Submitted by FERNANDO PARRA DOS ANJOS LIMA null (engfernandoparra@gmail.com) on 2016-10-31T11:47:54Z No. of bitstreams: 1 Fernando Parra A. Lima.pdf: 3667307 bytes, checksum: 0d206b9c09566cdb11de101b84976228 (MD5) / Approved for entry into archive by Felipe Augusto Arakaki (arakaki@reitoria.unesp.br) on 2016-11-07T16:42:09Z (GMT) No. of bitstreams: 1 lima_fpa_dr_ilha.pdf: 3363973 bytes, checksum: d8849cdd159a11920c497d025a8ae16a (MD5) / Made available in DSpace on 2016-11-07T16:42:09Z (GMT). No. of bitstreams: 1 lima_fpa_dr_ilha.pdf: 3363973 bytes, checksum: d8849cdd159a11920c497d025a8ae16a (MD5) Previous issue date: 2016-09-01 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Esta pesquisa é dedicada ao desenvolvimento de uma metodologia para a realização do diagnóstico de distúrbios de tensão de sistemas de distribuição de energia elétrica, baseada no uso de sistemas imunológicos artificiais (SIA). Trata-se da proposição de um novo paradigma no ambiente dos SIA que confere o aprendizado de modo contínuo (plasticidade). Esta concepção permite compor um sistema de diagnóstico apto a aprender continuamente, contemplando novos tipos de distúrbios advindos da constante evolução do setor elétrico, sem a necessidade de reiniciar o processo de aprendizado. Neste contexto, empregam-se dois algoritmos imunológicos artificiais, sendo o algoritmo de seleção negativa, responsável pelo processo de reconhecimento de padrões, e o algoritmo de seleção clonal responsável pelo processo de aprendizado. A principal aplicação deste novo método é auxiliar na operação do sistema durante distúrbios, bem como, supervisionar o sistema de proteção, e estar apto a acompanhar a evolução dos sistemas elétricos adquirindo conhecimento continuamente. Para avaliar a eficácia e o desempenho deste novo método foram realizadas simulações de distúrbios de tensão em sistemas de distribuições de energia elétrica com 5, 33, 84 e 134 barras, no software ATP/EMTP. Os resultados obtidos com esta abordagem mostram robustez e eficiência quando comparados à literatura. / This work develops a methodology to realize voltage disturbance diagnosis in electrical distribution systems, based on Artificial Immune Systems (AIS). It is a proposition of a new paradigm in AIS environment, which provides a continuous learning (plasticity). This conception allows composing a diagnosis system able to continuous learn, when new disturbances appear due to the constant evolution of the power systems, without needing to reinitialize the learning. This way, two artificial immune algorithms are used, such as the negative selection algorithm executing the pattern recognition process, and the clonal selection algorithm, executing the learning process. The main application of this new method is to aid the system operation during disturbances, as well as, supervise the system protection and be able to carry on the evolution of the electrical systems acquiring knowledge continuously. To evaluate the efficiency and the performance of this new method, voltage disturbance simulations were executed in electrical distributions systems with 5, 33, 84 and 134-bus in ATP/EMTP software. Results show robustness and efficiency when compared with those in the literature.

