Diagnosis and Protection of Ground Fault in Electrical Systems: A Comprehensive Analysis

Fault diagnosis in electrical systems is crucial for preventing infrastructure damage, cascading faults, and user injuries. In addition, power converters have become essential due to the rise of renewable energy, electric mobility, HVDC systems, and other emerging technologies. However, they pose a challenge for fault diagnosis and protection due to high-frequency switching noise, different AC and DC stages in the same circuitry, and faster aging of insulation materials. Since ground faults are the most common type of electrical faults, it is of special interest to analyze and understand the complications and limitations of current diagnosis and protection systems. Although several techniques exist for detecting or localizing these types of faults, they require complex processes to carry out the diagnosis. This work provides a comprehensive analysis of the different conventional and advanced methods focused on ground fault diagnosis that are used, not only for distribution and transmission lines, but also for either AC, DC or hybrid systems, which comprise electrical machines, power electronics, drives, HVDC systems, energy storage systems, microgrids, etc. It gives a complete overview of each method, highlighting their advantages and disadvantages. Finally, a discussion is provided studying areas of improvement and the latest emerging trends in the field, where artificial intelligence (AI) and machine learning (ML) techniques are gaining momentum.