Orientadores: Sergio Valdir Bajay, Gilberto De Martino Jannuzzi / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Mecânica / Made available in DSpace on 2018-08-19T02:38:12Z (GMT). No. of bitstreams: 1 Paccola_JoseAngelo_M.pdf: 1894026 bytes, checksum: 35049932286f9b2cae4ab204843bfb40 (MD5) Previous issue date: 2007 / Resumo: Durante as últimas décadas, têm-se introduzido poucas modificações no setor elétrico brasileiro no que se refere às novas modalidades de tarifas para os consumidores finais. As mudanças mais radicais foram implantadas no início da década de oitenta com a introdução das tarifas horosazonais verde e azul. O objetivo desta dissertação é testar as possibilidades de se formular novas tarifas horosazonais no País, através de uma análise crítica da aplicação destas no Brasil e no mundo, e a realização de um estudo de caso envolvendo pesquisas de campo, entrevistas e medições em consumidores industriais atendidos na categoria tarifária A4. A análise da experiência nacional aborda a implantação e o desenvolvimento das tarifas horosazonais - verde, azul e amarela - e tarifas de fornecimento interruptível no País por meio de um histórico completo dos fatos mais importantes, ocorridos desde 1957 até 2006, e de uma avaliação dos resultados a que se chegou com a implantação destas tarifas. Na análise da experiência internacional, examinaram-se os casos de França, Canadá, Estados Unidos e Portugal. O estudo de caso envolveu três pesquisas de campo. A primeira delas teve como objetivo escolher os segmentos industriais com maior potencial para a modulação de carga. Com a segunda pesquisa, conseguiu-se conhecer melhor os consumidores dos segmentos de calçados e de móveis, em termos de perfil de demanda e de detalhes de seus processos produtivos, visando estabelecer, com mais segurança, suas possibilidades de modulação de carga. A terceira pesquisa de campo permitiu o levantamento dos dados técnicos e econômicos, necessários para se fazer simulações e uma avaliação quantitativa dos impactos econômicos decorrentes de um terceiro posto tarifário para estes consumidores, na madrugada. As análises custo/benefício, realizadas no estudo de caso, levaram em conta tanto a ótica do consumidor como a da concessionária. Os resultados das simulações realizadas mostraram que os custos com mão-de-obra chegam a ser 35 vezes maiores do que aqueles com a fatura de eletricidade nestes segmentos industriais. Isto desestimula um possível deslocamento de parte da produção para o período noturno, por conta do adicional noturno no custo da mão-de-obra, mesmo com tarifas de energia elétrica muito baixas neste período / Abstract: During the last decades, the Brazilian electric sector has introduced little changes in tariffs for end-use consumers. The most radical changes were implanted at the beginning of the eighties, with the introduction of a seasonal tariff structure called green and blue. The objective of this work is to test some possibilities to formulate a new seasonal tariff in Brazil, through a critical analysis of these tariffs in Brazil and in other countries. This was made through field researches, interviews and the measuring of some industrial consumers. The national experience analysis includes the implementation and development of the seasonal tariffs - green, blue and yellow -, the supply curtailable rate and a complete historical of the more important facts occurred since 1957 up to 2006, together with an evaluation of the impacts in the implementation of these tariffs. In the analysis of the international experience, it was examined tariffs in France, Canada, United States and Portugal. The case study involved three researches on the field. The first one was to choose the industrial segments with larger potential for load modulation. The second research explored both footwear and furniture industries in terms of theirs demand profile and productive processes, with the objective of establishing their load modulation change possibilities more accurately. The third field research obtained the necessary technical and economical data, to work with simulations in a quantitative analysis of the economic impacts of a third tariff position during night time. The cost-benefit analysis considered both consumers and Utility's point of views. The results of the simulations has shown that the cost of labor is sometimes 35 times higher than electricity bills in the footwear and furniture industries, rendering useless a possible load modulation change during the night. Furthermore, during night time the cost of labor increases / Mestrado / Planejamento de Sistemas Energeticos / Mestre em Planejamento de Sistemas Energéticos

MSc (Statistics) / Department of Statistics / Forecasting of electricity peak demand levels is important for decision makers in Eskom. The overall objective of this study was to develop medium term load forecasting models which will help decision makers in Eskom for planning of the operations of the utility company. The frequency table of hourly daily demands was carried out and the results show that most peak loads occur at hours 19:00 and 20:00, over the period 2009 to 2013. The study used generalised additive models with and without tensor product interactions to forecast electricity demand at 19:00 and 20:00 including daily peak electricity demand. Least absolute shrinkage and selection operator (Lasso) and Lasso via hierarchical interactions were used for variable selection to increase the model interpretability by eliminating irrelevant variables that are not associated with the response variable, this way also over tting is reduced. The parameters of the developed models were estimated using restricted maximum likelihood and penalized regression. The best models were selected based on smallest values of the Akaike information criterion (AIC), Bayesian information criterion (BIC) and Generalized cross validation (GCV) along with the highest Adjusted R2. Forecasts from best models with and without tensor product interactions were evaluated using mean absolute percentage error (MAPE), mean absolute error (MAE) and root mean square error (RMSE). Operational forecasting was proposed to forecast the demand at hour 19:00 with unknown predictor variables. Empirical results from this study show that modelling hours individually during the peak period results in more accurate peak forecasts compared to forecasting daily peak electricity demand. The performance of the proposed models for hour 19:00 were compared and the generalized additive model with tensor product interactions was found to be the best tting model. / NRF