Abstract
Non-intrusive load monitoring (NILM) or energy disaggregation
aims to estimate appliance-level energy consumption from the
aggregate consumption data of households. While there is significant interest from academia and industry, NILM techniques are
still not adopted widely across households. This is mainly because
the techniques developed for one household cannot be generalized and applied in other households (applicability), and require
tremendous hand-tuning to apply across households (scalability).
To overcome the above issues, we propose a novel semi-supervised
energy disaggregation framework – UniversalNILM. The key idea
of UniversalNILM is to model appliances in a few training houses,
which has detailed appliance-level data and transfer this learning
on to test houses (blind disaggregation), which has only aggregate
house consumption to derive fine-grained appliance energy consumption. UniversalNILM was empirically evaluated across datasets
and outperforms the reported accuracy from both state-of-the-art
supervised and unsupervised NILM techniques.
aims to estimate appliance-level energy consumption from the
aggregate consumption data of households. While there is significant interest from academia and industry, NILM techniques are
still not adopted widely across households. This is mainly because
the techniques developed for one household cannot be generalized and applied in other households (applicability), and require
tremendous hand-tuning to apply across households (scalability).
To overcome the above issues, we propose a novel semi-supervised
energy disaggregation framework – UniversalNILM. The key idea
of UniversalNILM is to model appliances in a few training houses,
which has detailed appliance-level data and transfer this learning
on to test houses (blind disaggregation), which has only aggregate
house consumption to derive fine-grained appliance energy consumption. UniversalNILM was empirically evaluated across datasets
and outperforms the reported accuracy from both state-of-the-art
supervised and unsupervised NILM techniques.
| Original language | English |
|---|---|
| Title of host publication | e-Energy '18: Proceedings of the Ninth International Conference on Future Energy Systems |
| Editors | Hartmut Schmeck, Veit Hagenmeyer |
| Publisher | Association for Computing Machinery |
| Pages | 610-617 |
| Number of pages | 7 |
| ISBN (Print) | 9781450357678 |
| DOIs | |
| Publication status | Published - 15 Jun 2018 |
Publication series
| Name | Proceedings of the Ninth International Conference on Future Energy Systems |
|---|---|
| Publisher | Association for Computing Machinery |
