Authors :
Lakshin Pathak; Khushi Vasava; Stuti Gulati; Shreya Bhatia
Volume/Issue :
Volume 9 - 2024, Issue 12 - December
Google Scholar :
https://tinyurl.com/y358v6s6
Scribd :
https://tinyurl.com/yc6t4htp
DOI :
https://doi.org/10.5281/zenodo.14576727
Abstract :
This paper explores the application of deep
learning techniques to predict smart grid stability. With
the growing adop- tion of renewable energy sources, the
unpredictability of energy supply and fluctuating
consumer demands pose challenges to grid stability. The
proposed framework utilizes Artificial Neural Networks
(ANNs) to analyze operational parameters, such as power
values and time constants, for classifying grid conditions
as stable or unstable. The dataset is preprocessed with
normalization techniques and trained using a feed-
forward neural network with ReLU and sigmoid
activation functions, optimized with the Adam optimizer.
The framework achieves high accuracy and robustness, as
demonstrated by cross-validation and performance
metrics like precision, recall, and F1-score. The results
highlight the potential of deep learning to enhance grid
reliability and support real-time decision-making. This
study contributes to the integration of AI technologies in
energy systems, ensuring efficient management and
sustainable use of renewable energy resources.
Keywords :
Smart Grid, Deep Learning, Stability Predic- tion, Power Systems, Neural Networks.
References :
- Bashir, Ali Kashif, et al.” Comparative analysis of machine learning algorithms for prediction of smart grid stability.” International Transac- tions on Electrical Energy Systems 31.9 (2021): e12706.
- Wang, X., et al.” Machine Learning in Power Systems: Stability Predic- tion.” IEEE Transactions on Power Systems, 2020.
- Zhou, Y., et al.” Deep Learning for Renewable Energy Integration in Smart Grids.” Energy, 2021.
- Li, Z., et al.” Artificial Intelligence in Smart Grid Management.” IEEE Transactions on Smart Grid, 2020.
- Dimitrijevic´, Marko A., et al.” Implementation of artificial neural networks based AI concepts to the smart grid.” Facta Universitatis, Series: Electronics and Energetics 27.3 (2014): 411-424.
- Omitaomu, Olufemi A., and Haoran Niu.” Artificial intelligence tech- niques in smart grid: A survey.” Smart Cities 4.2 (2021): 548-568.
- Shi, Zhongtuo, et al.” Artificial intelligence techniques for stability anal- ysis and control in smart grids: Methodologies, applications, challenges and future directions.” Applied Energy 278 (2020): 115733.
- Yan, J., et al.” Machine learning for the prediction of smart grid stability: A review.” International Journal of Electrical Power and Energy Systems 124 (2021): 106325.
This paper explores the application of deep
learning techniques to predict smart grid stability. With
the growing adop- tion of renewable energy sources, the
unpredictability of energy supply and fluctuating
consumer demands pose challenges to grid stability. The
proposed framework utilizes Artificial Neural Networks
(ANNs) to analyze operational parameters, such as power
values and time constants, for classifying grid conditions
as stable or unstable. The dataset is preprocessed with
normalization techniques and trained using a feed-
forward neural network with ReLU and sigmoid
activation functions, optimized with the Adam optimizer.
The framework achieves high accuracy and robustness, as
demonstrated by cross-validation and performance
metrics like precision, recall, and F1-score. The results
highlight the potential of deep learning to enhance grid
reliability and support real-time decision-making. This
study contributes to the integration of AI technologies in
energy systems, ensuring efficient management and
sustainable use of renewable energy resources.
Keywords :
Smart Grid, Deep Learning, Stability Predic- tion, Power Systems, Neural Networks.