A Reliable Intelligent System for Real-Time Dynamic Security Assessment of Power Systems

TLDR

The study introduces an intelligent system for real‑time dynamic security assessment of power systems. The system employs an ensemble of extreme learning machines with decision rules to rapidly learn and estimate result credibility, enabling pre‑fault assessment and fallback to simulation when needed, and is validated on an IEEE 50‑machine system and a real‑world grid equivalent. The system achieves 100 % classification accuracy, low prediction error, and demonstrates high efficiency, robustness, accuracy, and reliability, making it suitable for practical deployment.

Abstract

A new intelligent system (IS) is developed for real-time dynamic security assessment (DSA) of power systems. Taking an ensemble learning scheme, the IS structures a series of extreme learning machines (ELMs) and generalizes the randomness of single ELMs during the training. Benefiting from the unique properties of ELM and the strategically designed decision-making rules, the IS learns and works very fast and can estimate the credibility of its DSA results, allowing an accurate and reliable pre-fault DSA mechanism: credible results can be directly adopted while incredible results are decided by alternative tools such as time-domain simulation. This makes the IS promising for practical application since the potential unreliable results can be eliminated for use. Case studies considering classification and prediction are, respectively, conducted on an IEEE 50-machine system and a dynamic equivalent system of a real-world large power grid. The efficiency, robustness, accuracy, and reliability of the IS are demonstrated. In particular, it is observed that the IS could provide 100% classification accuracy and very low prediction error on its decided instances.