This thesis aims to study the relations between Turkey and Russia in the post-Soviet era by focusing on the nature of economic interdependence between these countries. Focusing on the energy, trade, tourism and investment sectors, the thesis evaluates the interaction between the political and economic factors in the relations between Turkey and the Russian Federation. Contrary to the general view which claims that the historically conflictual relations between these two countries have been replaced by more cooperative economic and political relations, the thesis argues that these relations are characterized by both cooperation and conflict due to the complex nature of their interdependence. In this sense, there is always a potential for tensions in these relations due to the strategic role of economic relations between Turkey and Russia. Following the introduction chapter, the second chapter examines the historical background of Turkish-Russian relations until the end of the Soviet era in 1991. The third and fourth chapters analyze the Yeltsin and Putin periods in Turkish-Russian relations by exploring the limits of economic interdependence in four main sectors of economic transaction / trade, energy, investment and tourism. The last chapter is the conclusion.

Neural Network models got the attention of the scientific community for their increasing accuracy in predictions and good emulation of some human tasks. This led to extensive enhancements in their architecture, resulting in models with fast-growing memory and computation requirements. Due to hardware constraints such as memory and computing capabilities, the inference of a large neural network model can be distributed across multiple devices by a partitioning algorithm. The proposed framework finds the optimal model splits and chooses which device shall compute a corresponding split to minimize inference time and energy. The framework is based on PipeEdge algorithm and extends it by not only increasing inference throughput but also simultaneously minimizing inference energy consumption. Another thesis contribution is the augmentation of the emerging technology Compute-in-memory (CIM) devices to the system. To the best of my knowledge, no one studied the effect of including CIM, specifically DNN+NeuroSim simulator, devices in a distributed inference. My proposed framework could partition VGG8 and ResNet152 on ImageNet and achieve a comparable trade-off between inference slowest stage increase and energy reduction when it tried to decrease inference energy (e.g. 19% energy reduction with 34% time increase) and when CIM devices were augmenting the system (e.g. 34% energy reduction with 45% time increase).

Orientador: Paulo de Barros Correia / Tese (doutorado) - Universidade Estadual de Campinas, Faculdade de Engenharia Mecânica / Made available in DSpace on 2018-08-27T01:59:39Z (GMT). No. of bitstreams: 1 Silva_ElisaBastos_D.pdf: 2776184 bytes, checksum: 20b2252b72c7204d062893f8dcb3d304 (MD5) Previous issue date: 2015 / Resumo: Leilões de novos empreendimentos de energia envolvem o compromisso de construí-los e o direito de explorá-los por meio de contratos de outorga. O leiloeiro, cujo objetivo é minimizar o pagamento pela energia contratada, buscando a redução de seu preço para os consumidores finais, fornece o direito de outorga da usina para o vencedor. O licitante é um investidor, e.g., uma empresa de geração que procura maximizar seu benefício com a venda de energia proveniente do empreendimento. Quando a natureza desses empreendimentos é complementar, torna-se possível proporcionar maiores benefícios aos licitantes, e maior eficiência ao leilão, caso sejam negociados em conjunto. Atualmente, o projeto de leilão instituído é composto por uma abordagem híbrida, sequencial e simultânea, que não permite a extração das sinergias entre empreendimentos. Esta tese examina duas metodologias híbridas de leilões reversos, considerando-se o ponto de vista do leiloeiro. O primeiro modelo, centralizado, é composto por duas fases: uma simultânea de lance aberto e outra combinatória de lance fechado. A fase simultânea incentiva a revelação do preço da energia, enquanto a fase combinatória oferece oportunidade aos licitantes de submeterem ofertas mais agressivas através de pacotes de empreendimentos complementares. O modelo centralizado é formulado como um problema de otimização inteiro e combinatório. A função-objetivo consiste em minimizar o pagamento, isso é, energia multiplicada pelo preço (lance) para todas as usinas. A estratégia de solução identifica os vencedores, resolvendo um problema de set-packing restrito. A segunda metodologia utiliza uma abordagem, também, em duas fases. A primeira é um projeto simultâneo de lance aberto, e a segunda fase um projeto combinatório descentralizado. Nesse modelo, a dificuldade do problema aumenta progressivamente à medida que os pacotes são ofertados. A dificuldade da alocação é distribuída entre os licitantes e, por isso, o leiloeiro não necessita resolver um problema de otimização. As metodologias propostas são aplicadas aos leilões de energia nova para o setor elétrico brasileiro. Os resultados mostram que a utilização de ambas as metodologias resolvem o problema de alocação com um tempo computacional aceitável / Abstract: Auctions for new power plants involve a commitment of constructing and the right of exploring them through power sales contracts. The auctioneer -- whose objective is to minimize the payment for the contracted energy, seeking to reduce prices for consumers -- provides the power plant's right for the winner. The bidder is an investor, for example, a generation company, which aims to maximize benefits of energy sales. When the power plant's nature is complementary, it is possible to provide more benefits to bidders and greater efficiency to the auction if these plants were traded together. Currently, the instituted auction design consists of a hybrid approach -- sequential and simultaneous -- which does not allow the extraction of synergies among plants. This thesis examines two hybrid methods of reverse auctions from the auctioneer's view point. The first model, centralized, consists of two phases: a simultaneous open bid and a combinatorial sealed bid. The simultaneous phase encourages the energy prices revelation. The combinatorial phase allows aggressive bidders to acquire bundles of complementary plants. The centralized model is formulated as an integer and combinatorial optimization problem. The objective function consists of minimizing the payment, that is, energy multiplied by the price (bid) for all plants. The solution strategy identifies the winners solving a restricted set-packing problem. The second method also uses a two phase approach. The first phase is a simultaneous open bid design and the second phase is a decentralized combinatorial design. In this model, the problem difficulty increases gradually. The allocation difficulty is distributed among the bidders; therefore, the auctioneer does not need to solve an optimization problem. The proposed methodologies are applied to new energy auctions on Brazilian electrical energy sector. The results show the use of both methods solving the problem of allocation with an acceptable computational time / Doutorado / Planejamento de Sistemas Energeticos / Doutora em Planejamento de Sistemas Energéticos

