NILM by using multiple-choice knapsack problem (MCKP).
This is a naïve prototype/demonstration that was initially used when exploring different concepts of UNILM.
If you use my code in your research please cite our UNILM research paper (IEEE example):
A. Rodriguez-Silva and S. Makonin, “Universal Non-Intrusive Load Monitoring (UNILM) Using Filter Pipelines, Probabilistic Knapsack, and Labelled Partition Maps,” p. 6, 2019.
Read more about UNILM on arXiv at https://arxiv.org/abs/1907.06299.
- You need to install the following packages via pip: prettytable, statistics, scipy, numpy
- Download the
house1_power_blk1.csvfile from the RAE dataset from Dataverse