Fast Fault Diagnosis for Power Systems Using Deterministic Learning

In this paper, a learning-based approach is proposed for fast diagnosis of faults in power systems. The key issue of the proposed diagnosis scheme is that through deterministic learning, knowledge of various fault dynamics in power systems can be extracted and a knowledge bank can be established. By utilizing learnt fault dynamics, a set of estimators are constructed. The memories of the transient fault dynamics can quickly be recalled to give a fast diagnosis of fault. Simulation studies are included to demonstrate the effectiveness of the approach.