Blockchain-based peer-to-peer (P2P) electricity markets received considerable attention in the past years, leading to a rich variety of proposed market designs. Yet, little comparability and consensus exists on optimal market design, also due to a lack of common evaluation and benchmarking infrastructure. This article describes LabChain, an interactive prototype as research infrastructure for conducting experiments in (simulated) P2P electricity markets involving real human actors. The software stack comprises: (i) an (open) data layer for experiment configuration, (ii) a blockchain layer to reliably document bids and transactions, (iii) an experiment coordination layer and (iv) a user interface layer for participant interactions. As evaluation environment for human interactions within a laboratory setting, researchers can investigate patterns based on energy system and market setup and can compare and evaluate designs under real human behavior allowing alignment of intentions and outcomes. This contributes to the evaluation and benchmarking infrastructure discourse.

Motivated by numerous drivers, blockchain-based peer-to-peer energy trade whitepapers surged in the past two years. Assuming disruption through blockchain technology, they envisioned a transformation of energy systems through technosocio- economic solutions. Few impartial and sober assessments of blockchain-based energy projects exist, and many publications praise disruptive potential without further examination. A more distant and critical perspective, however, is imperative for a responsible use of a novel, in particular disruptive, technology. This review aims at surveying the energy system envisioned by the projects through discussing the projects by their characteristics, their perspective on the transactive energy lifecycle and the energy ecosystem envisioned in the white papers. This review is descriptive and comparative in nature, and attempts to synthesize topics raised in the white papers through methods of grounded theory, as well as assessing the disruptive potential of blockchain technology in energy systems. Through this and a critical and neutral perspective, it strives to (soberly) contribute to a discussion on the digitization of elements of the energy system, and how blockchain-based use cases can contribute constructively to the problems at hand.

