Reproduce “Occupancy Detection Based on Electricity Consumption” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Thomas Brilland, Guillaume Matheron, Laetitia Leduc, Yukihide Nakada
Researchers from Hello Watt developed a methodology for detecting extended holiday periods in residential settings using only low-frequency smart meter electricity consumption data. The approach identifies unoccupied intervals and extracts the consumption patterns of devices that remain active, providing insights for energy disaggregation and waste reduction without requiring additional hardware.
Key insights
- Variance-based classifiers are more effective than mean-based ones for distinguishing holiday periods, particularly when comparing segment variance to the full signal.
- Combining Discrete Fourier Transform and Autocorrelation (AutoPeriod) reliably detects periodic consumption patterns during unoccupied periods.
- Iteratively extracting periodic spikes allows for the separation of multiple overlapping appliance consumption patterns during unoccupied times.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.