A Secure Fault Identification Approach for MMC-HVDC Network

Abstract

In high voltage direct current (HVDC) systems, the occurrence of short circuits results in a rapid rise in line current, adversely affecting the interconnected alternating current (AC) grid. Particularly in voltage source-based multimodular converter (MMC) HVDC networks, such transients pose a significant threat to power converter units. Traditional relaying algorithms prove inadequate for safeguarding AC-DC-linked HVDC networks. Both the direct current (DC) and alternating current (AC) segments of such networks demand robust protection mechanisms. Signal processing-based techniques offer valuable insights during fault events, yet challenges such as noise interference, mode missing, and harmonics generation during faults persist, leading to erroneous conclusions. To address this, we introduce Synchro Squeezed Transform (SST) in this study to mitigate ambiguity in relaying algorithm decisions. SST facilitates the extraction of amplitude and effective instantaneous frequency of AC signals. The proposed method employs the Rényi entropy of time-frequency representation (TFR) as the primary logic, followed by the estimation of the spectrum-based Teager–Kaiser Energy Operator (TKEO) for DC signals as the secondary logic. These combined logics enable the identification of various AC and DC faults in Voltage Source Converter (VSC)-based bipolar HVDC networks. Simulation results, including comparisons with existing approaches, demonstrate the efficacy of the proposed methodology in enhancing fault detection and classification accuracy in AC-DC-linked HVDC networks.