Energy Control Problem Code in MATLAB-GCNMF-s2k Group Constrained Non-negative Matrix Factorization with Sum-k Constraint for Load Disaggregation

Energy Control Problem Code in MATLAB: Non-Intrusive Load Monitoring (NILM) for HVAC Systems

This repository contains the dataset we collected for HVAC energy disaggregation, as well as the source code and demonstrations from our paper in IEEE Transactions on Power Systems. To the best of our knowledge, this is the first dataset collected for studying Non-Intrusive Load Monitoring (NILM) applied to Heating, Ventilation, and Air Conditioning (HVAC) systems.

Energy disaggregation or Non-Intrusive Load Monitoring (NILM) addresses the problem of extracting device-level energy consumption information by monitoring the aggregated signal at a single measurement point, without the need to install meters on each individual device. This can be framed as a source separation problem where the aggregated signal is represented as a linear combination of the basic vectors in a matrix factorization framework.

In this work, we utilize machine learning to predict the energy consumption pattern of each device over the course of a day. The project is part of our collaboration with [institution name].

Prerequisites:

• MATLAB R2015a

Datasets

(Temporarily unavailable. Will be available once the required permissions are granted. Apologies for the inconvenience!)

Experiments

We designed two different experiments to evaluate our proposed algorithm. The first experiment disaggregates the energy of the entire household into the energy consumption of all devices within the home.