Electricity Load Forecast using Data Streams Techniques. Pedro RODRIGUES, Joao GAMA.
La revue MODULAD, numéro 36, Juillet 2007

Abstract:

Sensors distributed all around electrical-power distribution networks produce streams of data at high-speed. From a data mining perspective, this sensor network problem is characterized by a large number of variables (sensors), producing a continuous flow of data, in a dynamic non-stationary environment. In this work we analize the most relevant data mining problems and issues: online learning and change detection. We propose an architecture based on an online clustering algorithm where each cluster (group of sensors with high correlation) contains a neural-network based predictive model. The goal is to maintain in real-time a clustering model and a predictive model able to incorporate new information at the speed data arrives, detecting changes and adapting the decision models to the most recent information. We present preliminary results illustrating the advantages of the proposed architecture.


Keywords: Electricity demand forecast, online clustering, incremental neural networks.

Download paper : Electricity Load Forecast using Data Streams Techniques

Download slides : Electricity Load Forecast using Data Streams Techniques