter, weaknesses in our knowledge were analysed and studied through specific experiments. In a first step, dynamics of primary inoculum and dispersal patterns (both ascospores and conidia) from overwintered grape mummies were investigated in an experimental vineyard during three years. In a second step, the effect of temperature and humidity on the formation of G. bidwellii pycnidia and the extrusion of cirri in grape leaf lesions, production and germination of conidia (secondary inoculum), and the length of the latency period were studied under both environmental and controlled conditions. In a third step, environmental-controlled studies were conducted to investigate the production course of G. bidwellii conidia on grape leaf lesions as influenced by repeated washing events and alternate dry and wet periods. The model developed in this thesis can be used by vinegrowers as a predictive tool for scheduling fungicide sprays in the vineyards.

No description available.

Dans la forêt de pin maritime (Pinus pinaster) des Landes de Gascogne (sud-ouest de France), la mortalité des pins causée par le champignon pourridié Armillaria ostoyae (Basidiomycète) a augmenté au cours des 30 dernières années. Les premiers cas de cette maladie ont été signalés quelques années après un changement majeur dans l'utilisation des terres, qui a eu lieu dans cette région suite au remplacement des landes et marais d'origine par une forêt plantée et gérée da façon intensive. Notre objectif était de comprendre les facteurs à l'origine de cette maladie émergente. Pour cela, nous avons étudié la distribution spatiale des dommages causés par le pathogène en relation avec des facteurs historiques, estimé la variabilité des traits fongiques liés au parasitisme et saprophytisme, et étudié l'histoire démographique d'A. ostoyae. La répartition actuelle de la mortalité induite par A. ostoyae est apparue dépendre de la présence des forêts préexistantes, ce qui suggère qu'A. ostoyae était fréquent dans ces zones forestières anciennes, qui ont agi comme un réservoir pour la colonisation des forêts plantées récentes. La production de rhizomorphes était significativement corrélée avec la virulence, suggérant que ce trait joue un rôle important dans le stade parasitaire d'A. ostoyae. Aucune relation significative entre le parasitisme et saprophytisme n'a été détectée, suggérant une absence de compromis évolutif entre ces traits. Enfin, le meilleur scénario démographique pour expliquer la structure de la population d'A. ostoyae dans la forêt des Landes est un scénario en deux étapes : il y aurait eu d'abord une diminution puis une expansion de la population fongique, qui semblait suivre la dynamique de la population d'hôtes. Le temps de génération d’A. ostoyae a été estimé entre 10 et 20 ans. / In the maritime pine (Pinus pinaster) forest of the Landes de Gascogne (south-western France), pine mortality due to the root rot fungus Armillaria ostoyae (Basidiomycete) has been increasing over the last 30 years. The first cases of this disease were reported a few years after a major change in land use which occurred in this region following the replacement of original moors by an intensively managed planted forest. Our aim was to understand the factors driving this disease emergence. For this, we investigated the spatial distribution of pathogen damage related to historical factors, estimated the variation in fungal traits related to parasitism and saprophytism and investigated the demographic history of A. ostoyae. The current distribution of A. ostoyae mortality appeared depending on the pre-existing forests, suggesting that A. ostoyae was commonly distributed in pre-existing forest areas which acted as a reservoir for the colonization of recent planted forests. The rhizomorphs production was significantly correlated with virulence, suggesting that this trait plays an important role in the parasitic stage of A. ostoyae, but no significant relationship between parasitism and saprophytism components was detected, which may suggest that there is no trade-off between these traits. Finally, the best demographic scenario to explain A. ostoyae population structure in the Landes forest is a two step scenario: there was first a decrease and then an expansion in the fungal population, which appeared to follow the dynamics of the host population. The generation time of A. ostoyae was estimated between 10 and 20 years.

