Shifting the load demand of a commercial office space to utilise off peak tariffs would lead to cost savings as power is cheaper at this time. To achieve this shift, chemical energy storage was considered using Lead Acid batteries, Lithium-ion batteries and Advance Lead Acid batteries. The output of these storage types is electricity. Current costs of storage do not support the option of reducing peak demand by adding chemical storage, as electricity from the grid is cheaper over the storage project life. A levelised cost analysis was completed and lithium-ion batteries proved to be the outright best choice for chemical storage in commercial office spaces. Town buildings were analysed and assessed for energy savings in order to reduce overall load demand. Incorporating chemical storage as a viable option was assessed based on cost. Heating, Cooling and lighting proved to be the highest load demands in the buildings. Cost savings in buildings can be better met by increasing the efficiency in buildings, rather than by reducing the cost by shifting the purchase of electricity from peak to off-peak tariffs. More suitable options to chemical energy storage would be to replace standard lights with energy efficient variations, installing an air heat pump to heat the space and ice thermal storage to cool the space. The use of a solar thermosyphon will meet the demand for heated water. Solar energy generation was assessed as an alternative as off-peak electricity stored using chemical storage proved to be too costly. A 50kW system would be suitably sized when peak sun hours were above the local areas average. Net metering could be used to offset costs during the winter months, when the peak sun hours fall below the average.

Uma vez que sistemas de inferência neuro-fuzzy adaptativos são aproximadores universais que podem ser usados em aplicações de aproximação de funções e de previsão, este trabalho tem por objetivo determinar seus melhores parâmetros e suas melhores arquiteturas com o propósito de se executar previsão de demanda de energia elétrica no curtíssimo prazo em subestações de distribuição. Isto pode possibilitar o desenvolvimento de controles automáticos de carga mais eficientes para sistemas elétricos de potência. As entradas do sistema são séries temporais de demanda de energia elétrica, compostas por dados mensurados em intervalos de cinco minutos ao longo de sete dias em subestações localizadas em cidades do interior do estado de São Paulo. Diversas configurações de entrada e diferentes arquiteturas foram examinadas para se fazer a previsão de um passo a frente. Os resultados do sistema de inferência neuro-fuzzy adaptativo frente às abordagens encontradas na literatura foram promissores. / Since adaptive neuro-fuzzy inference systems are universal approximators that can be used in functions approximation and forecasting applications, this work has the objective to determine their best parameters and best architectures with the purpose to execute very short term load forecasting in distribution substations. This can allow the development of more efficient load automatic control for power systems. The system inputs are load demand time series, which are composed of data measured at each five minutes interval, during seven days, from substations located in cities from São Paulo state countryside. Several input configurations and different architectures were examined to make a prediction aiming one step forecasting. The adaptive neuro-fuzzy inference system results in comparison with other approaches found in literature were promising.

Uma vez que sistemas de inferência neuro-fuzzy adaptativos são aproximadores universais que podem ser usados em aplicações de aproximação de funções e de previsão, este trabalho tem por objetivo determinar seus melhores parâmetros e suas melhores arquiteturas com o propósito de se executar previsão de demanda de energia elétrica no curtíssimo prazo em subestações de distribuição. Isto pode possibilitar o desenvolvimento de controles automáticos de carga mais eficientes para sistemas elétricos de potência. As entradas do sistema são séries temporais de demanda de energia elétrica, compostas por dados mensurados em intervalos de cinco minutos ao longo de sete dias em subestações localizadas em cidades do interior do estado de São Paulo. Diversas configurações de entrada e diferentes arquiteturas foram examinadas para se fazer a previsão de um passo a frente. Os resultados do sistema de inferência neuro-fuzzy adaptativo frente às abordagens encontradas na literatura foram promissores. / Since adaptive neuro-fuzzy inference systems are universal approximators that can be used in functions approximation and forecasting applications, this work has the objective to determine their best parameters and best architectures with the purpose to execute very short term load forecasting in distribution substations. This can allow the development of more efficient load automatic control for power systems. The system inputs are load demand time series, which are composed of data measured at each five minutes interval, during seven days, from substations located in cities from São Paulo state countryside. Several input configurations and different architectures were examined to make a prediction aiming one step forecasting. The adaptive neuro-fuzzy inference system results in comparison with other approaches found in literature were promising.

