Les méthodes de décomposition sont une application du concept de diviser pour régner en optimisation. L'idée est de décomposer un problème d'optimisation donné en une séquence de sous-problèmes plus faciles à résoudre. Bien que ces méthodes soient les meilleures pour un grand nombre de problèmes de recherche opérationnelle, leur application à des problèmes réels de grande taille présente encore de nombreux défis. Cette thèse propose des améliorations méthodologiques et algorithmiques de méthodes de décomposition. Notre approche est basée sur l'analyse convexe et l'optimisation non-différentiable. Dans la décomposition par les contraintes (ou relaxation lagrangienne) du problème de planification de production électrique, même les sous-problèmes sont trop difficiles pour être résolus exactement. Mais des solutions approchées résultent en des prix instables et chahutés. Nous présentons un moyen simple d'améliorer la structure des prix en pénalisant leurs oscillations, en utilisant en particulier une régularisation par variation totale. La consistance de notre approche est illustrée sur des problèmes d'EDF. Nous considérons ensuite la décomposition par les variables (ou de Benders) qui peut avoir une convergence excessivement lente. Avec un point de vue d'optimisation non-différentiable, nous nous concentrons sur l'instabilité de l'algorithme de plans sécants sous-jacent à la méthode. Nous proposons une stabilisation quadratique de l'algorithme de Benders, inspirée par les méthodes de faisceaux en optimisation convexe. L'accélération résultant de cette stabilisation est illustrée sur des problèmes de conception de réseau et de localisation de plates-formes de correspondance (hubs). Nous nous intéressons aussi plus généralement aux problèmes d'optimisation convexe non-différentiable dont l'objectif est coûteux à évaluer. C'est en particulier une situation courante dans les procédures de décomposition. Nous montrons qu'il existe souvent des informations supplémentaires sur le problème, faciles à obtenir mais avec une précision inconnue, qui ne sont pas utilisées dans les algorithmes. Nous proposons un moyen d'incorporer ces informations incontrôlées dans des méthodes classiques d'optimisation convexe non-différentiable. Cette approche est appliquée avec succès à desproblèmes d'optimisation stochastique. Finalement, nous introduisons une stratégie de décomposition pour un problème de réaffectation de machines. Cette décomposition mène à une nouvelle variante de problèmes de conditionnement vectoriel (vectorbin packing) où les boîtes sont de taille variable. Nous proposons des heuristiques efficaces pour ce problème, qui améliorent les résultats de l'état de l'art du conditionnement vectoriel. Une adaptation de ces heuristiques permet de construire des solutions réalisables au problème de réaffectation de machines de Google. / Decomposition methods are an application of the divide and conquer principle to large-scale optimization. Their idea is to decompose a given optimization problem into a sequence of easier subproblems. Although successful for many applications, these methods still present challenges. In this thesis, we propose methodological and algorithmic improvements of decomposition methods and illustrate them on several operations research problems. Our approach heavily relies on convex analysis and nonsmooth optimization. In constraint decomposition (or Lagrangian relaxation) applied to short-term electricity generation management, even the subproblems are too difficult to solve exactly. When solved approximately though, the obtained prices show an unstable noisy behaviour. We present a simple way to improve the structure of the prices by penalizing their noisy behaviour, in particular using a total variation regularization. We illustrate the consistency of our regularization on real-life problems from EDF. We then consider variable decomposition (or Benders decomposition), that can have a very slow convergence. With a nonsmooth optimization point of view on this method, we address the instability of Benders cutting-planes algorithm. We present an algorithmic stabilization inspired by bundle methods for convex optimization. The acceleration provided by this stabilization is illustrated on network design andhub location problems. We also study more general convex nonsmooth problems whose objective function is expensive to evaluate. This situation typically arises in decomposition methods. We show that it often exists extra information about the problem, cheap but with unknown accuracy, that is not used by the algorithms. We propose a way to incorporate this coarseinformation into classical nonsmooth optimization algorithms and apply it successfully to two-stage stochastic problems.Finally, we introduce a decomposition strategy for the machine reassignment problem. This decomposition leads to a new variant of vector bin packing problems, where the bins have variable sizes. We propose fast and efficient heuristics for this problem that improve on state of the art results of vector bin packing problems. An adaptation of these heuristics is also able to generate feasible solutions for Google instances of the machine reassignment problem.

