Identification and classification of fault using S-transform in an unbalanced network

In this paper, S-Transform (ST) based fault classification rules are introduced in case of overhead transmission line. The power system model is simulated in MATLAB Simulink environment with unbalanced loading. The proposed technique requires voltage and current signals to be extracted at the sending and receiving end of the network. The current and voltage signals are processed by ST to produce complex S-matrices. Four types of features are obtained from the absolute value of S-matrix involving simple calculation and less computational time. It is noticed from the simulations that these features enjoy a valuable advantage to be a prime choice as parameters for detection of the type of fault and the affected phase on the basis of some threshold values. The classification rules based on these parameters have been established on the basis of 5220 simulations of fault conditions. The multiple fault conditions have been obtained by changing fault resistance, fault location and fault inception angle. The rules are also tested with signals impregnated by synthetic noise. The proposed scheme has been conveniently programmed in MATLAB and the output result is obtained fast and accurate. A computationally fast version of ST is intended to be implemented in future.