A Bayesian Optimization-Based AutoML Framework for Non-Intrusive Load Monitoring

TL;DR

This paper presents a novel AutoML framework using Bayesian Optimization for energy disaggregation in NILM, enabling easier application of machine learning without advanced expertise and providing an open-source toolkit for practitioners.

Contribution

It introduces a flexible AutoML framework tailored for NILM, integrating Bayesian Optimization for model selection and hyperparameter tuning, and offers an extensible open-source toolkit.

Findings

Supports 11 algorithms with hyperparameter tuning

Facilitates easier application of machine learning in NILM

Open-source toolkit for industry and research

Abstract

Non-Intrusive Load Monitoring (NILM), commonly known as energy disaggregation, aims to estimate the power consumption of individual appliances by analyzing a home's total electricity usage. This method provides a cost-effective alternative to installing dedicated smart meters for each appliance. In this paper, we introduce a novel framework that incorporates Automated Machine Learning (AutoML) into the NILM domain, utilizing Bayesian Optimization for automated model selection and hyperparameter tuning. This framework empowers domain practitioners to effectively apply machine learning techniques without requiring advanced expertise in data science or machine learning. To support further research and industry adoption, we present AutoML4NILM, a flexible and extensible open-source toolkit designed to streamline the deployment of AutoML solutions for energy disaggregation. Currently, this framework supports 11 algorithms, each with different hyperparameters; however, its flexible design allows for the extension of both the algorithms and their hyperparameters.