Orientador: Carlos Roberto Minussi / Resumo: Esta pesquisa é dedicada ao desenvolvimento de uma metodologia para a realização do diagnóstico de distúrbios de tensão de sistemas de distribuição de energia elétrica, baseada no uso de sistemas imunológicos artificiais (SIA). Trata-se da proposição de um novo paradigma no ambiente dos SIA que confere o aprendizado de modo contínuo (plasticidade). Esta concepção permite compor um sistema de diagnóstico apto a aprender continuamente, contemplando novos tipos de distúrbios advindos da constante evolução do setor elétrico, sem a necessidade de reiniciar o processo de aprendizado. Neste contexto, empregam-se dois algoritmos imunológicos artificiais, sendo o algoritmo de seleção negativa, responsável pelo processo de reconhecimento de padrões, e o algoritmo de seleção clonal responsável pelo processo de aprendizado. A principal aplicação deste novo método é auxiliar na operação do sistema durante distúrbios, bem como, supervisionar o sistema de proteção, e estar apto a acompanhar a evolução dos sistemas elétricos adquirindo conhecimento continuamente. Para avaliar a eficácia e o desempenho deste novo método foram realizadas simulações de distúrbios de tensão em sistemas de distribuições de energia elétrica com 5, 33, 84 e 134 barras, no software ATP/EMTP. Os resultados obtidos com esta abordagem mostram robustez e eficiência quando comparados à literatura. / Doutor

Am Beispiel moderner Automatisierungssysteme wird deutlich, dass die Steuerung und optimale Führung der technischen Prozesse eng verbunden ist mit der Verfügbarkeit eines möglichst exakten Prozessmodells. Steht jedoch kein Modell des zu steuernden Systems zur Verfügung oder ist das System nicht ausreichend genau analytisch beschreibbar, muss ein adäquates Modell auf der Basis von Beobachtungen (Messdaten) abgeleitet werden. Erschwerend wirken sich hierbei starke Nichtlinearitäten sowie der zeitvariante Charakter der zu identifizierenden Systeme aus. Die Zeitvarianz, beispielsweise durch Alterung oder Verschleiß hervorgerufen, erfordert zusätzlich eine schritthaltende Adaption an den sich verändernden Prozess. Das einmalige, zeitlich begrenzte Erstellen eines Modells ist somit nicht ausreichend. Stattdessen muss zeitlich unbegrenzt "nachtrainiert" werden, was dementsprechend als "Kontinuierliches Lernen" bezeichnet wird. Auch wenn das Ableiten eines Systemmodells anhand von Beobachtungen eine typische Aufgabenstellung für Neuronale Netze ist, stellt die Zeitvarianz Neuronale Netze dennoch vor enorme Probleme. Im Rahmen der Dissertation wurden diese Probleme identifiziert und anhand von unterschiedlichen Neuronalen Netzansätzen analysiert. Auf den sich hieraus ergebenden Ergebnissen steht anschließend die Entwicklung eines neuartigen Neuronalen Netzansatzes im Mittelpunkt. Die besondere Eigenschaft des hybriden ICE-Lernverfahrens ist die Fähigkeit, eine zur Problemkomplexität adäquate Netztopologie selbstständig zu generieren und diese entsprechend des zeitvarianten Charakters der Zielfunktion dynamisch adaptieren zu können. Diese Eigenschaft begünstigt insbesondere schnelles Initiallernen. Darüber hinaus ist das ICE-Verfahren in der Lage, parallel zur Modellausgabe Vertrauenswürdigkeitsprognosen für die aktuelle Ausgabe zur Verfügung zu stellen. Den Abschluss der Arbeit bildet eine spezielle Form des ICE-Ansatzes, bei der durch asymmetrische Aktivierungsfunktionen Parallelen zur Fuzzy-Logik hergestellt werden. Dadurch wird es möglich, automatisch Regeln abzuleiten, welche das erlernte Modell beschreiben. Die "Black-Box", die Neuronale Netze in der Regel darstellen, wird dadurch transparenter. / One of the main requirements for an optimal industrial control system is the availability of a precise model of the process, e.g. for a steel rolling mill. If no model or no analytical description of such a process is available a sufficient model has to be derived from observations, i.e. system identification. While nonlinear function approximation is a well-known application for neural networks, the approximation of nonlinear functions that change over time poses many additional problems which have been in the focus of this research. The time-variance caused for example by aging or attrition requires a continuous adaptation to process changes throughout the life-time of the system, here referred to as continuous learning. Based on the analysis of different neural network approaches the novel incremental construction algorithm ICE for continuous learning tasks has been developed. One of the main advantages of the ICE-algorithm is that the number of RBF-neurons and the number of local models of the hybrid network have not to be determined in advance. This is an important feature for fast initial learning. The evolved network is automatically adapted to the time-variant target function. Another advantage of the ICE-algorithm is the ability to simultaneously learn the target function and a confidence value for the network output. Finally a special version of the ICE-algorithm with asymmetric receptive fields is introduced. Here similarities to fuzzy logic are intended. The goal is to automatically derive rules which describe the learned model of the unknown process. In general a neural network is a "black box". In contrast to that an ICE-network is more transparent.

