Electricity consumption prognosis with the combination of smart metering and artificial neural networks

This work is an effort in order to predict in house sector one hour ahead the electricity consumption (EC). For this purpose, the combination of a Smart Meter (SM) with Automated Meter reading technology (AMR), online meteorological data and Artificial Neural Network (ANN) models were used in an area of Athens city, Greece. Concretely, a SM was used to record the EC in a residence in the Moschato municipality, which is located in the south of Athens city. Simultaneously, through the web portal Metar which is under the auspices National Observatory of Athens, online meteorological data concerning the area of Moschato were collected. Finally, an ANN forecasting model was developed and applied in order to predict the energy demand in a residence house, one hour ahead. Results showed that the combination of a SM and ANN model is a very promising tool for better management of electricity demand in the future.