Development of a machine learning application for IoT platform to predict energy consumption in smart building environment in real time

• Κύριος συγγραφέας: Χουσιάδας, Χρήστος
• Άλλοι συγγραφείς: Κουμπιάς, Σταύρος
• Μορφή: Thesis
• Γλώσσα: English
• Έκδοση: 2020
• Θέματα: Machine learning; LSTM; Smart building; Energy consumption; Μηχανική μάθηση; Κατανάλωση ενέργειας

The aim of this study is the development and implementation of an advanced application and a Web Services for an IoT platform. The aforementioned application is able to predict the short-term energy consumption of a household based on the huge amount of data, which are collected from the smart meters that was connected around the household’s appliances. Also, our platform is capable of extracting usage patterns of appliances based on the activity (ON/OFF states). These methods provide services for people, who want to control financial expenses (i.e. electricity billing), acquire higher energy awareness, etc. This topic has recently been receiving a lot of attention due to the availability of data providing from the Wireless Sensors and the increasing to integrate artificial intelligent computational systems with IoT platforms. It is a great technological interest for electrical engineers, researchers and electric utility companies.