15-10-2012, 04:37 PM
Detection of Internal Fault in Differential Transformer Protection Based on Fuzzy Method
Detection of Internal Fault.pdf (Size: 465.52 KB / Downloads: 45)
Abstract
Power transformers are vital and expensive components of electric power systems
and it is necessary to protect them from internal faults. Differential protection is based on
differential current. Occurrence of some phenomena such as inrush current, over excitation,
saturation of current transformers and mismatch of current transformers cause the
differential relay operates incorrectly. In this paper fuzzy logic is used as a powerful
mathematic tool. In this method of protection, algorithm is based on ruled out non-internal
fault phenomena. The results of simulations show that fuzzy protection system is able to
detect occurrence of fault in less than half a cycle and this method improves protection
system, satisfactorily.
Introduction
Differential protection is one of the most important and basic protection of power transformers. It is
necessary to enhance speed, sensitivity and accuracy of operation of differential protection. A
differential relay is based on the difference between current of low voltage and high voltage of the
transformer. In normal operation mode, there is no significant differential current but while an internal
fault occurs, the differential current will increase. Thus increase the differential current can be used as
a sign of the internal fault. But Occurrence of some phenomena such as inrush current, over excitation,
saturation of current transformers and mismatch of current transformers cause the differential current
increase and the differential relay operates incorrectly. Therefore using of digital differential protection methods is developed in recent decades. These methods improve reliability, flexibility and
performance of differential relay.
Differential Protection Based on Fuzzy Method
This section describes the previously defined criteria for power transformer protection, their settings,
and the aggregation method. In fuzzy method there are number of criteria which are being used to
differentiate between a fault and non-internal fault cases. In this method the symptoms qualify a case to
the group "internal fault" with the membership functionμ . If μ is zero, it is certainly no internal fault.
If μ = 1 it is certainly an internal fault. If 0<μ <1 it is a doubtful case. Fuzzy method became a very
powerful mathematical tool to solve this problem. With reference to the protection flow chart (fig. 2) at
first Relaying Signals are measured and then Criteria Quantities are computed. The aggregation of
membership function( ,..., ) 1 13 μ μ gives the fuzzy output for each non-fault phenomena. The tripping
signal is obtained by defuzzyfication of final fuzzy output using a constant threshold.
Conclusion
In this paper, the design and implementation of a differential transformer protection based on fuzzy
method is presented. The criteria have been aggregated and combined in order to generate more
reliable tripping signal. This Differential relay applied for digital protection of a transformer .