Power swing detection for correct distance relay operation using S-transform and neural networks

This paper presents an advanced technique for detecting power swings for distance relay operation. It uses a derivative of voltage and signal processing technique called as S-transform for feature extraction. Then an artificial neural network (ANN) is deployed in detecting the unstable swing in power systems. This approach overcomes the traditional relay scheme drawback, by distinguishing a fault, stable swing and unstable swing. To illustrate the effectiveness of the proposed approach, simulations were carried out on the IEEE 39 bus test system using the PSS/E software. Test results show that the proposed approach can effectively differentiate the fault, stable swing and unstable swing with a good accuracy.