An Edge-Based Intelligent System for Smart Building Energy Optimization

Abstract

The largest consumption of energy across the globe comes from residential building demographics. According to International Energy Agency nearly these buildings are responsible for around 30-40% of energy consumption across the world. One of the major issues related to energy waste in buildings is due to operation of traditional building systems at fixed timings without considering their occupancy in real-time. To address the above mentioned issue, the current paper proposes an Edge-enabled Digital Twin approach for optimal energy management in buildings. In the proposed framework, IoT-based sensors, edge computing, pre-diction of occupancy through machine learning, and use of digital twin dashboard helps in monitoring and optimizing energy usage through the indoor environment. All of the sensed data such as occupancy rate, temperature, and light intensity are processed locally using edge computing. Based on the above gathered information, HVAC systems and illumination systems are operated to save energy without making the occupant feel uncomfortable. The digital twin offers a visual representation of device status and building conditions which enables users to keep track of energy performance easily. The simulation results show that this approach can reduce approximately 20–30% of energy consumption compared to schedule-based methods. This high-lights the effectiveness of combining IoT, edge intelligence, and digital twin technology for smart building energy optimization.