Predictive Modeling of Power System Contingencies with SVMs and FACTS Devices for Enhanced Stability

Abstract

In modern power systems, ensuring stability and reliability is paramount. This study proposes a novel approach to the predictive modelling of power system contingencies using Support Vector Machines (SVMs) in conjunction with Flexible A.C. Transmission System (FACTS) devices. The integration of SVMs aids in accurately forecasting potential contingencies by analyzing historical data and identifying patterns. Additionally, FACTS devices can dynamically control power flow and enhance system stability. The proposed methodology involves two main phases: training the SVM model using historical data and simulating the impact of various contingencies with and without FACTS intervention. Comparative analysis demonstrates the effectiveness of the SVM-based predictive model in identifying critical contingencies. Moreover, incorporating FACTS devices showcases their potential to mitigate stability issues through real-time control actions. This combined approach offers an advanced tool for power system operators to anticipate and minimize contingencies effectively, ultimately leading to an enhanced and resilient power grid.