Isolates of Monilinia fructicola (Wint.) Honey obtained from stone fruit orchards in Hawkes Bay, North Island and from Californian fruit exported to New Zealand, were tested for resistance to methyl benzimidazole carbamate (MBC). Resistant isolates from the North Island had EC₅₀ values of >30,000, and most isolates from the imported fruit had of values approximately 1.5 mg a.i./l carbendazim. Sensitive isolates failed to grow on 1 mg a.i./l carbendazim. A detached peach shoot system was used in controlled conditions for estimation of values for incubation period, latent period and rate of spore production on flowers (cv Glohaven). The same variables and the rate of colonisation of host tissue were measured on fruit (cv Fantasia) in controlled conditions. An inoculum density of 1x10⁴ spore/flower or fruit greatly increased fitness in vivo compared to an inoculum density of 1x10² spore/flower (fruit). Isolates varied considerably, but there was no consistent relationship between the degrees of resistance and fitness. This was in contrast to earlier studies with dicarboximide resistant strains of M. fructicola. The survival in the field of 10 isolates resistant or sensitive to MBC or dicarboximide fungicides on twig cankers and mummified fruit was compared. The ability to produce conidia on twig cankers inoculated in late spring 1989 was maintained by all sensitive and MBC resistant isolates for at least 1 year. The production of conidia on mummified fruit inoculated in February 1990 decreased after 2-3 months in the field but some conidia were still produced on all fruit in the following spring. Dicarboximide resistant isolates produced less conidia than either the MBC resistant and the sensitive isolates. The pathogenicity and fitness of all isolates were similar to the original values after survival for 1 year. A technique was developed to produce apothecia reliably from inoculated peach (cv Black Boy) and nectarine (cv Fantasia) fruit in controlled conditions in the laboratory. The fruit were inoculated with resistant or sensitive isolates, or combinations, and were incubated for 8 weeks at 25°C (±1°C) with 12 hours photoperiod of fluorescent light (Sylvania 2x65 W, daylight) to produce mummified fruit. The fruit were then buried in moist autoclaved peat moss for 10 weeks at 25°C (±1°C) in the dark to form stromata. These fruit were then hydrated with running tap-water (total hardness (CaCO₃) = 47 g/m³ and conductivity at 20°C = 12.7 mS/m) for 72 hours. The hydrated mummified fruit were placed in moist peat moss and were incubated for 13-14 weeks at 8°C (±0.5°C) in the dark. At the end of this period, stipe initials were visible. Differentiation of stipe initials into mature apothecia occurred within 15-20 days after transfer to 12°C (±2 °C) with a 12 hour photoperiod of fluorescent and incandescent light. All isolates produced apothecia when treated in this way. A technique for isolation of ascospore sets in linear arrangement was developed for tetrad analysis of the inheritance of resistance. At least 3 hours of fluorescent and incandescent light at 12°C (±2°C) was essential to allow ascospore ejection from individual asci taken from apothecia previously maintained in a 12 hour photoperiod at 12°C (±1°C). A water film on the surface of water agar was necessary to hold a set of ejected ascospores in linear sequence. Single ascospores were obtained in sequence with the aid of a micromanipulator. Genetic analysis of MBC resistant isolates was carried out on ascospores derived from apothecia produced in the laboratory. Analysis of ascospore sets in linear arrangement and ascospore populations indicated that resistance to >30,000 mg a.i./l carbendazim (high-resistant) is governed by a single major gene and is affected by gene conversion mechanisms. Crossing over was frequent, suggesting that recombination of resistance with other characters, such as pathogenicity and fitness, may occur readily. The segregation ratio (1:1) from most resistant isolates revealed that heterokaryons containing both resistant and sensitive alleles were common in resistant populations and that resistance is dominant. Allozyme analysis of ascospore progeny through electrophoresis revealed a narrow genetic base of M. fructicola in New Zealand. The technique for reliable apothecial production in controlled conditions developed in this study provided an important step for the determination of the biology of M. fructicola strains resistant to MBC fungicides, and the complexity of its life cycle. Genetic heterogeneity in field populations can be conserved in one isolate through heterokaryosis, thus providing for adaptability of the pathogen to the changing environmental conditions. Knowledge on genetic variability, overwintering ability, pathogenicity and fitness factors may be useful for future management strategies of stone fruit brown rot. Special emphasis should be made in particular to prevent primary infection on blossoms, which would delay the establishment of recombinant strains of M. fructicola and the onset of brown rot epidemics.

Ce travail est une application de la théorie des problèmes inverses à la reconstruction 3-D des artères coronaires avec compensation du mouvement à partir de données Rot-X. Dans un premier temps nous étudions le problème inverse en dimension finie et infinie. En dimension finie nous nous concentrons sur la modélisation du problème inverse en tomographie en définissant la notion de base de voxels et de matrice de projection, en vue de pouvoir se ramener à une formulation matricielle du problème inverse. Nous étudions aussi la notion de tomographie dynamique et les problèmes qui lui sont liés. Notre formulation discrète permet grâce aux bases de voxels d'inclure n'importe quelle déformation de support étant un difféorphisme dans le problème inverse matriciel, dès lors que cette déformation est connue a priori. Dans le dernier chapitre, nous présentons une méthode d'estimation du mouvement utilisant les projections Rot-X synchronisées par rapport à l'ECG, basée sur un modèle déformable 3-D des artères coronaires. Le mouvement est modélisé par des fonctions B-splines.<br />Une fois le mouvement estimé, la reconstruction tomographique à un instant de référence est effectuée par une optimisation aux moindres-carrés qui inclut le mouvement ainsi qu'un terme de pénalité qui favorise les valeurs d'intensités fortes pour les voxels au voisinage de la ligne centrale 3-D, et les faibles valeurs pour les autres. Cette méthode a été testée sur des données simulées basées sur des lignes centrales 3-D préalablement extraites de données MSCT.