大地起伏值為正高與橢球高的差異量，如果取得高精度的大地起伏值，可以利用衛星定位測量施測橢球高並計算得到高精度的正高，其成本低廉，可望取代傳統的水準測量。而大地起伏值可以分為幾何法或重力法的大地起伏值，其中幾何法的大地起伏值計算方法簡易且精度高，可以利用曲面擬合方法取得之。但是幾何法的大地起伏值會受到地形起伏的影響，大範圍的曲面擬合會降低其精度。台灣的地形起伏大，難以進行大範圍曲面擬合。 於是本研究利用環域方法搜尋待測點位鄰近的水準點參與曲面方程式擬合大地起伏，試圖找到最適合的大地起伏擬合範圍。成果顯示：環域的範圍從10公里至30公里，利用二次曲面方程式擬合大地起伏在台灣平地區域能夠達到預測精度與內部精度同時低於5公分。另外由於衛星定位測量橢球高的誤差較高，需進行資料品質評估並進行粗差偵測。針對粗差偵測提出新的方法，利用最佳化演算法中的量子行為粒子群演算法計算最小二乘平差法中的權矩陣，期望能夠將粗差觀測量的權重降低，達到粗差偵測的效果。成果顯示最佳化權矩陣演算法，能夠將粗差對平差系統的影響量降到最低。 本研究建立一套台灣地區的大地起伏擬合作業程序：利用環域搜尋鄰近水準點、曲面方程式及環域範圍選擇與資料的粗差偵測，可獲得高品質的大地起伏。 / The geoidal undulation is the difference of ellipsoid height and orthometric height. We can obtain high accuracy of orthometric height by existing high accuracy of geoidal undulation and the ellipsoidal height measuring by GPS. It expected to replace the traditional leveling survey due to the less cost. This study uses buffer method to search the leveling benchmarks around the object point, attempts to find the proper range of fitting geoidal undulation to curve surface. Experimental results shows that it can archive 5cm level on both prediction error and internal precision by fitting geoidal undulation on 2nd curve surface model where the buffer range is from 10 km to 30 km. In this study, also uses the quantum-behaved particle swarm optimization to calculate the weight matrix of least square adjustment, the purpose is to down-weighting the suspicious outlier, and detect the outlier. Experimental results shows that the optimal weight matrix algorithm can reduce the influence of outlier. This study establish a procedure of fitting geoidal undulation: using buffer analysis to search the adjacent leveling benchmark, selecting the proper buffer range and surface equation and detecting outlier in data.

Pattern discovery in protein structures is a fundamental task in computational biology, with important applications in protein structure prediction, profiling and alignment. We propose a novel approach for pattern discovery in protein structures using Particle Swarm-based flying windows over potentially promising regions of the search space. Using a heuristic search, based on Particle Swarm Optimization (PSO) is, however, easily trapped in local optima due to the sparse nature of the problem search space. Thus, we introduce a novel fitness-based stagnation detection technique that effectively and efficiently restarts the search process to escape potential local optima. The proposed fitness-based method significantly outperforms the commonly-used distance-based method when tested on eight classical and advanced (shifted/rotated) benchmark functions, as well as on two other applications for proteomic pattern matching and discovery. The main idea is to make use of the already-calculated fitness values of swarm particles, instead of their pairwise distance values, to predict an imminent stagnation situation. That is, the proposed fitness-based method does not require any computational overhead of repeatedly calculating pairwise distances between all particles at each iteration. Moreover, the fitness-based method is less dependent on the problem search space, compared with the distance-based method. The proposed pattern discovery algorithms are first applied to protein contact maps, which are the 2D compact representation of protein structures. Then, they are extended to work on actual protein 3D structures and interaction networks, offering a novel and low-cost approach to protein structure classification and interaction prediction. Concerning protein structure classification, the proposed PSO-based approach correctly distinguishes between the positive and negative examples in two protein datasets over 50 trials. As for protein interaction prediction, the proposed approach works effectively on complex, mostly sparse protein interaction networks, and predicts high-confidence protein-protein interactions — validated by more than one computational and experimental source — through knowledge transfer between topologically-similar interaction patterns of close proximity. Such encouraging results demonstrate that pattern discovery in protein structures and interaction networks are promising new applications of the fast-growing and far-reaching PSO algorithms, which is the main argument of this thesis. / Thesis (Ph.D, Computing) -- Queen's University, 2014-04-21 12:54:03.37

