Intelligent Edge Computing for IoT-Based Energy Management in Smart Cities

TLDR

Smart cities increasingly rely on green energy management systems, yet long‑term efficiency remains a challenge despite IoT‑enabled monitoring. This study designs an IoT‑based energy management system that leverages edge computing and deep reinforcement learning. The authors describe the IoT energy‑management architecture, propose a software model for edge computing, and introduce a deep‑reinforcement‑learning scheduling algorithm. Experimental results demonstrate the proposed scheme’s effectiveness in improving energy scheduling.

Abstract

In recent years, green energy management systems (smart grid, smart buildings, and so on) have received huge research and industrial attention with the explosive development of smart cities. By introducing Internet of Things (IoT) technology, smart cities are able to achieve exquisite energy management by ubiquitous monitoring and reliable communications. However, long-term energy efficiency has become an important issue when using an IoT-based network structure. In this article, we focus on designing an IoT-based energy management system based on edge computing infrastructure with deep reinforcement learning. First, an overview of IoT-based energy management in smart cities is described. Then the framework and software model of an IoT-based system with edge computing are proposed. After that, we present an efficient energy scheduling scheme with deep reinforcement learning for the proposed framework. Finally, we illustrate the effectiveness of the proposed scheme.