La recherche de moyens alternatifs pour la propulsion d'un aéronef est primordiale tant la dépendance au pétrole est forte. Le travail proposé dans cette thèse s'inscrit dans le besoin de soutenir les concepteurs de systèmes de propulsion innovants utilisant l'électricité. En partant du constat qu'il est primordial de s'appuyer sur l'analyse de l'aéronef dans sa globalité, nous proposons une approche basée sur les modèles, faisant appel aux modèles d’ingénierie système, aux modèles comportementaux basés sur la physique et à la simulation numérique, et collaborative, car la conception implique de nombreux métiers.L'objectif de cette approche est la création d'un produit virtuel qui est un modèle global multidisciplinaire exécutable du produit pour en faciliter la conception. Une méthodologie est alors construite pour se concentrer sur la relation entre le produit virtuel, ses modèles constitutifs et leur obtention. Le fil directeur de la méthodologie correspond à une recherche d'informations, au moyen de l'analyse des interactions et impacts multidisciplinaires qui apparaissent dans le système, puis l'application de cette information pour la construction d'un modèle d'intention qui permet la requête de modèles comportementaux auprès d'experts. C'est finalement le lien manquant entre la conception globale conduite par l'ingénierie système et la conception basée sur la physique du monde réel qui est implicitement traité dans ces travaux. Pour réaliser l'ensemble de la méthodologie, un nouveau rôle a été défini, le rôle d'architecte de simulation. Cette thèse présente la méthodologie de manière théorique, incluant ses rôles et concepts, puis cette dernière est démontrée sur un cas d'étude correspondant à l'hybridation d'un drone de type hélicoptère. / The research of alternative aircraft propulsion system is mandatory because oil dependence is too strong. The work proposed in this thesis is oriented to support electric based innovative propulsion system designers. Considering that it is important to consider entire aircraft analysis, we propose a model based, with systems engineering models, physics-based behavioral models and numerical simulation, and collaborative, because design require numerous business expertise. The objective of this approach is to build a virtual product, which means a global multidisciplinary executable model of the product under design in order to facilitate its design. A methodology is then developed, focused on the relation between the virtual product, its constitutive models and their acquisition. The methodology director wire corresponds to information research, with multidisciplinary interactions and impacts in the system, and then the application of those pieces of information to build a model of intention which allows requesting a behavioral model from experts. Finally, it is the missing link between global design driven with systems engineering and real physics based design which is implicitly at stake. To perform the methodology, a new role has been defined, the simulation architect. This thesis presents theoretically the methodology, including roles and concepts, and then this methodology is demonstrated on a helicopter based drone study-case.

La recherche de moyens alternatifs pour la propulsion d'un aéronef est primordiale tant la dépendance au pétrole est forte. Le travail proposé dans cette thèse s'inscrit dans le besoin de soutenir les concepteurs de systèmes de propulsion innovants utilisant l'électricité. En partant du constat qu'il est primordial de s'appuyer sur l'analyse de l'aéronef dans sa globalité, nous proposons une approche basée sur les modèles, faisant appel aux modèles d’ingénierie système, aux modèles comportementaux basés sur la physique et à la simulation numérique, et collaborative, car la conception implique de nombreux métiers.L'objectif de cette approche est la création d'un produit virtuel qui est un modèle global multidisciplinaire exécutable du produit pour en faciliter la conception. Une méthodologie est alors construite pour se concentrer sur la relation entre le produit virtuel, ses modèles constitutifs et leur obtention. Le fil directeur de la méthodologie correspond à une recherche d'informations, au moyen de l'analyse des interactions et impacts multidisciplinaires qui apparaissent dans le système, puis l'application de cette information pour la construction d'un modèle d'intention qui permet la requête de modèles comportementaux auprès d'experts. C'est finalement le lien manquant entre la conception globale conduite par l'ingénierie système et la conception basée sur la physique du monde réel qui est implicitement traité dans ces travaux. Pour réaliser l'ensemble de la méthodologie, un nouveau rôle a été défini, le rôle d'architecte de simulation. Cette thèse présente la méthodologie de manière théorique, incluant ses rôles et concepts, puis cette dernière est démontrée sur un cas d'étude correspondant à l'hybridation d'un drone de type hélicoptère. / The research of alternative aircraft propulsion system is mandatory because oil dependence is too strong. The work proposed in this thesis is oriented to support electric based innovative propulsion system designers. Considering that it is important to consider entire aircraft analysis, we propose a model based, with systems engineering models, physics-based behavioral models and numerical simulation, and collaborative, because design require numerous business expertise. The objective of this approach is to build a virtual product, which means a global multidisciplinary executable model of the product under design in order to facilitate its design. A methodology is then developed, focused on the relation between the virtual product, its constitutive models and their acquisition. The methodology director wire corresponds to information research, with multidisciplinary interactions and impacts in the system, and then the application of those pieces of information to build a model of intention which allows requesting a behavioral model from experts. Finally, it is the missing link between global design driven with systems engineering and real physics based design which is implicitly at stake. To perform the methodology, a new role has been defined, the simulation architect. This thesis presents theoretically the methodology, including roles and concepts, and then this methodology is demonstrated on a helicopter based drone study-case.