It is more than a fact that electrical energy is a main production factor of every economic activity. Since electrical power is not easy to store, it needs to be consumed as it is generated in order to keep a constant balance between supply and demand. As a result, for developing an efficient energy market it is significant to create a method for accurately forecasting the electricity consumption. This thesis describes a method for analyzing data provided by the ENTSO-E transparency platform. The ENTSO-E (European Network of Transmission System Operators) is a network of electricity operators from 36 countries across Europe. Its main objective is to provide transparency concerning data of electricity generation and consumption in Europe in order to promote the development of efficient and competitive electricity markets. By using the method described in this thesis, one may use historical data provided by ENTSO-E to forecast the electricity consumption of an EU country for the years to come. As an example, data of electricity consumption in Greece during the years 2015-2018 have been used in order to calculate the average load demand of a weekday during the year 2030. On the other hand, in order to correctly predict the electricity demand of a specific region over the next decade, one should take into account some crucial parameters that may influence not only the evolution of the load demand, but also the fuel mix that will be used in order to cover our future electricity needs. Advances in power generation technologies, evolution of fuel prices, expansion of electricity grid and economic growth are a subset of parameters that should be taken into account for an accurate forecast of the electricity consumption in the long run. Particularly for Greece, a set of parameters that may affect the electricity consumption are being computationally analyzed in order to evaluate their contribution to the load demand curve by the year 2030. These include the interconnection of Greek islands to the mainland, the development of Hellinikon Project and the increase of the share of electric vehicles. The author of this thesis has developed code in Python programming language that can be found in the Appendix. These scripts and functions that implement most of the calculations described in the following chapters can also be used for forecasting the load demand of other EU countries that are included in the ENTSO-E catalogue. The datasets used as input to these algorithms may also be used from the readers to identify more patterns for predicting the load demand for a specific region and time. A sustainable energy system is based on consumers with environmental awareness. As a result, citizens living inside the European Union should become a member of a community that promotes energy saving measures, investments in renewable energy sources and smart metering applications.

