Earlier detection of risk of blackout by real-time dynamic security assessment based on Extreme Learning Machines

The lack of real-time tools capable of detecting risk of blackouts is one of the contribution factors to the recent large blackouts occurred around the world. In terms of dynamic security assessment (DSA), artificial intelligence and data mining techniques have been widely applied to facilitate very fast DSA for enhanced situational awareness of insecurity. However, many of the current state-of-the-art models usually suffer from excessive training time and complex parameters tuning problems, leading to their inefficiency for real-time implementation. In this paper, a new DSA method using Extreme Learning Machine (ELM) is proposed, which has significantly improved the learning speed and can therefore provide earlier detection of the risk of blackout. The proposed method is examined on the New England 39-bus test system, and compared with other state-of-the-art methods in terms of computation time and accuracy. The simulation results show that the ELM-based DSA method possesses superior computation speed and acceptably high accuracy.