Made available in DSpace on 2016-06-02T19:50:05Z (GMT). No. of bitstreams: 1 2552.pdf: 3720513 bytes, checksum: fef1e4e66d1cfc987b00426d1ca179d2 (MD5) Previous issue date: 2009-04-24 / The main concern of this work is related to the development of an aggregate production planning model of a cooperative society of sugar and alcohol milling companies. This mathematical model is based on a hierarquical approach between the annual planning problem of the cooperative and the tactical planning horizon of the sugarcane mills. In the cooperative level the main questions are related to the allocation of production goals to each mill and the management of inventory and dynamic demands. In the milling companies level a process selection model aims at helping the decision makers to determine the quantity of sugarcane crushed, the selection of sugarcane suppliers, the selection of sugarcane transport system suppliers, the selection of industrial process used in the sugar, alcohol, molasses and energy production. Besides that, this work presents an analysis of the impact of uncertainties in the aggregate planning problem parameters, using robust optimization techniques. To solve the linear and mixed integer mathematical problem found in this modeling, we applied a state of the art modelling language with an optimization solver. A case study was developed in a cooperative of sugar and alcohol milling companies located in the state of Alagoas- Brazil and at Santa Clotilde mill, located in the city of Rio Largo-AL. The results of this case study helped us to verify the applicability of the proposed models in the aggregate production planning of the studied organizations. Computational results are presented and analyzed with real data application. / O objetivo deste trabalho é desenvolver modelos de programação matemática para o planejamento agregado da produção em usinas cooperadas do setor sucroenergético. Os modelos desenvolvidos devem considerar a relação hierárquica existente entre o planejamento anual de toda a cooperativa e o planejamento tático de safra de uma das usinas cooperadas. No nível de decisão da cooperativa o modelo deve indicar a meta de produção de cada usina e definir a política de estocagem e de atendimento da demanda. No nível de decisão da usina o modelo deve sugerir a quantidade de cana-de-açúcar colhida por fornecedor, a quantidade de cana transportada por prestador de serviço, a seleção dos processos de produção de açúcar, álcool, melaço e energia elétrica. Além disso, esta tese explora a aplicação de técnicas de otimização robusta para tratar incertezas inerentes aos parâmetros utilizados no processo decisório da cooperativa e de cada usina. Para resolver os modelos de programação linear e programação inteira mista, utilizou-se uma linguagem de modelagem algébrica e um solver de última geração de programação matemática. Um estudo de caso foi realizado na cooperativa regional dos produtores de açúcar e álcool do estado de Alagoas e na usina cooperada Santa Clotilde, localizada no município de Rio Largo-AL. Neste estudo, foi possível verificar a adequação dos modelos propostos quando aplicados para apoiar decisões envolvidas no planejamento agregado da produção das organizações estudadas. Resultados computacionais são apresentados e analisados, comparando o planejamento executado pelas empresas e os resultados obtidos com a modelagem.

