Deep Learning Models for Predicting Stability in Smart Grids Using the Electrical Grid Stability Simulated Dataset

The stability of the smart grid system is a fundamental requirement to achieve effective and seamless energy distribution. This research investigates the prediction potency of deep learnings in the stability forecasting of smart grids by utilizing an extensive dataset of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{6 0, 0 0 0}$</tex> samples consisting of time constants, power levels and other related metrics. We examined several deep learning models, including Recurrent Neural Networks (RNNs), Long Short-Term Memory networks (LSTM), Gated Recurrent Unit (GRU) networks, and Multi-layer Perceptron's (MLP), as well as Transformer-based models. In addition, in our analysis, we included visualization to examine correlations and the balance between classes of variables while the feature importance of variables assesses by using Random Forest. The experimental findings indicated that the validation accuracy (97.03 %) of the hyperparametric trained GRU model on the validation accuracy maximizing dataset was the highest compared to the other models.