Transmission lines Fault Detection and Classification Using Deep Learning Neural Network

With the advancements in artificial intelligence techniques, deep learning has attracted colossal international attention. The deep learning approach has been used in a variety of power system applications and has shown to be effective. Using various deep learning techniques, several endeavours have been produced to classify faults on transmission lines. This paper presents a fault detection and classification method based on Long Short-Term Memory networks are a special type of recurrent neural network proficient at acquiring long-term dependencies. This unique method can classify raw process data directly without any feature extraction or classifier design. The three phase currents of one end are taken as input in the proposed scheme. In this paper a deep learning tool accessible in the MATLAB was used for training and testing purpose. This proposed technique is capable of detecting and classifying No Fault, as well as all sorts of line and ground faults with satisfactory results. These different faults are simulated with various parameters to check the adaptability of the method. In study this, an IEEE 9-bus system designed in MATLAB/Simulink environment for analysis.

Automated summary

With the advancements in artificial intelligence techniques, deep learning has attracted international attention. This paper presents a fault detection and classification method based on Long Short-Term Memory networks, which can classify raw process data directly without feature extraction or classifier design and has shown satisfactory results.