Neural network application for household electricity consumption forecast. Special emphasis on renovation

Abstract

The building sector accounts for 25-40% of the final energy consumption in OECD countries, space heating being the largest proportion of energy consumption in both residential and commercial buildings. The particular microclimate conditions of the urban area have in fact a significant influence on the thermal balance of buildings. The paper is mainly based on Italian experience on neural network application for HVAC systems energy consumption forecasting. It presents a forecasting model based on Elman's Artificial Neural Network (ANN) for the short time prediction of the household electricity consumption related to the area. The model mainly used for Mediterranean area energy consumption forecasting can also be applied for heating energy consumption forecasting due to renovation works and renovation efficiency estimation in Lithuania.