Worldwide concern about the environmental problems and a possible energy crisis has led to increasing interest in clean and renewable energy generation. Among various renewable energy sources, wind power is the most rapidly growing one. Therefore, how to provide efficient, reliable, and high-performance wind power generation and distribution has become an important and practical issue in the power industry. In addition, because of the new constraints placed by the environmental and economical factors, the trend of power system planning and operation is toward maximum utilization of the existing infrastructure with tight system operating and stability margins. This trend, together with the increased penetration of renewable energy sources, will bring new challenges to power system operation, control, stability and reliability which require innovative solutions. Flexible ac transmission system (FACTS) devices, through their fast, flexible, and effective control capability, provide one possible solution to these challenges. To fully utilize the capability of individual power system components, e.g., wind turbine generators (WTGs) and FACTS devices, their control systems must be suitably designed with high reliability. Moreover, in order to optimize local as well as system-wide performance and stability of the power system, real-time local and wide-area coordinated control is becoming an important issue. Power systems containing conventional synchronous generators, WTGs, and FACTS devices are large-scale, nonlinear, nonstationary, stochastic and complex systems distributed over large geographic areas. Traditional mathematical tools and system control techniques have limitations to control such complex systems to achieve an optimal performance. Intelligent and bio-inspired techniques, such as swarm intelligence, neural networks, and adaptive critic designs, are emerging as promising alternative technologies for power system control and performance optimization. This work focuses on the development of advanced optimization and intelligent control algorithms to improve the stability, reliability and dynamic performance of WTGs, FACTS devices, and the associated power networks. The proposed optimization and control algorithms are validated by simulation studies in PSCAD/EMTDC, experimental studies, or real-time implementations using Real Time Digital Simulation (RTDS) and TMS320C6701 Digital Signal Processor (DSP) Platform. Results show that they significantly improve electrical energy security, reliability and sustainability.

Made available in DSpace on 2016-12-12T20:25:12Z (GMT). No. of bitstreams: 1 Maico Jeremias da Silva.pdf: 5502697 bytes, checksum: 292822b79334828b19d2d99087aae96b (MD5) Previous issue date: 2015-07-28 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / This work discusses the behaviour of the convection heat transfer coefficient in plane fins with variation of air flow velocity and fin spacing. The differential equation that governs heat transfer in fins is discretized by the finite volume method, which enables computation of the fin thermal characteristics, such as temperature distribution, heat transfer and fin efficiency. The determination of the convection coefficient, important parameter for thermal analysis, is performed by applying a heuristic optimization method, known as Particle Swarm Optimization, which combines data measured from experimental analysis conducted in wind tunnel with the aforementioned heat transfer numerical approximation of fins. / Este trabalho analisa os efeitos da velocidade do escoamento de ar e do espaçamento entre aletas no coeficiente de convecção forçada sobre as superfícies de aletas planas. As equações diferenciais que definem a transferência de calor em aletas são discretizadas pela técnica de volumes finitos possibilitando a determinação das características térmicas da aleta, tais como perfil de temperatura, calor transferido e eficiência térmica. A determinação do coeficiente de convecção, parâmetro fundamental para análises térmicas, é realizada mediante a aplicação de um método de otimização heurístico, conhecido como Método do Enxame de Partículas, que combina os dados obtidos da análise experimental realizada em túnel de vento com a aproximação numérica da troca de calor na aleta descrita acima

Made available in DSpace on 2014-12-17T15:47:45Z (GMT). No. of bitstreams: 1 GivanaldoRS.pdf: 1524067 bytes, checksum: d73e18e4ae3a0bffab7711efc808bffa (MD5) Previous issue date: 2006-05-19 / Combinatorial optimization problems have the goal of maximize or minimize functions defined over a finite domain. Metaheuristics are methods designed to find good solutions in this finite domain, sometimes the optimum solution, using a subordinated heuristic, which is modeled for each particular problem. This work presents algorithms based on particle swarm optimization (metaheuristic) applied to combinatorial optimization problems: the Traveling Salesman Problem and the Multicriteria Degree Constrained Minimum Spanning Tree Problem. The first problem optimizes only one objective, while the other problem deals with many objectives. In order to evaluate