Fault Classification for Protection in MMC-HVDC Using Machine Learning Algorithms

The problems in MMC-HVDC protection systems are categorized in this study using machine learning algorithms. The voltage and current data were utilized to determine the classification's features. With the use of the features derived from the voltage and current, machine learning (ML) and artificial machine learning (ML) have produced a defect locator that is accurate enough. Using this data, simulations of various fault types and unknown locations at different system points were run to anticipate the outcomes. Metrics including specificity, accuracy, and sensitivity were used to evaluate the efficacy of the fault classification system; the results showed 98.22%, 97.41%, and 97.23%, respectively.