O objetivo de um sistema elétrico de potência é gerar energia elétrica e fornecer continuamente esta energia aos usuários finais dentro de padrões de qualidade aceitáveis. Neste contexto, diferentes aplicações usando algoritmos genéticos (AGs) para resolver problemas relacionados a sistemas elétricos de potência são apresentadas neste trabalho. A estimação de harmônicos, estudos relativos aos relés de freqüência e aplicações da proteção de distância são os assuntos investigados nesta tese. Para análise harmônica foram consideradas freqüências em sistemas de potência de até a vigésima quinta ordem, as quais foram estimadas pelos AGs e comparadas àquelas resultantes da análise quando da aplicação da Transformada Discreta de Fourier (TDF). Com respeito aos relés de freqüência, o objetivo foi estimar a amplitude, freqüência e ângulo de fase para diversas situações de formas de ondas utilizando uma nova estrutura que possa ser implementada dispondo em FPGAs (Field Programmable Gate Array). Finalmente, aplicados à proteção de distância, o principal propósito dos AGs foi identificar os fasores fundamentais da tensão e corrente e, então, calcular a impedância da linha medida pelo relé de distância associado. Cabe ressaltar que estes resultados também foram comparados ao método clássico da TDF. Todas as três abordagens foram formuladas como problemas de otimização, com objetivo de minimizar o erro estimado das variáveis envolvidas. Os resultados dessas aplicações mostram que o desempenho global dos Algoritmos Genéticos foi altamente satisfatório considerando-se as precisões das respostas encontradas. / The aim of electric power systems is to generate electricity and to deliver it continuously to the end-user in an acceptable standard. Therefore, different applications using genetic algorithms (GAs) to solve electric power system problems are presented in this work. Harmonic estimation, frequency relays and distance protection are the subjects investigated in this thesis. Concerning harmonic analysis, the ones up to the 25th order were considered in this work. They were estimated by GAs and compared to results obtained from the Discrete Fourier Transform (DFT) technique. Concerning frequency relays, the main objective was to estimate the amplitude, frequency and phase angle of waveforms using a new structure capable to be implemented in a FPGA (Field Programmable Gate Array). Finally, concerning distance protection, the main purpose was use the GAs to identify the fundamental voltage and current phasors and to calculate the line impedance seen by the distance relay. These results were also compared to those obtained from the classic DFT method. These three problems were formulated as an optimization problem, and the objective was to minimize the estimation error. The results from these applications show that the global performance of the genetic algorithms was highly satisfactory concerning the precision of the responses.

