Submitted by Renata Lopes (renatasil82@gmail.com) on 2017-03-02T11:42:31Z No. of bitstreams: 1 ericadacostareiscarvalho.pdf: 5557018 bytes, checksum: f6ffd53d6329e89b519786974a1b85e0 (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2017-03-06T19:32:37Z (GMT) No. of bitstreams: 1 ericadacostareiscarvalho.pdf: 5557018 bytes, checksum: f6ffd53d6329e89b519786974a1b85e0 (MD5) / Made available in DSpace on 2017-03-06T19:32:37Z (GMT). No. of bitstreams: 1 ericadacostareiscarvalho.pdf: 5557018 bytes, checksum: f6ffd53d6329e89b519786974a1b85e0 (MD5) Previous issue date: 2014-02-13 / CAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Nos últimos anos, várias meta-heurísticas têm sido adotadas para a solução de problemas de otimização com restrições. Uma dessas meta-heurísticas que se torna cada vez mais popular é a Otimização por Enxame de Partículas (Particle Swarm Optimization - PSO). O PSO é baseado na metáfora de como algumas espécies compartilham informações e, em seguida, usam essas informações para mover-se até os locais onde os alimentos estão localizados. A população é formada por um conjunto de indivíduos denominado partículas que representa possíveis soluções dentro de um espaço de busca multidimensinal. Neste trabalho, são analisados problemas clássicos de otimização com restrições onde um algoritmo PSO os trata como sendo sem restrições através da introdução de um método de penalização adaptativa (Adaptive Penalty Method - APM). O APM adapta o valor dos coeficientes de penalização de cada restrição fazendo uso de informações coletadas da população, tais como a média da função objetivo e o nível de violação de cada restrição. Diversos experimentos computacionais são realizados visando avaliar o desempenho do algoritmo considerando vários problemas testes encontrados na literatura. / In recent years, several meta-heuristics have been adopted for the solution of constrained optimization problems. One of these meta-heuristic that is becoming increasingly popular is the Particle Swarm Optimization - PSO. PSO is based on the metaphor of how some species share information and then use this information to move to the places where food is located. The population is formed by a group of individuals called particles representing possible solutions within a space multidimensional search. In this thesis, classical problems of constrained optimization where a PSO algorithm treats them as being unconstrained by introducing a method of adaptive penalty (Adaptive Penalty Method - APM) are analyzed. The APM adjusts the value of the penalty coeffcients of each constraint using the information collected from the population, such as the average of the objective function as well as the level of violation of each constraint. Several computational experiments are conducted to assess the performance the algorithm tests considering various problems found in the literature.

A Wireless Sensor Network (WSN) is a group of spatially dispersed sensors that monitor the physical conditions of the environment and collect data at a central location. Sensor deployment is one of the main design aspects of WSNs as this a ffects network coverage. In general, WSN deployment methods fall into two categories: planned deployment and random deployment. This thesis considers planned sensor deployment of a Mobile Wireless Sensor Network (MWSN), which is defined as selectively deciding the locations of the mobile sensors under the given constraints to optimize the coverage of the network. Metaheuristic algorithms are powerful tools for the modeling and optimization of problems. The Krill Herd Algorithm (KHA) is a new nature-inspired metaheuristic algorithm which can be used to solve the sensor deployment problem. A Multi-Agent System (MAS) is a system that contains multiple interacting agents. These agents are autonomous entities that interact with their environment and direct their activity towards achieving speci c goals. Agents can also learn or use their knowledge to accomplish a mission. Multi-agent systems can solve problems that are very difficult or even impossible for monolithic systems to solve. In this work, a modification of KHA is proposed which incorporates MAS to obtain a Multi-Agent Krill Herd Algorithm (MA-KHA). To test the performance of the proposed method, five benchmark global optimization problems are used. Numerical results are presented which show that MA-KHA performs better than the KHA by finding better solutions. The proposed MA-KHA is also employed to solve the sensor deployment problem. Simulation results are presented which indicate that the agent-agent interactions in MA-KHA improves the WSN coverage in comparison with Particle Swarm Optimization (PSO), the Firefly Algorithm (FA), and the KHA. / Graduate

