The results demonstrate that the proposed framework can provide an accurate and interpretable data-driven intelligence layer for energy-efficiency assessment and decision support in IoT-enabled smart grids.
Abstract
Energy efficiency assessment is fundamental to the reliable and sustainable operation of modern smart grids, particularly with the increasing deployment of Internet of Things (IoT) technologies and renewable energy resources. This study presents a data-driven framework for Energy Efficiency Score prediction that integrates Extreme Gradient Boosting Regression (XGBR) and Histogram-based Gradient Boosting Regression (HGBR) with Metaheuristic Optimization Algorithm (MOA), Particle Swarm Optimization (PSO), and Harris Hawks-inspired Learning Optimization Algorithm (HLOA) to optimize model parameters and improve predictive performance. Principal Component Analysis (PCA) is employed to reduce feature redundancy, while time-aware validation and cross-validation are used to improve the reliability of model evaluation. SHAP analysis is further incorporated to quantify feature contributions and improve model interpretability. The proposed framework achieved an R² of 0.9971 and an RMSE of 0.8744, demonstrating highly accurate estimation of the Energy Efficiency Score. The optimized hybrid configurations outperformed their corresponding baseline boosting models, confirming the benefit of metaheuristic parameter optimization for structured smart-grid data. The SHAP analysis further identified grid frequency, environmental conditions, reactive power, active power, and power consumption as among the most influential predictive variables. The results demonstrate that the proposed framework can provide an accurate and interpretable data-driven intelligence layer for energy-efficiency assessment and decision support in IoT-enabled smart grids. The framework therefore offers a scalable approach for predictive energy analytics while complementing, rather than replacing, physics-based grid optimization and operational models.
This study proposes a data-driven dynamic reconfiguration and energy-saving optimization framework for building energy-consuming microgrids. By integrating multi-source sensing data, load forecasting, and dynamic topology optimization, the framework establishes a coupling mechanism between building energy consumption and microgrid operation. A hierarchical architecture consisting of data acquisition, intelligent prediction, optimization decisionmaking, and execution control is developed to support adaptive energy management under varying load conditions. Load forecasting is performed using a long short-term memory network, while an enhanced particle swarm optimization algorithm is employed to determine optimal microgrid reconfiguration strategies under operational constraints. Case studies based on commercial building scenarios demonstrate that the proposed framework effectively improves load regulation capability, enhances renewable energy utilization, and supports stable microgrid operation. The framework can be readily integrated with wireless sensing infrastructures, antenna-enabled monitoring platforms, and edge–cloud collaborative systems, providing an engineering-oriented solution for real-time energy management and intelligent microgrid optimization in modern building environments.
S. L. Liu, D. T. Liu· Advanced Electromagnetics· 0 citations
A new hybrid BMS for SOC estimation based on a transaction-oriented adaptive Kalman Filter that merges traditional voltage-based SOC estimation has been developed in this study and demonstrates that it serves as an efficient and effective alternative for achieving long-term energy optimization in IoT-enabled smart agriculture applications.
Mohd Kamir Yusof, Nur Yasmin Salleh, Senny Luckyardi et al.· International Journal of Adv...· 0 citations
This review explores emerging smart technics in enhancing energy efficiency in commercial and residential buildings using systematic and bibliometric approaches from 1990 to 2024. According to the findings, the increase in internet of things applications, like AI and especially machine-learning applications, has enabled smarter buildings in recent years. The advancements of modern data analytics, predictive modelling, and real-time monitoring create a fair base for advancing into new paradigms for energy management. Energy yield prediction and building performance enhancements are ensured through machine-learning techniques, such as ensemble learning, neural networks, and support vector regression. The study found that deep reinforcement learning and fuzzy logic constitute those technologies that automate the consumption behaviors while perfectly balancing efficiency and comfort of occupants. According to the results, smart technologies offer better options toward energy efficiency but encounter major hurdles like poor internet availability, social acceptance, regulatory issues, high upfront cost, scaling issues, and data privacy. Real-time data coupled with smart technology systems should be combined to develop hybrid machine-learning models and predictive energy consumption models. For the attainment of energy efficiency goals, standardization of energy-efficient buildings and greening people's energy practices are key.
