Submitted by Renata Lopes (renatasil82@gmail.com) on 2016-05-16T17:51:25Z No. of bitstreams: 1 guilhermedeoliveiraalves.pdf: 1293169 bytes, checksum: a76074780b2af177b66be7c6435b16d1 (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2016-06-28T12:25:31Z (GMT) No. of bitstreams: 1 guilhermedeoliveiraalves.pdf: 1293169 bytes, checksum: a76074780b2af177b66be7c6435b16d1 (MD5) / Made available in DSpace on 2016-06-28T12:25:31Z (GMT). No. of bitstreams: 1 guilhermedeoliveiraalves.pdf: 1293169 bytes, checksum: a76074780b2af177b66be7c6435b16d1 (MD5) Previous issue date: 2015-09-18 / CAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / O presente trabalho tem por objetivo apresentar uma nova metodologia para estimação estática de estados em sistemas de distribuição de energia elétrica que estima as correntes nos ramos como variáveis de estado utilizando medições de tensão e corrente de ramo fasoriais oriundas de unidades de medição fasorial (Phasor Measurement Units - PMUs). A metodologia consiste em resolver um problema de otimização não linear minimizando uma função objetivo quadrática associada com as medições e estados estimados sujeito às restrições de carga das barras da rede que não apresentam PMUs instaladas baseadas em dados históricos, sendo esta a principal contribuição deste trabalho. Uma proposta de alocação de PMUs também é apresentada e que consiste em alocar duas unidades em cada ramificação do sistema, uma no começo e outra no final do trecho, procurando utilizar o menor número possível e que não comprometa a qualidade dos estados estimados. A resolução do problema de otimização é realizada de duas formas, através da ‘toolbox fmincon’ do software Matlab, que é uma ferramenta muito utilizada na resolução de problemas de otimização, e através da implementação computacional do Método de Pontos Interiores com Barreira de Segurança (Safety Barrier Interior Point Method - SFTB - IPM) proposto na literatura utilizada. Durante o processo de estimação de estados são utilizadas medidas obtidas através de um fluxo de potência que simulam as PMUs instaladas nos sistemas analisados variando o carregamento de cada sistema em torno da sua média histórica de carga até atingir os limites superior e inferior estabelecidos, sendo verificado o comportamento do estimador de estados perante a ocorrência de ruídos brancos nas medidas de todos os sistemas analisados. Foram analisados um sistema de distribuição tutorial de 15 barras e três sistemas encontrados na literatura contendo 33, 50 e 70 barras respectivamente. No sistema tutorial e no de 70 barras foram incluídas unidades de geração distribuída para se verificar o comportamento do estimador de estados. Todos os resultados do processo de estimação de estados são obtidos com os dois métodos de resolução apresentados e são comparados o desempenho de cada método, principalmente em relação ao tempo computacional. Todos os resultados obtidos foram validados usando um programa de fluxo de potência convencional e apresentam boa precisão com valor de função objetivo baixo mesmo na presença de ruídos nas medidas refletindo de maneira confiável o real estado do sistema de distribuição, o que torna a metodologia proposta atraente. / This work aims at presenting a new methodology for static state estimation in electric power distribution systems which estimates the branch currents as state variables using voltage measurements and current phasor branch obtained from phasor measurement units (Phasor Measurement Units - PMUs). The methodology consists of solving a nonlinear optimization problem minimizing a quadratic objective function associated with the estimated measurements and states, subject to load constraints for the non monitored loads based on historical data, which is the main contribution of this work. A PMU allocation strategy is presented which consists of allocating two PMUs for each system branch, one at the beginning and another at the end, trying to use as little PMUs as possible in such a way that the quality of the estimated states are not compromised. The solution of the optimization problem is obtained through two ways, the first is the toolbox ‘fmincon’ from Matlab solver software which is a widely used tool in the optimization problem. The second is a computer implementation of interior point method with security barrier (SFTB - IPM) proposed in the literature. Comparisons of computing times and results obtained with both methods are shown. A power flow program is used to obtain the voltages and branch currents in order to emulate the PMUs data in the state estimation process. Additionaly the non monitored loads are varied from the minimum bounds to their maximum, allowing white noise errors from the PMUs measurements. A tutorial test system of 15 buses is fully explored and three IEEE test systems of 33, 50 and 70 buses are used to show the effectiveness of the proposed methodology. For the tutorial and 70 bus systems, distribued generation units were included to see the state estimator behavior. All results from the state estimation process are obtained considering the two presented solving methods and the computing times performance compared. The results obtained were validated using a conventional power flow program and have good accuracy with low objective function value even in the presence of white noise errors in the measurements reflecting the reliability of the proposed methodology, making it very attractive for distribution system monitoring.

