A Machine Learning based Smart Grid for Home Power Management using Cloud-Edge Computing System

In recent years, a huge amount of data has been communicated between internet of things (IoT) devices. Apart from these dependencies, many fields such as smart grids, smart cities, and smart homes are incorporating IoT devices. This work focuses on refining issues faced with existing power grids, such as unidentified fault detection, prediction of power generation, and utilisation at the consumer side of existing power grids. To overcome this problem, a machine-learning-based smart grid using computing systems is designed to control renewable energy power generation, power prediction, fault detection, and utilisation at the consumer side. This is achieved using cloud-edge computing. As per the simulation analysis, the proposed work shows improved performance in comparison to convolution neural network (CNN) based cloud computing with respect to power utilisation, throughput, and fault detection of 0.25%, 0.12%, and 0.35%, respectively.