A primeira parte deste trabalho lida com o problema de posicionar um grupo de agentes retransmissores (relays) de forma a dar conectividade a um segundo grupo de agentes ativos (pursuers). A primeira abordagem apresentada consiste em modelar o cenário como um problema de programação quadrática (PPQ) com restrições lineares, usando uma estrutura de conectividade fixa. Para resolver o modelo proposto, foi implementada uma rede neural recorrente a qual converge rapidamente para a solução ótima do problema, mesmo em instâncias razoavelmente grandes. Como forma de avaliação, realizou-se um comparativo entre o solver de PPQ da plataforma Matlab e a rede proposta, também implementada na mesma plataforma. Na segunda abordagem foi proposto o uso de uma estimativa da conectividade algébrica do grafo de proximidade gerado pela rede, para direcionar o grupo de relays e pursuers, usando-se apenas as informações da vizinhança de cada agente. Nesta abordagem a estrutura do grafo é dinâmica, além disso, como proposta de paralelização, a esta solução distribuída foi acoplada um algoritmo de escalonamento por reversão de arestas (SER). Além do mais, as metaheurísticas Simulated Annealing, Genetic Algorithms, Particle Swarm Optimization e Backtracking Search Algorithm foram implementadas como alternativas de soluções ao problema. As soluções são avaliadas em um cenário de perseguição de alvos, os quais podem possuir comportamentos reativos (tais como fugir dos perseguidores). A segunda parte deste trabalho investiga o problema de rastreamento de múltiplos objetos em tempo real. Como solução, foi proposto um algoritmo que se baseia em memória de curto e longo prazo usando redes neurais sem peso. / The first part of this work deals with the problem of positioning a swarm of relay agents with the objective of providing connectivity to a second group of active agents (pursuers). The first approach consists of modelling the considered scenario as a quadratic programming problem (QP) with linear restrictions, using a fixed graph structure. In order to solve such model, a recurrent neural network is proposed with fast convergence rate to the optimal solution, even with reasonably big size instances. In addition, a comparison with the Matlab QP solver has been conducted in some experimental simulations. In the second approach, it is proposed an estimation of the algebraic connectivity of the underlying graph generated by the network. Over this estimation, it is proposed a metric to direct the group of relays and pursuers, using only local neighbourhood information of each agent. On this approach, the graph structure is dynamic and it is also proposed the use of the schedule by edge reversal (SER) as a solution to ordering the parallelization of the robot positioning computation. Moreover, the meta-heuristics Simulated Annealing, Genetic Algorithms, Particle Swarm Optimization and Backtracking Search Algorithm have been applied as alternative solution providers. The proposed solutions have been applied in a target pursuit scenario, for which the targets are deployed in different spots and may have some reactive behaviours (such as escape from the pursuers).The second part of this work investigates the visual tracking of shape shift objects in real time. As a solution, it is proposed a short- and long-time memories tracker which uses a weightless neural network for training and retraining the objects patterns.

O objetivo de um sistema elétrico de potência é gerar energia elétrica e fornecer continuamente esta energia aos usuários finais dentro de padrões de qualidade aceitáveis. Neste contexto, diferentes aplicações usando algoritmos genéticos (AGs) para resolver problemas relacionados a sistemas elétricos de potência são apresentadas neste trabalho. A estimação de harmônicos, estudos relativos aos relés de freqüência e aplicações da proteção de distância são os assuntos investigados nesta tese. Para análise harmônica foram consideradas freqüências em sistemas de potência de até a vigésima quinta ordem, as quais foram estimadas pelos AGs e comparadas àquelas resultantes da análise quando da aplicação da Transformada Discreta de Fourier (TDF). Com respeito aos relés de freqüência, o objetivo foi estimar a amplitude, freqüência e ângulo de fase para diversas situações de formas de ondas utilizando uma nova estrutura que possa ser implementada dispondo em FPGAs (Field Programmable Gate Array). Finalmente, aplicados à proteção de distância, o principal propósito dos AGs foi identificar os fasores fundamentais da tensão e corrente e, então, calcular a impedância da linha medida pelo relé de distância associado. Cabe ressaltar que estes resultados também foram comparados ao método clássico da TDF. Todas as três abordagens foram formuladas como problemas de otimização, com objetivo de minimizar o erro estimado das variáveis envolvidas. Os resultados dessas aplicações mostram que o desempenho global dos Algoritmos Genéticos foi altamente satisfatório considerando-se as precisões das respostas encontradas. / The aim of electric power systems is to generate electricity and to deliver it continuously to the end-user in an acceptable standard. Therefore, different applications using genetic algorithms (GAs) to solve electric power system problems are presented in this work. Harmonic estimation, frequency relays and distance protection are the subjects investigated in this thesis. Concerning harmonic analysis, the ones up to the 25th order were considered in this work. They were estimated by GAs and compared to results obtained from the Discrete Fourier Transform (DFT) technique. Concerning frequency relays, the main objective was to estimate the amplitude, frequency and phase angle of waveforms using a new structure capable to be implemented in a FPGA (Field Programmable Gate Array). Finally, concerning distance protection, the main purpose was use the GAs to identify the fundamental voltage and current phasors and to calculate the line impedance seen by the distance relay. These results were also compared to those obtained from the classic DFT method. These three problems were formulated as an optimization problem, and the objective was to minimize the estimation error. The results from these applications show that the global performance of the genetic algorithms was highly satisfactory concerning the precision of the responses.

