nilmtk-device
simulates a monitoring device installed in the home connected to the mains power supply.
To simulate this device in a real world setting using the REDD dataset run:
python simulate.py <device-id>
- currently set to simulate a monitoring device with a delay between measurements of
0.02
seconds, this must be set in the.env
file, e.g.DELAY_IN_MEASUREMENT_FREQUENCY_REALTIME_SAMPLING=0.02
.
git clone https://github.com/beautifwhale/nilmtk-device.git
cd nilmtk-device
conda env create -f environment.yml
python convert.py
Convert the data data/REDD/low_freq
to data/redd.h5
. This is sample mains data. The monitoring device will record and generate this .h5 file data for disaggregation.
python simulate.py <device-id>
id
The device ID used to register the device to a user.
Uses raw monitoring data for building 1 in the REDD dataset. Analyses this data and reports any abnormalities to the user using push notifications. Assumes disaggregation has be done on mains and individual appliance level metrics are available.