Detection and Classification of Transmission Line Faults Based on Unsupervised Feature Learning and Convolutional Sparse Autoencoder

We present in this paper a novel method for fault detection and classification in power transmission lines based on convolutional sparse autoencoder. Contrary to conventional methods, the proposed method automatically learns features from a dataset of voltage and current signals, on the basis of which a framework for fault detection and classification is created. Convolutional feature mapping and mean pooling are implemented in order to generate feature vectors with local translation-invariance for half-cycle multi-channel signal segments. Fault detection and classification are achieved by a softmax classifier using the feature vectors. Further, the proposed method is tested under different sampling frequencies and signal types. The generalizability of the proposed method is also verified by adding noise and measurement errors to the data. Results show that the proposed method is fast and accurate in detecting and classifying faults, and is practical for online transmission line protection for its high robustness and generalizability.

I. Introduction

Fault detection and classification are two important aspects of power transmission line protection. Over the years, researchers have been seeking to realize fast and accurate detection and classification of faults in transmission lines using various methods, so that the faulted system can be protected from possible destructive effects caused by the fault. Further, the information provided by fault detection and classification can greatly facilitate the location of fault, thus reducing the fault clearing time.