IntDEM: an intelligent deep optimized energy management system for IoT-enabled smart grid applications

The need for electricity has increased rapidly due to social and economic developments. Avoiding global warming and changing the structure of domestic energy usage are two advantages associated with reliable energy demand forecasting. There is a dearth of recent research examining energy management in supervised Internet of Things (IoT) networks, despite the fact that sophisticated load forecasting is crucial for optimum energy management (EM) in smart grid applications. The objective of this research is to develop intelligent deep optimized energy management (IntDEM), a novel and unique framework for Internet of Things (IoT)-enabled smart grid systems. It employs a novel deep learning methodology based on the Stacked Convoluted Bi-Directional Gated Attention Network (SCon-BGAN) to accurately estimate the energy load from the provided smart grid datasets. Moreover, a cutting-edge optimization method called Hybrid Darts Seagull Optimizer is used for learning rate estimation that improvises the prediction process with a lower error rate. Additionally, the performance outcomes and findings of the proposed IntDEM model are assessed and analyzed using a variety of well-known datasets, such as ISO-NE, SGSC, and IHEPC. The results of this investigation show that, when data-handling procedures are followed correctly, the IntDEM model successfully lowers the prediction error rate up to 0.3.