<![CDATA[The residential sector accounts for a significant portion of global energy consumption. The integration of the Internet of Things (IoT) and Artificial Intelligence (AI) presents a promising solution to mitigate energy waste through Smart Home Energy Management Systems (HEMS). This Systematic Literature Review (SLR) adheres to the PRISMA protocol to analyze 20 primary experimental studies published between 2021 and 2025. This study aims to evaluate the effective hardware architecture, the most accurate machine learning/optimization algorithms, and the measurable economic impact of these systems. The findings reveal that hybrid deep learning models, such as LSTM+GRU and Wavelet-LSTM-SVR, achieve high prediction accuracies exceeding 95%. For demand response scheduling, hybrid metaheuristic algorithms like HGPO and MMGO can reduce electricity costs by up to 57.8% and peak-to-average ratio (PAR) by 74.68% in cluster dwellings. Additionally, the integration of Large Language Models (LLMs) significantly enhances user compliance and sustained energy optimization. Overall, IoT and AI implementation in smart buildings is capable of reducing daily operational costs significantly. Future research should address data privacy and the scalability of peer-to-peer energy trading.]]>
Copyrights © 2026