Diese Dissertation umfasst drei Studien. Die erste Studie untersucht Overprecision („Überpräzision“). Ich untersuche, wie Menschen Spannweitenschätzungen vornehmen. Teilnehmende müssen die Größe der Spannweite (Präzision) mit der Wahrscheinlichkeit, dass sie den wahren Wert einschließt (Richtigkeit) balancieren. Die Ergebnisse zeigen, dass Menschen inhärente individuelle Vorlieben für Präzision zu haben scheinen. Gleichzeitig werden vorhersagbar zusätzliche Informationen genutzt, um die Richtigkeit der Schätzungen zu erhöhen. Dafür wird entweder Präzision geopfert, oder die Spannweitenschätzung insgesamt verschoben. Die Richtigkeit der Schätzung wird jedoch nicht maximiert, sondern ein Teil für höhere Präzision aufgegeben. Die zweite Studie untersucht, wie sich die Übersetzung von Finanzberichterstattung auf die Wahrnehmung einer Firma als attraktives Investment auswirkt. Sie beleuchtet drei verschiedene Kanäle: Lesbarkeit, Stimmung, und Präzision der Veröffentlichung. In einem Umfrageexperiment lesen Kleinanleger deutsche und englische Prognoseberichte deutscher Firmen. Die Ergebnisse zeigen, dass die deutschen Berichte als besser lesbar wahrgenommen werden. Im Gegensatz zu vorheriger Literatur ist die Lesbarkeit nicht mit einer höheren Investmentattraktivität korreliert. Allein die Stimmung des Textes zeigt eine Korrelation mit höherer Attraktivität. Die dritte Studie untersucht, wie das Angebot von formativen Onlinetests die Leistungen von Studierenden in der Klausur beeinflusst. Sie untersucht zudem, ob die Leistung sich unterscheidet, je nachdem ob die Studierenden zeitlich begrenzten Zugang zu den Tests haben oder ob sie jederzeit auf die Tests zugreifen können. Ein Experiment, welches es ermöglicht den kausalen Intention-to-treat-Effekt zu bestimmen, zeigt, dass die formativen Onlinetests die Studienleistung erhöhen können, allerdings nur für Studierende, welche sich nicht freiwillig für die Tests gemeldet hatten und in der kontinuierlichen Lerngruppe waren. / This dissertation comprises three papers. The first study examines overprecision. I examine how people provide range estimates, a challenging task that requires people to balance the width of the range (i.e., its precision) with the probability of the range covering the true value (i.e., accuracy). I find that people appear to have inherent individual preferences for a certain level of precision. At the same time, they appear to predictably incorporate additional information in order to increase accuracy by either sacrificing precision or shifting their ranges altogether. Still, they do not seem to maximise accuracy, but are willing to expend some of it to provide more precise estimates. The second study examines how the translation of financial disclosures changes investors’ perceptions of firms as an attractive investment. It examines three possible channels: readability, tone, and precision of the underlying disclosure. In a survey experiment, retail investors read forecast reports of German firms, provided in German and English. The findings indicate that the German versions are easier to read. Contrary to prior literature, the easier readability does not translate into higher investment attractiveness. Solely tone appears to be correlated with investment attractiveness. The third study analyses how offering formative online assessments influences student performance in the final exam. It further examines whether students perform differently depending on whether they have time-restricted access to the assessments, or whether they can access the assessments at any time. An experiment which allows for the identification of the causal intention-to-treat effect shows that offering formative online assessments can enhance student performance, but only for students who do not opt for taking the test voluntarily and who are in a continuous learning environment.

