Random forest based fault analysis method in IEEE 14 bus system

In order to obtain an uninterruptible power supply system, it is of paramount importance for power system to identify different types of faults as quickly as possible to protect it from complete blackouts using intelligent techniques. This paper introduces a novel method for fault classification based on Hilbert transform of Wavelet transform coefficients of the fault signals and Random Forest classifier. The voltage signals have been taken from a standard IEEE 14 bus system simulation in Matlab/Simulink. The time frequency resolution of discrete Wavelet transform gives us different coefficients and then we apply Hilbert transform on those coefficients. Parameters from the results are extracted in the form of statistical features and used as input feature for the classification using Random Forest. The main objective of this paper is to assess the worth of random forest classifier on power system fault diagnosis. The obtained results compared with the existing artificial intelligence techniques.