Artificial intelligence, IoT, and solar PV-integrated home energy management systems: A review

The rapid growth of solar photovoltaic (PV) systems, residential energy storage systems (ESS), Artificial Intelligence (AI) and Internet of Things (IoT)-enabled sensing devices has increased the demand for smart Home Energy Management Systems (HEMS) capable of optimizing energy use in smart buildings. This review presents a structured synthesis of recent research on AI and IoT-integrated HEMS, focusing on forecasting methods, optimization strategies, appliance scheduling, and demand side management (DSM). The analysis reveals that advanced neural network variants and hybrid AI approaches can achieve high accuracy in forecasting; however, the practical deployment is often constrained by computational complexity, limited generalization, and low technology readiness. Reinforcement learning (RL) shows strong potential for adaptive real-time control; however, sample inefficiency and the simulation-to-reality gap remain major challenges. Across the reviewed literature, fragmented IoT communication standards, cybersecurity vulnerabilities, limited prosumer participation, and insufficient validation through hardware-in-the-loop and field testing emerge as key barriers to large-scale adoption. Based on these findings, the paper outlines a future research roadmap emphasizing hybrid AI-optimization frameworks, edge-computing architectures, blockchain-enabled peer-to-peer (P2P) energy trading, and standardized validation protocols. The review provides actionable insights to support the development of scalable, secure, and interoperable HEMS, contributing to the realization of smart, sustainable, and net-zero residential energy systems.