Advances to Energy Informatics: On the Collection, Processing, and Privacy Protection of Electricity Consumption Data

Abstract

Electrical power grids around the world are currently being equipped with an increasing number of monitoring and actuation components, as part of their conversion to “smart grids”. This results in an improved visibility of the power grid’s real-time status as well as a fine-grained tracking of events. These features represent enabling elements for the adaptive and fine-grained decentralized control, targeting to sustain or even increase the power grid’s reliability. Besides catering to this purpose, an unprecedented volume of voltage, current, and/or power data becomes available from the distributed sensing devices. The analysis of such data enables a variety of use cases. These include the disaggregation of household consumption, the prediction of future electricity demand, or the detection of anomalies in the use of electrical power. The methodological collection and analysis of energy-related data is one of the major research challenges of the emerging “energy informatics” discipline. This thesis proposes a delineation of the scope of energy informatics, with a focus on electrical power grids. It furthermore presents selected contributions to the field of energy informatics, including approaches for the methodological data collection, energy analytics algorithms for the data analysis and interpretation, as well as ways to provide user-centric services based on such data. Lastly, the thesis highlights possible directions for future research in the field of energy informatics.