Submitted by Renata Lopes (renatasil82@gmail.com) on 2017-01-09T11:36:05Z No. of bitstreams: 1 brauliocesardeoliveira.pdf: 2150243 bytes, checksum: 62faa254539b7873aa1393d8cd8f1bf2 (MD5) / Approved for entry into archive by Diamantino Mayra (mayra.diamantino@ufjf.edu.br) on 2017-01-31T11:23:24Z (GMT) No. of bitstreams: 1 brauliocesardeoliveira.pdf: 2150243 bytes, checksum: 62faa254539b7873aa1393d8cd8f1bf2 (MD5) / Made available in DSpace on 2017-01-31T11:23:24Z (GMT). No. of bitstreams: 1 brauliocesardeoliveira.pdf: 2150243 bytes, checksum: 62faa254539b7873aa1393d8cd8f1bf2 (MD5) Previous issue date: 2016-03-08 / O presente trabalho tem por objetivo apresentar uma metodologia para estimação de estados em sistemas de distribuição de energia elétrica. São utilizadas como variáveis de estado as correntes nos ramos. As medições são obtidas por meio de medições fasoriais sincronizadas(PhasorMeasurementUnits-PMUs),sendoqueostiposdemedidasadvindos desses equipamentos são as tensões nodais e as correntes nos ramos. A abordagem é trifásica, portanto representa as características próprias de um sistema de distribuição. A metodologia consiste em resolver um problema de otimização não linear cuja função objetivo associa o erro quadrático das medidas em relação aos estados estimados sujeito às restrições de carga das barras da rede que não possuem PMUs instaladas baseadas em estimativas de cargas obtidas para o instante “t-1”, partindo-se da premissa que em curtos intervalos de tempo a carga não sofre grandes variações, sendo esta em conjunto com a abordagem trifásica as principais contribuições deste trabalho. Outra contribuição do trabalho é a descentralização, com esta técnica pode-se dividir uma determinada rede em vários subsistemas que podem ser resolvidos de forma separada e independente. Isso torna o processo mais rápido do ponto de vista computacional além de permitir o uso do processamento paralelo, visto que já existe um paralelismo natural entre as tarefas que devem ser resolvidas. Outra vantagem da divisão em subsistemas reside no fato do monitoramento de áreas de interesse. Para utilizar a descentralização foi proposta uma alternativa de alocação de PMUs que consiste em posicionar duas unidades em cada ramificação do sistema, uma no começo e outra no final do trecho, procurando utilizar o menor número possível e que não comprometa a qualidade dos estados estimados. A resolução do problema de otimização é realizada através da implementação computacional do Método de Pontos Interiores com Barreira de Segurança (Safety Barrier Interior Point Method - SFTB - IPM) proposto na literatura especializada. As medidas das PMUs foram obtidas através de um Fluxo de Potência Trifásico via Injeção de Correntes (FPTIC). Foram realizadas diversas simulações variando-se o percentual da carga e os resultados obtidos foram comparados com outra metodologia existente na literatura e com os valores verdadeiros que foram obtidos através do FPTIC para as barras não monitoradas. Foram tambémcomparadosotempocomputacionalentreaexecuçãoserialeaexecuçãoutilizando o processamento paralelo. Os testes mostraram bons resultados o que torna a metodologia proposta aplicável na supervisão de sistemas de distribuição. / This work aims to present a methodology for static state estimation in electric power distribution systems. Branch currents are used as state variables. Measurements are obtained by means of Phasor Measurement Units (PMUs), in which voltage and current branches measurements are used. The approach is three-phase, thus represents the distribution system characteristics. The methodology consists of solving a nonlinear optimization problem minimizing a quadratic objective function associated with the estimated measurements and states subject to load constraints for the non monitored loads based on estimated load obtained from the ‘t-1’ instant, starting from the assumption that in short time intervals the load does not have large variations, which together with the the three-phase approach are the main contributions of this work. Another contribution of this work is the descentralided approach, with this assumption the network can be divided into several subnetworks that can be solved separately and independently. This speeds up the process of being solved from a computational point of view and allows the use of parallel processing, since there is already a natural parallelism among tasks to be solved. Another advantage of the division into subsystems is the fact that the monitoring areas of interest. With the aim of allowing the decentralization was proposed PMUs allocation strategy that consists of allocating two units for each lateral feeder, one at the beginning