[pt] O Brasil possui um Sistema Interligado Nacional (SIN) que se baseia na geração de energia elétrica por meio de usinas hidrelétricas, térmicas, solares fotovoltaicas e eólicas. O planejamento e operação deste sistema é efetuado com base em previsões efetuadas em curto, médio e longo prazo a fim de evitar imprevistos que possam afetar o suprimento da demanda de energia elétrica em território nacional. Uma das informações consideradas fundamentais em cada uma das etapas do planejamento da operação é a carga, ou seja, a demanda por energia elétrica. Quando trabalhada em curto prazo, esta é importante para a programação diária da operação, garantindo um cenário ótimo para uso dos recursos disponíveis e, em cenário mais atual, determinação do Preço de Liquidação das Diferenças a cada hora. Quando trabalhada em médio prazo, esta funciona como base para manutenções de redes e negociações de contrato. Já em longo prazo, a previsão é importante para fornecer informações usadas como base para estratégias de expansão do Sistema. Normalmente a previsão em longo prazo é trabalhada de maneira a escalonar a curva histórica anual, mas as constantes alterações no hábito de consumo da população e a inserção de novas fontes ocasionam relevantes alterações no perfil da curva de carga diária em longo prazo, tornando necessário o planejamento não somente da expansão do sistema, mas também a forma com que este poderá ser programado. Assim, com o objetivo de propor uma ferramenta de suporte ao mercado brasileiro de energia, este trabalho propõe uma Metodologia para Geração de Cenários de Carga de Longo Prazo. O método proposto propõe uma abordagem bottom-up para previsão anual da demanda utilizando premissas de trabalhos acadêmicos recentes, propõe um método de geração de perfis específicos para suprir a escassez de dados horários detalhados no Brasil e propõe a utilização da Abordagem de Decomposição Parcial a fim de transformar as previsões anuais de demanda em curvas de carga horária. Finalizando a aplicação da Metodologia para Geração de Cenários de Longo Prazo, diferentes resultados gerados são utilizados para aplicação de simulação por Monte Carlo, sendo os intervalos de confianças gerados com base na resposta, possíveis cenários de comportamento da carga no futuro, transformando um método de previsão previamente determinístico em um previsor de cenários. Com o objetivo de demonstrar resultados da método, a Metodologia é aplicada para geração de cenários de longo prazo para a região sudeste brasileira até 2020 com base na curva histórica de 2016, apesar de ser capaz de gerar previsões para horizontes maiores, demonstrando verdadeiro potencial para se adaptar a possíveis alterações na curva de carga. / [en] Brazil has a National Interconnected System which produces and transmits electrical energy through a hydro-thermo-wind system. The planning and operation of this system is based on short, medium and long term on forecasts in order to avoid unforeseen that may affect the electricity supply in national territory. The short-term forecast is important for daily schedule of operation, certifying the resource use optimal scenario and, in a current scenario, the determination of Settlement Price for Differences at each hour. The medium-term forecast is used as a basis for network maintenance and contract negotiations. The long-term forecast is important to provide information used as basis for system expansion strategies. Usually, the long-term forecast is made staggering the annual load curve, however, the constant changes on people electrical consumption habits and insertion of new electrical generation sources cause relevant changes in daily load curve profile over the long term, making necessary not only the expansion planning, but also the way it can be programmed on long-term horizon. Thus, in order to propose a support tool to the Brazilian energy market, this work presents a Scenarios Generation Methodology. Such procedure proposes bottom-up approach as an annual demand projection provider, using assumptions of recent academic works, proposes a specific profile generation method as a way to overcome the lack of specific hourly data in Brazil. Not only that, the method also proposes Partial Decomposition Approach to adapt annual electricity demand into hourly load curves. Concluding the Scenarios Generation Methodology, future scenarios are developed by Monte Carlo simulation applied over different obtained results and confidence intervals calculated based on response are possible values of load behavior in the future, thus turning a deterministic forecasting method into a scenarios generation methodology. In order to demonstrate the Methodology application, it is used to generate long-term scenarios for the southeast Brazilian region by 2020 based on historical load curve from 2016, although it is capable of generating forecasts for larger horizons, proving true potential to adapt to possible changes on load curve.

This thesis addressed the need for improved analysis and interpretation of primary meter half-hourly energy consumption data. The current work offers a novel benchmarking technique that was tested for 6 types of municipal buildings. This approach is different from conventional annual benchmarking mainly because it uses electricity and gas data in half-hourly periods, together with outside temperature data. A survey to European local authorities’ metering and monitoring practices was conducted in order to assess municipal energy managers' current procedures and needs in terms of data analysis to assess building energy performance and to identify potential energy saving opportunities. The benchmarking approach was developed considering the energy managers’ needs, but also the state-of the art in terms of building energy monitoring techniques, particularly building energy signatures, and the analysis techniques used on electricity grid demand forecasting. The benchmarking approach is based on the use of a metric composed of several indicators that are related to the load demand shape profile and the building energy signature. The comparison of indicators for buildings of the same type using standard scores identifies uncommon load demand profile characteristics and/or gas dependency on outside temperature in specific buildings. The metric is able to support the identification of potential energy wastage, which is linked to the detection of opportunities to save energy. The benchmarking technique was tested in 81 municipal building owned by Leicester City Council. This methodology can be applied to any non-domestic building equipped with primary meters for registering half-hourly electricity and gas consumption. In theory, this approach can also be applied to residential buildings, and to other short time series data types, for example quarter-hourly or 10 minutes interval data. The main contribution of this thesis is to improve the objectivity of building primary meter half-hourly electricity and gas consumption data analysis and interpretation by using quantitative parameters, instead of subjective visualisation techniques. The interpretation of building consumption data in short time series periods can now be streamlined, automated and perhaps incorporated in existing energy analysis software. This thesis raises questions that can lead to future research projects aiming to improve the metric and also to enlarge the scope of its application to national and European scale, to other building types and to other utilities.