The successes or failures one experiences in life depend greatly on the decisions that one makes. This is not only true in one’s personal life but is also the case in the business environment. In this modern world, simple and complex decisions are the key elements for a business to be successful in the competitive global environment. Effective decision making is an intricate process. In Eskom it is important to integrate the technological and business aspects to support the decision making process. Research methodology provides one with the necessary tools to support this decision making. The main focus of the study is the development of an Optimisation Model for the Management of Transformers in the Eskom East Grid Transmission Sysytem. In the Eskom Transmission system, there are a large number of power transformers which are the most expensive and strategically important component of the Transmission system. There were three main objectives. The first objective included the forecasting of transformer failures using the forecasting techniques of moving average, weighted moving average, exponential smoothing and regression analysis. The second objective focused on the investigation of the re – location of the tap changer maintenance team to a new area using the mathematical and statistical methods of simulation and decision tree analysis. The third objective included the investigation of an inventory control and management model where the optimum number of spares pertaining to transformer units that should be made available was determined using the mathematical and statistical model called the economic order quantity. These objectives were then used to investigate the development of the optimisation model pertaining to transformers. The results of the study concluded that the operating research techniques which included the forecasting methods and the economic order quantity models were suitable for research in Eskom. However it is important to note that the Eskom system and the environment in which transformers operate in is dynamic and has some factors that cannot be controlled. These factors must be taken into consideration when the various models are used in the investigation of the optimisation model. The introduction of these external factors is beyond the scope of this study and is not included. It was concluded that the simulation and decision tree analysis could be used as an integral part of the optimisation model successfully. The limitations that were highlighted included the integrity of the secondary data (sample size, the source of the secondary data, data quality and data governance), the limitations associated with forecasting, the limitations of the operations research, mathematical and statistical models and the fact that the Eskom network is dynamic. . The recommendations included the application of the forecasting techniques and the inventory control model to a larger population size which was that of the transformers in the entire Transmission system. It was indicated that Eskom resources, time and money must be used to support the business’s strategy to train and develop employees to an acceptable competency level. Future studies should include the factors that affect forecasting and the implementation of the inventory control model to increase the accuracy of the results. The introduction of these external factors is beyond the scope of this study and was not included in the models. / Thesis (MBA)-University of KwaZulu-Natal, Westville, 2009.

Submitted by NADYA KALACHE null (nadyakalache@gmail.com) on 2017-02-02T18:30:28Z No. of bitstreams: 1 Nadya Kalache.pdf: 2308527 bytes, checksum: 9cdc9b520310cc767a416203f4a8451d (MD5) / Approved for entry into archive by LUIZA DE MENEZES ROMANETTO (luizamenezes@reitoria.unesp.br) on 2017-02-06T16:12:06Z (GMT) No. of bitstreams: 1 kalache_n_dr_ilha.pdf: 2308527 bytes, checksum: 9cdc9b520310cc767a416203f4a8451d (MD5) / Made available in DSpace on 2017-02-06T16:12:06Z (GMT). No. of bitstreams: 1 kalache_n_dr_ilha.pdf: 2308527 bytes, checksum: 9cdc9b520310cc767a416203f4a8451d (MD5) Previous issue date: 2016-12-09 / O gerenciamento econômico de transformadores de distribuição é de fundamental importância no planejamento econômico das concessionárias de energia, pois estão presentes em grande quantidade no sistema elétrico. No novo conceito de redes inteligentes, novas considerações sobre as perdas nos transformadores devem ser feitas e outras possibilidades de gerenciamento podem ser exploradas. A primeira afirmação se deve ao fato do aumento da não-linearidade no perfil da carga no transformador, proveniente do aumento de cargas eletrônicas e de unidades de geração distribuída que utilizam conversores eletrônicos para conexão com a rede. A segunda afirmação é justificada pela infraestrutura avançada de medição presente nas redes inteligentes, o que possibilita acesso remoto e dinâmico a uma maior quantidade de informações fundamentais para análise das condições operacionais dos transformadores. Este trabalho analisa como utilizar um sistema de monitoramento em redes inteligentes em conjunto com o cálculo de perdas para identificação de transformadores sobrecarregados. Além disso, com o conhecimento das curvas de carga diária, dos índices econômicos vigentes, desenvolveu-se um aplicativo para gerenciamento de unidades transformadoras em operação, ferramenta esta que poderá ser utilizada no sistema de gestão de ativos das concessionárias. / Economic management of distribution transformers has fundamental importance in the economic planning of utilities, since they are present in large quantities in the electrical system. In the new concept of smart grids, new considerations about losses in transformers must be made and other management possibilities can be explored. The first statement is due to the increase of non-linearity in the load profile transformer, from the increase of electronic loads and distributed generation units that use electronic converters for connection to the grid. The second statement is justified by the advanced metering infrastructure available in smart grids, which enables remote and dynamic access to a greater amount of essentials information for analyzing the transformers operating conditions. This paper analyzes how to use a monitoring system in smart grids with the calculation of losses for identification of overloaded transformers. Also, by the knowledge of daily load curves, the current economic indices, an application was developed to manage transforming units in operation, a tool that can be used in the assets management system of dealers.

Orientador: Luis Carlos Origa de Oliveira / Resumo: O gerenciamento econômico de transformadores de distribuição é de fundamental importância no planejamento econômico das concessionárias de energia, pois estão presentes em grande quantidade no sistema elétrico. No novo conceito de redes inteligentes, novas considerações sobre as perdas nos transformadores devem ser feitas e outras possibilidades de gerenciamento podem ser exploradas. A primeira afirmação se deve ao fato do aumento da não-linearidade no perfil da carga no transformador, proveniente do aumento de cargas eletrônicas e de unidades de geração distribuída que utilizam conversores eletrônicos para conexão com a rede. A segunda afirmação é justificada pela infraestrutura avançada de medição presente nas redes inteligentes, o que possibilita acesso remoto e dinâmico a uma maior quantidade de informações fundamentais para análise das condições operacionais dos transformadores. Este trabalho analisa como utilizar um sistema de monitoramento em redes inteligentes em conjunto com o cálculo de perdas para identificação de transformadores sobrecarregados. Além disso, com o conhecimento das curvas de carga diária, dos índices econômicos vigentes, desenvolveu-se um aplicativo para gerenciamento de unidades transformadoras em operação, ferramenta esta que poderá ser utilizada no sistema de gestão de ativos das concessionárias. / Doutor