and one at the end, trying to use as little PMUs as possible in such a way that the quality of the estimated states are not compromised. The resolution of the optimization problem is done through a computer implementation of Interior Point Method with Security Barrier (SFTB - IPM) proposed in the literature. The PMUs measurements were emulated using a Three-PhasePowerFlowusingtheCurrentInjectionmethod(FPTIC).Severalsimulations were performed varying the load percentage and the results obtained were compared with other existing methodology in literature and also the true values that were obtained from the FPTIC to non monitored loads. The computational time using serial and parallel processing were also compared. Results show good results which makes the proposed methodology applicable in monitoring distribution systems.

Fundação de Apoio a Pesquisa e à Inovação Tecnológica do Estado de Sergipe - FAPITEC/SE / The growth of the electric energy consumption in the last years has generated the need of the increase in the amount of power sources, making the electricity sector undergo some large changes. This has provided the search for tools that promotes a better efficiency and security to the electrical power systems. A planning problem that is considered important in the daily operation of the power systems is the Unit Commitment, where the time schedule of the operation is defined, determining which machines will be online or offline, and which are the operating points. Those units must operate by load variation, respecting the operative and security constraints. This research proposes the resolution of the problem for the short-term planning, taking a set of constraints associated with the thermal generation and the power system. Among them, we can highlight the output power variation constraints of the machines and the security restrictions of the transmission system, avoided in most Unit Commitment studies. This problem is nonlinear, mixed-integer and has a large scale. The methodology used involves the utilization of an Adaptive Genetic Algorithm, for the Unit Commitment problem, and the Interior-Point Primal- Dual Predictor–Corrector Method, for DC power flow resolution in economic dispatch problem. Furthemore, this research proposes the implementation of cross-over and mutation operators of Genetic Algorithm based on a ring methodology applied in Unit Commitment matrix. The results were obtained through simulations in a mathematical simulation software, using the IEEE test systems with 30 bus and 9 generators, and another with 24 bus and 26 generators. The validation of the algorithm was done by comparing the results with other works in the literature. / O crescimento do consumo de energia elétrica nos últimos anos vem gerando a necessidade de um aumento na quantidade de fontes geradoras, fazendo com que o setor elétrico passe por grandes mudanças. Isso tem proporcionado a busca por ferramentas que ofereçam maior eficiência e segurança aos sistemas de potência. Um problema considerado de extrema importância na operação diária dos sistemas elétricos é o planejamento da Alocação das Unidades Geradoras, onde define-se a programação horária das unidades do sistema, determinando quais máquinas deverão estar ligadas ou desligadas, e quais serão seus respectivos pontos de operação. Essas unidades geradoras devem operar de forma eficaz, mediante a variação da carga, respeitando restrições operativas e de segurança do sistema. Este trabalho propõe a resolução do problema para o planejamento de curto prazo, levando em consideração uma série de restrições relacionadas a geração térmica e ao sistema elétrico. Entre elas, podemos destacar as restrições de variação de potência de saída das máquinas e as restrições de segurança do sistema de transmissão, evitadas na maioria dos estudos de Alocação de Unidades Geradoras. Este problema tem característica não-linear, inteiro-misto e de grande escala. A metodologia utilizada para resolução do problema envolve a utilização de um Algoritmo Genético Adaptativo, para Alocação das Unidades, e o Método de Pontos Interiores Primal-Dual Preditor-Corretor, para a resolução do Fluxo de Potência Ótimo DC no problema do Despacho Econômico. Além disso, este trabalho propõe a implementação dos operadores de cross-over e mutação do Algoritmo Genético com base em uma metodologia anelar aplicada na matriz de alocação de unidades. Os resultados foram obtidos através de simulações em um software de simulação matemática, utilizando os sistemas testes do IEEE de 30 barras com 9 geradores e 24 barras com 26 geradores, e a validação do algoritmo foi feita comparando os resultados obtidos com os outros trabalhos da literatura.