Made available in DSpace on 2016-03-15T19:30:58Z (GMT). No. of bitstreams: 1 Leticia Rodrigues da Fonseca.pdf: 1750631 bytes, checksum: 386ec52022d302b9f42a054db2a46380 (MD5) Previous issue date: 2013-02-19 / Faculdade Cenecista de Varginha / The Brazilian market for software and services has grown considerably in recent years. According to the Brazilian Association of Software Companies (ABES, 2011), Brazil occupies the eleventh position on the world stage, moving 19.04 billion dollars, equivalent to 1.0% of gross domestic product (GDP), this total, 6.74 billion refer to software production, representing 2.2% of the world market. However, fierce competition manifests itself in the same proportion opportunities. The national software compete with themselves and with those from other countries, because there are not yet many barriers or legal restrictions against marketing of international products (TONINI et al, 2008). This sector is still characterized by evolving at a rapid pace, which causes frequent changes related to products, services, processes and technology itself (MAURER et al, 2002; NIASI, 2009). Therefore, it is believed to be fundamental for Software Development Organizations (ODSs) create a work environment that supports continuous learning to cope with the challenges of the new economy and remain competitive in the market. In this context, this study aims to understand how learning occurs continuously of ODSs that have deployed MPSBR and identify which of its aspects favor the individual and collective learning. By this, we carried out a qualitative study of type multicase in four ODSs three classified at Level C and a level F. It was used as a method of data collection to semi-structured interview. We interviewed individuals involved in software development and representatives of the group responsible for the implementation of MPSBR. The analyzes were performed using software for qualitative analyzes that allowed Weft_QDA from the transcript of the interviews, the establishment of analytical categories for the constructs and verifying relationships between their categories. By the results was possible to understand how MPSBR can contribute to the establishment of a software process quality and affordable, understand how MPSBR can encourage continuous learning through their assumptions and developers understand how these companies learn at level individual and collective. / O mercado brasileiro de software e serviços tem crescido consideravelmente nos últimos anos. Segundo a Associação Brasileira das Empresas de Software (ABES, 2011), o Brasil ocupa a décima primeira posição no cenário mundial, movimentando 19,04 bilhões de dólares, o equivalente a 1,0% do produto interno bruto (PIB) brasileiro; desse total, 6,74 bilhões referem-se à produção de software, o que representa 2,2% do mercado mundial. Contudo, a concorrência acirrada manifesta-se na mesma proporção das oportunidades. Os softwares nacionais competem entre si e com aqueles provenientes de outros países, pelo fato de ainda não existirem muitas barreiras ou restrições legais contra a comercialização de produtos internacionais (TONINI et al, 2008). Esse setor é ainda caracterizado pela evolução em ritmo acelerado, o que ocasiona mudanças frequentes relacionadas aos produtos, serviços, processos e à própria tecnologia (MAURER et al, 2002; NIASI, 2009). Portanto, acredita-se que seja fundamental para as Organizações de Desenvolvimento de Software (ODSs) instituir um ambiente de trabalho que apóie o aprendizado contínuo, para lidar com os desafios da nova economia e manterem-se competitivas no mercado. Diante desse contexto o presente estudo propõe-se a compreender como ocorre a aprendizagem contínua no ambiente de trabalho de ODSs que implantaram o MPSBR e identificar quais de seus aspectos favorecem a aprendizagem individual e coletiva. Para isso, realizou-se uma pesquisa de abordagem qualitativa do tipo multicaso em quatro ODSs, sendo três classificadas no Nível C e uma no nível F. Utilizou-se como método de coleta de dados a entrevista semi-estruturada. Entrevistaram-se os indivíduos envolvidos no processo de desenvolvimento dos softwares e os representantes do grupo responsável pela implantação do MPSBR. As análises foram realizadas por meio do software para análises qualitativas Weft_QDA que permitiu, a partir da transcrição das entrevistas, o estabelecimento de categorias analíticas para os construtos e a verificação de relações entre as suas categorias. Ao final, foi possível compreender como o MPSBR pode contribuir para o estabelecimento de um processo de software de qualidade e a um custo acessível, compreender como o MPSBR pode incentivar a aprendizagem contínua por meio de suas premissas e entender como os desenvolvedores dessas empresas aprendem no nível individual e coletivo.

No description available.