E. B. Agyekum, B. Tarawneh, S. Praveenkumar et al.· Energy Exploration & Exp...· 0 citations
The significance of Smart Charging Optimisation in EV Network is underscored by the fact that the fast uptake of PEV offers a game-changing chance to lessen reliance on oil-based fuels and encourage the use of low-carbon energy. Charging solutions that are both efficient and effective reduce operational costs while simultaneously improving grid stability and environmental sustainability. By utilising preprocessing approaches such KNN-based imputation, Z-score outlier identification, and Min-Max scaling to guarantee data quality, this study develops an advanced optimisation framework utilising the EV Smart Charging Dataset. Integrating smart grid technology, finding the best location for stations, making the most of renewable energy, and considering regulatory consequences are all important considerations that are analysed using PCA. To further enhance adaptability compared to traditional static systems, a new energy-aware model based on FLRL is presented. This model takes into consideration the rate of energy consumption and residual battery energy as inputs and employs a dynamic membership function. Energy efficiency and charge optimisation are both much improved by the suggested dynamic FLRL method, according to the simulation findings. The model's usefulness for intelligent EV network management is validated by experimental data, which show that it outperforms traditional FL, RL, and GNN techniques with an accuracy of 95.34% and a MAE of 4.5%.
V. Naresh, Chennuri Manve, Kodigela Vaishnavi et al.· 2026 7th International Confe...· 0 citations
Non-technical losses (NTLs) are a major problem in modern smart grids, damaging revenue and operational functionality. This study proposes an integrated IoT-edge-cloud framework to improve fraud detection, analyze electricity usage patterns, and enhance data reliability in distributed smart grids. The approach extracts multiple features from smart meter data and uses a hybrid machine learning method that combines classification and clustering. To test the system’s robustness, the study included simulated fraud scenarios in difficult circumstances. Results show that the system achieved high detection accuracy (96.4%) and an AUC of 0.98. The framework also reduced false alarms by 86% compared to traditional rule-based methods, improving consistency and productivity. It supports near-real-time operation with response times around 125 ms and is scalable for larger smart grid environments. Behavioral segmentation further improved reliability by identifying differences in electricity consumption and reducing incorrect classifications. Overall, the study shows that combining data quality management, behavioral analysis, and distributed processing yields a more reliable and resilient solution for operational smart grid systems.
F. Otosi, Celestine A. Udie, F. Faithpraise· E3S Web of Conferences· 0 citations
The growing penetration of renewable energy resources, electric vehicles (EVs) and distributed energy systems, demand-side management (DSM) in modern smart grids has become increasingly complex. Conventional DSM methods usually show lower predictive capability and ineffective load scheduling in a tight operating environment, which leads to increased operational costs, hindered grid stability. This paper presents a hybrid PSO–RNN framework for intelligent freedom of management for power systems. The proposed framework combines an RNN for precise short-term load forecasting with PSO for effective optimal load scheduling, EV charging coordination, and energy distribution by utilizing data on historical electricity consumption, outdoor weather conditions, renewable energy generation variables, electricity pricing data of local suppliers in Singapore,, average grid load from the previous day (24 hours) as well as #datetimeindex. The experimental evaluation shows that the proposed PSO–RNN model can reach 98.24% forecasting accuracy, reduce RMSE to 0.081 and MAE to 0.061 value, while achieving over 24.76 % of peak-load reduction; saving about 21.35 % of electricity cost; using approximately by up to 91.42 % renewable energy as well, compared with ANN, SVM, DRL and RNN standalone models. These features increase reliability, lower operation costs and assist in sustainable energy recovery. The designed framework can effectively serve as a scalable, adaptive and computationally efficient next-generation intelligent smart grid and real-time electricity demand side management solution.
Rajendra B. Sadaphale, Prakash Burade· International journal of com...· 0 citations