Short Circuit Fault Classification and Location in Transmission Lines Using A Combination of Wavelet Transform and Support Vector Machines

In this paper, a modern synthetic framework which has the capability to rapidly Classify and locate short circuit faults over transmission lines is presented. The proposed algorithm singles out short circuit faults based on the measured voltage waveform and three-phase current when fault events occur in power transmission lines. The values resulting from the three-phase currents and the three-phase voltages wavelet transform are used to fault classification algorithm. Then, fault location algorithm is activated as the result of fault classification. Different kinds of methods such as multilevel wavelet transform and support vector machine have been combined in a set to determine fault classification and location at every time. This paper lays out the fundamental concept of the proposed framework and introduces a pattern recognition approach via wavelet transform, statistical processing techniques, neural network (NN) and a joint decision-making mechanism. Voltage and the recorded current values in measurement devices from the fault moment to a quarter of post-fault have been used for preprocessing and training in the support vectors machines.