Les Stations de Transfert d’Energie par Pompage (STEP) sont des structures importantes pour stabiliser le réseau électrique.Les technologies à vitesse variable peuvent améliorer l'efficacité et la souplesse d’utilisation des STEPs. Les technologies actuelles telles que les machines alimentées en fréquence variable, ou les Machines Asynchrone à Double Alimentations (MADA) présentent des inconvénients. Dans ce travail, nous nous concentrons sur une structure non conventionnelle de machine à vitesse variable : la Machine à Induction à Double Alimentation sans Balais (BDFM). Les objectifs sont de mieux comprendre le fonctionnement de cette machine pour la dimensionner, l’optimiser et finalement la comparer aux structures existantes.Après un examen des techniques de dimensionnement et d'optimisation des machines classiques, des approches similaires sont étudiées pour les BDFM. Deux méthodes différentes pour des simulations éléments finis plus rapides de la BDFM sont présentées : une première, la plus rapide mais qui ne considère pas le phénomène de saturation et une seconde basée sur le principe des simulations magnétoharmoniques. Une analyse harmonique minutieuse combinée à une comparaison de cas en charges (couplage-croisés) entre des simulations FE et les résultats obtenus avec des schémas équivalents obtenus dans la littérature conduit à une modification du circuit équivalent de la BDFM. Une nouvelle méthode pour déterminer les paramètres de ce circuit grâce à des essais électriques est ensuite présentée. La détermination des paramètres à partir de la géométrie est également abordée pour aboutir à un modèle semi-analytique dérivable. Un tel modèle, couplé à un algorithme d'optimisation du 1er ordre pourrait être extrêmement puissant lors du dimensionnement d'une BDFM. Le potentiel d'une telle approche est montré dans ce travail grâce à l'optimisation d'une machine à induction. / Pumped-hydro storage plants (PSP) are important assets to stabilize electric grids.Variable speed technologies can improve the cycle efficiency and the power adjustability of PSPs. Current technologies such as fully-fed machines or Doubly-Fed Induction Machines (DFIM) have drawbacks. In this work, the focus will be on an unconventional design of variable speed machine: the Brushless Doubly-Fed Induction Machine (BDFM). The objectives are to better understand the working principles of this machine so as to size it, optimize it, and compare it to other types of variable speed machines.Following a review of sizing and optimization techniques for conventional machines, similar approaches are investigated for BDFMs. Two different methods for faster Finite-Element (FE) simulations of the BDFM are presented: a fast one without saturation considerations and another one based on the principle of magneto-harmonic simulations. A careful harmonic analysis combined with a comparison of cross-coupling tests between FE simulations and results of equivalent circuit found in the literature will lead to a modification of the BDFM equivalent circuit. A new method to determine the parameters of thisequivalent circuit from electrical tests is presented. The parameters determination from the geometry will also be considered for the elaboration of a derivable semi-analytical model. Such a model, paired with a 1st order optimization algorithm could be extremely powerful during the sizing of a BDFM. The potential of such an approach is shown in this work with the optimization of an Induction Machine.

