IoT‐Based Smart Green Building Energy Management System

Le résumé fourni par la source

Controlling and monitoring energy use requires accurate data, and this is what energy management systems (EMS) deliver. Using Internet of things (IoT)-based energy monitoring technology, these EMS may be vastly improved and upgraded, resulting in more energy savings. This research provides support for the use of real-time IoT for energy management in eco-friendly smart structures. Taking readings of energy use, making forecasts of future energy use, and recognizing people's faces are the three cornerstones of the proposed system. Predictions were made using a method called short-term load forecasting (STLF) that is based on the K-nearest neighbor (KNN) algorithm. Line A current, line B current, line C current, line voltage A, line volt B, and line volt C are the six digital power meter (DPM) parameters that must be utilized as data to serve as training for the prediction algorithms. The smart building's energy usage for the subsequent hours on the same day is calculated based on the predicted outcome. Based on the predicted outcome, the active, reactive, and seeming abilities may be determined. Using facial recognition technology, a smart building's administrators may restrict access to restricted areas. The Viola-Johns algorithm is the foundation of modern facial recognition technology. The system has a total accuracy of 91% in face detection and recognition, as measured by the Viola-Johns classifier's use of Haar characteristics. The results showed that the true negative rate (TNR), positive predictive value (PPV), and false discovery rate (FDR) each averaged 51%, whereas the PPV averaged 70.5% and the FDR averaged 31.6%.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.