An Edge Intelligence Framework for Elegant Power Management in IoT-enabled Power Grids

Abstract

The Internet of Things (IoT) is a concept that has the potential to attract new audiences in fields as diverse as manufacturing, healthcare, and more. IoT devices included in the sensor were the primary drivers of the massive data collection. To successfully combine, assess, and comprehend all program objects, thus, self-adaptive algorithms based on AI are necessary. The proliferation of both massive datasets and resource-intensive IoT devices makes stringent power management essential. The proliferation of both massive datasets and resource-intensive Internet of Things devices makes stringent energy management essential. Combining IoT with AI-based techniques is crucial for equitable power distribution to compact mobile devices. To this end, we offer an efficient way to communicate between power utilities and end users by forecasting future power usage over short periods of time. Innovations include a revolutionary convolutional recurrent model for a lightweight prediction method with low duration intricacy and minimum margins of error, as well as massive energy administration for edge devices via a centralized cloud-based data supervisory server. To maintain the power consumption and supply paradox efficiently, the suggested scheme has mobile nodes interact with a central remote server via an IoT network and then on to the corresponding power grid. We use a number of preparation methods to accommodate the varied electrical data, and then we construct a powerful decision-making engine for quick prediction on devices with limited resources.