Abstract

Abstract
Forecasting heating and cooling energy demand in buildings plays a critical role in supporting building management and operation. Thus, analysing the energy consumption pattern of a building could help in the design of potential energy savings and also in operation fault detection, while contributing to provide proper indoor environmental conditions to the building's occupants.This paper aims at presenting the main results of a study consisting in forecasting the hourly heating and cooling demand of an office building located in Lisbon, Portugal, using machine learning models and analysing the influence of exogenous variables on those predictions. In order to forecast the heating and cooling demand of the considered building, some traditional models, such as linear and polynomial regression, were considered, as well as artificial neural networks and support vector regression, oriented to machine learning. The input parameters considered in the development of those models were the hourly heating and cooling energy historical records, the occupancy, solar gains through glazing and the outside dry-bulb temperature.The models developed were validated using the mean absolute error (MAE) and the root mean squared error (RMSE), used to compare the values obtained from machine learning models with data obtained through a building energy simulation performed on an adequately calibrated model.The proposed exploratory analysis is integrated in a research project focused on applying machine learning methodologies to support energy forecasting in buildings. Hence, the research line proposed in this article corresponds to a preliminary project task associated with feature selection/extraction and evaluation of potential use of machine learning methods. © 2020 IEEE.

Abstract
O sistema de distribuição de energia elétrica é a parcela do sistema de potência mais vulnerável aos eventos de interrupção, originados por fatores naturais externos ou intrínsecos aos equipamentos elétricos. Na mitigação dos impactos desses eventos, simuladores digitais em tempo real são utilizados na obtenção dos transitórios do sistema elétrico, todavia, essa técnica demanda grande esforço computacional. Nesse contexto, propõe-se uma técnica híbrida para simulação do transitório em sistemas de distribuição, combinando a alta taxa de amostragem dos modelos no domínio do tempo para monitoramento de tensão e corrente com a velocidade de processamento dos algoritmos que operam os modelos fasoriais em regime quase-estacionário, ou permanente. A metodologia proposta também permite considerar diferentes tecnologias de geradores distribuídos acoplados na rede. Os resultados dos testes realizados indicam a consistência da metodologia proposta, representando o comportamento do transitório no sistema de distribuição de energia elétrica. Todas as simulações realizadas são comparadas com valores amostrais obtidos usando um software comercial especializado.

Abstract

Abstract
Lubricant oil is an essential element in wind and hydropower generation. We present a lifecycle assessment (LCA) of the lubricant oils (mineral, synthetic and biodegradable) used in hydropower and wind power generation. The results are given in terms of energy used, associated emissions and costs. We find that, for the oil turbines and regulation systems considered here, biodegradable oil is a better option in terms of energy and CO2 equivalent emissions than mineral or synthetic oils, from production and recycling through to handling. However, synthetic and mineral oils are better options due to the potential risks associated with the use of biodegradable oil, generally when it comes into contact with water. There are also significant savings to be made in the operation of wind turbines when using an improved type of synthetic oil.

Abstract