Made available in DSpace on 2016-06-02T19:52:15Z (GMT). No. of bitstreams: 1 DissRPOP.pdf: 1876402 bytes, checksum: 300c17ebe260a12d247647b5184b3700 (MD5) Previous issue date: 2006-02-22 / The main concern of this work is related with the development of an aggregate production planning model of a sugar and alcohol milling company. This mathematical model is based on the process selection model and the production lotsizing model, and aims to help the decision makers in the production planning and control process of determining the quantity of sugarcane crushed, the selection of sugarcane suppliers, the selection of sugarcane transport system suppliers, the selection of industrial process used in the sugar, alcohol and molasses production and the storage decisions related with this final products. The decisions are taken in a week bases and the planning horizon is the whole sugarcane harvesting season. To solve the mixed integer mathematical problem found in this modeling, we applied the GAMS modeling language and the CPLEX solver. A case study was developed in a sugar and alcohol milling company located in Rio Largo, state of Alagoas, Brazil. The results of this case study helped us to verify the applicability of the proposed model in the aggregate production planning of a milling company. Computational results are presented for some scenarios of one illustrative example and some scenarios in a real data application. / O objetivo deste trabalho é desenvolver um modelo de planejamento agregado da produção que seja capaz de apoiar parte das principais decisões envolvidas no planejamento e controle da produção de uma usina de açúcar e álcool. Este modelo se baseia nos modelos clássicos de seleção de processos e dimensionamento de lotes de produção para representar o sistema de produção de açúcar, álcool, melaço e subprodutos, desde as principais decisões da etapa agrícola até as decisões de estocagem do produto final, passando pelas decisões da fase de corte, carregamento e transporte e, principalmente, as decisões de moagem e escolha do processo produtivo. As decisões são tomadas em períodos semanais e o horizonte de planejamento é o período de safra. Para resolver o modelo de programação linear inteira mista resultante, utilizou-se a linguagem de modelagem GAMS e o solver CPLEX. Um estudo de caso foi realizado numa usina de açúcar e álcool localizada no município de Rio Largo, estado de Alagoas. Neste estudo de caso, foi possível verificar a adequação do modelo proposto quando aplicado às decisões envolvidas no planejamento agregado da produção de uma empresa deste tipo. Resultados computacionais são apresentados, resolvendo-se alguns cenários de um exemplo ilustrativo e outros cenários de uma aplicação com dados reais.

Submitted by Rafael Ramos de Souza null (rr.souza@live.com) on 2016-08-09T15:45:13Z No. of bitstreams: 1 VERSÃO_ENTREGUE.pdf: 1452852 bytes, checksum: ae6aa21d2282113ac3abaade8414218e (MD5) / Approved for entry into archive by Ana Paula Grisoto (grisotoana@reitoria.unesp.br) on 2016-08-11T12:16:39Z (GMT) No. of bitstreams: 1 souza_rr_me_bauru.pdf: 1452852 bytes, checksum: ae6aa21d2282113ac3abaade8414218e (MD5) / Made available in DSpace on 2016-08-11T12:16:39Z (GMT). No. of bitstreams: 1 souza_rr_me_bauru.pdf: 1452852 bytes, checksum: ae6aa21d2282113ac3abaade8414218e (MD5) Previous issue date: 2016-06-10 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / O problema de Fluxo de Potência Ótimo tem por objetivo a otimização de um critério de desempenho elétrico sujeito ao atendimento das demandas de potência ativa e reativa em cada barra e de restrições técnico-operacionais dos sistemas de geração e transmissão. É um problema de otimização, não-linear, não-convexo e de grande porte. Neste trabalho é explorado o problema de Fluxo de Potência Ótimo Reativo com o objetivo de minimizar as perdas de potência ativa na transmissão e para resolvê-lo é proposto um método primal-dual de pontos interiores/exteriores barreira logarítmica modificada com estratégias de teste quadrático e determinação de direções de busca combinadas. O teste quadrático é proposto como alternativa ao procedimento de Cholesky na verificação da positividade da matriz hessiana do problema, que, se definida positiva, garante direções de descida para o método. As novas direções de busca são determinadas através de combinações das direções dos procedimentos previsor e corretor, determinadas através da análise das condições de complementaridade das variáveis primais e duais do problema. O método proposto foi implementado em Matlab e aplicado aos sistemas elétricos 9 e 39 barras e aos sistemas IEEE 14, 30, 57 e 118 barras. O desempenho do método com as estratégias propostas é avaliado em termos do número de iterações e do tempo computacional. Os resultados são promissores e permitem a aplicação do presente método, com as estratégias propostas, para resolver o problema de Fluxo de Potência Ótimo Reativo com maior dimensão do que os sistemas testados. / The reactive optimal power flow problem is concerned with the optimization of a specific criterion associated with the transmission system while enforcing the power balance in each transmission bus, as well as operational and physical constraints associated with generation and transmission systems. It is a nonlinear, non-convex and large optimization problem. In this work we consider the active losses minimization in the transmission system as a criterion for the optimal power flow problem. The solution of the problem is investigated by proposing a modified log-barrier primal-dual interior/exterior point method with a quadratic test strategy and new search direction procedures. The quadratic test is proposed as an alternative strategy to the Cholesky procedure for calculating the positivity of the Hessian matrix of the problem.The new search directions investigated in the paper are determined by combining the search directions calculated in the predictor and corrector steps, respectively, and also by using information associated with the complementarity conditions. The method proposed is implemented in Matlab and applied to solving the reactive optimal power flow problem for 9 and 39-bus systems, as well as for the IEEE 14, 30, 57 and 118-bus test systems. The performance of the method with the proposed strategies for search directions is evaluated in terms of the number of iterations and computational times. The results are promising and allow the application of the present method with the proposed search strategies for solving problems of larger dimensions.