The goal of this thesis is two-fold. On the one hand, it pursues to provide a contribution to the conjugate duality by proposing a new duality concept, which can be understood as an umbrella for different meaningful perturbation methods. On the other hand, this thesis aims to investigate minimax location problems by means of the duality concept introduced in the first part of this work, followed by a numerical approach using epigraphical splitting methods. After summarizing some elements of the convex analysis as well as introducing important results needed later, we consider an optimization problem with geometric and cone constraints, whose objective function is a composition of n+1 functions. For this problem we propose a conjugate dual problem, where the functions involved in the objective function of the primal problem are decomposed. Furthermore, we formulate generalized interior point regularity conditions for strong duality and give necessary and sufficient optimality conditions. As applications of this approach we determine the formulae of the conjugate as well as the biconjugate of the objective function of the primal problem and analyze an optimization problem having as objective function the sum of reciprocals of concave functions. In the second part of this thesis we discuss in the sense of the introduced duality concept three classes of minimax location problems. The first one consists of nonlinear and linear single minimax location problems with geometric constraints, where the maximum of nonlinear or linear functions composed with gauges between pairs of a new and existing points will be minimized. The version of the nonlinear location problem is additionally considered with set-up costs. The second class of minimax location problems deals with multifacility location problems as suggested by Drezner (1991), where for each given point the sum of weighted distances to all facilities plus set-up costs is determined and the maximal value of these sums is to be minimized. As the last and third class the classical multifacility location problem with geometrical constraints is considered in a generalized form where the maximum of gauges between pairs of new facilities and the maximum of gauges between pairs of new and existing facilities will be minimized. To each of these location problems associated dual problems will be formulated as well as corresponding duality statements and necessary and sufficient optimality conditions. To illustrate the results of the duality approach and to give a more detailed characterization of the relations between the location problems and their corresponding duals, we consider examples in the Euclidean space. This thesis ends with a numerical approach for solving minimax location problems by epigraphical splitting methods. In this framework, we give formulae for the projections onto the epigraphs of several sums of powers of weighted norms as well as formulae for the projection onto the epigraphs of gauges. Numerical experiments document the usefulness of our approach for the discussed location problems.

Investors constantly seek information that provides an edge over the market. One of the conventional methods is to find factors which can predict asset returns. In this study we improve the Fama and French Five-Factor model with Regime-Switches, student's t distributions and copula dependencies. We also add price momentum as a sixth factor and add a one-day lag to the factors. The Regime-Switches are obtained from a Hidden Markov Model with conditional Student's t distributions. For the return process we use factor data as input, Student's t distributed residuals, and Student's t copula dependencies. To fit the copulas, we develop a novel approach based on the Expectation-Maximisation algorithm. The results are promising as the quantiles for most of the portfolios show a good fit to the theoretical quantiles. Using a sophisticated Stochastic Programming model, we back-test the predictive power over a 26 year period out-of-sample. Furthermore we analyse the performance of different factors during different market regimes.