the performance of the algorithms proposed, they are compared, in terms of the quality of the solutions found, to other approaches / Os problemas de otimiza??o combinat?ria t?m como objetivo maximizar ou minimizar uma fun??o definida sobre um certo dom?nio finito. J? as metaheur?sticas s?o procedimentos destinados a encontrar uma boa solu??o, eventualmente a ?tima, consistindo na aplica??o de uma heur?stica subordinada, a qual tem que ser modelada para cada problema espec?fico. Este trabalho apresenta algoritmos baseados na t?cnica de otimiza??o por nuvem de part?culas (metaheur?stica) para dois problemas de otimiza??o combinat?ria: o Problema do Caixeiro Viajante e o Problema da ?rvore Geradora M?nima Restrita em Grau Multicrit?rio. O primeiro ? um problema em que apenas um objetivo ? otimizado, enquanto o segundo ? um problema que deve lidar com m?ltiplos objetivos. Os algoritmos propostos s?o comparados a outras abordagens para o mesmo problema em quest?o, em termos de qualidade de solu??o, a fim de verificar a efici?ncia desses algoritmos

Ces travaux de recherche portent sur la synthèse systématique et l’optimisation de correcteurs numériques à structure polynomiale RST par approches métaheuristiques. Les problèmes classiques de placement de pôles et de calibrage des fonctions de sensibilité de la boucle fermée RST sont formulés sous forme de problèmes d’optimisation multi-objectif sous contraintes pour lequel des algorithmes métaheuristiques de type NSGA-II, MODE, MOPSO et epsilon-MOPSO sont proposés et adaptés. Deux formulations du problème de synthèse RST ont été proposées. La première approche, formulée dans le domaine temporel, consiste à minimiser des indices de performance, de type ISE et MO, issus de la théorie de la commande optimale et liés essentiellement à la réponse indicielle du système corrigé. Ces critères sont optimisés sous des contraintes non analytiques définis par des gabarits temporels sur la dynamique de la boucle fermée. Dans la deuxième approche de synthèse RST, une formulation dans le domaine fréquentiel est retenue. La stratégie proposée consiste à définir et calculer une fonction de sensibilité de sortie désirée en satisfaisant des contraintes de robustesse de H∞. L’utilisation de parties fixes dans la fonction de sensibilité de sortie désirée assurera un placement partiel des pôles de la boucle fermée RST. L’inverse d’une telle fonction de sensibilité désirée définira le filtre de pondération H∞ associé. Un intérêt particulier est porté à l’approche d’optimisation par essaim particulière PSO pour la résolution des problèmes multi-objectif de commande reformulés. Un algorithme MOPSO à grille adaptative est proposé et puis perfectionné à base des concepts de l’epsilon-dominance. L’algorithme epsilon-MOPSO obtenu a montré, par comparaison avec les algorithmes MOPSO, NSGA-II et MODE, des performances supérieures en termes de diversité des solutions de Pareto et de rapidité en temps de convergence. Des métriques de type distance générationnelle, taux d’erreurs et espacement sont toutefois considérées pour l’analyse statistique des résultats de mise en œuvre obtenus. Une application à la commande en vitesse variable d’un moteur électrique DC est effectuée, également pour la commande en position d’un système de transmission flexible à charges variables. La mise en œuvre par simulations numériques sur les procédés considérés est également présentée dans le but de montrer la validité et l’efficacité de l’approche de commande optimale RST proposée / This research focuses on the systematic synthesis and optimization of digital RST structure based controllers thanks to global metaheuristics approaches. The classic and hard problems of closed-loop poles placement and sensitivity functions shaping of RST control are well formulated as constrained multi-objective problems to be solved with proposed metaheuristics algorithms NSGA-II, MODE, MOPSO and especially epsilon-MOPSO. Two formulations of the metaheuristics-tuned RST problem have been proposed. The first one, which is given in the time domain, deals with the minimization of several performance criteria like the Integral Square Error (ISE) and the Maximum Overshoot (MO) indices. These optimal criteria, related primarily to the step response of the controlled plant, are optimized under non-analytical constraints defined by temporal templates on the closed-loop dynamics. In the second approach, a formulation in the frequency domain is retained. The proposed strategy aims to optimize a desired output sensitivity function satisfying H∞ robustness constraints. The use of a suitable fixed part of the optimized output sensitivity function will provide partial pole placement of the closed-loop dynamics of the digital