The aim of this study is to conduct a thorough theoretical study on the relevant aspects involved in knowledge management and organisational learning, and to assess the maturity level of knowledge management within the South African steel industry. Various aspects of knowledge, knowledge management and organisational learning with specific relation to sustainable competitive advantage are discussed in the literature study. During the literature research several factors which either promote or hinder the effective management of knowledge were identified and a list of lessons learned by other successful knowledge focused companies, are discussed. The critical success factors for a successful knowledge management program are also discussed. Knowledge can create a sustainable competitive advantage within an organisation, if successfully applied to make value adding decisions and to enable learning, and if it is applied to make decisions which are superior to that of its rivals across the supply chain. An integrated approach needs to be followed when KM is pursued and the knowledge must be applied to make value added decisions and facilitate learning across all processes in the value chain. The focus must be to retain an organisation’s tacit knowledge as this is a key success factor to ensure a sustainable competitive advantage. The study includes research on whether knowledge management is effectively used as a sustainable competitive advantage in the South African steel industry. The maturity level of the application of knowledge and learning principles implemented within the South African steel industry is assessed and compared to the maturity level of ArcelorMittal, Monlevade, located in Brazil. A survey was designed and distributed to determine the knowledge management and organisational learning maturity levels at two steel facilities of ArcelorMittal in South Africa and one facility in Brazil. The key problem areas as identified through the empirical research are discussed and it is concluded that South African facilities do not effectively use knowledge management as a sustainable competitive advantage. The maturity level of knowledge management in ArcelorMittal, South Africa is low compared to the maturity at Monlevade and rival companies such as Tata and Posco steel. A significant effort needs to be made in order to allow the effective creation, acquisition, sharing and leveraging of knowledge within the South African steel facilities. The key factors which constrain effective knowledge management is related to ineffective Human Resource policies, organisation structure, lack of knowledge exchange forums, collaboration and communication, coaching, and a lack of incentives to share tacit knowledge. It is also evident that knowledge is not seen as a sustainable competitive advantage by many respondents and that they perceive they do not have the time or capacity to transfer knowledge. Ten practical design principles were constructed and a knowledge management framework was developed to guide South African steel companies during the design and execution of a knowledge management programme which will ensure that knowledge management will result into a sustainable competitive advantage. / Thesis (M.B.A.)--North-West University, Potchefstroom Campus, 2011.

The aim of this study is to conduct a thorough theoretical study on the relevant aspects involved in knowledge management and organisational learning, and to assess the maturity level of knowledge management within the South African steel industry. Various aspects of knowledge, knowledge management and organisational learning with specific relation to sustainable competitive advantage are discussed in the literature study. During the literature research several factors which either promote or hinder the effective management of knowledge were identified and a list of lessons learned by other successful knowledge focused companies, are discussed. The critical success factors for a successful knowledge management program are also discussed. Knowledge can create a sustainable competitive advantage within an organisation, if successfully applied to make value adding decisions and to enable learning, and if it is applied to make decisions which are superior to that of its rivals across the supply chain. An integrated approach needs to be followed when KM is pursued and the knowledge must be applied to make value added decisions and facilitate learning across all processes in the value chain. The focus must be to retain an organisation’s tacit knowledge as this is a key success factor to ensure a sustainable competitive advantage. The study includes research on whether knowledge management is effectively used as a sustainable competitive advantage in the South African steel industry. The maturity level of the application of knowledge and learning principles implemented within the South African steel industry is assessed and compared to the maturity level of ArcelorMittal, Monlevade, located in Brazil. A survey was designed and distributed to determine the knowledge management and organisational learning maturity levels at two steel facilities of ArcelorMittal in South Africa and one facility in Brazil. The key problem areas as identified through the empirical research are discussed and it is concluded that South African facilities do not effectively use knowledge management as a sustainable competitive advantage. The maturity level of knowledge management in ArcelorMittal, South Africa is low compared to the maturity at Monlevade and rival companies such as Tata and Posco steel. A significant effort needs to be made in order to allow the effective creation, acquisition, sharing and leveraging of knowledge within the South African steel facilities. The key factors which constrain effective knowledge management is related to ineffective Human Resource policies, organisation structure, lack of knowledge exchange forums, collaboration and communication, coaching, and a lack of incentives to share tacit knowledge. It is also evident that knowledge is not seen as a sustainable competitive advantage by many respondents and that they perceive they do not have the time or capacity to transfer knowledge. Ten practical design principles were constructed and a knowledge management framework was developed to guide South African steel companies during the design and execution of a knowledge management programme which will ensure that knowledge management will result into a sustainable competitive advantage. / Thesis (M.B.A.)--North-West University, Potchefstroom Campus, 2011.