Artificial Intelligence and IoT-Based Smart Home Energy Management Systems: A Comprehensive Review of Architectures, Security Challenges, and Future Research Directions

Home Energy Management Systems have evolved from conventional energy monitoring tools into intelligent, inter- connected platforms that integrate IoT infrastructure, Artificial Intelligence algorithms, and advanced communication technologies. This review presented a comprehensive analysis of IoT-based architectures, machine learning and reinforcement learning techniques, cybersecurity challenges, and privacy- preserving mechanisms within modern HEMS. The comparative evaluation of representative studies reveals substantial advancements in real-time monitoring, load forecasting, and adaptive scheduling. However, interoperability limitations, cybersecurity vulnerabilities, computational complexity, and limited large-scale deployment remain significant challenges. Future research should focus on unified AIoT-based frame- works that combine intelligent optimization, secure communication, privacy preservation, and smart grid integration. The development of scalable, secure, and energy-efficient HEMS will play a vital role in achieving sustainable residential energy management and advancing next-generation smart home ecosystems.

[1]    M. Rehman et al., “Future of Energy Management Models in Smart Homes: A Systematic Literature Review,” 2025. [2]    “Home Energy Management Systems: A Review of the Concept, Architecture, and Scheduling Strategies.” [3]    A. Kormpakis et al., “Home Energy Management Systems: Challenges, Heterogeneity and Integration Architecture,” 2024. [4]    R. Sureshkumar et al., “Internet-of-Things-Based Smart Home Energy Management System with Multi-Sensor Data Fusion,” 2024. [5]    M. Agyeman et al., “Design and Implementation of an IoT-Based Energy Monitoring System,” 2019. [6]    “A Smart Energy Monitoring System Using ESP32 Microcontroller,” 2024. [7]     “IoT-Based Smart Energy Meter Using ESP32,” 2021. [8]    “Smart Home Energy Management: Sequence-to-Sequence Load Fore- casting and Q-Learning for Demand Response,” 2021. [9]     “AIoT-Based Smart Home Energy Management System,” 2023. [10]    “Integration of Multi-Mode Preference into HEMS Using Deep Reinforcement Learning,” 2025.