GA-ANN Short-Term Electricity Load Forecasting

This paper presents a methodology for short-term load forecasting based on genetic algorithm feature selection and artificial neural network modeling. A feed forward artificial neural network is used to model the 24-h ahead load based on past consumption, weather and stock index data. A genetic algorithm is used in order to find the best subset of variables for modeling. Three data sets of different geographical locations, encompassing areas of different dimensions with distinct load profiles are used in order to evaluate the methodology. The developed approach was found to generate models achieving a minimum mean average percentage error under 2 %. The feature selection algorithm was able to significantly reduce the number of used features and increase the accuracy of the models.

Consumer behavior after the Brazilian power rationing in 2001

In June 2001, after a dry period, the level of the water reservoirs in Brazil was below their operational levels. This situation, combined with other historical factors, led the country into a period of power rationing. As expected, power consumption lowered during this period. After December 2001, when the power rationing ended, electrical utilities expected to return to their normal power consumption in a matter of months, but the level of power consumption only returned to its level around years 2004 2005. Consumer behavior went through a change during this period, and the consumers kept this behavior after, leading to electrical and economical consequences until today. This paper presents an analysis of several factors that led to these events, including historical consumption data and comparisons with similar situations. The objective of this analysis is to give helpful information to electrical utilities, that could deal with similar situations, in their load forecasting studies. © 2006 IEEE.

Evolutionary heuristic to determine future land use

In the spatial electric load forecasting, the future land use determination is one of the most important tasks, and one of the most difficult, because of the stochastic nature of the city growth. This paper proposes a fast and efficient algorithm to find out the future land use for the vacant land in the utility service area, using ideas from knowledge extraction and evolutionary algorithms. The methodology was implemented into a full simulation software for spatial electric load forecasting, showing a high rate of success when the results are compared to information gathered from specialists. The importance of this methodology lies in the reduced set of data needed to perform the task and the simplicity for implementation, which is a great plus for most of the electric utilities without specialized tools for this planning activity. © 2008 IEEE.

Aplicações de sistemas multiagentes na previsão espacial de demanda elétrica em sistemas de distribuição

Pós-graduação em Engenharia Elétrica - FEIS

Sistema inteligente para previsão de carga multinodal em sistemas elétricos de potência

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Previsão espacial de demanda em sistemas de distribuição com uma base reduzida de dados

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Desenvolvimento de redes neurais para previsão de cargas elétricas de sistemas de energia elétrica

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Previsão de demanda de cargas elétricas por seleção de variáveis stepwise e redes neurais artificiais

Pós-graduação em Engenharia Elétrica - FEIS

Alueellinen sähkökuormien kehitysennuste Rovaniemen Verkko Oy:n jakelualueella

Tämän työn tavoitteena on skenaarioiden avulla luoda pitkän aikavälin alueellinen sähkökuormien kehitysennuste Rovaniemen Verkko Oy:lle. Pitkän aikavälin kuormitusennusteet ovat välttämättömiä verkon kehittämisen pohjalle, jotta verkko voidaan mitoittaa vastaamaan kuormitusta pitkälle tulevaisuuteen tekniset ja taloudelliset vaatimukset huomioiden. Kuormitusennusteen onkin jatkossa tarkoitus toimia apuvälineenä verkon strategisessa kehittämisessä. Pohjana kuormitusennusteissa on tilastokeskuksen ja Rovaniemen kaupungin väestö- ja työpaikkaennusteet. Väestöennusteiden ja erilaisten rakentamistilastoiden avulla arvioidaan uudisrakentamisen määrä tulevaisuudessa. Uudisrakentamisen kuormitusvaikutuksiin päästään työssä määritettyjen paikallisten ja rakennustyyppikohtaisten sähkön ominaiskulutuksien avulla. Kuormituksien alueellinen sijoittautuminen arvioidaan kaavoituksen ja kaupungin maankäytön toteuttamisohjelman avulla. Työssä tutkitaan myös tulevaisuudessa sähkönkäytössä tapahtuvien useiden muutosten vaikutusta alueelliseen kuormitukseen. Näitä muutoksia ovat muun muassa sähköautojen, hajautetun tuotannon, lämpöpumppujen ja kysynnän jouston lisääntyminen. Myös rakennusten jatkuvasti parantuva energiatehokkuus aiheuttaa muutoksia sähkön kulutukseen.

A Bayesian Approach for Forecasting Heat Load in a District Heating System

The growing population in cities increases the energy demand and affects the environment by increasing carbon emissions. Information and communications technology solutions which enable energy optimization are needed to address this growing energy demand in cities and to reduce carbon emissions. District heating systems optimize the energy production by reusing waste energy with combined heat and power plants. Forecasting the heat load demand in residential buildings assists in optimizing energy production and consumption in a district heating system. However, the presence of a large number of factors such as weather forecast, district heating operational parameters and user behavioural parameters, make heat load forecasting a challenging task. This thesis proposes a probabilistic machine learning model using a Naive Bayes classifier, to forecast the hourly heat load demand for three residential buildings in the city of Skellefteå, Sweden over a period of winter and spring seasons. The district heating data collected from the sensors equipped at the residential buildings in Skellefteå, is utilized to build the Bayesian network to forecast the heat load demand for horizons of 1, 2, 3, 6 and 24 hours. The proposed model is validated by using four cases to study the influence of various parameters on the heat load forecast by carrying out trace driven analysis in Weka and GeNIe. Results show that current heat load consumption and outdoor temperature forecast are the two parameters with most influence on the heat load forecast. The proposed model achieves average accuracies of 81.23 % and 76.74 % for a forecast horizon of 1 hour in the three buildings for winter and spring seasons respectively. The model also achieves an average accuracy of 77.97 % for three buildings across both seasons for the forecast horizon of 1 hour by utilizing only 10 % of the training data. The results indicate that even a simple model like Naive Bayes classifier can forecast the heat load demand by utilizing less training data.

