Critical Events Forecasting in Power Grids Operating Close to Margin

Strategic investment in infrastructure will be vital for ensuring the reliability of future electrical grids, as the expected surge in demand for electric power places significant pressure on existing systems. In this regard, two strategies have emerged: investing in infrastructure expansion or leveraging existing infrastructure by developing a software-based toolset to enhance network reliability which is more cost-effective. Accurately forecasting critical events that arise from passing a predefined threshold in the electrical grid is a prerequisite for undertaking operation and maintenance (O\&M) remedial measures which increase reliability. This paper introduces a new framework for predicting critical events using artificial intelligence (AI) to improve the reliability of electrical networks that operate near their maximum capacity using existing infrastructure. The framework integrates techniques such as wavelet transformation, particle swarm optimization (PSO) algorithm, and convolutional neural network (CNN) model. The proposed framework achieved a high level of accuracy, as demonstrated by the evaluation results.