Name
Abstract | ||
|---|---|---|
Smart meter time series often show time features and cycles that relate well to their key underlying determinants. Yet, the time series are somewhat stochastic due to the extreme variability in occupants' behaviours, occupancy, the presence of electrical appliances, varying weather conditions and the specifics of the home's building envelope. This makes information gain a challenge when it comes to compression of big smart meter datasets, which otherwise would be overwhelming. This paper examines a method of modeling smart meter time series in the state space, so that the information gain is maximised. Some theories are discussed using a large residential smart meter dataset, from the Smart Grid Smart City project in Australia. The hypothetical outcomes and an account of the future works are also included.
|
Year | DOI | Venue |
|---|---|---|
2019 | 10.1145/3307363.3307394 | Proceedings of the 11th International Conference on Computer Modeling and Simulation |
Keywords | DocType | ISBN |
Load Profile, Smart Meter Time Series, State Space Reconstruction, Time Series Clustering, Time Series Compression | Conference | 978-1-4503-6619-9 |
Citations | PageRank | References |
0 | 0.34 | 0 |
Authors | ||
2 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Omid Motlagh | 1 | 0 | 0.34 |
| Jiaming Li | 2 | 0 | 0.34 |