Submitted by Renata Lopes (renatasil82@gmail.com) on 2016-07-01T11:43:52Z No. of bitstreams: 1 germandavidyagimoromisato.pdf: 4216499 bytes, checksum: a1b6dec404f94fd91a0a919755636775 (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2016-07-13T16:00:06Z (GMT) No. of bitstreams: 1 germandavidyagimoromisato.pdf: 4216499 bytes, checksum: a1b6dec404f94fd91a0a919755636775 (MD5) / Made available in DSpace on 2016-07-13T16:00:06Z (GMT). No. of bitstreams: 1 germandavidyagimoromisato.pdf: 4216499 bytes, checksum: a1b6dec404f94fd91a0a919755636775 (MD5) Previous issue date: 2012-08-02 / CAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Este trabalho tem como objetivo apresentar uma nova metodologia baseada em Programação Dinâmica Dual Determinística (PDDD) para o cálculo da Energia Firme de sistemas energéticos. A Energia Firme tem uma relação direta com os certificados de energia garantida atribuídos às usinas hidráulicas, os quais representam o limite superior para os contratos de energia estabelecidos com os consumidores (distribuidores e consumidores livres). Neste contexto, este trabalho possui uma importância relevante para o cenário atual do Setor Elétrico Brasileiro (SEB). Os resultados são comparados com aqueles obtidos pela metodologia em vigor no SEB, o qual é baseado em métodos heurísticos. / The objective of this work is to introduce a new methodology based in The Deterministic Dual Dynamic Programming (DDDP) to calculate the firm energy of energetic systems. The firm energy is directly related to the guaranteed energy certificates assigned to hydraulic power plants. These energy certificates represent the limits of energy contracts that can be established with consumers (energy distributors and free consumers). In this context, this work has a relevant importance to the current scenario of the Brazilian Electric Sector (BES). The results are compared to those obtained by the BES approved computational model based in heuristic methods.

[pt] O objetivo desta dissertação é desenvolver uma metodologia para oferta estratégia de uma empresa geradora em ambiente de mercado com múltiplas usinas hidrelétricas, levando em consideração múltiplos estágios e a incerteza nas afluências, e ilustrar a aplicação da mesma em sistemas realistas. Mostra-se inicialmente que o problema de oferta estratégica pode ser formulado como uma recursão de programação dinâmica estocástica (PDE), onde as variáveis de estado são os níveis de armazenamento dos reservatórios no início de cada estágio e as afluências observadas nos estágios anteriores. Entretanto, a dificuldade computacional dos algoritmos de PDE restringe sua aplicação a sistemas com poucos reservatórios, limitando bastante a aplicação da técnica a sistemas realistas. Assim, a abordagem proposta nesta dissertação é estender a metodologia de programação dinâmica dual estocástica (PDDE), até então aplicada a problemas de minimização de custos, ao problema de otimização da oferta. Isto é feito através de dois passos principais: (i) Uso de uma estratégia de oferta por quantidade somente (análogo a um modelo de Cournot em problemas de equilíbrio econômico) e (ii) a recursão de PDDE, que por ser baseada numa aproximação por hiperplanos requer que o problema seja convexo, o que não ocorre necessariamente no caso da oferta estratégica. A abordagem proposta consiste em aproximar a cada estágio a função de benefício futuro (FBF) por sua envoltória côncava (concave hull). Com isso, a técnica de PDDE pode ser aplicada para resolver o problema de ofertas multi- estágio e estocástico de uma empresa hidroelétrica