Machine learning-Wavelet protection Analysis for SVC Controlled Wide Area Network in presence of Wind Energy Source

Abstract

The power system network is one of the most extensively dispersed electrical engineering systems designed to transport the majority of electricity over distances of several kilometers from one end of the country to the other. New wind production units and balancing equipment are often added to an existing power system network as part of the integration of power projects. The system's topology and dynamics have changed, necessitating a reevaluation of the protection plan. In the event of disability, the lines' spread across various topographies and geographic regions makes them particularly susceptible to various atmospheric disasters, which frequently result in significant short circuits among the intermediate connected energy sources. To link the electrical system with smart environments based on Internet-of-Things technologies, quick detection methods are now required. In essence, Wavelet (WT) analysis fault transient signals at various frequencies and breaks down the waveform into successive precise and approximative coefficients, which are vital for determining the location and kind of fault. Machine learning has traditionally been used with great effectiveness in a variety of defect analysis fields. The implementation of mother wavelet detailed coefficients for fault detection and localization and the use of machine learning for fault location on transmission lines for transmission lines This paper offers a detailed explanation of the suggested approach for the diagnosis of system defects using an IoT-Wavelet-based mechanism that was created and put into use in the network of the SVC integrated power system and wind energy source with a machine learning approach.