A Machine Learning System for Predicting Severity Under Single Transmission Line Outages

• Main Authors: Prabha, Ankala Satya, Daram, Suresh Babu, Reddy, Chintam Jagadeeswar
• Format: Conference Proceeding
• Language: English
• Subjects: Line voltage stability index and Predictive modeling; Machine learning; Power system reliability; Power system stability; Power transmission lines; Predictive models; Single line transmission outages; Space weather; Stability criteria; Support vector machines; Transmission line measurements
• ISSN: 2769-2884
• DOI: 10.1109/ICRITO61523.2024.10522106

The electric power system requires a deep comprehension of possible dangers and effective mitigation measures. Space-weather-caused single transmission line outages (STLO) threaten electricity grid reliability. The Line Voltage Stability Index (LVSI) is crucial for assessing power interruptions. Space weather events and their STLOs are unpredictable, making it difficult for power system operators to recognize and react to threats. This study develops a support vector machine (SVM)-based machine learning structure to overcome the following difficulties: This study discusses LVSI predictions of power system severity under STLO, considering various load situations for a complete ranking. The method can reliably forecast transmission line outage severity ratings via SVM with the Radial Basis Function (RBF) kernel. The study uses transmission line metrics, load features, and LVSI values under normal and outage conditions. Training and testing the SVM model on a large dataset proved its STLO severity classification accuracy.