Made available in DSpace on 2016-04-27T16:59:04Z (GMT). No. of bitstreams: 1 Custodio Thomaz Kerry Martins.pdf: 1658321 bytes, checksum: f7c50429b02bebeb4fa4f2c178e7525d (MD5) Previous issue date: 2010-05-11 / The research was planned through the design and implementation of a four activities sequence, for small student groups, beginning a Bachelor of Computer Science course. This sequence of activities has been through two trials, the first in the second half of 2008 and second in the second half of 2009. The interest focus is the disciplines learning process in introduction to algorithms and programming, contained in the higher education courses curriculum as Computer Science and Engineering. This research aims to investigate how students produce the perception and the dynamic aspects knowledge, belonging to algorithms, and the reflection of this domain in the algorithms understanding activities and development. The Semiotic Representations Records theory (Duval, 2008), the Tool-Object Dialectic ideas (Maranhão, 2008) and the Didactic Engineering notion (Artigue, 2009), compose the main theoretical-methodological contributions to research. This is a qualitative research, with axis in a Didactic Engineering, composed of a four activities sequence. The domain of the concept of dynamic process, and the use of this notion, were revealed in the speech, gestures and written records of students. The observations and analysis showed that the work with the activities, promoted the understanding and mastery of the algorithms dynamic process notions, which contributed to the development learning of algorithms and programs / A pesquisa foi planejada com o projeto e a realização de uma seqüência de quatro atividades por pequenos grupos de alunos iniciantes em um curso de bacharelado de Ciência da Computação. Essa seqüência de atividades passou por duas experimentações, a primeira no segundo semestre de 2008 e a segunda no segundo semestre de 2009. O foco de interesse é o processo de aprendizagem em disciplinas de Introdução aos Algoritmos e Programação que figuram na grade curricular de cursos superiores como Ciência da Computação e Engenharia. O alvo estabelecido foi investigar como o estudante revela, trata e domina a noção de processo dinâmico inerente a um algoritmo ou a um programa, e o reflexo desse domínio nas atividades de entendimento e elaboração de algoritmos. A teoria dos Registros de Representação Semiótica (Duval, 2008), as idéias da Dialética Ferramenta-Objeto (Maranhão, 2008) e a noção de Engenharia Didática (Artigue, 2009) compõem os aportes teórico-metodológicos principais da pesquisa. Trata-se de uma pesquisa qualitativa, com eixo em uma Engenharia Didática composta por uma seqüência de quatro atividades. O domínio da noção de processo dinâmico, e o emprego dessa noção, foram revelados nas falas, nos gestos e nos registros escritos dos estudantes. As observações e análises permitiram concluir que o trabalho com as atividades favoreceram o entendimento e o domínio da noção de processo dinâmico dos algoritmos, e isso contribuiu para o aprendizado de elaboração e entendimento dos algoritmos e programas

The project deals with artificial neural networks. After designing and debugging the test data set and the training sample set, we created a multilayer perceptron network in the Neural NetworkToolbox (NNT) of Matlab. When creating networks, we used different training algorithms and algorithms improving the generalization of the network. When creating a radial basis network, we did not use the NNT, but a specific source code in Matlab was written. Functionality of neural networks was tested on simple training and testing patterns. Realistic training data were obtained by the simulation of twelve monoconic antennas operating in the frequency range from 2 to 6 GHz. Antennas were located inside a mathematical model of Octavia II. Using CST simulations, electromagnetic fields in a car were obtained. Trained networks are described by regressive characteristics andthe mean square error of training. Algorithms improving generalization are applied on the created and trained networks. The performance of individual networks is mutually compared.