Abstract
Smart Homes (SHs) represent the human side of a Smart Grid (SG). Data mining and analysis of energy data of electrical devices in SHs, e.g., for the dynamic load management, is of fundamental importance for the decision-making process of energy management both from the consumer perspective by saving money and also in terms of energy redistribution and reduction of the carbon dioxide emission, by knowing how the energy demand of a building is composed in the SG. Advanced monitoring and control mechanisms are necessary to deal with the identification of appliances. In this paper, a model for their automatic identification is proposed. It is based on a set of 19 features that are extracted by analyzing energy consumption, time usage and location from a set of device profiles. Then, machine learning approaches are employed by experimenting different classifiers based on such model for the identification of appliances and, finally, an analysis on the feature importance is provided.
Highlights
Through the integration and interconnection of software-centered devices, traditional power system, which are typically centered on mechanical and electrical components, are supported by more complex equipment which enable more advanced functionalities in terms of management and control
The rationale behind them relies on extracting and using information that is directly derivable or measurable from a device, such as its energy consumption, and to combining it with further information related to the way of using a particular device—for example, by considering differences of the use of a device in specific daily time slots, weekly or seasonal, relationships with other devices such as their use in sequence or in parallel, as well as by distinguishing whether a device is used in a specific area or if it is used, with a certain frequency, in different places
The trace-base data have been grouped into three categories in accordance with the process they have been collected: “Full-day traces”: which contains complete traces that have been recorded for a time period over 24 h; “Incomplete traces”: which contains traces with missing measurements in different instants of time over the day; “Synthetic traces”: which contains trace fragments of devices that have been manually completed with values corresponding to zero consumption readings, when the real values were not available
Summary
Through the integration and interconnection of software-centered devices, traditional power system, which are typically centered on mechanical and electrical components, are supported by more complex equipment which enable more advanced functionalities in terms of management and control. The result of such evolution which makes those systems more intelligent is named as Smart Grid (SG) [1,2]. The advantage of having software components in the network enables the introduction of more sophisticated mechanisms for monitoring the SG as well as to support, in a more effective way, the decision process for the dynamic energy distribution according to the current use of energy, resource state and weather conditions [3,4,5,6].
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call