Particle Swarm Optimization Algorithm with Adaptive Inertia Applied to Residential Load Identification

Abstract

Electricity demand has been constantly increasing, mainly in the last decade, as well as the solutions for the efficient use of electrical energy. Therefore, it is of great interest to the scientific community and energy sector the development of Nonintrusive Load Monitoring methodologies. In this sense, some NILM methodologies can be characterized as an optimization problem, being this the focus of the present research project. For this purpose, it will be considered an approach based on Particle Swarm Optimization algorithm with adaptive inertia. Thus, the optimization algorithm will be responsible to estimate the load vector with its respective consumptions for a determined temporal window, i.e., it will disaggregate the residential consumption for each appliance. In order to analyze and compare the proposed methodology with the state-of-the-art ones, a public benchmark dataset will be used, namely as UK-DALE (United Kingdom -- Domestic Appliance-Level Electricity).