A Review on IoT and ML Enabled Smart Grid for Futurestic and Sustainable Energy Management

Abstract

The Smart Grids (SG) are the upgraded version of classical power grid, which involve the communication infrastructure, big data, and machine learning technologies to improve the productivity and management of power demand and supply. The use of machine learning empowers the smart grids to proactively deal with the emergency situations. In this context, a review to explore the utilization of ML techniques in SGs have been provided. Next, the collected literature has identified the research opportunities and also studied the relevant solutions. Finally, the objectives for future studies have been proposed. Among them it has been tried to establish our initial objectives of studying the ML algorithms and the application of ML is smart grid. In addition, an experimental performance study among three machine learning algorithms namely Support Vector Machine (SVM), Artificial Neural Network (ANN) and Linear Regression (LR) has been carried out. The aim of employing these algorithms is to predict the appliances power demand in Home Area Network (HAN). The experimentation of variable size of datasets shows that the ANN is beneficial for deal with the large amount of data and superior than the SVM and LR based approach in prediction accuracy and training time requirements.