Adaline for Fault Detection in Electrical High Voltage Transmission Line

The application of neural networks to power systems has been extensively reported. Neural networks based protection techniques have been proposed by a number of authors. However, almost all the studies have so far employed the back-propagation neural network structure with supervised learning. This paper presents an on line method for fault identification in Electrical High Voltage (EHV) transmission line. This approach utilizes linear adaptive neuron, which is called Adaline. The Adaline neural network is generally used for prediction and identification problems and is rarely used for power system protection. Using current signals, the Adaline process has a strong tracking capability and is fast due to its simple construction, which makes it more suitable for the implementation. Our Adaline approach is compared with a multilayer perceptron in order to see the influence of the fault resistance over the fault time detection.