Submitted by Nadir Basilio (nadirsb@uninove.br) on 2016-06-21T15:07:12Z No. of bitstreams: 1 Mauricio Gayubas.pdf: 4547233 bytes, checksum: 9a10b3048bfb27a539b8a17481b2b509 (MD5) / Made available in DSpace on 2016-06-21T15:07:12Z (GMT). No. of bitstreams: 1 Mauricio Gayubas.pdf: 4547233 bytes, checksum: 9a10b3048bfb27a539b8a17481b2b509 (MD5) Previous issue date: 2016-03-04 / Remanufacturing auto parts is an important activity from the point of view of the triple bottom line (economic, social and environmental) because its components can be reuse and not only has to be sending to be recycled. This case study sought to identify the members of the distribution channel (remanufacturer, distributor and fleet owners) how they influence the willingness to pay in a B2B relation. In addition, regarding all members influences the other members of this distribution channel in making production, distribution and application decisions of alternators and starter motors remanufactured applied in urban and intercity buses. Studyinglocal governmental regulations in Brazil, it was not identified obligation regarding producers to collect alternators and starter motors at the end of their life cycle. This situation would help the manufacturer to develop a more efficient reverse logistics strategy, with higher product volume of remanufactured production and increased activity in volume within the factory and thus invest the market in a vision at the triple bottom line point of view, as well as a better vision of the product by environmentally sustainable point. The remanufacturer produce their products identifying this activity as a business opportunity in the same way that the product is distributed wholesale with the same vision for the distributor and applied in bus fleets as a more expensive product that the reconditioned or repaired in maintenance shops. However, this price is offseted by the benefit cost of quality of the remanufactured products. Despite efforts by the trade associations and remanufacturers, albeit timidly, the information from the point of view that remanufacturing is an environmentally and socially sustainable activity. Customers in the B2B relationship does not identify the remanufacturing activity in this way, leaving an opportunity for the remanufacturer to manifest and develop information tools for green marketing information regarding the pillars of sustainability, bringing the remanufacturer a competitive advantage in environmental awareness among the members of the distribution channel, as well as greater awareness in society. It is an opportunity to promote environmentally sustainable pillar vision for the fleet owners and society as a curriculum in the development of environmental awareness for future generations. / A remanufatura de autopeças é uma atividade importante do ponto de vista das dimensões da sustentabilidade (econômico, social e ambiental) por reutilizar os componentes de um produto com defeito e não somente enviar esse mesmo produto para reciclagem, assim, este estudo de caso buscou identificar nos membros do canal de distribuição (fabricante remanufaturador, distribuidor de autopeças e frotista) a relação B2B que um membro influencia o outro membro desse canal de distribuição na tomada de decisão de produção, distribuição e aplicação dos alternadores e motores de partida veiculares remanufaturados aplicados em ônibus urbanos e rodoviários. Mesmo com as diretrizes mundiais e as leis locais, no Brasil não foi identificada a obrigatoriedade do recolhimento dos alternadores e motores de partida no final do ciclo de vida, o que contribuiria para o fabricante poder desenvolver uma estratégia de logística reversa mais eficiente, com maiores volumes de produtos e maior atividade em volumes dentro da fábrica e assim, investir perante o mercado uma visão do produto pelo ponto de vista ambientalmente sustentável. O fabricante remanufatura os seus produtos identificando essa atividade como uma oportunidade de negócio, da mesma forma que o produto é distribuído no atacado com essa mesma visão pelo distribuidor e aplicado nas frotas de ônibus como um produto mais caro que o recondicionado ou consertado nas oficinas de manutenção dentro das suas próprias garagens. Porém, esse preço é compensado pelo custo benefício da qualidade em que é produzido. Mesmo com esforços por parte das entidades de classe e dos fabricantes, para informar do ponto de vista de que a remanufatura é uma atividade ambientalmente e socialmente sustentáveis, os clientes dessa relação B2B não identificam a atividade de remanufatura dessa maneira, deixando uma excelente oportunidade para o fabricante se manifestar e desenvolver ferramentas de divulgação pelo marketing verde sobre a informação das dimensões da sustentabilidade, trazendo para o remanufaturador uma vantagem competitiva sobre a consciência ambiental junto aos membros da cadeia de distribuição, como também maior conscientização junto à sociedade. É a oportunidade de fomentar a visão da dimensão ambientalmente sustentável para os frotistas e para a sociedade como currículo no desenvolvimento da conscientização ambiental.

The thesis discusses the use of biomass for electricity island operation. The first chapter explains the concept of biomass and its distribution. It also shows the potential and availability of biomass. The second chapter describes the transformation of biomass into a suitable fuel form. The third to fifth chapter describes the conversion of biomass as a fuel into electrical energy through combustion, Stirling and steam engine. In these chapters there are brought the following parameters and prices of these motors and single-phase and three-phase alternators that can be connected to these engines. Other chapters describe the production of electricity in line with consumption, where the cost per hour to run engines for the various types of fuel and price of the electricity produced in kilowatt hours is calculated. In the last chapter of the thesis there is described the possibility of accumulation of electrical energy.

Artificial Intelligence in Automotive Industry / Cost reduction is one of the main concerns in industry. Companies invest considerably for better performance in end-of-line fault diagnosis systems. A common strategy is to use data obtained from existing instrumentation. This research investigates the challenge of learning from historical data that have already been collected by companies. Machine learning is basically one of the most common and powerful techniques of artificial intelligence that can learn from data and identify fault features with no need for human interaction. In this research, labeled sound and vibration measurements are processed into fault signatures for vehicle starter motors and alternators. A fault detection and identification system has been developed to identify fault types for end-of-line testing of motors. However, labels are relatively difficult to obtain, expensive, time consuming and require experienced humans, while unlabeled samples needs less effort to collect. Thus, learning from unlabeled data together with the guidance of few labels would be a better solution. Furthermore, in this research, learning from unlabeled data with absolutely no human intervention is also implemented and discussed as well. / Thesis / Master of Applied Science (MASc)