Residential loads labelling through gramian angular field and recurrence plots: a comparative analysis using the UK-DALE dataset

• Grant number: 22/00750-9
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: April 01, 2022
• End date: March 31, 2023
• Field of knowledge: Engineering - Electrical Engineering - Electrical, Magnetic and Electronic Measurements, Instrumentation
• Principal Investigator: Ricardo Augusto Souza Fernandes
• Grantee: Juan Salin Corrêa
• Host Institution: Centro de Ciências Exatas e de Tecnologia (CCET). Universidade Federal de São Carlos (UFSCAR). São Carlos , SP, Brazil

Abstract

In recent years, the demand for electricity has been increasing rapidly and, for this reason, there is a need to efficiently use it. This way, the identification of residential loads enables such use for consumers and, moreover, is crucial for demand response programs. Therefore, the development of labeling algorithms is of great interest to the scientific community and the productive sector. Due to the variety of electrical/electronic loads in homes and their dynamic behavior, the search for patterns that better explain and allow the correct labeling of temporal windows becomes a challenging task, since a window may contain more than one load. In this sense, the present research proposes the transformation of time series into two-dimensional images, more specifically using Gramian Angular Field and Recurrence Plots. Thus, the images resulting from this process can effectively be labeled using techniques considered advanced in the scope of machine learning, as is the case of Convolutional Neural Networks. For this purpose, data from UK-DALE (United Kingdom -- Domestic Appliance-Level Electricity) will be considered, as it is a public dataset. The results will make it possible to define the effectiveness of the algorithms for transforming the time series into two-dimensional images, in addition to demonstrating the robustness and generalization capacity of the deep learning technique.(AU)