Signal Processing and Artificial Intelligence Based Vsc Hvdc Network Protection

Abstract

In this modern era, the smart grid (SG) is progressively empowered for direct newlinecurrent (DC) power transmission, either at high voltage (HV) or at medium voltage (MV). newlineThe high voltage direct current (HVDC) transmission systems have a significant impact on newlinethe SG operation even though the integration of add-on green and renewable-based power newlinegeneration (RPG). In this RPG era, the application of HVDC can be seen in the gridconnected newlineremote offshore wind or photovoltaic (PV) based large RPG, and newlineinterconnections amongst the states or countries. In such a context, a smart grid inevitably newlineis ridden with technical complexities (both control and protection viewpoints), and thus newlinesignals processing (SP) techniques along with artificial intelligence (AI) methods are newlineconsidered very much important to understand, effectively plan, draw up plans, and drive newlinethe multifaceted electric power system. This thesis presents an extensive study of HVDC newlinenetwork protection systems with special attention to SP and AI applications. newlineThis initial two chapters presents an extensive review of the suggested techniques newlinein the past three decades and discussed the pros and cons of each method. Apart from that newlinecritical findings are discussed and analyzed along with the future scope of research toward newlinethe development of HVDC protection. There chapters tries to focus on the gap between the newlineexisting protection schemes and topology with the smart grid-based power system newlineperspective and convey to the power engineers and researchers the possibilities of further newlineresearch as a solution to the associated issues and challenges. From this analysis it can be newlineannalysed that there is a critical need for an advanced, data-driven HVDC protection newlineframework that can leverage both the power of signal processing for feature extraction and newlinethe capabilities of artificial intelligence (AI) and deep learning (DL) for pattern recognition newlineand fault prediction. Such a framework must be capable of handling the diverse fault types newlineand conditions that occur in HVDC systems while