RST controller. The opposite of such desired sensitivity function will define the associated H∞ weighting filter. The Multi-Objective Particle Swarm Optimization (MOPSO) technique is particularly retained for the resolution of all formulated multi-objective RST control problems. An adaptive grid based MOPSO algorithm is firstly proposed and then improved based on the epsilon-dominance concepts. Such proposed epsilon-MOPSO algorithm, with a good diversity of the provided Pareto solutions and fast convergence time, showed a remarkable superiority compared to the standard MOPSO, NSGA-II and MODE algorithms. Performance metrics, such as generational distance, error rate and spacing, are presented for the statistical analysis of the achieved multi-optimization results. An application to the variable speed RST control of an electrical DC drive is performed, also for the RST position control of a flexible transmission plant with varying loads. Demonstrative simulations and comparisons are carried out in order to show the validity and the effectiveness of the proposed metaheuristics-based tuned RST control approach, which is formulated in the multi-objective optimization framework

Economic dispatch is an important optimization challenge in power systems. It helps to find the optimal output power of a number of generating units that satisfy the system load demand at the cheapest cost, considering equality and inequality constraints. Many nature inspired algorithms have been broadly applied to tackle it such as particle swarm optimization. In this dissertation, two improved particle swarm optimization techniques are proposed to solve economic dispatch problems. The first is a hybrid technique with Bat algorithm. Particle swarm optimization as the main optimizer integrates bat algorithm in order to boost its velocity and to adjust the improved solution. The second proposed approach is based on Cuckoo operations. Cuckoo search algorithm is a robust and powerful technique to solve optimization problems. The study investigates the effect of levy flight and random search operation in Cuckoo search in order to ameliorate the performance of the particle swarm optimization algorithm. The two improved particle swarm algorithms are firstly tested on a range of 10 standard benchmark functions and then applied to five different cases of economic dispatch problems comprising 6, 13, 15, 40 and 140 generating units. / Electrical and Mining Engineering / M. Tech. (Electrical Engineering)

Fundação de Amparo à Pesquisa do Estado do Rio de Janeiro / Diversas das possíveis aplicações da robótica de enxame demandam que cada robô seja capaz de estimar a sua posição. A informação de localização dos robôs é necessária, por exemplo, para que cada elemento do enxame possa se posicionar dentro de uma formatura de robôs pré-definida. Da mesma forma, quando os robôs atuam como sensores móveis, a informação de posição é necessária para que seja possível identificar o local dos eventos medidos. Em virtude do tamanho, custo e energia dos dispositivos, bem como limitações impostas pelo ambiente de operação, a solução mais evidente, i.e. utilizar um Sistema de Posicionamento Global (GPS), torna-se muitas vezes inviável. O método proposto neste trabalho permite que as posições absolutas de um conjunto de nós desconhecidos sejam estimadas, com base nas coordenadas de um conjunto de nós de referência e nas medidas de distância tomadas entre os nós da rede. A solução é obtida por meio de uma estratégia de processamento distribuído, onde cada nó desconhecido estima sua própria posição e ajuda os seus vizinhos a calcular as suas respectivas coordenadas. A solução conta com um novo método denominado Multi-hop Collaborative Min-Max Localization (MCMM), ora proposto com o objetivo de melhorar a qualidade da posição inicial dos nós desconhecidos em caso de falhas durante o reconhecimento dos nós de referência. O refinamento das posições é feito com base nos algoritmos de busca por retrocesso (BSA) e de otimização por enxame de partículas (PSO), cujos desempenhos são comparados. Para compor a função objetivo, é introduzido um novo método para o cálculo do fator de confiança dos nós da rede, o Fator de Confiança pela Área Min-Max (MMA-CF), o qual é comparado com o Fator de Confiança por Saltos às Referências (HTA-CF), previamente existente. Com base no método de localização proposto, foram desenvolvidos quatro algoritmos, os quais são avaliados por meio de simulações realizadas no MATLABr e experimentos conduzidos em enxames de robôs do tipo Kilobot. O desempenho dos algoritmos é avaliado em problemas com diferentes topologias, quantidades de nós e proporção de nós de referência. O desempenho dos algoritmos é também comparado com o de outros