Energy Management for Internet of Things-Based Smart Buildings Using a Novel Deep Learning Technique

On a global scale, the main challenges encountered by many sectors, including industrial and residential sectors, are to energy use and conservation. This study introduces an innovative approach that integrates deep learning and Internet of Things (IoT) technologies to efficiently control the operation of electrical systems, with the objective of reducing energy consumption. To achieve a somewhat ambitious goal, we have developed a people recognition system that utilises deep learning techniques and employs the YOLOv3 algorithm to precisely ascertain the number of individuals inside a certain area. Hence, the effective administration of electrical equipment operations may be attained inside a smart building. Furthermore, the number of persons and the operational status of the electrical equipment units are made publicly available through the internet and presented on the dashboard of the Internet of Things (IoT) platform. The technology being evaluated enhances the decision-making process for energy utilisation. To assess the effectiveness and practicality of the suggested approach, a set of comprehensive test scenarios is carried out inside a specifically designed smart building setting, with consideration given to the existence of electrical equipment units. The findings from the simulation demonstrate the efficacy of the recognition algorithm based on deep learning in accurately identifying the quantity of humans present in a certain area.