This paper presents a new approach to differential protection in power transformer using fuzzy-neuro technology enhances fuzzy logic systems on learning capability. The proposed neuro fuzzy, was trained by data from simulation of a power system under different conditions, and tested by data with different training data. Details of the design procedure and the results of performance studies with the proposed relay are given in the paper. Performance study results show that the proposed algorithm is very good performance to recognize the various fault types in the power transformers. The AI methods are stable against phenomenon as inrush current. Nevertheless, the proposed method has an outstanding preference over the previous schemes in term of elimination the under load tap changer error meanwhile operation of the differential relay. It is clearly shown that with applying this integrated approach, the accuracy in differential relay is significantly improved over other techniques based on a conventional algorithm.
[1]
Eugeniusz Rosolowski,et al.
Multi-objective optimization of a neural network based differential relay for power transformers
,
1999,
1999 IEEE Transmission and Distribution Conference (Cat. No. 99CH36333).
[2]
Chuen-Chien Lee.
FUZZY LOGIC CONTROL SYSTEMS: FUZZY LOGIC CONTROLLER - PART I
,
1990
.
[3]
M. S. Sachdev,et al.
Using a neural network for transformer protection
,
1995,
Proceedings 1995 International Conference on Energy Management and Power Delivery EMPD '95.
[4]
M. S. Sachdev,et al.
Design, implementation and testing of an artificial neural network based fault direction discriminator for protecting transmission lines
,
1995
.
[5]
P. Bastard,et al.
Neural network-based algorithm for power transformer differential relays
,
1995
.