Prediction of Faults in Power Grids Using ANN and Decision Tree-Based Models for Enhanced Grid Reliability and Performance

Electric Power Grids (EPGs) are vital infrastructure worldwide, requiring reliability and resilience to avoid disruptions. EPGs face rising risks, their repercussions, and ways to reduce them in this survey. Smart grids with improved control and monitoring aim to increase power distribution reliability and sustainability. Understanding the causes and effects of failures is critical to improving electrical power system resilience as failures become more frequent and severe. Various methods were used to reduce derived features and improve system efficiency. Model classification performance was assessed using accuracy, specificity, sensitivity, and precision. Data balance greatly improved DT-ANN classifier specificity. With a dataset test case, the best classification models demonstrated their remarkable detection and categorisation abilities. The results show that the suggested solution outperforms existing methods, improving a framework for recognising and mitigating electrical power network vulnerabilities. These findings can help build more resilient and sustainable electricity distribution networks.