DISTANCE AND SIMILARITY MEASURES FOR SENSORSSELECTION IN HEAVILY INSTRUMENTED BUILDINGS:APPLICATION TO THE INCAS PLATFORM

Abstract

Energy management in residential buildings is taking an increasing role in the construction workflows. It entails understanding thermal processes at stake in the buildings and quantifying energy consumption, which meets inhabitants comfort requirements. Experimental platforms such as INCAS aim at providing experts with a practical way to study such problems in real conditions. These heavily equipped buildings yield huge amounts of real-time data (sampling rates, number and types of sensors) for which new automatic approaches could be useful to thermal experts. Generic similarity measures from data-mining could therefore provide comprehensive analysis tools to thermal experts. This paper focuses on the ability of some distance and similarity measures to organize millions of data from homogeneous and heterogeneous sensors into coherent clusters. Simplifying data interpretations to thermal experts in highly equipped buildings, this approach could also stand as a basis for studying smart grids of less equipped domestic houses studies. Two types of similarity measures are explored. The first one consists of a set of three distances, and accounts for differences in terms of amplitude scaling and shifting between pairs of measurements. It relies on the comparison of homogeneous sensors by quantifying the relative proximity of their amplitude in terms of mean value, variance and time shift. The second type of similarity measure employs a pre-processing step transforming continuous signals into binary events. The resulting spike trains are then compared by quantifying the amount of unitary transformations (events moves or events deletions/additions) needed to align pairs of events sequences. These proximity measures are computed on real data from experimental buildings of the INCAS platform. It comprises three experimental buildings (with different construction types) dedicated to testing various approaches regarding systems, control and energy-saving policies. These geometrically identical buildings are equipped with hundreds of sensors measuring temperature, humidity, differential pressure, and others data at various positions of the structures with sampling rates of one measurement per minute. Simulation-based temperatures are integrated in the sensors set providing a comparison between real and simulated data. Results illustrate the contribution of the applied methods when dealing with large amounts of measurements related to instrumented buildings behaviors. Actually results show that coherent clusters regarding distinct signal properties are automatically generated. These clusters can be used for dimensionality reduction (clusters of sensors could be summarized by a single virtual measurement), or relative comparisons between sensors or between real and simulated datasets.