DSUALM

This is the repository that constitutes the second experiment with ozm v1 using time-coupled data. The time coupling tells us that there is synchronization in time between the different measurements of the different meters associated with the commonly used synchronization and the main common meter. Only the usual measurements are included in this experiment (so we will not use voltage, current, and power harmonics).

The OZM v1 used to obtain the measurements that we will analyze in this repository is a single-phase electric energy meter (although there is already a three-phase version), which is also a power quality analyzer. This device is both open source and open hardware, and has been developed jointly by the Universities of Almeria and Granada, also having IoT capabilities, which not only allows us to measure a wide range of electrical variables at a high sampling rate of 15625 Hz (voltage, current, active power, reactive power,  Total harmonic distortion or THD, power factor and harmonics of both intensity and voltage and power up to the order 50), but also allows us to capture and process all those measurements.

We used 6 OZM v1 type meters applied to 5 commonly used household appliances. This is the list of devices:

1- Mains

2-Boiler

3-Fan

4-Freezer

5- TV

6-Vacuum cleaner

This repository discusses the impact of NOT taking harmonics to disaggregate electricity consumption using OZM v1 and the NILMTK Toolkit. In this we specify the number of measurements supported by the different OZMs, such as active, apparent and reactive power, frequency, voltage, current and power factor. We do NOT add the harmonics of voltage (50), current (50) and power (50), that is, the harmonics are ignored.

The contents of the metadata yaml files are attached in a separate directory within the root, as well as the new converter that also disregards harmonics. It should be noted that in the Jupyter Notebooks there is not only the code in Python but also the results of running on the dataset DSUALM0H, Also, mention that to execute this code we need to have the NILMTK toolkit installed (also available on Github), as well as have the new dataset that already includes the harmonics.