algoritmos de localização, tendo apresentado resultados 40% a 51% melhores. Os resultados das simulações e dos experimentos demonstram a eficácia do método proposto. / Many applications of Swarm Robotic Systems (SRSs) require that a robot is able to discover its position. The location information of the robots is required, for example, to allow them to be correctly positioned within a predefined swarm formation. Similarly, when the robots act as mobile sensors, the position information is needed to allow the identification of the location of the measured events. Due to the size, cost and energy source restrictions of these devices, or even limitations imposed by the operating environment, the straightforward solution, i.e. the use of a Global Positioning System (GPS), is often not feasible. The method proposed in this work allows the estimation of the absolute positions of a set of unknown nodes, based on the coordinates of a set of reference nodes and the distances measured between nodes. The solution is achieved by means of a distributed processing strategy, where each unknown node estimates its own position and helps its neighbors to compute their respective coordinates. The solution makes use of a new method called Multi-hop Collaborative Min-Max Localization (MCMM), herein proposed, aiming to improve the quality of the initial positions estimated by the unknown nodes in case of failure during the recognition of the reference nodes. The positions refinement is achieved based on the Backtracking Search Optimization Algorithm (BSA) and the Particle Swarm Optimization (PSO), whose performances are compared. To compose the objective function, a new method to compute the confidence factor of the network nodes is introduced, the Min-max Area Confidence Factor (MMA-CF), which is compared with the existing Hops to Anchor Confidence Factor (HTA-CF). Based on the proposed localization method, four algorithms were developed and further evaluated through a set of simulations in MATLABr and experiments in swarms of type Kilobot robots. The performance of the algorithms is evaluated on problems with different topologies, quantities of nodes and proportion of reference nodes. The performance of the algorithms is also compared with the performance of other localization algorithms, showing improvements between 40% to 51%. The simulations and experiments outcomes demonstrate the effectiveness of the proposed method.

Submitted by Maike Costa (maiksebas@gmail.com) on 2017-06-29T14:15:20Z No. of bitstreams: 1 arquivototal.pdf: 1397036 bytes, checksum: 303111e916d8c9feca61ed32762bf54c (MD5) / Made available in DSpace on 2017-06-29T14:15:20Z (GMT). No. of bitstreams: 1 arquivototal.pdf: 1397036 bytes, checksum: 303111e916d8c9feca61ed32762bf54c (MD5) Previous issue date: 2017-01-31 / The Maximu m Diversity Problem (MDP) is a problem of combinatorial optimization area that aims to select a pre-set number of elements in a given set so that a sum of the differences between the selected elements are greater as possible. MDP belongs to the class of NP-Hard problems, that is, there is no known algorithm that solves in polynomial time accurately. Because they have a complexity of exponential order, require efficient heuristics to provide satisfactory results in acceptable time. However, heuristics do not guarantee the optimality of the solution found. This paper proposes a new hybrid approach for a resolution of the Maximum Diversity Problem and is based on the Particle Swarm Optimization (PSO) and Tabu Search (TS) metaheuristics, The algorithm is called PSO_TS. The use of PSO_TS achieves the best results for known instances testing in the literature, thus demonstrating be competitive with the best algorithms in terms of quality of the solutions. / O Problema da Diversidade Máxima (MDP) é um problema da área de Otimização Combinatória que tem por objetivo selecionar um número pré-estabelecido de elementos de um dado conjunto de maneira tal que a soma das diversidades entre os elementos selecionados seja a maior possível. O MDP pertence a classe de problemas NP-difícil, isto é, não existe algoritmo conhecido que o resolva de forma exata em tempo polinomial. Por apresentarem uma complexidade de ordem exponencial, exigem heurísticas eficientes que forneçam resultados satisfatórios em tempos aceitáveis. Entretanto, as heurísticas não garantem otimalidade da solução encontrada. Este trabalho propõe uma nova abordagem híbrida para a resolução do Problema da Diversidade Máxima e está baseada nas meta-heurísticas de Otimização por Nuvem de Partículas (PSO) e Busca Tabu(TS). O algoritmo foi denominado PSO_TS. Para a validação do método, os resultados encontrados são comparados com os melhores existentes na literatura.

