Transformers are major elements of any power systems.
Normally they must be properly protected by differential relays.
This protection system should be precise and reliable via
implementation of strong algorithms that able to differentiate
between faulted and unfaulted condition to fully grantee of power
continuity. This system should be able to detect the non-faulted
condition, such as inrush currents which should not be activated
in this condition meanwhile; it must be activated in internal fault
conditions as fast as possible. This paper presents an approach for
differential protection of power transformers this uses wavelet
transform (WT) and adaptive network-based fuzzy inference
system (ANFIS) to discriminate internal faults from inrush
currents.
The proposed algorithm has been designed based on the
differences between both amplitudes of wavelet transform
coefficients in a specific frequency band and rising and decaying
duration generated by faults and inrush currents. The
performance of this simulated model is demonstrated by
simulation of different faults and switching conditions on a power
transformer using Matlab/Simulink software Package.