Optimization problems are vital in physical sciences, commercial and finance matters. In a nutshell, almost everyone is the stake-holder in certain optimization problems aiming at minimizing the cost of production and losses of system, and also maximizing the profit. In control systems, the optimal configuration problems are essential that have been solved by various newly developed methods. The literature is exhaustively explored for an appropriate optimization method to solve such kind of problems. Particle Swarm Optimization is found to be one of the best among several optimization methods by analysing the experimental results. Two novel PSO variants are introduced in this thesis. The first one is named as N State Markov Jumping Particle Swarm Optimization, which is based on the stochastic technique and Markov chain in updating the particle velocity. We have named the second variant as N State Switching Particle Swarm Optimization, which is based on the evolutionary factor information for updating the velocity. The proposed algorithms are then applied to some widely used mathematical benchmark functions. The statistical results of 30 independent trails illustrate the robustness and accuracy of the proposed algorithms for most of the benchmark functions. The better results in terms of mean minimum evaluation errors and the shortest computation time are illustrated. In order to verify the satisfactory performance and robustness of the proposed algorithms, we have further formulated some basic applications in power system operations. The first application is about the static Economic Load Dispatch and the second application is on the Dynamic Economic Load Dispatch. These are highly complex and non-linear problems of power system operations consisting of various systems and generator constraints. Basically, in the static Economic Load Dispatch, a single load is considered for calculating the cost function. In contrast, the Dynamic Economic Load Dispatch changes the load demand for the cost function dynamically with time. In such a challenging and complex environment the proposed algorithms can be applied. The empirical results obtained by applying both of the proposed methods have substantiated their adaptability and robustness into the real-world environment. It is shown in the numerical results that the proposed algorithms are robust and accurate as compared to the other algorithms. The proposed algorithms have produced consistent best values for their objectives, where satisfying all constraints with zero penalty.

Games have been investigated as computationally complex problems since the inception of artificial intelligence in the 1950’s. Originally, search-based techniques were applied to create a competent (and sometimes even expert) game player. The search-based techniques, such as game trees, made use of human-defined knowledge to evaluate the current game state and recommend the best move to make next. Recent research has shown that neural networks can be evolved as game state evaluators, thereby removing the human intelligence factor completely. This study builds on the initial research that made use of evolutionary programming to evolve neural networks in the game learning domain. Particle Swarm Optimisation (PSO) is applied inside a coevolutionary training environment to evolve the weights of the neural network. The training technique is applied to both the zero sum and non-zero sum game domains, with specific application to Tic-Tac-Toe, Checkers and the Iterated Prisoners Dilemma (IPD). The influence of the various PSO parameters on playing performance are experimentally examined, and the overall performance of three different neighbourhood information sharing structures compared. A new coevolutionary scoring scheme and particle dispersement operator are defined, inspired by Formula One Grand Prix racing. Finally, the PSO is applied in three novel ways to evolve strategies for the IPD – the first application of its kind in the PSO field. The PSO-based coevolutionary learning technique described and examined in this study shows promise in evolving intelligent evaluators for the aforementioned games, and further study will be conducted to analyse its scalability to larger search spaces and games of varying complexity. / Dissertation (MSc)--University of Pretoria, 2005. / Computer Science / unrestricted

After the historic chess match between Deep Blue and Garry Kasparov, many researchers considered the game of chess solved and moved on to the more complex game of soccer. Artificial intelligence research has shifted focus to creating artificial players capable of mimicking the task of playing soccer. A new training algorithm is presented in this thesis for training teams of players from zero knowledge, evaluated on a simplified version of the game of soccer. The new algorithm makes use of the charged particle swarm optimiser as a neural network trainer in a coevolutionary training environment. To counter the lack of domain information a new relative fitness measure based on the FIFA league-ranking system was developed. The function provides a granular relative performance measure for competitive training. Gameplay strategies that resulted from the trained players are evaluated. It was found that the algorithm successfully trains teams of agents to play in a cooperative manner. Techniques developed in this study may also be widely applied to various other artificial intelligence fields. / Dissertation (MSc)--University of Pretoria, 2013. / Computer Science / unrestricted

This dissertation focuses on daily volt/var control in distribution networks with feeder capacitors, substation capacitors and transformers equipped with on-load tap changers. A hybrid approach is proposed to solve the daily volt/var control problem. To reduce the computational requirements of the problem, this approach combines two methods, namely heuristic and optimal scheduling for the substation and feeder sub-problems respectively. The feeder capacitor dispatch schedule is determined based on a heuristic reactive power setpoint method. At this stage the objective is to minimize the reactive power flow through the substation bus in every time-interval. And as such, mathematical modeling of the distribution network components is adapted to suit time-varying conditions. Furthermore, an optimization model to determine a proper dispatch schedule of the substation devices is formulated. The objective of this model is to minimize the daily total energy loss and voltage deviations. Additionally, the reference voltage of the substation secondary bus and the transformer tap position limits are modified to adapt to given load profiles. The optimization model is solved with a discrete particle swarm optimization algorithm, which incorporates Newton’s method to determine the power-flow solution. The proposed method is applied to a time-varying distribution system and evaluated under different operational scenarios. It is also compared to on-line volt/var control with various settings. Simulation results show that the proposed approach minimizes both the voltage deviations and the total energy loss, while on-line control prioritizes one objective over the other depending on the specified settings. / Dissertation (MEng)--University of Pretoria, 2015. / Electrical, Electronic and Computer Engineering / Unrestricted

