A K-NN algorithm based fault locating system for HVDC transmission lines

High voltage direct current power transmission HVDC is nowadays in full expansion in the world for different reasons; economic, technical and environmental ones. On the other hand, more recently, the energy transition has boosted this technology enormously for the integration of renewable energies and carbon neutrality strategy these last times. As any system, the HVDC are subject to different faults which can affect their operation functioning. The aim of this work is to use a classifier based on one of the artificial intelligence methods to localize these faults. We have chosen the k-NN classifier, the theory of statistical learning.in k-NN classification, the result is a membership class. An input object is classified according to the majority result of the membership class statistics of its k nearest neighbors then the proposed approach has the ability to help in the field of classification of defects because there is no restriction on the number of features. It's one of the supervised classifications and the result are a different fault location in dc cable