Contingency Analysis of Transmission Lines Using Artificial Neural Networks (ANNs): A Case Study of the Ethiopian Southern Region 230 and 132 kV Transmission Network

ABSTRACT

In power system operation, contingency analysis (CA) is the systematic, proactive evaluation of credible, predefined events (such as the unplanned outage of a transmission line) to assess their potential impact on system security and reliability before they occur. In interconnected networks, unplanned line outages significantly affect the operating limits of remaining components, making CA an essential tool for power system operation. This study presents an artificial neural network (ANN)‐based contingency analysis approach using a radial basis function neural network (RBFNN) to predict transmission line outage performance in the Ethiopian Southern Region's 230 and 132 kV power network. Load flow data were generated using the Newton–Raphson (NR) method under various loading conditions. The RBFNN model achieves a mean square error (MSE) of , a mean absolute percentage error (MAPE) of 1.7%, and a correlation coefficient R = 0.99997, demonstrating excellent predictive accuracy. The most severe contingency identified is the outage of the Hawassa I‐Yirgalem I 132 kV line, which causes complete blackout of five downstream substations (total load loss 26 MW). The Wolkite‐Wolkite I 230 kV line outage ranks highest among 230 kV contingencies with a performance index (PI) of 1.1118. The proposed RBFNN predicts contingency PI values nearly identical to the NR method but with dramatically faster computation (0.03 s vs. 2.1 s per contingency). Upgrading the Hawassa I‐Yirgalem I corridor to double circuit is recommended as the most effective mitigation strategy.