article · Energy Reports
Rising electricity demand requires efficient power distribution to minimise energy loss across networks. Smart grids address this challenge, and integrating Internet of Things devices alongside artificial intelligence enhances the forecasting of grid performance and consumer demand. This research assesses several advanced machine learning techniques, including neural networks, decision trees, support vector machines, and k-nearest neighbours, for predicting dynamic stability in smart grids following operational disturbances. The focus centres on hyperparameter optimisation for gradient boosting algorithms. Combining the dipper throated optimisation algorithm with gradient boosting yielded the highest predictive performance, achieving an accuracy of 99.32 percent, sensitivity of 99.16 percent, and specificity of 99.54 percent. Diagnostic and regression analyses confirm that tactical hyperparameter tuning substantially improves prediction reliability, offering a robust computational framework for monitoring grid stability.
As global electricity demand rises, smart distribution networks must withstand operational disturbances to prevent widespread blackouts and energy loss. Accurate dynamic stability prediction allows network operators to identify potential failures before they escalate. Demonstrating high-accuracy machine learning methods helps ensure that modern power grids remain resilient and efficient as renewable energy sources and connected devices expand.
This algorithmic approach could be integrated into energy management systems and grid monitoring software used by electrical utilities and grid operators. By accurately forecasting whether a network will recover from disturbances, it supports real-time automated network management. The research is computational and algorithmic, representing early-stage to applied testing based on model evaluation, meaning further integration and field testing within live grid environments would be needed before deployment.
AI-generated from the published abstract. Always read the original work before citing.
With the surge in global population and economic expansion, there's been a marked increase in electricity demand. This necessitates the efficient distribution of electricity to both residential and industrial sectors to minimize energy loss. Smart Grids (SG) emerge as a promising solution to reduce power dissipation in distribution networks. The application of machine learning and artificial intelligence in SGs has significantly improved the precision of predicting consumer electricity needs. This paper presents a novel approach to improving the stability prediction of Internet of Things (IOT)-driven SGs using different advanced machine learning models. This study explores multiple advanced machine-learning techniques, including Gradient Boosting (GB), K-Nearest Neighbor (KNN), Support Vector Machine (SVM), Neural Networks, and the Decision Tree classifier, focusing on the stability prediction of SGs. This study explores the efficiency of hyperparameter-optimized GB models in predicting SG dynamic stability that encompasses the ability of the system to return to a stable operating point following a disturbance. Focusing on various models, it identifies the Dipper Throated Optimization Algorithm DTO+GB model as the standout, exhibiting unparalleled accuracy and reliability across critical performance metrics such as accuracy (99.32 %), sensitivity (99.16 %), and specificity (99.54 %). Diagnostic and regression analyses further emphasize its better predictive power and the need for hyperparameter optimization to improve the model. This paper highlights the capabilities of advanced machine learning algorithms in conjunction with tactical hyperparameter optimization in enhancing SG stability prediction, introducing a new baseline for future technological and methodological developments in this application.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.1016/j.egyr.2024.06.034
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
