Towards Context-aware Power Forecasting in Smart-homes

Forecasting future power consumption in residential buildings is important to optimize the power grid, to assist inhabitants in everyday activities, and to save energy. Several machine learning methods have been proposed to predict future electricity consumption in smart homes based on the history of past consumption data acquired from smart meters. However, the increasing availability of smart home sensors can provide insights about the routines and activities of inhabitants, that may be exploited to provide more accurate predictions. In this paper, we propose a machine learning approach to forecast future energy consumption considering not only past consumption data, but also context data such as inhabitants' actions and activities, use of household appliances, interaction with furniture and doors, and environmental data. We performed an experimental evaluation with real-world data acquired in an instrumented environment from a large set of users. The results of a comparison with two baseline methods show that our approach is promising.