An Intelligent Home Energy Management System Using Deep Reinforcement Learning

Abstract

Nowadays, Home energy management systems (HEMS) have more pathways to reduce energy usage due to the growing interest in power and the development of smart networks. With the establishment of intelligent meters, it is feasible to perform load control utilizing Demand Response (DR) empowered machines. This paper proposed an intelligent framework for home energy management by utilizing deep reinforcement learning. Intelligent frameworks give a normalized systemic way to deal with and tackle significant and genuinely complex issues and get predictable and dependable outcomes over the long haul. In this research, the proposed model detects faults such as system overrides, sensor location, nuisance alarms, and improper maintenance accurately by using deep reinforcement learning and in the future may be performance improved by using transfer learning.