com múltiplas usinas. Exemplos e estudos de caso serão ilustrados com os sistemas reais da Romênia e El Salvador, ilustrando a aplicabilidade da metodologia proposta em estudos e análises de poder de mercado. / [en] The objective of this work is to present a methodology for the strategic bidding (or bid-based) problem of a hydropower based company, taking into account multiple hydro plants, time-coupling, multiple inflow scenarios and illustrate its application for real case studies. It is initially show that the bid-based dispatch for a hydro plant can be formulated as a stochastic dynamic programming (SDP) recursion scheme, where the state variables are the storage levels and the past inflows. As widely known, the computational effort of the SDP algorithms restricts its applications for systems with just a few reservoirs, which is not the case of the real world systems. Therefore, the approach proposed in this thesis is to extend the stochastic dual dynamic programming (SDDP) scheme, usually applied to cost minimization problems, to the strategic bidding problem. This is done through two main steps: (i) use of a quantity-only bidding scheme (similar to the Cournot model of economic equilibria); (ii) SDDP recursion, which is based on a linear approximation by piecewise linear segments and thus requires that the underlying problem to be convex. This is not necessarily observed in the strategic bidding problem. Thus, the proposed approach consists in approximating, at each stage, the future benefit function (FBF) by its concave hull, which then assures that the SDDP scheme can be applied to solve the multi-stage and stochastic strategic bidding problem of a company with a portfolio of several hydro plants. The proposed approach is illustrated with examples and case studies from real hydro systems from Rumania and El Savador, where market power analysis will be presented.

Cette thèse a pour objet la résolution exacte d’un problème d’optimisation des flux de puissance (OPF) dans un réseau électrique. Dans l’OPF, on doit planifier la production et la répartition des flux de puissances électriques permettant de couvrir, à un coût minimal, la consommation en différents points du réseau. Trois variantes du problème de l’OPF sont étudiées dans ce manuscrit. Nous nous concentrerons principalement sur la résolution exacte des deux problèmes (OPF − L) et (OPF − Q), puis nous montrerons comment notre approche peut naturellement s’´étendre à la troisième variante (OPF − UC). Cette thèse propose de résoudre ces derniers à l’aide d’une méthode de reformulation que l’on appelle RC-OPF. La contribution principale de cette thèse réside dans l’étude, le développement et l’utilisation de notre méthode de résolution exacte RC-OPF sur les trois variantes d’OPF. RC-OPF utilise également des techniques de contractions de bornes, et nous montrons comment ces techniques classiques peuvent être renforcées en utilisant des résultats issus de notre reformulation optimale. / Alternative Current Optimal Power Flow (ACOPF) is naturally formulated as a non-convex problem. In that context, solving (ACOPF) to global optimality remains a challenge when classic convex relaxations are not exact. We use semidefinite programming to build a quadratic convex relaxation of (ACOPF). We show that this quadratic convex relaxation has the same optimal value as the classical semidefinite relaxation of (ACOPF) which is known to be tight. In that context, we build a spatial branch-and-bound algorithm to solve (ACOPF) to global optimality that is based on a quadratic convex programming bound.

