A Survey For Real-Time Energy Monitoring and Management Using IoT and Machine Learning

Abstract

The growing demand for energy, coupled with
rising costs and sustainability concerns, has motivated extensive
research into efficient energy monitoring and management
systems. This review consolidates findings from sixteen
prominent studies spanning IoT-enabled smart homes, realtime
monitoring, non-intrusive load monitoring (NILM),
energy forecasting, and microgrid management. These works
demonstrate diverse approaches, including machine learning,
artificial intelligence, middleware design, graph signal processing,
and low-cost IoT protocols. A comparative analysis
highlights advancements in real-time anomaly detection,
appliance-level disaggregation, and hierarchical energy management
systems. The review also identifies current challenges
such as interoperability, data scarcity, system scalability,
and user adaptability, while outlining promising research
directions toward affordable, reliable, and sustainable energy
solutions