Premium
The functional latent block model for the co‐clustering of electricity consumption curves
Author(s) -
Bouveyron Charles,
Bozzi Laurent,
Jacques Julien,
Jollois FrançoisXavier
Publication year - 2018
Publication title -
journal of the royal statistical society: series c (applied statistics)
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.205
H-Index - 72
eISSN - 1467-9876
pISSN - 0035-9254
DOI - 10.1111/rssc.12260
Subject(s) - cluster analysis , computer science , electricity , block (permutation group theory) , consumption (sociology) , inference , stochastic block model , functional data analysis , data mining , mathematical optimization , artificial intelligence , machine learning , mathematics , engineering , social science , geometry , sociology , electrical engineering
Summary As a consequence of recent policies for smart meter development, electricity operators nowadays can collect data on electricity consumption widely and with a high frequency. This is in particular so in France where the leading electricity company Électricité de France will be able soon to record the consumption of its 27 million clients remotely every 30 min. We propose in this work a new co‐clustering methodology, based on the functional latent block model (LBM), which enables us to build ‘summaries’ of these large consumption data through co‐clustering. The functional LBM extends the usual LBM to the functional case by assuming that the curves of one block live in a low dimensional functional subspace. Thus, the functional LBM can model and cluster large data sets with high frequency curves. A stochastic expectation–maximization–Gibbs algorithm is proposed for model inference. An integrated information likelihood criterion is also derived to address the problem of choosing the number of row and column groups. Numerical experiments on simulated and original Linky data show the usefulness of the methodology proposed.