Over the last 150 years the natural nighttime environment has been drastically altered by the proliferation of artificial light. The amount of artificial light at night is on the increase, and there is a current trend to replace older lighting with more energy efficient types such as light emitting diodes (LEDs) or ceramic metal halide; in Cornwall, UK, there has been a relatively recent replacement of the street lighting, from low pressure sodium to ceramic metal halide. Alongside the increasing amount of artificial nighttime light, recent research has highlighted declines in macro moth numbers. Given the well-known ‘flight-to-light’ behaviour of moths, and the negative effects this behaviour can have, alongside other known and potential ways in which nighttime light can affect moths, the increasing amount of artificial light in the environment is a suspected contributor to the declines. It is particularly important to understand how modern lighting technologies will impact upon moths, as different spectra of light are known to vary in terms of how attractive they are. As a means to determine the potential impact of different street lighting types on moths, particularly the ceramic metal halide lighting rolled out in Cornwall, UK, we compared the attractiveness to macro moths, of a number of increasingly used, energy efficient, street lighting types. We found that shorter wavelength metal halide lighting attracted significantly more individuals and species of moth than longer wavelength high pressure sodium lighting. In a second experiment, we also found ceramic metal halide lighting to be more attractive to macro moths than LED lighting. Reduced emissions of short wavelength UV light was deemed the likely reason behind the fewer macro moths attracted to the high pressure sodium and LED lighting. Interestingly, we also found striking differences in the relative attractiveness of the different lighting types to different moth groups. The metal halide lighting attracted significantly more Noctuidae than high pressure sodium lighting, whereas both high pressure sodium and metal halide lighting were equally attractive to Geometridae. Understanding accurately the extent to which different groups of moth are attracted to different wavelengths of light could be useful in determining the impact of artificial light on moth populations. In addition to impacting moths through attraction, artificial light has the potential to alter the day length as perceived by organisms, which at mid- to high latitudes is utilised by certain species as an abiotic cue to ensure the coincidence of development with favourable environmental conditions. Due to a paucity of knowledge on how raised ambient nighttime light levels affect moths and the trophic levels with which they interact, we carried out analyses into the impact of nighttime light on the winter moth and its host plant oak; a well-studied model system, where synchrony between moth egg hatch and oak budburst is important for the moth’s survival. Firstly we carried out an analysis looking at the relationship between the amount of nighttime light and the date of oak budburst. Spatially referenced budburst dates were matched with satellite imagery of nighttime lighting and average spring temperature data, and the relationship between the variables was analysed. Model predictions suggested that oak budburst occurs earlier in brighter areas. In addition, the predicted advance of budburst in brighter areas was still apparent when analysing only the data points that fell outside of large urban areas, where the urban heat island effect is likely reduced. The findings suggested that artificial nighttime light may be causing an advance in oak budburst. To follow up the spatial analysis we carried out a field experiment. We used light cages that simulated various nighttime lighting scenarios to test whether oak budburst and winter moth egg hatch were affected by low intensity light at night. In contrast to the spatial analysis, there was no significant relationship found between light treatment and the phenology of either oak budburst or winter moth egg hatch. However, there was a suggestion in the data that the higher buds of the oak saplings emerged earlier in the yellow light treatment, highlighting the need for further research into the potential impact of artificial nighttime light on phenology and species interactions. In conclusion, the findings of this research project provide information useful to those seeking ecologically sensitive lighting solutions, and also highlight a potential tool to assist in determining whether light at night is a causative factor behind apparent moth declines. In addition, they suggest that artificial light at night may be affecting the phenology of an ecological system at a national scale. Finally, this research project has highlighted the complexity of the ecological impacts of artificial light at night, and also a need for further research.