Smart Energy Management System Using Non-intrusive Load Monitoring

Abstract

Energy Management is a problem faced by many around the world. The ever-rising demand for energy is putting a strain on the worldwide resources. Additionally, during the pandemic, it was observed that there was a lot of discrepancy in the electricity bills. To do our part in addressing the issue, the combination of Internet of Things (IoT) and Machine Learning (ML) has been used in creating a solution which will help measure, monitor and visualize daily energy consumption of a household. Additionally, using the concept of Non-Intrusive Load Monitoring (NILM), a single hardware setup can be used to measure the energy consumption of each appliance in the household. This hardware setup with the use of certain ML algorithms like Factorial Hidden Markov Model (FHMM) and Combinatorial Optimization (CO) disaggregates the combined household energy readings to device specific values. These values then get sent to a cloud database and are presented to the user through a Dashboard like visual interface. Therefore, the system in whole offers a combined solution to the user with minimal setup and cost to give a generic idea based on the energy usage, pattern, consumption. Such monitoring and systems can help efficient and responsible energy usage and can go a long way in ensuring sustainability.