A participação da energia eólica na matriz energética brasileira tem aumentado significativamente nos últimos três anos, dada a contratação da fonte em leilões organizados pelo governo, que contabilizaram cerca de 7,0 GW contratados no final de 2012. No entanto, frente às vantagens como fonte renovável, a geração eólica possui desvantagens como fonte regular de energia devido à sazonalidade e dependência do regime de ventos. Assim, considerando o cenário de contratação da fonte via leilões, foi desenvolvido um modelo que, integrando cenários para a geração observada no parque eólico (associando as incertezas de geração) e para os preços do mercado de energia elétrica (associando a alta volatilidade dos preços), submetidos às regras de contratação pactuadas, avalia os riscos na comercialização da energia gerada por um parque eólico nos quais, o agente gerador está exposto. Tendo sido o modelo aplicado para dois parques eólicos reais no âmbito dos Leilões de Energia Nova (tipo A-3) e Leilões de Energia de Reserva, verificou- se que no Brasil o esquema de contratação de energia eólica considera a produção média ao longo dos anos, permitindo reajustes e compensações, notadamente nos leilões para a contratação de Energia de Reserva para todo o sistema brasileiro. Considerando a busca pela máxima rentabilidade associada ao menor risco, o agente gerador foi indicado a comercializar valores muito próximos ao total da energia garantida pela geração dos parques registrada nos leilões. Em termos de riscos, a variação é mais acentuada no Leilão A-3, no qual se constatou a exposição do gerador aos preços de mercado na liquidação das receitas. Tendo sido avaliados também cenários de venda da energia gerada pelo parque eólico no Ambiente de Contratação Livre, constatou-se que nesse, o risco de comercialização é assumido integralmente pelo agente gerador, cujos ganhos e perdas são valorados aos preços do mercado de curto prazo e ainda, na negociação há a dificuldade de obtenção de mecanismos de proteção (hedge) para gerenciar adequadamente as incertezas sobre a geração. Comparando com o cenário dos leilões no qual, além das regras pactuadas evidenciarem mecanismos claros de mitigação de riscos, importante parcela do risco é assumida pelas distribuidoras, que possuem o direito de repassar à tarifa do consumidor final as suas perdas. / The participation of wind energy in the Brazilian electric matrix has risen significantly over the past three years due to the purchase of the source in auctions held by the federal government, accounting for about 7 GW by the end of 2012. However, in spite of the advantages of being a renewable source, wind generation has disadvantages as a regular energy source due to its seasonality and dependence on wind regime. Thus, considering the purchase scenario of the this source in the auctions, a model that integrates the scenarios for the generation observed in the wind farm (associating generation uncertainties) and scenario for the prices of the electric power market (associating the high volatility of the prices), submitted to the established purchase agreements, assess the risks in trading the energy generated by a wind farm, where the generating agent is exposed. Having applied this model to two real wind farms comprising New Energy Auctions (A-3 type) and Reserve Energy Actions, it was possible to observe that in Brazil the wind energy purchase arrangement considers the average production along the years, allowing adjustments and compensations, mainly in the auctions aiming at the purchase of Reserve Energy for the entire Brazilian system. Considering the search for the maximum profitability associated to the smallest risk, the generating agent was advised to trade values that are very close to the total amount of energy guaranteed by the generation of the farms that was registered in the auctions. In terms of risks, the variation is sharper in A-3 Auctions, where it was possible to observe the exposure of the generator to the market prices at the liquidation of the incomes. Also, having assessed the sales scenarios of the energy generated the wind farms in the Free Market, it was observed that here, the risk of trade is completely taken by the generating agents, whose gains and losses are equivalent to the prices of the short term market. Also, it is difficult to find protection mechanisms, hedge, to properly manage the uncertainties of the generation. Comparing to the auction scenario, where, besides the established show clear mechanisms to mitigate the risks, an important part of the risk is taken by the distributors, which have the right to compensate their losses in the tariffs paid by the end-consumers.

A participação da energia eólica na matriz energética brasileira tem aumentado significativamente nos últimos três anos, dada a contratação da fonte em leilões organizados pelo governo, que contabilizaram cerca de 7,0 GW contratados no final de 2012. No entanto, frente às vantagens como fonte renovável, a geração eólica possui desvantagens como fonte regular de energia devido à sazonalidade e dependência do regime de ventos. Assim, considerando o cenário de contratação da fonte via leilões, foi desenvolvido um modelo que, integrando cenários para a geração observada no parque eólico (associando as incertezas de geração) e para os preços do mercado de energia elétrica (associando a alta volatilidade dos preços), submetidos às regras de contratação pactuadas, avalia os riscos na comercialização da energia gerada por um parque eólico nos quais, o agente gerador está exposto. Tendo sido o modelo aplicado para dois parques eólicos reais no âmbito dos Leilões de Energia Nova (tipo A-3) e Leilões de Energia de Reserva, verificou- se que no Brasil o esquema de contratação de energia eólica considera a produção média ao longo dos anos, permitindo reajustes e compensações, notadamente nos leilões para a contratação de Energia de Reserva para todo o sistema brasileiro. Considerando a busca pela máxima rentabilidade associada ao menor risco, o agente gerador foi indicado a comercializar valores muito próximos ao total da energia garantida pela geração dos parques registrada nos leilões. Em termos de riscos, a variação é mais acentuada no Leilão A-3, no qual se constatou a exposição do gerador aos preços de mercado na liquidação das receitas. Tendo sido avaliados também cenários de venda da energia gerada pelo parque eólico no Ambiente de Contratação Livre, constatou-se que nesse, o risco de comercialização é assumido integralmente pelo agente gerador, cujos ganhos e perdas são valorados aos preços do mercado de curto prazo e ainda, na negociação há a dificuldade de obtenção de mecanismos de proteção (hedge) para gerenciar adequadamente as incertezas sobre a geração. Comparando com o cenário dos leilões no qual, além das regras pactuadas evidenciarem mecanismos claros de mitigação de riscos, importante parcela do risco é assumida pelas distribuidoras, que possuem o direito de repassar à tarifa do consumidor final as suas perdas. / The participation of wind energy in the Brazilian electric matrix has risen significantly over the past three years due to the purchase of the source in auctions held by the federal government, accounting for about 7 GW by the end of 2012. However, in spite of the advantages of being a renewable source, wind generation has disadvantages as a regular energy source due to its seasonality and dependence on wind regime. Thus, considering the purchase scenario of the this source in the auctions, a model that integrates the scenarios for the generation observed in the wind farm (associating generation uncertainties) and scenario for the prices of the electric power market (associating the high volatility of the prices), submitted to the established purchase agreements, assess the risks in trading the energy generated by a wind farm, where the generating agent is exposed. Having applied this model to two real wind farms comprising New Energy Auctions (A-3 type) and Reserve Energy Actions, it was possible to observe that in Brazil the wind energy purchase arrangement considers the average production along the years, allowing adjustments and compensations, mainly in the auctions aiming at the purchase of Reserve Energy for the entire Brazilian system. Considering the search for the maximum profitability associated to the smallest risk, the generating agent was advised to trade values that are very close to the total amount of energy guaranteed by the generation of the farms that was registered in the auctions. In terms of risks, the variation is sharper in A-3 Auctions, where it was possible to observe the exposure of the generator to the market prices at the liquidation of the incomes. Also, having assessed the sales scenarios of the energy generated the wind farms in the Free Market, it was observed that here, the risk of trade is completely taken by the generating agents, whose gains and losses are equivalent to the prices of the short term market. Also, it is difficult to find protection mechanisms, hedge, to properly manage the uncertainties of the generation. Comparing to the auction scenario, where, besides the established show clear mechanisms to mitigate the risks, an important part of the risk is taken by the distributors, which have the right to compensate their losses in the tariffs paid by the end-consumers.

