Decision Tree Based Multi-Terminal VSC-HVDC Transmission Line Protection Scheme

Abstract

HVDC transmission systems are gaining widespread popularity over HVAC transmission because of increased flexibility and robust capability. The detection of external and internal faults in the HVDC transmission system is a massive issue and is highly challenging. Despite these challenges faced; faults must be cleared within a fraction of a period. This paper presents a decision tree classifier based fault detection and classification for a multi terminal HVDC system. The main objective of this paper is to extract the DC voltage and current from the relays present in the HVDC transmission lines. The faults are internal DC faults, external DC faults, and external AC faults from which 14 features were extracted for the experimental proposal. The decision tree classifiers are applied to the extracted features for fault detection. The new approaches allow rapid detection of various faults (internal and external) and quicker fault restoration. The suggested method is carefully evaluated for different possible fault conditions simulated with diverse operating condition on the transmission system. This experimental technique significantly decreases the complexity and the amount of time needed to detect the faults at different locations on the HVDC grid with significant accuracy.