A demonstration of reproducible state-of-the-art energy disaggregation using NILMTK

Batra, NipunNipunBatraKukunuri, RithwikRithwikKukunuriPandey, AyushAyushPandeyMalakar, RaktimRaktimMalakarKumar, RajatRajatKumarKrystalakos, OdysseasOdysseasKrystalakosZhong, MingjunMingjunZhongMeira, PauloPauloMeiraParson, OliverOliverParson2025-08-312025-08-312019-11-13[9781450370059]10.1145/3360322.33609992-s2.0-85077293639https://d8.irins.org/handle/IITG2025/24360Non-intrusive load monitoring (NILM) or energy disaggregation involves separating the household energy measured at the aggregate level into constituent appliances. The NILM toolkit (NILMTK) was introduced in 2014 towards making NILM research reproducible. NILMTK has served as the reference library for data set parsers and reference benchmark algorithm implementations. However, few publications presenting algorithmic contributions within the field went on to contribute implementations back to the toolkit. This work presents a demonstration of a new version of NILMTK [2] which has a rewrite of the disaggregation API and a new experiment API which lower the barrier to entry for algorithm developers and simplify the definition of algorithm comparison experiments. This demo also marks the release of NILMTK-contrib: a new repository containing NILMTK-compatible implementations of 3 benchmarks and 9 recent disaggregation algorithms. The demonstration covers an extensive empirical evaluation using a number of publicly available data sets across three important experiment scenarios to showcase the ease of performing reproducible research in NILMTK.falseEnergy disaggregation | Non-intrusive load monitoring | Smart metersA demonstration of reproducible state-of-the-art energy disaggregation using NILMTKConference Paper358-35913 November 201913cpConference Proceeding10