Energy consumption rates have been dramatically increasing on a global scale within the last few decades. A significant role in this increase is subjected by the recent high temperature levels especially at summer time which caused a rapid increase in the air conditioning demands. Such phenomena can be clearly observed in developing countries, especially those in hot climate regions, where people depend mainly on conventional air conditioning systems. These systems often show poor performance and thus negatively impact the environment which in turn contributes to global warming phenomena. In recent years, the demand for urban or district cooling technologies and networks has been increasing significantly as an alternative to conventional systems due to their higher efficiency and improved ecological impact. However, to obtain an efficient design for district cooling systems is a complex task that requires considering a wide range of cooling technologies, various network layout configuration possibilities, and several energy resources to be integrated. Thus, critical decisions have to be made regarding a variety of opportunities, options and technologies. The main objective of this thesis is to develop a tool to obtain preliminary design configurations and operation patterns for district cooling energy systems by performing roughly detailed optimizations and further, to introduce a decision-making approach to help decision makers in evaluating the economic aspects and environmental performance of urban cooling systems at an early design stage. Different aspects of the subject have been investigated in the literature by several researchers. A brief survey of the state of the art was carried out and revealed that mathematical programming models were the most common and successful technique for configuring and designing cooling systems for urban areas. As an outcome of the survey, multi objective optimization models were decided to be utilized to support the decision-making process. Hence, a multi objective optimization model has been developed to address the complicated issue of decision-making when designing a cooling system for an urban area or district. The model aims to optimize several elements of a cooling system such as: cooling network, cooling technologies, capacity and location of system equipment. In addition, various energy resources have been taken into consideration as well as different solar technologies such as: trough solar concentrators, vacuum solar collectors and PV panels. The model was developed based on the mixed integer linear programming method (MILP) and implemented using GAMS language. Two case studies were investigated using the developed model. The first case study consists of seven buildings representing a residential district while the second case study was a university campus district dominated by non-residential buildings. The study was carried out for several groups of scenarios investigating certain design parameters and operation conditions such as: Available area, production plant location, cold storage location constraints, piping prices, investment cost, constant and variable electricity tariffs, solar energy integration policy, waste heat availability, load shifting strategies, and the effect of outdoor temperature in hot regions on the district cooling system performance. The investigation consisted of three stages, with total annual cost and CO2 emissions being the first and second single objective optimization stages. The third stage was a multi objective optimization combining the earlier two single objectives. Later on, non-dominated solutions, i.e. Pareto solutions, were generated by obtaining several multi objective optimization scenarios based on the decision-makers’ preferences. Eventually, a decision-making approach was developed to help decision-makers in selecting a specific solution that best fits the designers’ or decision makers’ desires, based on the difference between the Utopia and Nadir values, i.e. total annual cost and CO2 emissions obtained at the single optimization stages. / Die Energieverbrauchsraten haben in den letzten Jahrzehnten auf globaler Ebene dramatisch zugenommen. Diese Erhöhung ist zu einem großen Teil in den jüngst hohen Temperaturniveaus, vor allem in