A Deep Learning Approach for Prediction of Smart Grid Stability

Smart Grids (SGs) have significant potential for mitigating power losses during power distribution processes. A plethora of work is available to tackle the issues of power loss and make things automatic. Among these solutions, Machine learning and artificial intelligence techniques have been effectively integrated into SG systems to enhance the accuracy of customer demand prediction. To mitigate this issue, various deep learning algorithms may be used to determine the suitable one for application in SGs stability. In this research work, we have employed a state-of-the-art deep-learning algorithm named Convolutional Neural Networks (CNNs) to predict the stability of SGs. The dataset has utilized in this research work is sourced from the publicly available SGs dataset collected from the UC Irvine (UCI) machine learning repository. We have achieved 84% accuracy, 80% precision, and 76% recall values. This study may help practitioners to use deep learning models for SG stability.