Prior Flow Variational Autoencoder: A density estimation model for Non-Intrusive Load Monitoring
This work provides a competitive solution for improving appliance-by-appliance power demand estimation in industrial settings, which is significant for energy management and efficiency.
This paper addresses Non-Intrusive Load Monitoring (NILM) by proposing a conditional density estimation model, the Prior Flow Variational Autoencoder (PFVAE), which combines a Conditional Variational Autoencoder with a Conditional Invertible Normalizing Flow. The model estimates individual appliance power demands simultaneously and achieved significant improvements in normalized disaggregation error (NDE) from 28% to 81% and signal aggregated error (SAE) from 27% to 86% for six of eight machines in a poultry feed factory dataset.
Non-Intrusive Load Monitoring (NILM) is a computational technique to estimate the power loads' appliance-by-appliance from the whole consumption measured by a single meter. In this paper, we propose a conditional density estimation model, based on deep neural networks, that joins a Conditional Variational Autoencoder with a Conditional Invertible Normalizing Flow model to estimate the individual appliance's power demand. The resulting model is called Prior Flow Variational Autoencoder or, for simplicity PFVAE. Thus, instead of having one model per appliance, the resulting model is responsible for estimating the power demand, appliance-by-appliance, at once. We train and evaluate our proposed model in a publicly available dataset composed of power demand measures from a poultry feed factory located in Brazil. The proposed model's quality is evaluated by comparing the obtained normalized disaggregation error (NDE) and signal aggregated error (SAE) with the previous work values on the same dataset. Our proposal achieves highly competitive results, and for six of the eight machines belonging to the dataset, we observe consistent improvements that go from 28% up to 81% in NDE and from 27% up to 86% in SAE.