der Sommerzeit, begründet, die einen starken Anstieg der Nachfrage nach Klimaanlagen verursachen. Solche Ereignisse sind deutlich in Entwicklungsländern zu beobachten, vor allem in heißen Klimaregionen, wo Menschen vor allem konventionelle Klimaanlagensysteme benutzen. Diese Systeme verfügen meist über eine ineffiziente Leistungsfähigkeit und wirken sich somit negativ auf die Umwelt aus, was wiederum zur globalen Erwärmung beiträgt. In den letzten Jahren ist die Nachfrage nach Stadt- oder Fernkältetechnologien und -Netzwerken als Alternative zu konventionellen Systemen aufgrund ihrer höheren Effizienz und besseren ökologischen Verträglichkeit satrk gestiegen. Ein effizientes Design für Fernkühlsysteme zu erhalten, ist allerdings eine komplexe Aufgabe, die die Integration einer breite Palette von Kühltechnologien, verschiedener Konfigurationsmöglichkeiten von Netzwerk-Layouts und unterschiedlicher Energiequellen erfordert. Hierfür ist das Treffen kritischer Entscheidungen hinsichtlich einer Vielzahl von Möglichkeiten, Optionen und Technologien unabdingbar. Das Hauptziel dieser Arbeit ist es, ein Werkzeug zu entwickeln, das vorläufige Design-Konfigurationen und Betriebsmuster für Fernkälteenergiesysteme liefert, indem aureichend detaillierte Optimierungen durchgeführt werden. Zudem soll auch ein Ansatz zur Entscheidungsfindung vorgestellt werden, der Entscheidungsträger in einem frühen Planungsstadium bei der Bewertung städtischer Kühlungssysteme hinsichtlich der wirtschaftlichen Aspekte und Umweltleistung unterstützen soll. Unterschiedliche Aspekte dieser Problemstellung wurden in der Literatur von verschiedenen Forschern untersucht. Eine kurze Analyse des derzeitigen Stands der Technik ergab, dass mathematische Programmiermodelle die am weitesten verbreitete und erfolgreichste Methode für die Konfiguration und Gestaltung von Kühlsystemen für städtische Gebiete sind. Ein weiteres Ergebnis der Analyse war die Festlegung von Mehrzieloptimierungs-Modelles für die Unterstützung des Entscheidungsprozesses. Darauf basierend wurde im Rahmen der vorliegenden Arbeit ein Mehrzieloptimierungs-Modell für die Lösung des komplexen Entscheidungsfindungsprozesses bei der Gestaltung eines Kühlsystems für ein Stadtgebiet oder einen Bezirk entwickelt. Das Modell zielt darauf ab, mehrere Elemente des Kühlsystems zu optimieren, wie beispielsweise Kühlnetzwerke, Kühltechnologien sowie Kapazität und Lage der Systemtechnik. Zusätzlich werden verschiedene Energiequellen, auch solare wie Solarkonzentratoren, Vakuum-Solarkollektoren und PV-Module, berücksichtigt. Das Modell wurde auf Basis der gemischt-ganzzahlig linearen Optimierung (MILP) entwickelt und in GAMS Sprache implementiert. Zwei Fallstudien wurden mit dem entwickelten Modell untersucht. Die erste Fallstudie besteht aus sieben Gebäuden, die ein Wohnviertel darstellen, während die zweite Fallstudie einen Universitätscampus dominiert von Nichtwohngebäuden repräsentiert. Die Untersuchung wurde für mehrere Gruppen von Szenarien durchgeführt, wobei bestimmte Designparameter und Betriebsbedingungen überprüft werden, wie zum Beispiel die zur Verfügung stehende Fläche, Lage der Kühlanlage, örtliche Restriktionen der Kältespeicherung, Rohrpreise, Investitionskosten, konstante und variable Stromtarife, Strategie zur Einbindung der Solarenergie, Verfügbarkeit von Abwärme, Strategien der Lastenverschiebung, und die Wirkung der Außentemperatur in heißen Regionen auf die Leistung des Kühlsystems. Die Untersuchung bestand aus drei Stufen, wobei die jährlichen Gesamtkosten und die CO2-Emissionen die erste und zweite Einzelzieloptimierungsstufe darstellen. Die dritte Stufe war ein Pareto-Optimierung, die die beiden ersten Ziele kombiniert. Im Anschluss wurden nicht-dominante Lösungen, also Pareto-Lösungen, erzeugt, indem mehrere Pareto-Optimierungs-Szenarien basierend auf den Präferenzen der Entscheidungsträger abgebildet wurden. Schließlich wurde ein Ansatz zur Entscheidungsfindung entwickelt, um Entscheidungsträger bei der Auswahl einer bestimmten Lösung zu unterstützen, die am besten den Präferenzen des Planers oder des Entscheidungsträgers enstpricht, basierend auf der Differenz der Utopia und Nadir Werte, d.h. der jährlichen Gesamtkosten und CO2-Emissionen, die Ergebnis der einzelnen Optimierungsstufen sind.

