Timezone: »
Motivated by electricity consumption reconstitution, we propose a new matrix recovery method using nonnegative matrix factorization (NMF). The task tackled here is to reconstitute electricity consumption time series at a fine temporal scale from measures that are temporal aggregates of individual consumption. Contrary to existing NMF algorithms, the proposed method uses temporal aggregates as input data, instead of matrix entries. Furthermore, the proposed method is extended to take into account individual autocorrelation to provide better estimation, using a recent convex relaxation of quadratically constrained quadratic programs. Extensive experiments on synthetic and real-world electricity consumption datasets illustrate the effectiveness of the proposed method.
Author Information
Jiali Mei (EDF R&D & Université Paris-Sud)
Yohann De Castro (LMO)
Yannig Goude (EDF Lab Paris-Saclay)
Georges Hébrail (EDF Lab Paris-Saclay)
Related Events (a corresponding poster, oral, or spotlight)
-
2017 Talk: Nonnegative Matrix Factorization for Time Series Recovery From a Few Temporal Aggregates »
Mon Aug 7th 04:24 -- 04:42 AM Room C4.4
More from the Same Authors
-
2019 Poster: Target Tracking for Contextual Bandits: Application to Demand Side Management »
Margaux Brégère · Pierre Gaillard · Yannig Goude · Gilles Stoltz -
2019 Oral: Target Tracking for Contextual Bandits: Application to Demand Side Management »
Margaux Brégère · Pierre Gaillard · Yannig Goude · Gilles Stoltz