ABSTRACT
Reliable and efficient operation of modern power systems is crucial, especially with integrating renewable energy sources, energy storage systems, and advanced monitoring devices. Accurate forecasting is necessary to prevent grid instabilities, blackouts, and equipment failures. To address these challenges, advanced forecasting techniques, such as neural network-based models, are required to capture the complex relationships and non-linear behaviour of the smart grid. This chapter highlights the importance of smart grid stability forecasting and proposes a fractional feed-forward neural network model to improve its accuracy. By incorporating fractional-order derivatives, the model converts conventional tangential activation functions into fractional-order activation functions, which results in more precise predictions of smart grid stability. The proposed model’s performance is evaluated against a conventional model using widely accepted performance metrics, such as mean square error and coefficient of determination. This evaluation provides valuable insights into the effectiveness of fractionalorder activation functions in improving the accuracy of grid stability predictions.