Ein großer Anteil des Datenverkehrs in „Internet Service Provider“ (ISP)-Netzen wird heutzutage von „Content Delivery Networks“ (CDNs) verursacht. Betreiber von CDNs verwenden Lastverteilungsmechanismen um die Auslastung ihrer CDN-Infrastruktur zu vergleichmäßigen (Load Balancing). Dies geschieht ohne Abstimmung mit den ISP-Betreibern. Es können daher große Verkehrslastverschiebungen sowohl innerhalb eines ISP-Netzes, als auch auf den Verbindungsleitungen zwischen ISP-Netz und CDNs auftreten. In der vorliegenden Arbeit wird untersucht, welche nicht-kooperativen Möglichkeiten ein ISP hat, um Verkehrslastverschiebungen, welche durch Lastverteilungsmechanismen innerhalb eines CDNs verursacht werden, entgegenzuwirken bzw. abzumildern. Die Grundlage für diese Untersuchung bildet die Analyse des Serverauswahlverhaltens des YouTube-CDNs. Hierzu ist ein aktives Messverfahren entwickelt worden, um das räumliche und zeitliche Verhalten der YouTube-Serverauswahl bestimmen zu können. In zwei Messstudien wird die Serverauswahl in deutschen und europäischen ISP-Netzen untersucht. Auf Basis dieser Studien wird ein Verkehrsmodell entwickelt, welches die durch Änderungen der YouTube-Serverauswahl verursachten Verkehrslastverschiebungen abbildet. Das Verkehrsmodell wiederum bildet die Grundlage für die Bestimmung optimaler Routen im ISP-Netz, welche hohe Robustheit gegenüber CDN-bedingte Verkehrslastverschiebungen aufweisen (Alpha-robuste Routingoptimierung). Für die Lösung des robusten Routing-Optimierungsproblems wird ein iteratives Verfahren entwickelt sowie eine kompakte Reformulierung vorgestellt. Die Leistungsfähigkeit des Alpha-robusten Routings wird anhand von drei Beispielnetztopologien untersucht. Das neue Verfahren wird mit alternativen robusten Routingverfahren und einem nicht-robusten Verfahren verglichen. Neben der robusten Routingoptimierung werden in der Arbeit drei weitere Ideen für nicht-kooperative Methoden vorgestellt (BGP-, IP-Präix- und DNS-basierte Methode), um CDN-bedingten Verkehrslastverschiebungen entgegenzuwirken.

With the increasing demand for renewable energy sources, new systems are being developed to sustain future infrastructure, accommodating these new energy sources. One of the proposed solutions is to incorporate distributed energy resources to different households in order to provide local energy demands effectively. To enable large-scale integration of flexible energy resources, it is crucial to reduce end-user energy and power costs, which can be done by designing an optimization model objected to minimize the total electricity bill. In the scope of this Master thesis, the interest lies in investigating a control strategy to operate batteries, heat pumps, and other assets by producing the optimal setpoints using the designed optimization algorithm that takes, amongst others, market and weather data as well as customer behavior into account. The applied method for producing these setpoints is sensitivity analysis in linear programming, and heat pump scheduling has been investigated for performance evaluation of this technique. The results show that applying this method produces the optimal setpoints over the non-controllable electricity load range by utilizing a low number of optimizations, i.e. high computation-efficiency, and high accuracy. Consequently, the controller by having the given setpoints as the input can easily adjust the heat pump output power based on the real-time non-controllable electricity load without creating any peaks and extra costs for the customers. / Med en ökad efterfrågan på förnybara energikällor utvecklas nya system för att upprätthålla framtida infrastruktur vilket kommer säkra dessa nya energikällor. En av de föreslagna lösningarna är att integrera distribuerade energiresurser till olika hushåll för att effektivt kunna tillgodose lokala energikrav. För att möjliggöra en storskalig integrering av flexibla energiresurser det avgörande är att man kan minska slutkundens energi och effektkostnader. Detta kan nås genom att utforma en optimeringsmodell av problemet som tar hänsyn till olika resourses begränsningar osv. för att minska elkosnaden hos slutkunden. Syftet med detta examensarbete är att undersöka en kontrollstrategi för att använda batterier, värmepumpar och andra tillgångar på ett optimalt sätt, genom att producera de optimala börvärdena med hjälp av den utformade optimeringsalgoritmen som tar hänsyn till bland annat marknads och väderdata samt kund beteende. För att producera dessa börvärden användes methoden känslighetsanalys som är en del inom linjär programmering och fokus har varit styrningen av värmepumpar. Resultaten visar att tillämpningen av denna metod leder till att de optimala börvärdena över det icke-kontrollerbara elektriska lasten erhålles, med ett lågt antal optimeringar, dvs att metoden har hög beräknings-effektivitet samt noggrannhet. Följaktligen kan regulatorn med de givna börvärdena som ingång enkelt justera värmepumpens utgångseffekt baserat på realtids icke-kontrollerbar elektriska lasten, utan att skapa några toppar och extra kostnader för kunderna.