Monitoring Electricity Consumption Based on Time Series Analysis

• RebecaP. DIAZ REDONDO ^[1] ; Ana FERNANDEZ VILAS ^[1] ; Alberto Estevez Caldas ^[1]
  1. [1] Universidade de Vigo

     Universidade de Vigo

     Vigo, España

     
• Localización: Intelligent Environments 2020 Workshop Proceedings of the 16th International Conference on Intelligent Environments / Carlos Ángel Iglesias Fernández (ed. lit.), José Ignacio Moreno Novella (ed. lit.), Alessandro Ricci (ed. lit.), Diego Rivera Pinto (ed. lit.), Dumitru Roman (ed. lit.), 2020, ISBN 978-1-64368-090-3, págs. 321-330
• Idioma: inglés
• Texto completo no disponible (Saber más ...)
• Resumen

  • Smart meters provide fine-grained accurate readings that can be analysed for a more efficient energy consumption monitoring. Using the energy readings as input data, we seek to obtain building energy consumption patterns, to detect anomalies in these patterns and to compare patterns from different buildings. With this aim, we introduce a hybrid methodology that combines several techniques based on time series analysis: (i) statistical and visualization approaches, (ii) techniques for decomposing time series, (ii) algorithms for checking the consistency and fluctuation of the data and (iv) algorithms for comparing time series shapes. Our experiments used a large data set that gathers the energy readings from different buildings in a university campus.