Current differential protection principle is superior in terms of sensitivity and speed of operation in comparison with other protection principle used in power systems. From the last five decades, various current differential protection schemes are widely used to protect busbars, transformers, and short-transmission lines. The deployment of high capacity microwave and optical fiber technologies redefined the line protection systems by facilitating the use of current differential protection schemes for long transmission lines. The common application issue of these schemes is mis-operation due to current transformer (CT) saturation during close-in external faults. Moreover, transformer differential protection schemes face mis-trip due to inrush current during energization. The techniques presented in the literature to address those issues, de-sensitize protection function and increase the time of operation. A comprehensive fault discrimination algorithm and an inrush current detection algorithm are highly demanded for current differential protection schemes. The purpose of this dissertation is to optimize the performance of differential schemes applied to protect busbar, transformer and line. This research derives the mathematical model of saturated secondary current of CT and introduces the concept of Partial Operating Current (POC). Based on these mathematical developments, the characteristics of POC are identified for all three types of differential zones like busbar, transformer and line protection. A new inrush current blocking algorithm is developed for transformer differential protection. A new time-domain CT saturation detection algorithm is also proposed. Based on these new developments, three separate differential schemes are designed for busbar, transformer, and line protection, respectively. The proposed schemes provide complete immunity against the mis-operations due to CT saturation during close-in external faults and transformer inrush current without sacrificing the sensitivity for internal faults. The speed of operation is also improved. The model for each scheme is built in Matlab platform and the performance is validated using the test system simulated in Electro-Magnetic Transient Program (EMTP) for all possible fault scenarios. Documented results show the improved performance of the proposed schemes when compared to traditional differential schemes in terms of reliability, sensitivity, selectivity, and speed

<p>The use of relays is seen as a promising way to extend cell coverage and increase rates in LTE Advanced networks. Instead of increasing the number of base stations (BS), relays with lower cost could provide similar gains. A relay will have a wireless link to the closest BS as only connection to the core network and will cover areas close to the cell edge or other areas with limited rates.</p><p>Performing transmissions in several hops (BS-relay & relay-user) requires more radio resources than using direct transmission. This thesis studies how the available radio resources should be allocated between relays and users in order to maximize throughput and/or fairness. Time and frequency multiplexed backhaul is investigated under a full buffer traffic assumption. It is shown that the system will be backhaul limited and that the two ways of multiplexing will perform equally when maximising throughput and/or fairness. The analysis results in a set of throughput/fairness suboptimal solutions, dependant on how many relays are used per cell. The results are verified by simulations, which also show the limiting effects on throughput caused by interference between relays.</p><p>It is also analysed how the resource allocation should be done given non-fullbuffer traffic. A resource allocation that minimises packet delay given a certain number of relays per cell is presented. The analysis is based on queuing theory.</p><p>Finally some different schedulers and their suitability for relay networks are discussed. Simulation results are shown, comparing the throughput and fairness of Round Robin, Weighted Round Robin, Proportional Fairness and Weighted Proportional Fairness schemes. It is shown that allocating the resource among the relays according to the number of users served by the relays improves the fairness.</p>

<p>We are very thirsty in terms of everything to fulfil our needs in a sophisticated way, and this leads me choose the so called master thesis titled “Multihop Concept in Cellular Systems”.</p><p>This thesis introduces an approach towards the integration of relaying or multihop scheme in the next generation of cellular networks. In a multihop cellular architecture, the users send their data to the base station via relay station or with direct communication to the base station. These relay stations can either be the nomadic, fixed at specific location or users’ mobile station (i.e. mobile relay station). The main objective of this paper is to compare the difference between the relaying network architecture with different channel bandwidth as well as their performance gain. For this we integrate the relay station into conventional cellular networks using IEEE 802.16j (One of the standard introduced relay station concept in WiMAX) OFDMA (Orthogonal Frequency Division Multiple Access is a transmission technique that is based on many orthogonal subchannels (set of carriers) that transmits simultaneously). The results show that under certain conditions the throughput and coverage of the system has been increased with the introduction of the relay station in to cellular base station zone.</p>