Made available in DSpace on 2016-12-08T15:56:17Z (GMT). No. of bitstreams: 1 Denilso Palaoro.pdf: 3446834 bytes, checksum: a842ffb16a48a459dc2b2e44efa303af (MD5) Previous issue date: 2015-02-24 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / The rubber industry has a great role in many areas of the economy, such as automotive, construction, footwear industry, hospital, etc. Rubber products are produced from complex mixtures of different raw materials, both natural and synthetic sources. From an industrial point of view, a major difficulty is to develop a formulation that meets the particular product requirements. This work seeks to develop an experimental methodology to obtain EPDM rubber compounds, using experimental design techniques coupled with computer numerical optimisation. A fractional factorial design was used to design and analyse the experiments, with factors the contents of calcium carbonate, paraffinic oil and vulcanizing accelerator (weight fractions). Twelve properties were measured (six of original samples, three of heat aged and three of processing). Statistical analyses enabled to find regression models for the properties and the cost of the formulation. A computer program was developed to minimise the cost function, subject to constraints on the properties. The results showed that it was possible to obtain formulations of EPDM rubber compounds which can be optimized at low cost, for example, of US$ 2.02/kg a 2.43/kg, for use in hoses and pads at different manufacturing processes such as compression moulding, transfer or injection. Selected compositions were analysed by FTIR, SEM and TGA, regards to their chemical and structural characteristics. Compositions with low vulcanization accelerator contents contribute to form cross-links with many sulphur- sulphur between the carbon chains which can damage mechanical properties in the original cured and aged samples. Using higher accelerator content, better properties are achieved probably due to lower content of the same sulphur- sulphur cross-links between polymer chains. The EPDM compounds studied may be used in cushions, hoses products which can withstand to hot air environments. Thus, the present study provides an experimental and scientific technique that allows developing rubber compounds with increased efficiency and reliability in research and development, taking into account the cost of the material. / A indústria da borracha possui um papel significativo em diversas áreas da economia, tais como: indústria automobilística, construção civil, indústria calçadista, hospitalar, etc. Artefatos de borracha são produzidos a partir de misturas complexas de diversas matérias-primas, naturais e sintéticas. Na indústria, uma grande dificuldade é desenvolver uma formulação que atenda aos requisitos de determinado produto. Assim, neste trabalho, busca se desenvolver uma metodologia experimental para obtenção de compostos de borracha de etileno propileno dieno (EPDM), utilizando-se técnicas de planejamento de experimentos aliado com otimização numérica computacional. Foi utilizado um planejamento fatorial fracionado 33-1 (três níveis e três fatores), sendo os fatores: teor de carbonato de cálcio, teor de óleo parafínico e teor de acelerador de vulcanização. Foram medidas no total, 12 propriedades (seis originais, três envelhecidas e três propriedades de processos). A partir de estudos estatísticos foram obtidos modelos de regressão para as propriedades e para o custo da formulação. Um programa computacional foi desenvolvido para minimizar a função custo, sujeita às restrições nas propriedades. Os resultados mostraram que foi possível obter formulações de compostos de borracha EPDM otimizadas a um custo variando entre US$ 2,02/kg a 2,43/kg, para aplicação em mangueiras e coxins e em processos de transformação diversos, tais como, moldagem por compressão, transferência ou injeção. Composições selecionadas foram escolhidas e analisadas, por meio de FTIR, MEV e TGA, quanto às suas características químicas e estruturais. Composições com baixos teores de acelerador de vulcanização contribuem para formar ligações cruzadas com cerca de 4 a 7 átomos de enxofre entre as cadeias carbônicas, prejudicando as propriedades mecânicas no vulcanizado original e envelhecido. Com um teor maior de acelerador, melhores propriedades são obtidas, tendo em vista que um número elevado de ligações cruzadas com uma quantidade de átomos de enxofre inferior a 4 na cadeia são formadas. Os compostos de EPDM estudados podem ser utilizados em produtos de coxins e mangueiras para aplicações resistentes ao calor. Assim, esta pesquisa apresenta uma metodologia experimental, para a pesquisa e desenvolvimento de compostos de borracha EPDM, levando em conta o custo do material e restrições nas propriedades.