Back to publications

A time-series clustering methodology for knowledge extraction in energy consumption data

Baca-Ruiz, LuisĀ G. and Pegalajar, M.C. and Arcucci, Rossella and Molina-Solana, Miguel
Expert Systems with Applications , under review, pp. (SUBMITTED)

Abstract:

One of the essential aims for incorporating intelligent systems in cities and buildings is the energy savings and pollution reduction that can be attained. To achieve this goal, energy modelling and better understanding of how energy is been consumed are key factors. As a result, a methodology for knowledge acquisition in energy-related data is proposed here by the use of Time-Series Clustering (TSC) techniques. In our experimentation, we utilize data from the buildings at University of Granada (Spain) and we compare several clustering methods for getting the best model as well as several algorithms for obtaining the best grouping. Thus, our methodology is able to obtain non-trivial knowledge from raw energy data. In contrast to previous studies in this field, an automatic recursive strategy is proposed here for searching and analysing periodicity in these series.

Links:

DOI:
PDF:

Bibtex:

@article{Baca2019b,
  author = {Baca-Ruiz, Luis~G. and Pegalajar, M.C. and Arcucci, Rossella and Molina-Solana, Miguel},
  title = {A time-series clustering methodology for knowledge extraction in energy consumption data},
  journal = {Expert Systems with Applications},
  year = {SUBMITTED},
  volume = {under review},
  doi = {},
  comment = {},
  timestamp = {23}
}