<p>Syftet med denna uppsats är att ta reda på hur vanligt förekommande smilisen är generellt på en chatt och vilka sorter som används. Som ett komplement till detta vill jag även ta reda på om enskilda personer använder många smilisar och hur många sorter dessa representerar.</p><p>Först redogör jag för tidigare forskning på området och då främst när det gäller ickeverbala signaler och deras motsvarande chattstrategier. Dessutom kategoriserar jag smilisar generellt efter orientering, föreställning, funktion och variation och får på detta sätt en grund att stå på i materialanalys och diskussion. Som motivering till mitt materialval har jag även placerat in olika chattar i tre olika kategorier: parinriktade, gruppinriktade och virtuella världar.</p><p>Mitt material består av sju dagars loggar från IRC-kanalen #Sweden och omfattar sammanlagt 16582 faktiska inlägg. För att inte påverka resultaten har jag insamlat detta i rollen som spion.</p><p>Det jag kommer fram till är att smilisar förekommer i 14,5 % av inläggen. Dessa är orienterade både som liggande och stående med en klar övervikt på den förra. Alla tre föreställningar är representerade men föremål och händelser står bara för 1 % av totalen. Även samtliga funktioner används men bara i orienteringen liggande. Slutligen har också variationen av de liggande smilisarna visat sig vara stor.</p><p>Resultaten komna ur analyser av tre chattare visar sig vara ett bra komplement. En mycket aktiv chattare visar sig endast använda ett fåtal smilisar av ett mycket litet antal sorter. Mindre aktiva chattare använder däremot en större andel smilisar.</p><p>Smilisen har alltså visat sig vara ett vanligt och ett värdefullt komplement till den reella texten. Den liggande smilisen är till synes helt klart användbarast då den är både enkel att skriva och lätt att förstå. Jag tycker mig också ana att de smilisar som används oftast kan förknippas med genrestilen medan de som endast förekommer sällsynt snarare är tecken på personlig stil.</p><p>Slutligen argumenterar jag för chattens och sålunda också smilisens fortsatta överlevnad.</p>

Thesis (Ph. D.)--Ohio State University, 2006. / Title from first page of PDF file. Includes bibliographical references (p. 175-177).

No description available.

This thesis studies a two-way relay network (TWRN), which consists of two single antenna source nodes and a multi-antenna relay node. The source nodes exchange information via the assistance of the relay node in the middle. The relay scheme in this TWRN is amplify-and-forward (AF) based analog network coding (ANC). A robust beamforming matrix optimization algorithm is presented here with the objective to minimize the transmit power at the relay node under given signal to interference and noise ratio (SINR) requirements of source nodes. This problem is first formulated as a non-convex optimization problem, and it is next relaxed to a semi-definite programming (SDP) problem by utilizing the S-procedure and rank-one relaxation. This robust beamforming optimization algorithm is further validated in a MATLAB-based orthogonal frequency-division multiplexing (OFDM) MIMO two-way relay simulation system. To better investigate the performance of this beamforming algorithm in practical systems, synchronization issues such as standard timing offset (STO) and carrier frequency offset (CFO) are considered in simulation. The transmission channel is modeled as a frequency selective fading channel, and the source nodes utilize training symbols to perform minimum mean-square error (MMSE) channel estimation. BER curves under perfect and imperfect synchronization are presented to show the performance of TWRN with ANC. It is shown that the outage probability of robust beamforming algorithm is tightly related to the SINR requirements at the source nodes, and the outage probability increases significantly when the SINR requirements are high.