Made available in DSpace on 2016-06-02T19:26:23Z (GMT). No. of bitstreams: 1 MEDOLAGO_Cesar_2013.pdf: 1874610 bytes, checksum: 3e19d8b8517c60b82a8deacf9f5614ce (MD5) Previous issue date: 2013-11-08 / This study aims to describe the pattern of moulting and reproduction, evaluating their temporal overlap in an assembly of birds in the Atlantic Forest. These events tend to have little or no overlap due to high energy costs involved, but some authors argue that in tropical regions, they may present a significant overlap, since the period of resource abundance would be longer in this region. We also noticed the amount of fat deposition, because this phenomenon is important in thermo-isolation, energy reserves and development of egg-yolk. It is possible that environmental variables act directly on the breeding period of birds, which in turn influences the moult, it is expected that this starts right after the breeding season, when the young leave their nests. There taking into account that ecological groups, such as trophic guilds, may show different patterns for the periods, since the supply of food resources varies temporally in a different way for each group. It were determined five areas in Carlos Botelho State Park , state of São Paulo ( 24 ° 06 ' 55'' , 24 º 14' 41'' S , 47 º 47 ' 18'' and 48 º 07' 17'' W), which were sampled from June 2012 to May 2013, once a month, during the daytime, using lines with ten mist nets (3x12m , mesh 36mm ). Each bird was received a numbered metal band provided by CEMAVE. With a total of 4650 mistnet-hours were held 700 catches and 130 were recaptures, totaling 54 species, all residents. The period of moult of flight concentrated from November to April, with its peak in February. Incubation began in August, with the highest percentage of individuals presenting brood patch occurred in the months of November and December, declining from February, when the percentage of young individuals in the assemblage began to increase. The highest percentage of individuals with fat deposition occurred in the months comprising the coldest period of the year. The incubation period began at the end of the dry season, increasing with the photoperiod, reaching its peak in November. Thus, the young individuals leave their nests in the beginning of the hot season, when the supply of food resources would be higher, which would support the new individuals in the community as well as the start of moult period. There was little difference in the incubation period and fat deposition between trophic guilds and no difference in their moult period. The overlap between the events found in this study was 7 %, which confirms the tendency to avoid the overlap of these cycles, even in tropical regions, such as the Atlantic Forest, due to high energy costs involved. / O presente estudo tem por objetivos descrever o padrão de muda de penas e reprodução, avaliando sua sobreposição temporal, em uma assembleia de aves na Mata Atlântica. Esses eventos tendem a apresentar pouca ou nenhuma sobreposição devido aos altos custos energéticos envolvidos; porém alguns autores defendem que em regiões tropicais, eles poderiam apresentar uma maior sobreposição, já que a o período de abundância de recursos é mais longo nessa região. Foi verificado também o período de acúmulo de gordura na região da fúrcula, pois esse fenômeno tem importância no isolamento térmico, reserva de energia e desenvolvimento da gema do ovo. É possível que as variáveis ambientais atuem diretamente sobre o período de reprodução das aves, que por sua vez, teriam influência sobre a muda de penas, pois é de se esperar que esse comece logo em seguida ao período reprodutivo, juntamente com a saída dos jovens dos ninhos. Há de levar-se em conta que grupos ecológicos, como as guildas tróficas, podem apresentar padrões distintos em relação a esses períodos, já que a oferta recursos alimentares varia temporalmente, de maneira diferente, para cada um desses grupos. Para isso foram determinadas cinco áreas no Parque Estadual Carlos Botelho, estado de São Paulo (24º 06 55 , 24º 14 41 S, 47º 47 18 e 48º 07 17 W), que foram amostradas de junho de 2012 a maio de 2013, uma vez por mês, durante o período diurno, utilizando-se linhas de dez redes de neblina (3x12m, malha 36mm). Foram utilizadas anilhas metálicas padrão CEMAVE para marcar os indivíduos. Com um total de 4650 horas-rede, foram realizadas 700 capturas, das quais 130 foram recapturas, totalizando 54 espécies, todas residentes. O período de muda de penas de voo concentrou-se de novembro a abril, apresentando seu auge em fevereiro. A incubação teve início em agosto, sofrendo influência do fotoperíodo e a maior porcentagem de indivíduos apresentando placa de incubação se deu nos meses de novembro e dezembro, declinando a partir de fevereiro, quando a porcentagem de indivíduos jovens começou a aumentar. A maior porcentagem de indivíduos com acúmulo de gordura se deu nos meses que compreende o período mais frio do ano. O período de incubação iniciou-se no final da estação seca, atingindo seu ápice em novembro. Assim, a saída dos jovens dos ninhos coincidiria com o início da estação mais quente, quando a oferta de recursos alimentares seria maior, o que suportaria os novos indivíduos na comunidade, bem como a realização da muda de penas. Houve uma pequena diferença no período de incubação e de acúmulo de gordura entre as guildas tróficas e não houve diferença entre seus períodos de muda. A sobreposição entre os eventos encontrada nesse estudo foi de 7%, o que corrobora a tendência em evitar a sobreposição desses ciclos, mesmo em regiões tropicais, como é o caso da Mata Atlântica, devido aos altos custos energéticos envolvidos.

