The rising necessity for intelligent and sustainable power management in modern electricity networks has led to the evolution of smart grids, supported by sophisticated data analytics and communication frameworks. This research introduces a deep learning-based method designed to anticipate energy requirements within IoT-enabled smart grids and to enhance the routing of data for greater energy conservation. Leveraging Long Short-Term Memory (LSTM) neural networks, the system forecasts energy usage trends using historical records from IoT devices and smart meters. Additionally, a dynamic, energy-aware routing strategy is implemented, which modifies transmission paths in real time based on the predicted energy consumption, thereby reducing energy use during data transfer. The approach is validated using real-world smart grid datasets through simulation, showing a marked reduction in transmission energy costs and overall grid enhancement. The findings emphasize that combining deep learning-based forecasting with adaptive routing can considerably boost efficiency, lower expenses, and increase the operational reliability of IoT-integrated smart grid systems.
4 July 2022 in Journal of Ambient Intelligence and Humanized Computing
Accurate road mark In intelligent transportation systems, traffic monitoring, and autonomous driving, detection is essential. The application of YOLOv8 (You Only Look Once, Version 8), a cutting-edge object detection model, for the accurate recognition and categorization of road markings is the main goal of this research study. Improved anchorfree detection and YOLOv8’s sophisticated neural architecture allow for real-time performance with increased efficiency and accuracy. The model is trained on a diverse dataset containing various road markings, including lane lines, pedestrian crossings, arrows, and other traffic symbols. The proposed system leverages YOLOv8’s superior feature extraction capabilities to detect road marks under diverse environmental conditions such as varying lighting, occlusions, and complex road textures. Important parameters including mean Average Precision (mAP), precision, recall, and inference speed are used to evaluate performance and show how well the model works in real-time applications. According to the experimental results, YOLOv8 performs better than earlier iterations and other object detection frameworks, enabling quicker and more precise road marking recognition.