For a linear array, the excitation coefficients of each element and its geometry play an important role, because they will determine the radiation pattern of the given array. Side Lobe Level (SLL) is one of the key parameters to evaluate the radiation pattern of the array. Generally speaking, we desire SLL to be as low as possible. For the linear array with uniform spacing, there are some classic methods to calculate the excitation coefficients to make the radiation pattern satisfy the given requirements. For the linear array with non-uniform spacing, linear programming and particle swarm optimization are proposed to calculate the excitation coefficients to make the array get minimum SLL in this thesis. They are demonstrated for symmetric and asymmetric array in the first part of this thesis. In the second part of this thesis, a simple method is proposed for correcting excitation coefficients of a linear phased array. This proposed method corrects the coefficients through using the Normalized Least Means Squares(NLMS) algorithm, dither signal and a near-field sensor being used for sensing the field emitted by the array. The advantage of this proposed method is that it avoids the problem of estimating the largest eigenvalue of the coefficient matrix to get optimal step size. Its robustness in different environments is demonstrated as well as the effect of noise with various Signal-to-Noise Ratio (SNR), and mutual coupling. In addition, the effect of using discrete dither signal to the array is considered, because the continuous dither signal cannot be generated in practice.

The theoretical part of this thesis deals with principles and characteristics of the surface acoustic wave filter. It explained the principle of magnetostriction and piezoelectric effect, which uses a filter. In the practical part in the program Comsol Multiphysics are made three simple models of SAW filter. Some modes are founded and are compared to previously known results. In the next phase of project is further studied using the parametric analysis. In the last phase of project is applied global optimization PSO at admittance characteristic from simple 2D structure. The results are compared with the commercially produced devices.

Les travaux présentés dans ce mémoire proposent une nouvelle méthode d'optimisation globale dénommée MPSO-SA. Cette méthode hybride est le résultat d'un couplage d'une variante d'algorithme par Essaim de particules nommé MPSO (Particle Swarm Optimization) avec la méthode du recuit simulé nommé SA (Simulted Annealing). Les méthodes stochastiques ont connu une progression considérable pour la résolution de problèmes d'optimisation. Parmi ces méthodes, il y a la méthode Essaim de particules (PSO° qui est développée par [Eberhart et Kennedy (1995)]. Quant à la méthode recuit simulé (SA), elle provient du processus physique qui consiste à ordonner les atomes d'un cristal afin de former une structure cristalline parfaite. Pour illustrer les performances de la méthode MPSO-SA proposée, une comparaison avec MPSO et SA est effectuée sur des fonctions tests connues dans la littérature. La métode MPSO-SA est utilisée pour la résolution des problèmes réels interopérabilité des systèmes d'information, ainsi qu'aux problèmes d'optimisation et de fiabilité des structures mécaniques. / The work presented in this PhD thesis contibutes to a new method for a modified particle swarm optimization algorith (MPSO) combined with a simulating annealing algorithm (SA). MPSO is known as an efficient approach with a high performance of solving optimization problems in many research fields. It is a population intelligence algorithm [Eberhart et Kennedy (1995)] inspired by social behavior simulations of bird flocking. Considerable research work on classical method PSO (Particle Swarm Optimization) has been done to improve the performance of this method. Therefore, the propose hybrid optimization algorithms MPSOSA use the combination of MPSO and simulating annealing SA. This method has the avantage to provide best results comparing with all heuristics methods PSO and SA. In this matter, a benchmark of eighteen well-known functions is given. These functions present different situations of finding the global minimum with gradual difficulties. Numerical results presented, in this paper, show the robustness of the MPSOSA algorithm. Numerical comparisons with three algorithms namely, Simulating Annealing, Modified Particle swarm optimization and MPSO-SA show that hybrid algorithm offers better results. This method (MPSO-SA) treats a wide range of optimization problems, in information systems interoperability and in structural optimization field.

The subjects of perfect information games, machine learning and computational intelligence combine in an experiment that investigates a method to build the skill of a game-playing agent from zero game knowledge. The skill of a playing agent is determined by two aspects, the first is the quantity and quality of the knowledge it uses and the second aspect is its search capacity. This thesis introduces a novel representation language that combines symbols and numeric elements to capture game knowledge. Insofar search is concerned; an extension to an existing knowledge-based search method is developed. Empirical tests show an improvement over alpha-beta, especially in learning conditions where the knowledge may be weak. Current machine learning techniques as applied to game agents is reviewed. From these techniques a learning framework is established. The data-mining algorithm, ID3, and the computational intelligence technique, Particle Swarm Optimisation (PSO), form the key learning components of this framework. The classification trees produced by ID3 are subjected to new post-pruning processes specifically defined for the mentioned representation language. Different combinations of these pruning processes are tested and a dominant combination is chosen for use in the learning framework. As an extension to PSO, tournaments are introduced as a relative fitness function. A variety of alternative tournament methods are described and some experiments are conducted to evaluate these. The final design decisions are incorporated into the learning frame-work configuration, and learning experiments are conducted on Checkers and some variations of Checkers. These experiments show that learning has occurred, but also highlights the need for further development and experimentation. Some ideas in this regard conclude the thesis. / Dissertation (MSc)--University of Pretoria, 2007. / Computer Science / MSc / Unrestricted