World energy systems are going through a continuous change. The focus has been shifted from large thermal or hydal power generation to small distributed generation, mainly based upon renewable energy systems. This transition is also backed by some governments. There have also been significant improvements in grid technology, and modern-day smart grid can provide real time bi-directional flow of data i.e. “real time energy deficit and surplus, and also real time prices to both producers and consumers. Smart grid can also accommodate intermittent small suppliers of electricity. This shift in energy generation policy and improvement in grid technology has opened ways for small scale energy producers and consumers to share energy with each other. It has also opened ways to purchase or sale energy to unknown peers over a smart grid. Need has been felt to store these transactions among peers in a secure, non-alterable yet quickly accessible way. Blockchain technology offers to provide this secure, unalterable yet quickly accessible ledger. In this study this transition process and role of blockchain technology for future energy systems has been historically reviewed. It has been found out that on top of keeping record of Peer to Peer transactions, blockchain technology can fill many other purposes. However, technology is still not matured for large scale projects, Research projects are underway to decrease the large time and energy consumption for block building computational processes yet keeping them safe and reliable.

In this paper I investigate the energy cooperation between the EU and Russia from the political and economic points of view. The relevance of the issue is due to the growing role of energy sector, the need for security of energy supply and demand and for political and economic balance of power in the region, and closer economic integration. Two theoretical approaches, geopolitical and bureaucratic, guiding the research, explain the behavior of protagonists. To better understand the present level of cooperation I analyze the background and dynamics of EU - Russia energy trade relations. As a case study, I investigate the Nord Stream project as an example of successful cooperation of Germany as an EU member state and Russia in this field. I discuss challenges for the healthy mutual partnership in the energy sector, namely, high politicization of the issue, monopolization of Russian energy sector and low sustainability of Russian economic development, and mistrust between the protagonists. I attempt to identify possible policy changes aimed to improve this cooperation and achieve fruitful partnership, security and stability. Key words: Russia, EU, energy, cooperation, Realpolitik, international relations, oil, gas, energy dialogue, ECT, security, balance of power, geopolitics, interest groups,...