The aim of this thesis is to present new results concerning duality in scalar optimization. We show how the theory can be applied to optimization problems arising in the theory of risk measures, portfolio optimization and machine learning. First we give some notations and preliminaries we need within the thesis. After that we recall how the well-known Lagrange dual problem can be derived by using the general perturbation theory and give some generalized interior point regularity conditions used in the literature. Using these facts we consider some special scalar optimization problems having a composed objective function and geometric (and cone) constraints. We derive their duals, give strong duality results and optimality condition using some regularity conditions. Thus we complete and/or extend some results in the literature especially by using the mentioned regularity conditions, which are weaker than the classical ones. We further consider a scalar optimization problem having single chance constraints and a convex objective function. We also derive its dual, give a strong duality result and further consider a special case of this problem. Thus we show how the conjugate duality theory can be used for stochastic programming problems and extend some results given in the literature. In the third chapter of this thesis we consider convex risk and deviation measures. We present some more general measures than the ones given in the literature and derive formulas for their conjugate functions. Using these we calculate some dual representation formulas for the risk and deviation measures and correct some formulas in the literature. Finally we proof some subdifferential formulas for measures and risk functions by using the facts above. The generalized deviation measures we introduced in the previous chapter can be used to formulate some portfolio optimization problems we consider in the fourth chapter. Their duals, strong duality results and optimality conditions are derived by using the general theory and the conjugate functions, respectively, given in the second and third chapter. Analogous calculations are done for a portfolio optimization problem having single chance constraints using the general theory given in the second chapter. Thus we give an application of the duality theory in the well-developed field of portfolio optimization. We close this thesis by considering a general Support Vector Machines problem and derive its dual using the conjugate duality theory. We give a strong duality result and necessary as well as sufficient optimality conditions. By considering different cost functions we get problems for Support Vector Regression and Support Vector Classification. We extend the results given in the literature by dropping the assumption of invertibility of the kernel matrix. We use a cost function that generalizes the well-known Vapnik's ε-insensitive loss and consider the optimization problems that arise by using this. We show how the general theory can be applied for a real data set, especially we predict the concrete compressive strength by using a special Support Vector Regression problem.

The aim of this thesis is to present new results concerning duality in scalar optimization. We show how the theory can be applied to optimization problems arising in the theory of risk measures, portfolio optimization and machine learning. First we give some notations and preliminaries we need within the thesis. After that we recall how the well-known Lagrange dual problem can be derived by using the general perturbation theory and give some generalized interior point regularity conditions used in the literature. Using these facts we consider some special scalar optimization problems having a composed objective function and geometric (and cone) constraints. We derive their duals, give strong duality results and optimality condition using some regularity conditions. Thus we complete and/or extend some results in the literature especially by using the mentioned regularity conditions, which are weaker than the classical ones. We further consider a scalar optimization problem having single chance constraints and a convex objective function. We also derive its dual, give a strong duality result and further consider a special case of this problem. Thus we show how the conjugate duality theory can be used for stochastic programming problems and extend some results given in the literature. In the third chapter of this thesis we consider convex risk and deviation measures. We present some more general measures than the ones given in the literature and derive formulas for their conjugate functions. Using these we calculate some dual representation formulas for the risk and deviation measures and correct some formulas in the literature. Finally we proof some subdifferential formulas for measures and risk functions by using the facts above. The generalized deviation measures we introduced in the previous chapter can be used to formulate some portfolio optimization problems we consider in the fourth chapter. Their duals, strong duality results and optimality conditions are derived by using the general theory and the conjugate functions, respectively, given in the second and third chapter. Analogous calculations are done for a portfolio optimization problem having single chance constraints using the general theory given in the second chapter. Thus we give an application of the duality theory in the well-developed field of portfolio optimization. We close this thesis by considering a general Support Vector Machines problem and derive its dual using the conjugate duality theory. We give a strong duality result and necessary as well as sufficient optimality conditions. By considering different cost functions we get problems for Support Vector Regression and Support Vector Classification. We extend the results given in the literature by dropping the assumption of invertibility of the kernel matrix. We use a cost function that generalizes the well-known Vapnik's ε-insensitive loss and consider the optimization problems that arise by using this. We show how the general theory can be applied for a real data set, especially we predict the concrete compressive strength by using a special Support Vector Regression problem.