Forecasting electricity load demand: analysis of the 2001 rationing period in Brazil

This paper studies the electricity load demand behavior during the 2001 rationing period, which was implemented because of the Brazilian energetic crisis. The hourly data refers to a utility situated in the southeast of the country. We use the model proposed by Soares and Souza (2003), making use of generalized long memory to model the seasonal behavior of the load. The rationing period is shown to have imposed a structural break in the series, decreasing the load at about 20%. Even so, the forecast accuracy is decreased only marginally, and the forecasts rapidly readapt to the new situation. The forecast errors from this model also permit verifying the public response to pieces of information released regarding the crisis.

Investigation of the relationship between load and loss factors for a Brazilian electric utility

With the considerable increase of the losses in electric utilities of developing countries, such as Brazil, there is an investigation for loss calculation methodologies, considering both technical (inherent of the system) and non-technical (usually associated to the electricity theft) losses. In general, all distribution networks know the load factor, obtained by measuring parameters directly from the network. However, the loss factor, important for the energy loss cost calculation, can only be obtained in a laborious way. Consequently, several formulas have been developed for obtaining the loss factor. Generally, it is used the expression that relates both factors, through the use of a coefficient k. Last reviews introduce a range of factor k within 0.04 - 0.30. In this work, an analysis with real life load curves is presented, determining new values for the coefficient k in a Brazilian electric utility. © 2006 IEEE.

Minimizing residential distribution system operating costs through intelligently scheduled plug-in hybrid electric vehicle charging

Rising fuel prices and environmental concerns are threatening the stability of current electrical grid systems. These factors are pushing the automobile industry towards more effcient, hybrid vehicles. Current trends show petroleum is being edged out in favor of electricity as the main vehicular motive force. The proposed methods create an optimized charging control schedule for all participating Plug-in Hybrid Electric Vehicles in a distribution grid. The optimization will minimize daily operating costs, reduce system losses, and improve power quality. This requires participation from Vehicle-to-Grid capable vehicles, load forecasting, and Locational Marginal Pricing market predictions. Vehicles equipped with bidirectional chargers further improve the optimization results by lowering peak demand and improving power quality.

O impacto energético e as emissões de CO2 dos veículos eléctricos em Portugal

Para a diminuição da dependência energética de Portugal face às importações de energia, a Estratégia Nacional para a Energia 2020 (ENE 2020) define uma aposta na produção de energia a partir de fontes renováveis, na promoção da eficiência energética tanto nos edifícios como nos transportes com vista a reduzir as emissões de gases com efeito de estufa. No campo da eficiência energética, o ENE 2020 pretende obter uma poupança energética de 9,8% face a valores de 2008, traduzindo-se em perto de 1800 milhões de tep já em 2015. Uma das medidas passa pela aposta na mobilidade eléctrica, onde se prevê que os veículos eléctricos possam contribuir significativamente para a redução do consumo de combustível e por conseguinte, para a redução das emissões de CO2 para a atmosfera. No entanto, esta redução está condicionada pelas fontes de energia utilizadas para o abastecimento das baterias. Neste estudo foram determinados os consumos de combustível e as emissões de CO2 de um veículo de combustão interna adimensional representativo do parque automóvel. É também estimada a previsão de crescimento do parque automóvel num cenário "Business-as-Usual", através dos métodos de previsão tecnológica para o horizonte 2010-2030, bem como cenários de penetração de veículos eléctricos para o mesmo período com base no método de Fisher- Pry. É ainda analisado o impacto que a introdução dos veículos eléctricos tem ao nível dos consumos de combustível, das emissões de dióxido de carbono e qual o impacto que tal medida terá na rede eléctrica, nomeadamente no diagrama de carga e no nível de emissões de CO2 do Sistema Electroprodutor Nacional. Por fim, é avaliado o impacto dos veículos eléctricos no diagrama de carga diário português, com base em vários perfis de carga das baterias. A introdução de veículos eléctricos em Portugal terá pouca expressão dado que, no melhor dos cenários haverão somente cerca de 85 mil unidades em circulação, no ano de 2030. Ao nível do consumo de combustíveis rodoviários, os veículos eléctricos poderão vir a reduzir o consumo de gasolina até 0,52% e até 0,27% no consumo de diesel, entre 2010 e 2030, contribuindo ligeiramente uma menor dependência energética externa. Ao nível do consumo eléctrico, o abastecimento das baterias dos veículos eléctricos representará até 0,5% do consumo eléctrico total, sendo que parte desse abastecimento será garantido através de centrais de ciclo combinado a gás natural. Apesar da maior utilização deste tipo de centrais térmicas para produção de energia, tanto para abastecimento das viaturas eléctricas, como para o consumo em geral, verifica-se que em 2030, o nível de emissões do sistema electroprodutor será cerca de 46% inferior aos níveis registados em 2010, prevendo-se que atinja as 0,163gCO2/kWh produzido pelo Sistema Electroprodutor Nacional devido à maior quota de produção das fontes de energia renovável, como o vento, a hídrica ou a solar.